Notes From The Bullpen

This Bird Has Flown

2005-04-04T10:39:00.000-07:00

I've had it with blogger's sketcy, unreliable service. I'm surprised that this Google-backed enterprise is of such consistently poor quality. I'm sure they will eventually put a couple of smart interns on it to fix it, but in the meanwhile I've had enough.

You can now find "Notes From The Bullpen" at http://www.livejournal.com/users/alexvollmer/.

Draft Day Results

2005-03-29T20:31:00.000-08:00

It's been a long time since my last post. We finally closed on both houses and we're deep in the middle of packing, painting and prepping to move. Fortunately I was able to grab enough time here and there to get my pre-draft analysis complete.

This year's analysis was much like the approach I took last year. I calculated a regression line for several offensive and defensive categories and then evaluated how far above or below the regression line (called a residual) each player was in those categories. Since this model is inherently biased towards players with major-league experience, it highlights past performances instead of estimating the future. This year I wrote some Python to pull down the projected depth charts from Baseball Prospectus and simply multiplied hitters' total residual scores by their projected plate appearances.

I decided not to do this with pitchers because multiplying the residual scores by projected innings pitched buried relievers far down on the chart. By not multiplying you could argue that relievers bubbled up a little too high on the list, but I was less likely to miss some reliever with the pitchers sorted in this order.

So without further ado, here is the rosters for your 2005 Swing Kids:

Jorge Posada (NYY - C) - My second overall pick and perhaps drafted a little too high. Last year I got killed with unproductive catchers. This year I vowed not to let that happen

K. Cash (TB - C) - OK, so Cash was one of the lame catchers I had last-year. He was a very late-round pickup and at that point all of the remaining catchers were awful.

Mi. Sweeney (KC - 1B) - God I hope they don't let him play in the field any more than he has to. If he sticks with being DH I figure there's that much less of a chance of him hurting himself. If I can get 400 plate appearances out of him I'll be happy.

J. Cantu (TB - 2B,3B) - A late-round pickup. Honestly, who would draft anyone from the D-Rays before round 15?

J. Crede (CWS - 3B) - This was a guy I drafted based entirely off of my computed draft sheet, not from any first-person knowledge. If you asked me to pick his mug out of a lineup, I couldn't do it.

Á. Berroa (KC - SS) - I really wanted to steer clear of any Devil Rays, Tigers and Royals as much as possible. But as a mid-round pick when all of the most productive middle-infielders were gone, Berroa could work out. I have to think that somewhere between his rookie campaign and his awful sophomore year lies the real Berroa.

J. Gibbons (Bal - 1B,OF) - Roster-filler

R. Adams (Tor - SS) - See Jay Gibbons

G. Sheffield (NYY - OF) - I know, I know. I said to myself I wasn't going to draft a bunch of old guys. As great as Sheffield was for me last year, he was hurt and he's on the wrong side of the age-curve. But there he was in round 6. I couldn't just let him go by? That's leaving .300/35/110 on the table! I couldn't let him go by?! Could I?

J. Damon (Bos - OF) - Here's hoping that Jesus, er, Johnny has another year in him like last year.

S. Finley (LAA - OF) - Yes I know that Mr. Finley is old. But he is also a freak of nature and has somehow managed to remain extremely productive for a long time. I was thrilled to get him

T. Nixon (Bos - OF) - Probably one of the best names in baseball. He was almost worth drafting for that alone.

F. Catalanotto (Tor - OF) - See Jay Gibbons

E. Durazo (Oak - Util) - I had to get at least one Athletic. Unfortunately it wasn't Eric Chavez.

R. Johnson (NYY - P) - My first pick. I had to get one super-star pitcher. Since our league does a down-and-back style draft I had the next immediate pick and spent an awfully long time thinking about picking up Curt Schilling to boot (see Jorge Posada)

T. Percival (Det - P) - In the real-world the Angels made the right move by letting Percival and his increasing salary and declining skills walk. In fantasy-baseball land he's still gotta be good for 30+ saves and a good pile of K's

K. Millwood (Cle - P) - A good-enough addition to the rotation and a mid-round pick I was happy to land.

B. Ryan (Bal - P) - Should be the closer in Baltimore. They should be close enough in enough games that he should get about 30 saves.

E. Guardado (Sea - P) - Yeah yeah yeah. Another old guy. Sure everyday-Eddie could turn out to be a bust and never pitch again. But if he can come back healthy enough to close for the M's I should have enough K's, WHIP, ERA and Saves from Ryan and Percival to have a good staff.

P. Byrd (LAA - P) - See Kevin Millwood

M. Timlin (Bos - P) - This year we added 'holds' as a new category in our fantasy league. Timlin is in the small group of guys who are practically guaranteed to get a pile of holds.

W. Miller (Bos - P) - Of the new additions to the Red Sox rotation, he is the weakest. I really wanted to get Matt Clement, but as a number five started Miller should do...or I'll cut him.

O. Hernández (CWS - P) - Sure El Duque ain't what he used to be, but he should be an adequate space-filler.

A Mover And A Shaker

2005-03-20T10:26:00.000-08:00

It's been quite a while since my last post--mostly due to the craziness that comes from buying and selling a house in the middle of a white-hot real estate market.

Getting our house ready has been a wonderful exercise in the types of lies adults tell one another. In order to present our current house as spacious and livable we have had to put in a Herculean effort over the last week to move most of our stuff off-site. Presenting our house is not about describing our lives, it's about describing a kind of life one could have in this living space. But the truth is that anyone who is buying a house in this market probably has lots of stuff--a consequence of being a participant in the modern consumer society. Realistically no one could really live a life so spartan, clean and presentable as the one we are portraying for today's open house.

Similarly, when we looked at the house we bought we tried to keep this in mind, but the presentation by the seller greatly enhanced our feeling that our new home was spacious. In the new house we could have a neater, more organized life in the new house instead of living in sprawling clutter.

The pessimist in me believes that it's only a matter of time before we acquire more possessions which will prompt us to look for yet a bigger house once again. But if there is one thing I've learned in the last week of hyper-accelerated home improvement, is that I've been pretty lazy getting our house in order. For an organizationally-oriented personality like me the investment of a little time each weekend to "take care of business" pays off big dividends. While I understand that the fictional life portrayed in open-houses is non-existant, there is a measure of satisfaction in getting close.

Quick Notes

2005-03-11T09:05:00.000-08:00

Some of my most creative baseball thinking happens in March. But after going to an open-house on Sunday any baseball content in my brain has been replaced almost wholesale by thoughts housing, contractors and financing. Yes I'm going to explore new levels of property-backed debt in an effort to pursue the American dream of a pot roast on every stove and 2.5 kids. Our offer was accepted on a new house that we absolutely love and so we are now in a mad scramble to prepare our current house for sale.

All of that means less time for baseball. Fortunately I have a very understanding wife who is okay with me disappearing for six to eight hours the day before the first open house to attend my fantasy baseball league's draft. And once we run out of boxes to pack for the evening and Audrey has gone to bed I have managed to get some quality time in on the laptop preparing for Draft Day. I'm optimistic that I will be able to show up with something prepared rather than a freshly-purchased fantasy guide from the local 7-11.

Act Your Age (Pt. III)

2005-03-03T21:20:00.000-08:00

My original question was "do players follow a general aging pattern with respect to playing time?" To answer this question we first looked at very general trends. In the last installment we explored this question along a temporal dimension, focusing on the last forty years of baseball. Today I want to see how aging differs over a positional dimension.

To start, I created some graphs of average games played, by age for each position. Each graph has a line for every five-year period explored. Here they are by position:
First basemen:

Second basemen:

Third basemen:

Shortstops:

Catchers:

Outfielders:

What is interesting is how wild the variation is for 1B and 2B versus the other positions. Between the ages of 26 and 34 all other positions, other than OF, don't show any particular trend over time--just noisy consistency. Now outfielders are another story. Note how far off of the bottom outfielders fall in average number of games since 1995. What can possibly explain the precipitous drop-off? One guess is the heavy increase in platooning outfielders. But is it enough to cause an average drop of twenty games?

This final chart looks at average playing time by position since 2000. Clearly designated hitters are in a category by themselves. Like outfielders in the previous graph, this probably reflects the DH platoon situation many clubs face. Additionally inter-league play eats into their playing time.

What can we conclude from all of these graphs and numbers? The only certain conclusion is that any trends about age only apply at a very gross level. We can say that players tend to get a steady amount of playing time between the ages 26 and 34 with a sharp drop-off thereafter.

How will this affect my strategy on Draft Day? I'm definitely going to be very wary of drafting anyone over 34. When looking at players like Gary Sheffield or Steve Finley I would think twice about the inevitable effects of aging. He's had a great career, but when will Father Time finally catch up? On the other hand, if we have learned anything from this exercise it is that baseball tends to have more exceptions than rules.

Why BP Rocks

2005-03-02T08:48:00.000-08:00

This morning the line at the local coffee joint was huge. I knew I had a good ten-minute wait from enqueueing to getting my morning go-juice so I cracked open my phone-book sized 2005 Baseball Prospectus to pass the time. Lucky for me the canned music covered up my snorting laughter when I read a passage in which the BP boys described the Angels decision to move Mo Vaughn to from DH to first base akin to "putting a glove on a jelly donut". That's funny.

BP!

2005-03-01T20:16:00.000-08:00

A Picture Share!
Originally uploaded by livollmers.

After a teeth-gnashing February, I managed to finally track down a copy of the elusive 2005 Baseball Propectus. Oh Amazon! How you have failed me! Oh Elliot Bay Books! How you have won me over again!

When I checked out I saw two other people pickup up their copy. If you don't have yours yet, you'd better get cracking!

Unit-Testing Hibenate with HSQLDB

2005-02-27T13:06:00.000-08:00

The Motivation

Since the beginning of my career as a software developer, I've ended up on projects that, one way or another, make use of relational databases. I've used lots of methods to schlep data between databases and my code; from hand-coded SQL to JDO to EJB. I've never found a method I liked particularly well. This has become especially acute since adopting test-driven development as a guiding philosophy. To support TDD it's vital to keep database access code away from the core object model and to test as much as possible without touching a real database.

This has often led me to one of two patterns. The first is externalizing all data access to domain objects and their relationships to separate classes or interfaces. This is the easiest to mock-out for unit-testing, but tends to leave your domain model objects as dumb bags of data. Ideally I'd like to be able to access child records directly from the parent object rather than handing the parent object to some data oracle (no pun intended) to determine the children.

The other method has been to have the domain objects have access to an interface into the data-mapping layer a la Martin Fowlers Data Mapper pattern. This makes a lot of sense and certainly seems to leave you with a better feeling about how your domain objects are factored.

The problem for me came down to trying to insert a layer of indirection while using the uber-powerful Hibernate as my data mapper. The extra layer was so flimsy that it felt embarassing to write it--there was no "there" there. The real deployed version was simply a pass-through to a Hibernate-specific implementation. Strangely enough my mock versions had more complexity in them than the real "production" version simply because it didn't have some of the basic object storage that came with Hibernate.

This left me with another problem which was that I had a fair degree of complexity in my Hibernate code that I wanted to unit-test. However, testing against a "real live" database was simply out of the question. I've gone down this path before with the best of intentions and have always ended up creating a maintenance nightmare for myself. Ideally individual test-methods on your test cases are independent from one another. If you're like me you tend to use the same obvious primary-key fields in constructing test fixture data. This requires lots of code to clean your database before each test case which can become a serious nightmare when lots of relationships are involved.

I settled on the idea of investigating in-memory databases that would allow me to test the Hibernate layer without having to startup a full-featured RDBMS external to my development environment. I can say with much satisfaction that this was a very worthy exercise. I found numerous bugs I would not have been able to catch without this technique.

Setting up HSQLDB

I'm using version 1.7.3.0 of the HSQLDB. To use an in-memory version of the database you need to invoke the static loader for the org.hsqldb.jdbcDriver. Then when you get a JDBC connection you use JDBC url of something like jdbc:hsqldb:mem:yourdb where 'yourdb' is the name of the in-memory database you want to use.

Since I'm using Hibernate (3.0 beta 4), I hardly ever need to touch real-live JDBC objects. Instead I can let Hibernate do the heavy lifting for me--including automatically creating the database schema from my Hibernate mapping files. Since Hibernate creates its own connection pool it will automatically load the HSQLDB JDBC driver based on the configuration code lives in a class called TestSchema. Below is the static initializer for the class.


public class TestSchema {

    static {
        Configuration config = new Configuration().
            setProperty("hibernate.dialect", "org.hibernate.dialect.HSQLDialect").
            setProperty("hibernate.connection.driver_class", "org.hsqldb.jdbcDriver").
            setProperty("hibernate.connection.url", "jdbc:hsqldb:mem:baseball").
            setProperty("hibernate.connection.username", "sa").
            setProperty("hibernate.connection.password", "").
            setProperty("hibernate.connection.pool_size", "1").
            setProperty("hibernate.connection.autocommit", "true").
            setProperty("hibernate.cache.provider_class", "org.hibernate.cache.HashtableCacheProvider").
            setProperty("hibernate.hbm2ddl.auto", "create-drop").
            setProperty("hibernate.show_sql", "true").
            addClass(Player.class).
            addClass(BattingStint.class).
            addClass(FieldingStint.class).
            addClass(PitchingStint.class);

        HibernateUtil.setSessionFactory(config.buildSessionFactory());
    }

Thanks to the smart guys over at Hibernate I have a couple of options for configuring the framework, including programmatic configuration. The code above sets up the connection pool. Note that the username 'sa' is required to use HSQLDB's in-memory database. Also be sure to specify a blank as the password.

The other imporant configuration option was setting the hibernate.hbm2ddl.auto property to 'create-drop'. This is the bit of magic that makes Hibernate create my database schema in the HSQLDB database auto-magically for me.

Testing In Practice

My project is crunching a bunch of baseball statistics so I add the four classes that I'm mapping. Finally I create a Hibernate SessionFactory and insert it into the HibernateUtil class which simply provides a single access method for my entire application for Hibernate sessions. The code for the HibernateUtil is below:


import org.hibernate.*;
import org.hibernate.cfg.Configuration;

public class HibernateUtil {

    private static SessionFactory factory;

    public static synchronized Session getSession() {
        if (factory == null) {
            factory = new Configuration().configure().buildSessionFactory();
        }
        return factory.openSession();
    }
    
    public static void setSessionFactory(SessionFactory factory) {
        HibernateUtil.factory = factory;
    }
}

Since all of my code (production code as well as unit-tests) get their Hibernate sessions from the HibernateUtil I can configure it in one place. For unit-tests the first bit of code to access the TestSchema class will invoke the static initializer which will setup Hibernate and inject the test SessionFactory into the HibernateUtil. For production code the SessionFactory will be initialized lazily using the standard hibernate.cfg.xml configuration mechanism.

So what does this look like in the unit-tests? Below is a snippet of a test that checks the getEligiblePositions method of the Player class:


    public void testGetEligiblePositions() throws Exception {
        Player player = new Player("playerId", null);
        TestSchema.addPlayer(player);
        
        FieldingStint stint1 = new FieldingStint("playerId", 2004, "SEA", Position.CATCHER);
        stint1.setGames(20);
        TestSchema.addFieldingStint(stint1);
        
        Set positions = player.getEligiblePositions(2004);
        assertEquals(1, positions.size());
        assertTrue(positions.contains(Position.CATCHER));
    }

I first create a new Player instance and add it to the TestSchema via the addPlayer() method. This step must occur first because the FieldingStint class has a foreign-key relationship to the Player class. If I didn't add this instance first I would get a foreign-key constraint violation when I try to add the FieldingStint. Below is the code for the addPlayer() method. You will notice that I don't use bare-metal JDBC, but instead leverage Hibernate to do the heavy lifting for me:


    public static void addPlayer(Player player) {
        if (player.getPlayerId() == null) {
            throw new IllegalArgumentException("No primary key specified");
        }

        Session session = HibernateUtil.getSession();
        Transaction transaction = session.beginTransaction();
        try {
            session.save(player, player.getPlayerId());
            transaction.commit();
        }
        finally {
            session.close();
        }
    }

One of the most important things in unit-testing is to keep your test-cases isolated. Since this method still involves a database, you need a way to clean your database prior to each test case. Since I only have four tables I simply wrote a reset() method on the TestSchema that removes all rows from my four tables using JDBC. Note because HSQLDB knows about foreign keys, the order in which the tables are deleted is important. Here is the code:


    public static void reset() throws SchemaException {
        Session session = HibernateUtil.getSession();
        try {
            Connection connection = session.connection();
            try {
                Statement statement = connection.createStatement();
                try {
                    statement.executeUpdate("delete from Batting");
                    statement.executeUpdate("delete from Fielding");
                    statement.executeUpdate("delete from Pitching");
                    statement.executeUpdate("delete from Player");
                    connection.commit();
                }
                finally {
                    statement.close();
                }
            }
            catch (HibernateException e) {
                connection.rollback();
                throw new SchemaException(e);
            }
            catch (SQLException e) {
                connection.rollback();
                throw new SchemaException(e);
            }
        }
        catch (SQLException e) {
            throw new SchemaException(e);
        }
        finally {
            session.close();
        }
    }

One dumb mistake I made here (mostly out of habits developed over writing tons of hand-crafted JDBC code), was to close the JDBC Connection. Since I configured Hibernate to create a connection pool with only one Connection I completely torpedoed any tests after the first one.

Since you're never sure what state the database may be in when your test class is running (imagine running all of your test cases), I put database cleanup in my setUp() method like so:


    public void setUp() throws Exception {
        TestSchema.reset();
    }

Conclusion

Being able to test against a real-live RDBMS without all of the hassles of trying to run tests against your deployed database is essential, even when working with sophisticated O/R mappers like Hibernate. The example I showed here is not exclusive to Hibernate and could probably be made to work with JDO or TopLink. Though Hibernate makes this kind of testing particularly easy since it has a built-in schema generation tool. With a setup like the one described above you don't ever have to leave the comfort of your IDE and still have extensive test coverage over your code.

Act Your Age (Pt. II)

2005-02-22T09:02:00.000-08:00

In my last post we looked at the gross trends for age as it relates to playing time. In general hitters peak around age 26 and pitchers about a year later. You may recall that pitchers showed some interesting spikes toward the far right-hand side of the graph which can be explained by some truly exceptional, long-lasting pitchers. It is interesting to note that no such spike was observed with hitters.

One of the important dimensions missing from the first part of our data analysis is the effects of sports medicine and improved conditioning on playing time. In this installment we will explore the observed playing times related to five-year "eras" (a misnomer to be sure), since 1960.

The graph below shows total AB by age by "era". In general, the trend is for AB to increase in each progressive era. The change is most radically apparent around age 28. Note that the gap between the 1960-1965 era and the 1965-1970 is larger than any other gaps. This is most likely due to the change from 154-game seasons to 162-game seasons in 1961. Similarly the 1980-1985 and 1990-1994 eras short-change themselves due to strike-shortened seasons. Lastly the 2000-2005 era is missing the last year of data so the arbitrary five-year sample period leaves a bit to be desired. These anomalies will appear again and again in the graphs below, so keep them in mind.

The next graph looks at the average number of AB by era.

In the meaty part of the graph the general trend is downward as you move forward across eras. For example, at age 28 the players playing between 1965 and 1970 as well as between 1975 and 1980 averaged around 172 AB. Conversely, players between 1995 and 2000 only averaged around 128 AB. However if you note the two bookend eras (1960-1965 and 2000-2005), you will see two very different trends. In the '60-'65 era players peaked in average AB around 33 and then begin a sharp decline. However in the 2000-2005 era players peak around 38 before they begin a similar decline perhaps highlighting the differences between player conditioning in the 1960's vs. the 21st century.

You will notice that the data between ages 18 and 21 as well as 33 to 45 are pretty noisy. Look below at the graph of batting samples. The 18-21 block is a very small part of the overall sample curve. While not quite as small, the 33-45 block in the curve are dwarfed by the main 22-32 block.

Because of these small samples sizes at the extremes, we need to look at the middle to get a good feel for the differences between eras. Remember, we are trying to understand differences in playing time through the years--not how long players played or how early phenoms make their presence felt in the majors.

While the total number of AB by age increased over time, the average AB tends to decrease. I suspect this is due to expansion more than anything else. More players are coming into the game (increasing overall ABs), but younger players are getting less ABs that similarly-aged players in previous years.

The next graph shows the average number of IPOuts (for IP just divide by 3) for each age by era.

Like the batting graphs we just looked at, you see a lot of noise and wild variation at the right-hand side of the graph starting at around age 34. Similarly there is some noise between ages 18 and 21, though not as much as is seen for the veterans.

It's interesting to note that in the meaty part of the graph (between ages 22 and 33) we see that the average for pitchers between 1960 and 1980 hardly changes. By 1980 we start to see the beginning of an overall downward trend (per age) for each era. This is most likely due to the widespread adoption of the five-man rotation as well as the emergence of the relief pitcher which ate into starting pitchers innings pitched.

The pitching graphs shows a much shallower climb (compared to hitters) from ages 23 to 34. Past that, the graph devolves into noise and outliers. It is interesting to note that hitters tend to display a noticeable downward trend before hitting the noisy portion of the graph. In contrast the pitchers seem to steadily increase their workload before all hell breaks lose and we can no longer determine any trends. The graph below shows the number of samples used. The small number under the curve from age 34 onward confirms our small sample-size theory.

What about that long right-hand tail for the 1960-1965 era? Who is that statistical outlier? Why that's none other than the king of baseball aphorisms, Satchel Paige.

To complete our look at aging, we need to explore the differences in playing time across positions. Common sense would dictate that catchers don't play nearly as long as say, first basemen (Carlton Fisk not withstanding). However I'm much more interested in data than common sense. Stay tuned to see what the numbers have to tell us next.

Act Your Age (Pt. I)

2005-02-15T21:53:00.000-08:00

In preparing for this year's Fantasy Baseball Season, I decided to re-evaluate last year's feet-first jump in the pool of Sabermetric analysis. While I was happy with the work I did last year, I really felt the pain of not accounting for the effects of aging and injury. With injuries to Magglio Ordonez and Kevin Brown I decided that this year's preparation would include a look at the effects of aging on playing time.

I started out by looking the last 130 years of recorded baseball history to get a rough feel for what the breakdown is for playing time by age. The two graphs below are simply a sum of at-bats (for hitters) and outs pitched (for pitchers) at a given age:

There's no real surprise here. Conventional baseball wisdom has always said that the peak age is around 27 or 28 years old. It's slightly interesting to note that the "peak" age for pitchers is a bit later than hitters, which is probably due to the slower rate at which pitching skills improve relative to batting.

The following graphs look at the same data but track a running average of at-bats or outs pitched as well as a running standard deviation:

The batting graph was no real surprise, but the pitching graph set off some alarm bells. The standard deviation drops over time, while the average outs pitched peaks at around age 34--six years later than the peak age for total samples. I believe this reflects the fact that the real work-horses of major-league staffs are mature pitchers that range from Pedro Martinez to Tim Wakefield to Randy Johnson, not all young-guns a la Bronson Arroyo.

Even more fascinating is the fact that around age 40 we see some serious upward-ticks in the average for outs pitched. I think this is due to the fact that pitchers who are still in the league at age forty are probably some of the best pitchers in the game. Below is a list of pitchers who have pitched beyond the age of 40 since 1960. You will note that this list stocked with either current or future Hall of Fame-quality pitchers, or "crafy"-type pitchers like Gaylord Perry, the Niekro brothers and Jamie Moyer

Larry Andersen

Doug Bair

Jim Bibby

Bert Blyleven

Wade Boggs

Hal Brown

Jim Bunning

Lew Burdette

Tom Burgmeier

John Candelaria

Tom Candiotti

Steve Carlton

Roger Clemens

Dave Concepcion

David Cone

Dennis Cook

Mike Cuellar

Danny Darwin

Rick Dempsey

Dennis Eckersley

Roy Face

Jeff Fassero

Mike Fetters

Chuck Finley

Mike Flanagan

Ken Forsch

Tony Fossas

Art Fowler

John Franco

Woodie Fryman

Gary Gaetti

Gene Garber

Bob Gibson

Ruben Gomez

Rich Gossage

Harvey Haddix

Dick Hall

Greg Harris

Bill Henry

Roberto Hernandez

Orel Hershiser

Joe Hoerner

Rick Honeycutt

Charlie Hough

Grant Jackson

Mike Jackson

Fergie Jenkins

Tommy John

Randy Johnson

Doug Jones

Jim Kaat

Johnny Klippstein

Jerry Koosman

Dennis Lamp

Dennis Martinez

Bob McClure

Lindy McDaniel

Tug McGraw

Don McMahon

Stu Miller

Mike Morgan

Jamie Moyer

Terry Mulholland

Joe Niekro

Phil Niekro

Diomedes Olivo

Jesse Orosco

Satchel Paige

Orlando Pena

Gaylord Perry

Jim Perry

Dan Plesac

Ron Reed

Rick Reuschel

Jerry Reuss

Robin Roberts

Rich Rodriguez

Kenny Rogers

Nolan Ryan

Scott Sanderson

Tom Seaver

Diego Segui

Lee Smith

Warren Spahn

Gerry Staley

Dave Stieb

Don Sutton

Frank Tanana

Billy Taylor

Kent Tekulve

Luis Tiant

Ed Vosberg

David Wells

Hoyt Wilhelm

Al Worthington

Early Wynn

Geoff Zahn

Next time we'll start looking at how the effects of aging have changed over time in baseball.

Where Starbuck Is Not Coffee

2005-02-10T08:41:00.000-08:00

I saw the good news today on Slashdot that the Sci-Fi Network is renewing Battlestar Galactica for another season. I was a big fan of the original show when I was a kid, but I can definitely see the warts as an adult. However the new series is surprisingly sophisticated. Even my wife admits that it's pretty engaging!

If you haven't yet checked this series out you should tune in. I'm not the world's biggest sci-fi fan and hardly ever watch the Sci-Fi network so this show passed through a lot of checkpoints for me before it made the regular Tivo rotation. I'm ashamed as an adult to tell everyone "I watch Battlestar Galactica". Now if they made some kind of cool mugs or something...

Super Bull

2005-02-06T06:55:00.000-08:00

Here we are at the end of another football season. I'm facing today's Super Bowl much like an already-full diner faces the final bite of an all-too-rich dessert. I've come this far on this gigantic empty-calorie cake called the NFL, I might as well finish it off.

My attention to the Super Bowl has varied over the years. As a kid my dad used to get together with co-workers and have Super Bowl parties which were very fun. I faithfully watched the Big Game until I hit college where I lost interest and became more fully devoted to baseball. After living in San Francisco I regained my love for the sport, helped greatly by the Super Bowl-winning 49ers of the 1994 season. Ever since I've tuned into every game.

However I face today's game with some trepidation. I'm not overly eager to watch, although I feel somewhat obligated to do so as an informed sports-fan. The glitzier-than-usual, non-stop, never-ending coverage of the game is beginning to sour me on the whole experience. From tamed-down commercials to watered-down half-time shows, to the exclusion of the Florida A&M marching band from the televised half-time show, to the sponsorship of every single moment of the game--I'm getting pretty fed up with the whole thing.

Don't try bring out the old saw about the "purity" of the college game. The players are less skilled and the institution is easily more corrupt. For all of its warts, I'll take the NFL any day.

However my enthusiasm not helped by the overwhelming feeling that the Patriots will again walk away as winners. Aren't we supposed to be in the age where dynasties are a thing of the past? When will salary-caps, luxury taxes and free agency catch up with these guys? I don't find the Eagles a particularly compelling team to root for other than they are not the Patriots.

Well, if the game gets boring I can turn to a copy of "Napolean Dynamite" that I haven't watched yet and I can comfort myself in the stack of baseball books I bought yesterday. Finally I can find solace in seven hopeful words: pitchers and catchers report in two weeks.

The End Of The Line For Your PC

2005-02-04T08:58:00.000-08:00

I found an interesting link on Slashdot today about a proposal from the House of Representatives for computer recycling. The proposal requires a $10 fee added to any hardware purchase. This would fund some sort of federal recycling program.

I definitely feel small pangs of guilt whenever I buy new computer equipment and try to figure out what to do with the old stuff. Given the high turnover of computer equipment in the marketplace one has to wonder where all of this toxic stuff is getting dumped. It seems like a serious problem that is growing and needs to be dealt with.

Of course this type of issue will line up parties on both side in their usual fighting positions. Environmentalists will accuse manufacturers and retailers of being concerned only with the Almighty Dollar at the expense of the environment. Business leaders will cry foul that their industry is being regulated over threats that don't yet exist. This is to say nothing of the fact that the majority of run-of-the-mill PC hardware is manufactured outside of the United States; how these types of fees, tariffs or laws would apply is open for debate.

The auto industry might be a place to look for a similar model. Each year auto makers put out new vehicles at a staggering rate. Consumers sell their old cars to purchase new ones, effectively dividing the auto-buying public into two categories: those who purchase the new cars coming into the system, and those who are taking over the used ones. We have to wonder then about what happens to the cars the second category of consumers are shedding. Land fill? Junk yards? Recycled for parts?

It seems that given the rate of technological advancement of PC internals, re-use can only cover so much. Consumers buy new machines because they are more powerful than the old ones that they have. These accelerations tend to be core components of the system leaving little behind that is re-usable except for basics like cases. That still leaves a measurable amount of toxic parts left behind that nobody wants. What happens to these items is going to be the challenge for the future.

Coke/Soda/Pop

2005-01-30T11:51:00.000-08:00

I stumbled onto this graphical breakdown of the regional differences in requesting a sugary, carbonated beverage. I have no idea how they got this data or if it is even real, but it's funny nonetheless.

There was also an article in this Sunday's Seattle Times about why the M's should bring 19-year old phenom Felix Hernandez up to the bigs. The fine folks over at the USS Mariner saw the article the same way I did, which is to say "huh?". Perhaps my favorite line from this article was:

...he is either ready to pitch in the big leagues or he isn't.

Ooh that's good. If he were an investment advisor he could give you equally helpful advice such as, "the stock will either go up or down this year." Thanks, but I think I can figure out the short list of possiblities on my own.

While we're on the subject of baseball here's a little funny story for you. I walked into the local bookstore on Friday in search of the baseball annuals that come out around this time of year (Athlon sports, Street & Smith's, etc). I picked up the least dog-eared copy I could find and happily took it back to the office. Over lunch I checked out their predictions for the season. To my surprise they projected the Mariners to take first place in the AL West. Intrigued, I flipped over to the Mariners section of the magazine and nearly fell out of my chair when I read the headline "Smart signings of Aurillia and Ibanez strengthen veteran ball club". What?

A quick check of the cover revealed that the magazine I had bought was for 2004, not 2005. Grrr. It looks like I'm going back there Monday to look for a refund.

While I don't know if anyone could predict the magnitude of the collapse the M's experienced in 2004, I will probably avoid buying a Street & Smith's preview again based on their 2004 predictions.

Flashdance

2005-01-28T22:11:00.000-08:00

What did you do on your Friday night? Go out for drinks with friends? Catch a show? Dinner and a movie? As for me, I watched three episodes of "West Wing" and got into the gritty details of hotplug and getting my USB flash drive to auto-mount when it's plugged in and unmount when unplugged.

After much searching around I finally seemed to get all the fiery hoops lined up. First you have to get your /etc/hotplug/usb.usermap lined up so that it can refer to a control script. Either I mis-read various sites or they steered me astray but I mistakenly thought the first hex code in this file indicated a bit mask of which fields I wanted to match on. This was incorrect. Instead it mapped to the same hex code matching the kernel module I was using that is specified in /etc/hotplug/usb.distmap.

Once I got that figured out I pretty much hi-jacked some other guys simple script to mount and unmount the flash drive. With a little /etc/fstab magic I can get the USB drive to always be owned by the user I always log into the laptop with.

Now I have a weird issue where my writes (e.g. touch /mnt/flash/foo) don't seem to persist between mounts. I'll have to look into this some more.

Preferential Treatment

2005-01-23T10:23:00.000-08:00

This new laptop has given me the opportunity to re-consider how I want the desktop to behave. At work I'm often under the pressure to be "doing something productive", and figuring out how to seamlessly integrate some esoteric feature into my desktop is not often a big priority. With the new laptop I have the luxury of getting everything setup just the way I want it.

I used KDE for a long time having always found it more stable than GNOME. At my current job we're a predominantly GNOME shop so I figured it was just easier to go with the flow and run GNOME on my work machine than swim upstream. However, for my laptop I've decided to run KDE again.

I've always believed that the KDE team was far ahead of the GNOME team in terms of total desktop integration and the overall "tightness" of the desktop environment. I know that there are lots of arguments about bloated libraries (Qt. vs. GTK) and such, but honestly I have never perceived any performance difference between the two. What I have noticed is GNOME going off the rails and crashing in ways that KDE doesn't seem to.

As an example of good integration, KDE has done a remarkable job integrating GPG into the desktop. You can simply popup the encryption tool from the system tray and encrypt something from the clipboard, decrypt something from the clipboard or pop up a text editor and save the file as encrypted text. Similarly you can encrypt and decrypt files using the file-browsing mode of Konqueror with the click of a mouse button. How many times have I had Nautilus simply crash or even fail to load?

Honestly I don't mind running split-brained. My other desktop machine is running GNOME and I'm not terribly inclined to move it over to KDE. I'm one of the few people that can keep lots of Bucky Bits in my head (I'm just about equally comfortable with vi and emacs too.) Some people focus their brains or storing massive amounts of information, others use it for complex abstraction. Mine is wasted on useless trivia and lots of keyboard shortcuts.

We Are The Champions

2005-01-21T20:41:00.000-08:00

I am posting this with a real-live 802.11 connection on my new laptop! Ladies and gentlemen this was not an easy task, but kind of satisfying. I even managed to get 128-bit WEP working. This was suprisingly difficult and I only found the answer when I stumbled upon a Linux/Wireless HOWTO written in German. Folks, my German is limited almost entirely to World War II military terminology, but thanks to the universal language of Linux commands I grokked the important setting.

It turns out that you can't do this:
iwconfig eth1 essid myessid

And then follow it up with:
iwconfig eth1 key xxxxxxxxxx

They are like ships passing in the night. You need to set the ESSID and WEP key all at the same time with:
iwconfig eth1 essid myessid key xxxxxxxxx.

While I find this a little irritating, at least I've overcome a major hurdle. Right now I'm using all 64-bit via the smart fellas over at Linuxant. I have a trial license that is still good for about a month so I'll probably still try to see if I can get this working with Ndiswrapper.

Laptop Post

2005-01-17T20:58:00.000-08:00

I am posting this blog from my brand-new AMD 64 laptop. While I don't have everything working yet, I'm pretty damn close. I'm holding out hope that one of the release candidates of ndiswrapper will create a kernel module that doesn't reboot my machine. But I've got Java, Eclipse, Python, zsh, sound and DVD support working.

I knew going into this project that there was going to be a lot of careful hand-crafting to get this box up and running. I decided to be extra-methodical and make small changes, one step at a time. I decided to make a CVS module where I put kernel configurations, Xorg files, zsh files and the like. At any point I could rollback to a known point and recover using the version history CVS provides.

It Boots!

2005-01-13T21:58:00.000-08:00

By god there are days when you think you can cure a rainy day. After a few initial hiccups I've finally got the new laptop booting up with a 2.6 kernel. Now it's time to give the ol' CPU a real work-out by emerging Xorg and Gnome. I figure if my house doesn't burn down in the middle of the night from cooling failures I'll be in good shape.

No Class In The Heartland

2005-01-13T08:45:00.000-08:00

Apparently the Vikings felt they hadn't damaged their image badly enough with Randy Moss' butt-rubbing at Lambeau Field last week. Now Vikes owner Red McCombs has demanded that Fox remove Joe Buck from the broadcast booth for calling Moss' action "a disgusting act".

I'm so completely floored by the temerity of this action I can hardly write. Where is the apology for Moss' behavior in front of some the NFL's best fans? Under what circumstances are his actions ever acceptable? (okay, maybe in Oakland).

While I rarely find myself on the side of the Fox Network, kudos to them for standing up for their guy. Joe Buck called it right and Red McCombs should be disciplining his players, not upbraiding a broadcaster for calling Randy Moss' demonstration what it was...a digusting act.

The New Laptop!

2005-01-12T20:51:00.000-08:00

The New Laptop!
Originally uploaded by livollmers.

Oh great day of days, the new laptop is here. I gave myself a real treat with this AMD 64-bit beauty! Right now I'm in the process of putting Gentoo Linux on it. We'll see just how much teeth-gnashing I have to go through to get this thing up and running. But I would like to say that I never booted this thing up with the copy of Windows XP that came with it. Now I just have to figure out how to get that damn decal off of there...

Playoffs

2005-01-10T08:55:00.000-08:00

Did I really see Randy Moss pretend to pull his pants down and rub his rear-end on the goal posts at Lambeau Field? Did I really see that, or was it just some awful nightmare? Say what you will about Terrell Owens' end-zone hijinks, I doubt he would pull off a class-less maneuver like that.

When you consider what it takes to complete a pass play in the NFL it's simply mind-boggling. First, all of the players have to know the play inside and out in order to make last-second adjustments. The offensive line, and perhaps the backfield, have to quickly read the defense as the ball is snapped to change blocking assignments according to the pass-rush. The receivers need to run their routes, throw fakes, shake off blocks and get open. The quarterback has to quickly read his progression and pick a receiver to throw to in an instant before getting sacked. The ball has to get downfield untouched by linemen and finally, if all these things happen, the receiver has to wrap his hands around the ball and hit the ground without dropping it.

In Sunday's Seahawks-Rams game all of these things happened except for the last bit. I won't argue that catching a football thrown by an NFL-caliber quarterback isn't difficult, but when everything else goes right so often and the last step fails so much, you have to wonder if you have the right personnel on the field.

Give credit to Matt Hasselbeck for not killing any of his receivers this season. He's said all the right things in the media and taken all the blame. The fact of the matter is that no one who calls themselves an NFL receiver should drop passes that hit them in the numbers as much as the 'Hawks receivers do. This problem is more than a statistical anomaly, it's a crisis of skill. You can't run the precision West Coast Offense (or the vertical game for that matter) if your QB can't rely on his receivers making catches.

Finally, can anyone tell me why Jerry Rice didn't even touch the ball on Saturday? I understand that he may have lost a step over the years, but he can still play.

Four Sweet Words

2005-01-08T10:53:00.000-08:00

A check of my inbox this morning revealed a very pleasant surprise--a message from HP with the subject "Your Order Has Shipped". I feel like a little kid who mail-ordered some toy after saving up a million cereal box tops. Only instead of a secret agent decoder ring, this is an AMD 64-bit laptop with lots (but not all) of the fixins.

All of this is part of a post-Christmas effort to scale down our life. I realize that this is a very common post-holiday psychological reaction, much like the urge people feel to buy exercise equipment after spending November and December stuffing their faces. But having experienced my 33rd Christmas, I feel like there are so few things I really need and there is so much in my house I don't need.

The goal of the laptop is to replace one of the desktop machines under my desk. I'll keep my Linux box for a server, but the 17" monitor and the XP box are either going to get sold or donated to the local Boys & Girls Club. This replacement is part of an overall plan I have to replace items in my collection of "stuff", not to enlarge it. For example, I went clothes shopping last weekend (oh the sales!) and for every shirt or pair of pants purchased, I pulled one out of the wardrobe and either put it in the rag bin or bagged it for a donation.

My laptop should be here in a week. Oh what fun we'll have getting it all setup and tuned! By the way if anyone wants an XP box with monitor, let me know.

There's Another Game On?

2005-01-05T08:40:00.000-08:00

Yesterday's public de-pantsing of the Oklahoma Sooners by USC gave me some small measure of satisfaction. The only thing worse than Oklahoma's play on the field was Ashlee Simpson's pathetic attempt to play live music. Honey, stick to recordings, you'll get booed less in the end.

It's good to see a Pac-10 team finally win the big one. Although given the BCS' proclivities for idiocy it wouldn't surprise me if they concocted some shared championship with Auburn also. While it's good to see the old Pac-10 get some respect, as a University of Oregon Ducks fan the Triumph of the Trojans can only mean more dark recruiting years ahead for the other nine teams in the conference (except maybe Cal).

In other news, the Hall Of Fame elected Wade Boggs and Ryne Sandberg as this years inductees. Looking at the list of players, these are probably the two guys I would have picked also, although I think Andre Dawson makes a good case. It is interesting though to consider stars of the 80's for the HOF. The 80's seem to be a somewhat forgotten era in baseball, sandwiched between the sunny days of the 70's and the monster-athlete days of the 90's. You never hear anyone wax poetically about the 80's, but maybe that's because we still can't get over the number of teams that incorporated robin's egg blue into their uniforms.

Now the college football season is over, and this weekend we turn our attention to the NFL playoffs. Every matchup this weekend should be a good one. Last night I turned on the USC-Oklahoma game and my wife asked incredulously, "is there another game on?". To which I replied, "Yes honey, but this is the last one...for college". Hopefully she got the hint and won't throttle me when I turn the TV on Saturday afternoon to watch the Seahawks and Rams.

Happy New Year!

2005-01-01T08:19:00.000-08:00

Normally I would be spending New Year's Day filling up on tortilla chips and beer and watching college football until my eyes liquefy and run from my skull. This year however, I have seriously curtailed the amount of college ball on my TV. I was ready to watch Cal pound Texas Tech and hold it up as another example of inequities of the BCS Alliance, but Cal decided to mail it in and ended up only lending credence to Mack Brown's Longhorns getting a trip to Pasadena. Now I'm just too tired to be mad about it or care.

This year I watched college football because it was a nice pre-cursor for the NFL games on the next day. But I'm very tired of the BCS and the idiocy, hypocrisy and hype that surrounds it. Don't get me wrong: I'll be watching USC and Oklahoma got at it on Monday, but I don't have to enjoy it.

Now What?

2004-12-30T07:26:00.000-08:00

I finished the first phase of the Great Fantasy Baseball Simulator a couple of weeks ago. The simulation can now draft and total like a champ, but now I have figure out what I want to measure. Since then I've been poring over statistical texts learning about sampling and polling theory.

My working theory is that there is a fixed distribution of talent as far as rotisserie categories go. However the distribution across categories may not be the same (e.g. stolen bases vs. home runs), so there is a weight that should be applied. This has led me to wonder, given a fixed talent pool, would winning teams exhibit a particular distribution of talent across categories? If we ran a simulation over and over would a pattern emerge for the archetypical winning team?

My gut tells me that this brute-force simulation approach is the right path. The question that now hangs over me is "how do you know when you have simulated enough?" This seems identical to the problem faced in sampling or polling, but the difference is that this isn't a simple yes/no kind of question. At first glance it seems that the sample data has multiple dimensions, but perhaps it can simply be reduced to category totals. Perhaps the simulation can be run until the variation in totals for each category for the winning team begin to fall within the margin of error. I would like to have the simulation simply run until it reaches this point and then spit out the final winning pattern.

Pyrrhic Victory

2004-12-28T07:46:00.000-08:00

A Picture Share!
Originally uploaded by livollmers.

One of the best presents I got this Christmas was a wooden-replica model of the famous H.M.S. Victory, which was Admiral Nelson's ship at the Battle of Trafalgar. Unfortunately it appears to missing some crucial pieces like the keel, rudder and bowsprit.

The way the model is engineered strikes me as particularly old-world Italian, compared to moden American or Japanese models. The latter tend to be by-products of the Industrial Revolution. Each piece is carefully engineered for consistent reproduction and efficiency. In contrast the Italian model assumes that the person putting the model together is less of an engineer and more of a craftsman. Instead of surgically removing carefully shaped pieces from plastic sprue, you cut various strips of wood to the desired lengths and shapes. This style seems to give the modeler a much greater degree of freedom--and of course a much greater possibility of screwing up.

The translations on the back of the box are quite humorous. I realize that translations that butcher English are probably just above puns in the hierarchy of humor, but I just can't help myself. Consider this:

Necessary In The School: Necessary because it has the essential characteristics of material used to didactic level, both in particulars already arranged and in which to get.

I think this is trying to extoll the educational virtues of model-making. I've noticed that a lot of hobbies tend to introduce themselves to newbies with this kind of language: "Insert random hobby here enhances brain development, aids in digestion and could bring about world peace in our time."

I'm not sure why hobbies feel they need to do this. Perhaps it's some desperate attempt to bring some legitimacy to hobbies that are often characterized as reclusive. I've been known to build a model or two in my time and have enjoyed the process immensely. But it is a good idea to put down the model-glue and go walk the dog, look at the sky and chat with the neighbors.

Merry Christmas!

2004-12-25T10:04:00.000-08:00

A Picture Share!
Originally uploaded by livollmers.

This is a Danish Kringle from Larsen's Bakery here in Seattle Washington. This might be one of the tastiest things I have ever eaten for breakfast. Sugar, and his good buddy butter, both put in strong performances here. Yum!

Mauling a cereal box

2004-12-21T09:14:00.000-08:00

Mauling a cereal box
Originally uploaded by livollmers.

Here is an example of a desperate parent trying to open a box of cereal while their 1-and-a-half-year-old is howling for breakfast. The box is now very good at opening, but doesn't perform so well in the closing department.

Speaking of asymmetric performance, I've been thinking about the various site syndication aggregators I've been playing with and I haven't found one that I really like yet. I want one that allows me to drill-down into a site's update as easily as possible, but not to load anything more than I want. I want a very low-impact UI that keeps me focused on one thing at a time. First I want a list of the sites I follow and a simple visual indication if they have new content or not. Then I want to be able to drill-down into a particular site and see what new content they have. Then I want to be able to hover over the topic and get a quick taste of the update with the ability to click it and be taken their in my browser.

The closest tool I've seen that does this is the Sage extension for Firefox. I just wish it didn't take up a full-height sidebar in the browser and instead of generating summary pages in the main browser window, just left it alone until I actually clicked a link in Sage.

Either I'll learn to live with Sage, figure out some configuration tricks, or start learning how to code GTK in Python so I can use the well-crafted feedparser RSS parser for Python.

Bitter Humor

2004-12-19T13:11:00.000-08:00

A Picture Share!
Originally uploaded by livollmers.

I caught this bumper-sticker this morning while we dashed out to get some last-minute holiday shopping done before the crowds converged. I've tried to get over my bitter election-day disappointment but on a day when it's announced that President Bush will be Time's Person Of The Year, this helps take the sting out.

Spending Spree

2004-12-17T08:22:00.000-08:00

ballpark
Originally uploaded by livollmers.

The big-money moves made by the M's in the last two days should make going to the ballpark more likely to revolve around watching baseball rather than the hydro races, the blooper reel or cap dance. If nothing else the M's have made a huge splash by getting both Richie Sexson and Adrian Beltre.

Beltre is a curious fellow. Here's a guy that had power projected throughout his young career but, until last year, had the tag of never-developed promise. You look at the guys number, isolated power, even using extra-base hits as a prognostication of future home-run power and no one could have predicted the monster year he had last-year. One important thing to remember about Beltre is that his home/road splits for home runs was nearly even. The dude hit twenty-three home runs in Dodger Stadium! If he can jack that many pitches in that canyon, surely he can hit a few out at Safeco.

In other news it looks like the BCS is beginning to formalize a proposal to consider creating a preliminary exploratory committee to possibly consider that the BCS alliance and its attendant bowls are not working. We can put a man on the moon but we can't figure out who the real national champion is?

It drives me crazy when people profess a greater love of college football over professional football because it's more honest and genuine. I would submit that it is exactly the opposite--because colleges are ostensibly about academics high-powered football schools have figured out ways to sneak poor academic athletes through the system. In addition the so-called amateurism of the sport should be called into question: both from the perspective that college players are being exploited by schools for money and in violation of existing amateur-status rules.

What really drives me crazy about college is that it is always the same schools in the hunt for national title contention every year. The parity of the NFL is one of the most exciting aspects about watching from year to year. You never know who is going to come out on top. In college football you can safely bet that the national champion will come out of a short-list of well-known football schools.

Back in the Saddle

2004-12-16T09:20:00.000-08:00

After two days of super-achey flu fun, I finally dragged my sorry rear-end back into work and got back in the saddle again. A couple of exciting news items. First, the Mariners have signed Richie Sexson, which I think is a good thing. I won't quibble over the salary or the deal, because it's nice to see the M's make a splash for once. Let's hope this isn't Kevin Mitchell all over again.

The other bit of cool news is that Firefox is running their full two-page ad in the New York Times today. I've said it before, and I'll say it again. If you're still running Internet Explorer, please for the love of God stop and use Firefox instead.

Eureka!

2004-12-12T21:07:00.000-08:00

I'm your typical left-wing, foreign-car driving, NPR fan. I love screwing up Arbitron ratings when I tell them that I listen exclusively to NPR. I'm also a big sports fan. Every Friday you get John Feinstein or Frank DeFord to come on Morning Edition and talk a wee bit o' sports. I always enjoy these pieces--I only wish there was more. So, what if NPR produced a weekly sports show?

I'm not talking about the usual sports-talk, call-in drivel that we are all familiar with. No rockin' guitar music at the break, no super-echoey low-pitched introductions, no beer ads. What I'm talking about is a George Plimpton-like sports-talk show where intelligent fans discuss sports and the social and economic issues around sports. Better yet it would have guests that are only peripherally associated with sports, but can talk eloquently about the game. The usual interviews with athletes, coaches and managers tend to devolve into simple platitudes (see "Bull Durham" if you can't remember how the routine goes).

Honestly NPR doesn't need another guess-the-real-definition-of-this-obscure-word game-show (though I do like them). Let's put some fun back into the NPR weekends and talk about sports!

Draft Strategies

2004-12-10T20:08:00.000-08:00

I ran across this whitepaper this week about the various draft-day strategies employed in classic Rotisserie Baseball leagues. Of particular interest is the notion of "information assymetry", which is an fancy economic term for the difference in information available to all parties in a transaction. Given the glut of information that is readily available at the click of a mouse to the average roto-league owner, relying on beating your opponents by beating them in the information war is a questionable strategy.

If decisions are guided by economics (i.e. valuing players in the most economically efficient manner), teams can maximize their draft day returns. In our league last year we removed a major component of draft day strategy which is bidding on the players. Teams simply selected players when their given draft slot opened up. When the value of a player is bid on (not just defined by when they were drafted) it adds a whole new dimension to the game.

It got me thinking about my particular profession--software development. Relying on "information assymetry" as a means of longevity in this industry seems like a fool-hardy strategy. The road is littered with lots of people who immersed themselves in the expertise of once-prized, now-dead technologies.

Testing Out Flickr

2004-12-10T19:13:00.000-08:00

Learn Unix!
Originally uploaded by livollmers.

One of my co-workers turned me onto flickr, an online photo storage site. Sure there are a ton of these (Ofoto, Yahoo, etc.) but these guys seem to be bringing the whiz-bang factor to online photos that Google brought to webmail.

I have to come clean and admit the biggest draw for me is that I can easily publish photos to this blog!

Sunday Sports Section

2004-12-05T07:44:00.000-08:00

Perusing the Sunday edition of the Seattle Times this morning I came across two football-related items worth mentioning. One was an advertisement by Bon-Macy's selling Jerry Rice Seahawks jerseys. No Koren Robinson. No Darrell Jackson. No Itula Mili. No Jerramy Stevens. Apparently the 'Hawks pass-catching inabilities are so bad that they are finally affecting retail decisions in the holiday season.

In other news, Tim Brown--Jerry's teammate in Oakland--was angry about the booing his QB, Brad Johnson, got at home last week. He was quoted as saying, "...when they boo Brad, it shows you're coming to the game to be entertained, not to be supportive." Uh, hello Tim? You guys aren't politicians, public servants or soldiers--you are entertainers. I would like to think that most NFL player are not this out-of-touch but Shaun Alexander's similar rant may prove me wrong.

Guess what guys? Fans dole out a lot of money to see you play. Fans want the team to win so badly. The team is bigger than you. When you suck, the team is hurt. When you continue to under-perform you get replaced. Booing is a form of public referendum of your on-field performance. If you don't like it, then don't play.

Tyler Tyler Tyler...

2004-12-01T08:41:00.000-08:00

There's sad news in the cycling world today, as it was announced that Tyler Hamilton was dropped by his team, Phonak. I had considered Hamilton's performance in the 2003 Tour de France one of the grittiest displays of personal courage I had ever witnessed in sports. Riding through the Tour with a broken collarbone suffered on the first stage, he managed an individual stage win as well as a finish in the top five. My wife and I even used Tyler as inspiration for suffering through pain when she was in labor with our child.

It's disappointing to see such a figure taken down by doping. The saddest thing about this story is that it removes what little hope I had for any clean performances in cycling. When you look at this story you have to think to yourself, who isn't doping?

GRUB Conquered

2004-11-30T20:42:00.000-08:00

Thanks to some help from some people at work who are much much smarter than me I finally beat GRUB into submission and got the damn Linux box to boot properly. I think GRUB is pretty nifty, but the documentation goes right into a bunch of low-level details and it's hard to get the big picture.

As I often do, I ended up relying on the frustrating experiences of others to get me out of technical jams. I simply don't have the patience to learn all of the vagaries of booting personal computers.

A Battle For The Future

2004-11-30T10:49:00.000-08:00

I stumbled upon this article about why Java will not be the language of the future. The author makes the assertion that because using XML is so painful in Java (which I won't disagree with), Java is a lousy language to do lots of text processing with. This misses the essential point of a good object-oriented language. A good object-oriented design should push the XML-specific concerns to the very ends of the request-response model. The bulk of the code should have a nice domain model that has XML transformations at the beginning of the request (into objects) and at the end of the response (into XML).

His assertion is based on the notion that XML (or XML-like structures) would pervade your code and be your domain model. This smacks of the kind of reductionism that views the world simply as lists or parallel arrays. The implementation and storage are intimately tied together completely removing any layer of abstraction. The fundamental problem with a lot Perl/PHP solutions is that there is little separation between the bare-metal shuttling of database data onto web-pages.

There is an argument to be made for avoiding an overly grand design. For certain web applications I can understand the frustration developers would have dealing with the behemoth that is the J2EE web container to build a simple interactive application. But somewhere in the continuum of complexity the cheap and easy, bare-metal solution won't scale. It becomes way too hard to maintain that kind of application because there are no layers of abstraction. These layers should be here not for the sake of high-minded academic design, but to serve as firewalls to change throughout the application.

It Hurts

2004-11-29T21:44:00.000-08:00

Sometimes I think technology is just a giant time-suck--a voracious obligation that does not improve the lot of humanity. Case in point: I have spent the past two weeks trying to upgrade my Linux box to a gigantic 160GB hard-drive. After numerous problems copying the damn files from the old disk to the new disk I ran into all kinds of booting problems.

Right now I've hacked together a GRUB boot-disk that seems to get me up off of the ground. But if I try to boot off of the new hard-drive I get GRUB GRUB GRUB GRUB GRUB when the disk should be booting. Very irritating. If I weren't afraid of waking my daughter I probably would have loudly broken a few things by now.

Grubbin' Around

2004-11-20T11:33:00.000-08:00

Among the eight-thousand to-dos on my list, one was putting a new 160GB hard-drive in my Linux box. Every three or four years I seem to reach a point where I've maxed out my memory/CPU/hard-drive and buy something bigger and badder than ever. And without fail I always think to myself , 'Ha! It will take forever for me to use this up'. And yet, here I am facing another upgrade.

I switched over to using GRUB as my boot-loader about a year ago, but never really took much advantage of it. Now I can see how helpful it is to have a boot shell where you can experiment with a bunch of options before committing. Nifty nifty stuff.

In Summary...

2004-11-12T22:24:00.000-08:00

I made a nice bit of progress on the simulation program. Tonight's big goal was to get League objects to understand how to rank their Teams. There is more to this than may appear at first glance. First, you are not simply ordering the teams 1st place, 2nd place, and so on. Instead you are awarding points to each team based on their position in the category. For a ten-team league, the first place team gets ten points, the second place team gets nine and so on.

Then it turns out that each scoring category has its own notion of what is "better" for the purposes of ranking. It turns out that this was relatively simple to express in the Category instances themselves. This kept the code in the League relatively straightforward and pushed a lot of the details about per-category ranking into the Category objects themselves.

By putting the power of the Collections package to use I simply had to stuff a few objects in a List and hand it a custom Comparator to sort them. Piece o' cake.

Now the next challenge is to start working on the innards of the Hibernate queries and get them plumbed into the PlayerPool. Once those are all running I'll be ready to run a maiden-voyage of the simulation. I still need to add a little more Hibernate plumbing to write-down the results. After that, then I can consider grabbing some graphing package and doing a little data-analysis. It's now over yet...

XDoclet

2004-11-11T08:36:00.000-08:00

For you non-geeks, just walk away now. There is nothing here you will want to continue to read and your opinion of my sense of humor will only drop. Please, just browse somewhere else...

A few weeks ago I ran across this cartoon on TheServerSide. I had heard of XDoclet and the notion of using "attribute-oriented programming" to get around the god-awful maintenance nightmare that is EJB deployment descriptors. We ran an experiment at work to start using EJBGen which was a very Weblogic-specific doclet kind of package. While we liked getting rid of deployment descriptor maintenance, we decided that since it's creator left BEA to work for Google we had better look for a tool that would be better supported. So a couple of days ago I started converting some of our EJBs over to using XDoclet. While it beats EJBGen in a lot of respects, it still appears to be a work in progress.

My chief complaint is the sheer amount of stuff you have to specify. On a 21" monitor at 1600x1400 resolution the class-level doclet stuff takes up an entire screen! While the tool make a nice differentiation between default EJB stuff and container-specific stuff, there could be some sensible defaults. For example, when I want to declare an EJB reference I have to include a doclet tag at the default EJB level (presumably for the ejb-jar.xml file) and another at the weblogic level (presumably for the weblogic-ejb-jar.xml file). These declarations are nearly identical and it would be nice if the container-specific doclets had reasonable defaults that could be gleaned from the EJB-level doclets.

Additionally it would be nice to not have to put a little doclet magic on each public method in the bean class simply to tell XDoclet to add it to the generated interface. Could I just have a class-level doclet tag that allowed me to declare all public methods eligible?

In XDoclet's defense, it beats EJBGen for handling container-managed relationships. I found that I had to do much fewer mental gymnastics to get the deployment descriptors to come out the way we wanted them. The real root of this evil is the inherent complexity of CMR so you could argue that it's a better solution to a lousy problem.

So while XDoclet is a nifty tool, it has a few warts. But compared with the intensive manual maintenance that would otherwise be required by EJBs, these are relatively minor offenses. The real issue is using EJBs to begin with. I wouldn't completely eschew them in future projects, but I would probably limit myself to SessionBeans only. Having played with Hibernate I believe it is a much cleaner solution to object/relational mapping.

Hinge Update

2004-11-10T07:02:00.000-08:00

Last weekend I successfully operated on my wife's laptop and restored the hinges. While the bezel case around the screen doesn't quite fit as well as before, the screen moves properly and it works and, most importantly, it only cost me about $90. All in all, it was a very satisfying DIY experience.

The Kids Are Alright

2004-11-10T06:48:00.000-08:00

The baseball season is almost a speck in the rear-view mirror and it's time to think about this years fantasy baseball season. Despite their youthful moniker, the Swing Kids were anything but young this year. In retrospect this isn't terribly surprising. My draft strategy was based entirely off of past performances, and while the system could account for varying playing histories, youth was still at a disadvantage in this system. I certainly couldn't account for the huge drop-off in performance that Edgar Martinez exhibited this year.

The Kids also got burned with the injury bug. Any baseball fan worth their salt will immediately recognize the correlation between injuries and the ageism stated previously. On Draft Day I was tickled to get Magglio Ordonez and knew he would be at the center of my team. He had a blockbuster year until he got hurt. Kevin Brown also missed playing time (yes I knew his injury history going in).

Another strange occurence was the number of saves Keith Foulke picked up. I thought for sure that joining the Red Sox I'd be looking at a 45+ save season, instead he only had 32. I don't have the analysis to back this up, but if memory serves I recall that the Sox either blew teams out a lot or were blown out in their losses. It would be interesting to graph a distribution of the Sox' win/loss margins against the league average.

Finally, I made almost no adjustments to the roster during the year other than picking up free agents for injuries. The winner of our league pulled himself up from deep in the standings to the top with a mass of savvy trades. I have to admit that I did not apply the same fury of analysis to roster moves as I did to drafting. Ultimately you need to be able to look at the players you are losing and what you are picking up and deciding if you are going to be better. I don't just want to feel that I'm getting better, I want to know that I'm getting better. I don't have a system that allows me to do that...yet.

My little simulation experiment will hopefully shed some light on defining what winning teams look like. If the data shows any sort of trends, this will answer the question of what categories are most important for winning. Next year I'll factor in youth a little better into the pre-season analysis.

The Red And The Blue

2004-11-10T06:34:00.000-08:00

It's been a while since I've posted. Lots of big world events have happened and my blog has lain abandoned like a beloved stuffed-animal in a confused adolescent's bedroom.

Since Nov. 2 I've had to fight both rising tides of dyspepsia and typical left-wing progressive snobbery. The fact of the matter is that I'm extremely disappointed in the results of the election. Even more so when I read that exit-polls showed that "moral values" trumped all other voter concerns--even when voters admitted that Iraq was a mess/mistake and the economy was in the toilet. To me the decision to go into Iraq with no evidence, to continue to alter the rationale for going in and to continue to prosecute this action are at the core of a tangible "moral values" debate. This administration's moral code seems ethereal at best.

Why did W win? I think most Americans are living in fear. Some of these fears may be justifiable. There's nothing of true substance this administration has done to prevent another 9/11 (don't get me started on the TSA--what a joke), events have simply unfolded that way. We could be attacked again in a similarly devastating manner any time. This administration has done a fine job fanning the flames of Muslim discontent throughout the world.

Some of these fears are at the core of the schism between the so-called red and blue states. Any discussion of legislation toward gay marriage draws both sides to their respective moral fortresses with a vast wasteland remaining as the middle-ground. I admit to being on the left side of this issue. I simply cannot understand how two people's decision to forge a personal bond with one another is a threat to someone else's marriage. Nevertheless, this issue seems to be an important one to the right and one, that when pressed, most voters on the right would admit to being a big part of the "moral values" issue.

I truly wish the best for this administration. I want them to do well, no, I need them to do well. I shudder at the possibility of a government as large and powerful as ours run by Bible verses and the mistaken notion of a mandate.

At the very least they shouldn't be able to screw up baseball much--even if we have to continue to suffer through that dreadful dirge known as "God Bless America" in the seventh inning.

Cross Your Fingers

2004-10-31T11:42:00.000-08:00

It's time to start handing out the uber-geek points because today I took the plunge to fix the broken hinges on my wife's Powerbook. I felt like a total tech-troglodyte, cracking the glue spots that hold the screen together to pull the old broken hinges out. For all of Apple's ultra-futuristic design, this one must have been done by one of the summer interns.

It is satisfying though to fix the thing when the equivalent service through Apple would have cost about $1200. It's really the same part of me that likes running Linux and hacking my Tivo. It's my equipment dammit! I'll fix it how I choose.

Hibernation

2004-10-29T07:41:00.000-07:00

After a nerve-racking pair of League Championship Series, the World Series turned out to be a bit of a snoozer--not that I minded. I was pulling for the Red Sox all the way. Has any team ever gotten hotter at a more critical moment in the post-season?

Sadly, the last out that put an end to Boston's suffering also means that there will be no more baseball in 2004. Yes there's much football to be watched (including this weekend's battle of the bantam-weights that I'm looking forward to), but my mind finds itself always drifting back to baseball.

Fortunately my little fantasy baseball simulation has gotten some attention and I'm starting to get somewhere with it. I got over the first hurdle which was learning enough about Hibernate to figure out how to get statistical data from a MySQL database into my Java objects.

I've spent the bulk of my career worrying about Java/relational-database interactions. I've used and invented lots of technologies for making the job easier. I haven't seen anything that looks as good as Hibernate. It wins for three very compelling reasons:

Hibernate-oriented concerns don't leak into my domain model. The Team, Player, BattingStint, PitchingStint and FieldingStint classes are pure as new-fallen snow
Hibernate makes the easy stuff easy and the hard stuff possible.
Hibernate has lots of sensible defaults

My next task is to start thinking about how drafting would work. I can't simply take the 500+ available players and randomly spread them out among the teams in the simulated league. I would end up with a much higher percentage of scrubs among the teams than a real league would. Even a fantasy baseball rookie would at least arrive at the draft with some notion of the big stars and some form of fantasy guide under their arm.

So how to filter out the bottom of the talent pool? The most obvious thing is to start with a minimum number of at-bats (for position players and DHs) and innings-pitched (for pitchers). Note that two IP minimums must be set: one for starters and one for relievers. Right now my data doesn't have an explicit notion of the difference between starters and closers. We all know that there are a handful of class-A closers. After they are snatched up the remaining teams usually put together a bullpen-by-committee of guys that end up with 12-15 saves. Where is the line drawn?

I have suspected all along that the responsiblity of the Team class in the object model is to do the drafting. A Team instance will need to use some kind of DraftStrategy to help guide their selections through the draft. The tricky part here is to remember that I'm trying to keep a lot of the randomness in the system, so having a super-clever DraftStrategy would taint the results. The point of this exercise is to determine what a winning team looks like given a 10-team league using the category scoring the Moneyballers used this year?

Stay tuned...

I'll Pick My Jaw Up Off The Floor Now

2004-10-20T21:48:00.000-07:00

Go Sox! 'nuff said.

Spread The Word!

2004-10-20T13:38:00.000-07:00

In case you hadn't heard the Firefox project is soliciting donations to take out a full-page add in the New York Times to announce the 1.0 release of its browser. I made a donation today to put my name and my wife's name in the add so when it comes out get your magnifyining glass out and look for us!

For those of you who are still using Internet Explorer, for pete's sake go download Firefox and rid yourself of security holes, popups and bad performance.

The Golden Age

2004-10-19T21:49:00.000-07:00

I know that the playoffs are getting excited when a co-worker of mine who normally cares nothing for baseball is talking about the previous night's game with me when I come into work in the morning. After a lopsided start, the Yankees-Red Sox series has shaped up to be all that it has promised. All the analogies to fifteen-round heavyweight bout can't do justice to the epic nature of this battle.

I have to admit that I get a bit sucked in emotionally and can't help turning this series into a sort of morality play. There are few institutions in life that I truly despise but, as any reader of this blog knows, I despise the New York Yankees. For me to frame this contest as a David vs. Goliath battle is a bit erroneous considering that the Red Sox are certainly not the David-figure that a team such as the Minnesota Twins are. But I can't help wanting to see the Yankees and George Steinbrenner finally get their come-uppance.

Tonight merely adds to my emotional connection with the series. To see Curt Schilling out there, in obvious pain, bleeding through his sock and shutting the Yankees down seems like an honest display of both courage and humility in the face of the arrogance of the cheap shot Alex Rodriguez took on Bronson Arroyo. I'm a big Curt Schilling fan (for baseball and geek reasons) and am thrilled with the performance he turned in tonight.

Baby-boomers love to wax poetically about the "Golden Age" of baseball--Mickey Mantle, Ted Williams, Stan Musial et. al. I would submit that we are seeing some of the finest major league baseball ever played. I've hardly talked about the Astros-Cardinals series, but that too has been an exciting series showcasing the awesome talents of Carlos Beltran and Albert Pujols. We should really savor this post-season as we really live in the new Golden Age.

What Makes A Winner?

2004-10-14T21:29:00.000-07:00

I've thought about doing a project where I simulate several hundred mock fantasy baseball seasons using, say, data from 2003. The hope is that looking at the average winner of each simulated league would give me some idea of what makes a winning team given our league setup. Right now I'm not considering any mid-season roster adjustments so who gets drafted is who is on the team for the season. Hopefully with enough sample data any outliers would not stand out.

For this project I've decided to delve into some new technology. I'm going to use Hibernate as a way to map relational data from a MySQL database into Java objects (which I dearly love). I'm also relying upon a couple of trusty tools like Eclipse for my Java IDE and Maven for my build system. In about five minutes I created a simple Maven project file, specified my Hibernate and MySQL dependencies, built my Eclispe project and classpath files and then imported my project into CVS. Man, Java development has sure come a long way from the days of editing my Java code by hand in XEmacs.

I am planning on posting a final review of what went right and what went wrong with the Swing Kids this year. I haven't really spent the time to formulate the various ideas that have popped into my head in the last month--but be sure I will address it.

C'mon Collapse Now

2004-10-11T21:27:00.000-07:00

I've watched a lot of sports in my life, but I'm not sure I've seen too many last-minute meltdowns like the one I watched the Seahawks performed on Sunday. How do you blow a 17-point lead late in the fourth quarter? How can a team that so thoroughly dominated its opposition suddenly evaporate so late in the game?

It seems that every generation of sports fan has their own version of the archetypal playing-not-to-lose game. In this type of game a team is winning (or dominating) in a way that is completely unexpected. As the end of the game approahces they turn to a very conservative game plan in order to consolidate their gains.

The Seahawks offense on Sunday became very conservative by running consecutive between-the-tackle plays resulting in two three-and-out series. Not only did this give the ball back to the Rams, but it gave the Seahawks defense little chance for rest. Similarly, the attack-from-any-angle (I'm big into hyphenated phrases today for some reason) that Ray Rhodes kept the Rams off-balance with evaporated into passive converage that only turned aggressive again when it was too late in the game (the winning touchdown in OT was against a two-deep blitz).

Like any sort of sporting analogy drawn up against life, this one should be taken with some caution. But it seems to me that there is an important lesson to be learned--never quit while you are ahead.

More Cheese Please

2004-10-08T09:04:00.000-07:00

Apparently I'm not the only one that found "Scooter" the talking baseball a surprise irritant in Fox's post-season baseball broadcasts. One quick Google search revealed and outpouring of hatred rapidly approaching Jar-Jar Binks levels for Fox's new mascot. While I support exposing fans the finer points of the game--such as the difference between a slider and cut-fastball--I would rather hear it from actual baseball players, not some high-pitched (no pun intended) animated baseball.

In other news my daughter ripped the screen of my wife's PowerBook clean off of its hinges. I don't know what we feed her but it made me feel like Barney Rubble and Bam-Bam. The laptop is still somewhat intact, but quite floppy. A quick check with the local Apple store revealed this to be about a $1000 fix. Ugh. Apparently this is not an uncommon problem. But, thanks to eBay, I found someone selling replacement hinges (and instructions!). God bless the Internet.

Bye Bob

2004-10-05T08:53:00.000-07:00

The Mariner's showed Bob Melvin the door yesterday. I'm not particularly surprised by the move, but I think it's interesting the way we have come to expect these sorts of decisions. Bill Bavasi was quite clear about not laying all of the blame for the pitiful 2004 season at Melvin's feet. But in the end Bavasi still has his job and Melvin is out on the street.

My personal opinion is that Melvin doesn't have the track-record (yet) to deal with the extremely young, raw ballclub the Mariners will be in 2005. He was a fine steward for a mostly veteran (bordering on ancient) team, but the M's are going to need a fire-breathing ass-chewer a la Lou Piniella. Let's see what Howard Lincoln and the boys do this time.

The End Of An Era

2004-10-04T08:25:00.004-07:00

I had the fortune of making it to Edgar Martinez' last professional baseball game. Edgar didn't get a hit, but the crowd cheered like mad at every plate appearance. In many ways, he is the essence of the Seattle mind-set: work hard, be polite and help your team. Edgar was also the centerpiece of the M's strategy of making the team about personalities, not just on-field performance. My wife really wanted to go this game because she was a big Edgar Martinez fan. She was the main force behind naming our dog after #11.

While I'm sad that the M's sad season is over, I'm very excited about the playoffs. We've got some new faces (Cardinals, Dodgers, Astros) and some familiar ones (Red Sox, Yankees, Braves). These all seem like "legitimate" teams--no Florida Marlins.

In unrelated news, I broke down and bought a copy of Madden for the XBox this weekend. I stayed up late, way past my bedtime, doing the mini-camps and practicing rushing plays between tackling dummies. This game is simply amazing. I learned more about football in two hours of playing Madden, than I have watching entire seasons of football.

Joe Montana once said that he didn't like watching football on TV because the camera shots were too focused on the lines and offensive backfield. To him the play was dictated by the secondary and, in particular, the safeties. When you play a game like Madden you really get to see the formations much better. When you're watching TV you have to quickly count the number of defenders, sort out linemen from linebackers from cornerbacks and then subtract that from eleven. Whatever is left over is what's in the defensive backfield that you can't see. Then as the plays progress you have to remember where these guys were so that when someone makes a tackle, you can judge whether or not the defense is doing its job.

May The Force Be With You

2004-09-29T21:28:00.000-07:00

With all of the hoopla surrounding the release of the original "Star Wars" trilogy on DVD, I thought I would dust off my trusty old VHS copy Episode IV and spend some time with an old friend. If you haven't been following the controversy surrounding the edited versions of these films, the basic issue is that George Lucas has made all sorts of changes to the film. The one that seems to raise the ire of most Star Wars fans is the revision of the Greedo vs. Han Solo scene. In the original film Han Solo dusts Greedo in the cantina before Greedo ever gets a shot off. In the theater re-release put out a couple of years ago, Greedo shoots first (and misses horribly) and then Solo blasts Greedo ending with the wonderfully laconic line, "sorry about the mess". Apparently in the new DVD version their shots are close to simultaneous.

Lots has been written on this topic so I'm not going to re-hash it. I'm a big Star Wars fans (of the original films). I'll admit that I've seen the first film probably somewhere around 120 times and know the dialog inside and out. But this sort of stuff just doesn't really seem to bother me. If George Lucas wants to work-over his films and re-release him--that's his deal. I have to admit that the call of a full disc of bonus features and cleaned up re-mastering is really appealing.

As much as I enjoyed my dog-eared copies of those films, I willing to part with sentimentality and watch a cleaned-up version of some of my favorite films of all-time.

One For The Constitution

2004-09-23T08:29:00.000-07:00

The NY Times (free subscription required) is reporting that the U.S. is letting Yaser Hamdi go after being locked away for over two years at Guantanamo Bay in Cuba. The Supreme Court ruled the erstwhile "enemy combatant" could not be held incommunicado without evidence. The court stated "a state of war is not a blank check for the president". It gives me a glimmer of hope that a body like the Supreme Court will stand up to this adminsitrations Machiavellian approach to the so-called "war on terror".

I'm no legal scholar but I'm pretty sure that due process, access to a lawyer and the burden of proof are the pillars of the American justice system. Since the term "enemy combatant" was coined by the administration we've seen these cherished principles trounced on again and again in the name of security.

While the U.S. criticized Vladimir Putin's power grab (and rightfully so) we should be reflecting on our own activities. Where Russia has Chechen terrorists, the U.S. has Iraqi suicide bombers. Where Russia has defacto state-controlled media, the U.S. has coroprate-controlled media. Where Putin has stripped Russia of gubernatorial elections the U.S. has stripped its own citizens of their constitutional rights.

At the center of this balance-of-power tipping by the executive branch is the notion that we are in a state of emergency and that desperate times call for desperate measures. We cannot afford to let a cabal of officials elected without a mandate to so radically alter our justice system and our standing in the world.

Random Stuff

2004-09-15T13:46:00.000-07:00

It's been a long time since I posted anything. The last month has been absolute hell for my sleep schedule and I've survived off of little but my wife's fortitude and an awful lot of coffee. However life is not all grey, sullen exhaustion.

I've been completely obsessed with The Vines new album "Winning Days" which I bought off of iTunes recently. I listen to little music off of any other format than my iPod and it was only three weeks later that I burned "Winning Days" to a CD and actually listened to it in the car. Anyway, it's a great album.

Other obsessions have included watching "The West Wing" with my wife after Audrey goes to bed. We bought the first two seasons on DVD and have been doing parallel-time by following the fifth season thanks to Tivo.

Work has brought new challenges--some fun, some taxing. This week we got into heavily-interactive web applications with JavaScript. I've done enough JS to enjoy it but also to know how frustrating the multi-browser support can be. I like to think of playing with JavaScript as being like riding a mo-ped: fun to do, but you would never want your friends to catch you doing it.

Gridiron Improv

2004-09-09T21:27:00.000-07:00

What a season opener! If you didn't catch the matchup between the Colts and Patriots, you missed one heck of a ballgame. While it did get a sloppy at the end, I can't help but feel like the offenses of pro football are going through a major evolutionary change.

Since the 80's the West-Coast Offense has dominated professional football. This scheme is characterized by genius coaches (all of whom fall from the Bill Walsh tree) that control their teams like a Formula One driver controls a finely-tuned car. But with the success of the Colts and their no-huddle but not-in-a-particular-hurry style of offense, teams are taking offensive decisions farther from the sidelines and closer to the end of the playclock.

If you will indulge a music analogy, traditional offenses are like orchestras with conductors on the sidelines and in the coaching booth above. Players focus on their particular job and the coaches coordinate the big picture. The new Payton Manning-style offenses are closer to improvisational jazz quartets (think Charlie Parker or John Coltrane) where each member of the group has a collection of "riffs" they can pull out at any time. Like the great jazz groups each player is a virtuoso in his own right, but his true strength lies in understanding when to apply his best licks.

When you watch the Colts work a defense it's like no other. Marvin Harrison will read the defense as Brandon Stokely goes in motion. Harrison breaks to the inside as Stokely takes his man down the sideline. Uh oh, there's the safety coming up to cover Harrison. No worries, the TE reads this an releases his block into the flat for quick pass that ends up getting fifteen yards.

These are not scripted plays--each player would have to know a thousand plays to cover the possibilities. What they now are the fundamental building blocks of great plays. What makes them fun to watch is how they assemble them together.

Is this a new era of offenses? Consider that one of New England's biggest off-season aquisitions was veteran Corey Dillon to shore up their running game and that their first eight player were no-huddle, four-wide pass plays with no backfield. He didn't make it into the game until the Pats were in the red zone on their first drive. I think that this sincerest form of flattery speaks volumes about the impact of the new QB-driven game.

All Hail The Pigskin

2004-09-05T08:00:00.000-07:00

It didn't take long for me to get excited about football season. I caught most of Saturday's exciting, but heartbreaking, game between Oregon State and LSU. What picture better portrays the agony of defeat than Alexis Serna throwing his helmet off after missing the PAT in overtime after Derek Anderson kept the Beavers alive with a fourth-down touchdown reception? Ladies and gentlemen, this is what football is all about!

Despite my lifelong love of the game (longer than baseball!), I've become increasingly jaded about the college and pro game. Every year another BCS screwup drives me into an apoplectic rage. The NCAA looks the other way about a game that is essentially the minor-leagues for the NFL. The pro game is so filled with prima donnas and big money that teams are essentially "bought", not developed organically.

But I'll tell ya what folks, as soon as I watched the OSU-LSU game my brain clicked-over into football mode. Quickly, count the number of defensive linemen and linebackers. Okay four up front and two behind. Looks like a nickel package. The football juices got flowing and now I can answer "yes" when Hank Williams Jr. asks me, "are you ready for some football?".

A Triumph of Mail

2004-09-04T10:42:00.000-07:00

GMail Rocks!

This week I got into the secret GMail society and I have to say that I'm awfully darned impressed by what those guys have been up to. I'm a keyboard guy--I use the mouse as a last resort but find it a clunky interface to the computer. GMail has all sorts of wonderful keyboard shortcuts that quickly make you forget you are using a web application.

GMail Epiphany

I'm very meticulous about keeping a clean inbox. My first impression was that it was strange how relatively difficult it was to send mail to trash as compared with other keyboard-enabled actions. Then it dawned on me that that's goal of the 1GB space given to each account. Don't worry about cluttering up your inbox--keep everything and label it. The real benefit is when you take advantage of the searching capabilities of GMail. Now I have the best of both worlds: an easily searchable archive of mail and a nice way to keep my inbox clean.

Migrating to GMail

I spent a good chunk of last night and this morning putting the finishing touches on a chain of freeware programs (fetchyahoo and GML) to automatically forward mail from my Yahoo account to my GMail account. Once the mail arrives in my GMail account it has a special "Yahoo Mail" label applied to it that makes it very easy to see where the mail originally came from.

In Defense Of Duke

2004-08-25T10:54:00.000-07:00

There was a post on Slashdot yesterday about why Java development isn't considered "cool". As a Java developer who is enthusiastic about the open-source world, this particular thread of conversation really hits home for me. I figured I'd avoid the flame-war on Slashdot and instead try to craft a response on this here blog.

There seem to be two main complaints from non-Java people: Java is too slow, and the Java language is too verbose. The first complaint comes from, at worst, C and C++ developers who think that "real" developers are braniacs that live in the world of subtle memory-management bugs and pointer arithmetic. Complaints by the latter are from, at worst, Perl acolytes who believe that those who win the Obfuscated Perl Contest are geniuses.

The usual response to the first argument is the recitation of Moore's Law--computing power is increasing so quickly that basing decisions on which language to use purley on performance is meritricious at best. Those, like myself, that love the Java language would hate to see it discounted in a project so easily for those reasons.

The second argument is that Java is too verbose. What takes two lines of Perl code requires ten to fifteen in Java. There's no argument that Perl beats Java in terseness but, like the performance argument, it is a cheap shot. Anyone using a modern IDE can reduce the verbosity problem to nothing.

While Perl is certainly terse it suffers from a "multiple-dialects" problem as a result from it's all-too-flexible "there's more than one way to do it" philosophy. This is where the cost of code maintenance is potentially much higher for Perl--unless each developer understands the same dialect and idioms, Perl's super-flexibility can produce some very difficult bugs. You have to tip your hat to the Python guys--they seemed to really get the terseness/object-orientation mix just right.

With the open-source world being so Unix/Linux slanted, it won't lose it's C-bias for a long time. Given the timing of Perl 5 and explosion of the Web we probably won't see Perl disappear anytime soon either. However I think it's unfair to portray the open-source world as entirely hostile towards Java. Despite some questions about Sun's handling of licensing, there is a vibrant open-source Java community. I guess that rather than getting into flame-wars, most of us Java guys just want to get the job done right.

Site Changes

2004-08-22T21:18:00.000-07:00

I got tired of the old look of the site so I thought I'd try out some other layout. I'm all fired up about web layout from reading some CSS/web-design books from work so I'm starting with a minimal layout that, time permitting, I'll get a chance to hack. Hope you like it.

Power Trip

2004-08-21T08:19:00.000-07:00

This past weekend the ultra-sleek design school-inspired power adaptor for my wife's laptop crapped out. We had to make a trip to the Apple Store, and $85 later we had a new power supply. While we were both a little shocked at the price of the power adaptor, I pointed out to my wife that while they may charge $85 for a power supply, the alternative was cheaper PC hardware (and $15 power supplies) with the albatross of Windows hanging around it's neck. We both agreed that the $85 power adaptor was a bargain at twice the price.

It led me to think about those faced with fewer options. People who can't afford more expensive hardware or the time to get down with Linux are left with no alternative but Windows. I don't know if Linux will ever really reach critical mass on the desktop, but it's too bad. There are some wonderful window managers out there and in this day and age there isn't much you can't do on a Linux box that you can do on a Windows box.

On a related note, I was talking about power adaptors with a co-worker this week and we both remarked at how few wall-warts play well with others. They tend to dominate both vertical and horizontal space making them difficult to use either in the wall or on your standard Ace Hardware power-strip.

Why is this still the case? Sure I've seen special power adaptors that accomodate larger computer-related wall-warts, but honestly--we've put a man on the moon for cryin' out loud! Can't we figure out how to make wall-warts get along?

You Ess Ehh

2004-08-16T08:32:00.000-07:00

Those of you who know me well know that I'm not the world's biggest basketball fan. For me, basketball lacks the poetic symmetry of baseball and the tactical mental-engagement of football. So it is with a bit of schadenfreude that I chuckled to myself when I learned of the U.S. Men's basketball team getting their pants pulled down by Puerto Rico.

To watch these guys swagger into the Parade of Nations for the Opening Ceremonies you would think they had better things to do. Looking totally non-plussed and way to cool for everyone else, it's nice to see some earnest nation smack 'em good across the face.

Apparently I'm not the only one that feels this way either.

A word to the wise you cocky upstarts: there will be no U.S. baseball team at the Olympics this year because they weren't good enough. Yes, even for a sport born in America, the U.S. has no birthright for an automatic bid to the games; you have to earn it. I would imagine that the so-called "Dream Team" players may be starting to catch on that this isn't simply a vacation to Greece.

Reinventing The Wheel

2004-08-13T08:07:00.000-07:00

NOTICE: Software-related post ahead...

I've seen this enough it's starting to drive me crazy so I'm going to get my frustration out in a blog post. I write software primarly using the Java programming language. For better or for worse, the Java APIs have a huge scope that cover a broad range of topics from web applications to 3D graphics to database access. One of the most important skills you can have as a Java developer is intimate familiarity with these APIs.

What drives me crazy is when somebody decides to build a "framework" (and I use that term derisively) that essentially re-maps the API to the way the framework-writer would have solved the problem had they written the API. This does not move the ball down the field any further. This does not add to collective intelligence of the developer community. Instead, it smacks of ego-driven development that is best left to people who think that staying up for a week straight on a diet of Coke and Ho-Hos and hacking Perl makes them gods.

There is, I believe, a certain amount of criteria a framework has to satisfy before a line of code is written:

Does it enhance the existing API?
Does it simplify the use of the existing API (perhaps by reducing boilerplate code)?
Does it clarify the solution the existing API is trying to solve in a better way?
Is the new framework no more (and hopefully less) invasive in your code, especially your domain objects?

I think if you can answer "yes" to all of these you can start writing your framework. Otherwise your framework probably doesn't really solve the problem. In fact, it's quite likely that if you can't hurdle this barrier, you don't really have a good grasp of the problem.

While I don't like every bit of each Java API, I believe that smart people spent a lot more time thinking about the problem than I have and deserve the benefit of the doubt. Software development and human/computer interaction will never move forward if we don't start leveraging (ewww, business-speak) what others have done and build on it. So check your ego at the door dude and RTFM.

Chandler In The Newsroom

2004-08-11T08:16:00.000-07:00

As I'm browsing this morning's ESPN baseball headlines I came across Jason Stark's column titled "Marlins so needs Beckett to Shine" (emphasis mine). I realize that this Chandler-ism has taken hold in daily conversation (I myself enjoy using this construct for comedic effect), but I would expect even a high-octane rag like ESPN to at least stick to proper journalistic vernacular.

Speaking of ESPN, I've subscribed to The Mag since it came out and either my tastes have change or the journalism has really gone downhill. I don't even bother to read them anymore, instead I pipeline them from the mailbox to the recycling bin. If I had the inclination I would call and cancel my subscription in an attempt to save a few trees. What happened? Of the 40% of the magazine I actually read (all the articles not about Donovan McNabb's crib) I realized that they all fit nicely into the following categories:

Struggling African-American athlete overcomes all odds and makes it big with a chip on their shoulder

The Misunderstood super-athlete who has finally shunned the media out of frustration. We are supposed to admire their tortured genius and pray they don't retire soon

The obligatory X Games plug. "Yo! Check Jonesy Half-Pipe the hottest Extreme Rubber-Band Rider on the planet dude!"

The grizzled veteran making one more attempt at glory

Now I may not be any great student of literature, but I recognize that sports has a limited set of stories to tell. What shocks me about "The Mag" is how little art is involved in distracting us from this fact. You have to give Sports Illustrated credit for craftily engaging us with the same stories time and again.

Link of the Day

2004-08-10T11:17:00.000-07:00

Here's an interesting opinion on the state-of-the-art of sabermetrics by Mr. Will Carroll.

Jewel of the Northwest

2004-08-10T11:15:00.000-07:00

I just came back from a long weekend in Vancouver B.C. and can't help but feel like that city really got urban growth right. Now I will admit that going to there in August one can't help but feel like it has to be the most beautiful place on earth. Frankly Portland and Seattle are the same way. But I will attest to the fact that the city doesn't loose it's charm in January either.

Goodbye Edgar

In other news, Edgar Martinez has announced his retirement. You hate to see the guy go out after such a lousy season--he deserves much better. This is perhaps the final swan song of the post-1995 Mariners. The new M's will have to re-invent themselves with new lovable personalities. My wife in particular will miss Edgar. Heck, we named our dog after him.

Pleasant Surprises

It's nice to go away on vacation and come back to find your fantasy team jump up three spots in the standings. Jeez, I might have to make a trade or something just to see how far I can get. It looks like somehow I've gone from a team of iron-hands to a team full of Gold Glovers. Of course it doesn't hurt that a bunch of teams around me seem to be sucking. It might be time for another opportunity/risk analysis.

Dazed And Confused

2004-07-29T08:40:00.000-07:00

On my way to work this morning I passed by a protest put on by the SEIU outside of the UW medical center. Most protests seem to ultimately boil down to one bullhorn-enabled slavedriver asking the same questions over and over in an increasingly hysterical tone in the hopes of getting increasing enthusiastic canned answers.

Q: What do we want?
A: Fair contracts!
Q: When do we want them?
A: Now!
...

What struck me as interesting is that there were only two questions asked of the protesters by their protest commander. Given that they were marching in a tight loop answering these same questions over and over, the protest took on the look of a religious ritual where chanting would bring the participants into a higher state on consciousness. This part however, did not seem to be happening.

While the lady with bullhorn shrieked her questions unceasingly, the crowd responded with a dwindling fury. I began to think about the archetypal protests of the the 60's.

In my limitied, pseudo-scientific survey of one, it seems that those protests (admittedly over big things like war and racial equality) evoked a kind of revivalist meeting where the protesters just might walk until their legs became stumps and their throats voiceless from shouting. The one I observed this morning appeared to be D.O.A. This begs the question, are we less enthusiastic about protesting? Do we think protesting is a dying means of political expression? Are the topics we choose to protest about/for/against less compelling that forty years ago? Were people in the 60's just more eager to yell slogans and march in circles with signs?

In this age of cynicism protests seem like quaint rituals akin to, say, party conventions. I have always been curious about the effectivity of protesting. Did it work in the 60's? Does it work now? I don't know, but I'd like to find out.

I'm not the protesting type but, lacking any evidence that tells me that protesting is harmful, I support the right of people to do so. The cynic in me thinks that such activity doesn't stand a chance against the mechanisms by which the powers-that-be shape our public policy and society. I guess until someone can show me otherwise, protesting seems akin to charging tanks on horseback with swords--honorable perhaps, but ultimately futile.

The Yankees of Cycling

2004-07-25T21:21:00.000-07:00

Sorry for the long dead space between posts. I finished the Seattle to Portland Classic last weekend and managed to give myself a whopping case of Acute Achilles Tendinitis. After gimping around for a few days I finally took heed of my wife's advice and saw a sports medicine doctor. One walking boot and prescription for physical therapy later I was good to go.

One benefit of my doctor's visit (which took forever) was that, being a sport medicine clinic, they were playing the Tour de France on a TV in the lobby. And speaking of the TDF...

Well, Lance has finally done it--a record sixth Tour de France victory. While his relations with European cycling fans has certainly improved over the years, I don't think he'll ever be a beloved rider. I can't help but think that, at this point, Lance is probably to cycling what the New York Yankees are to baseball.

Like the Yankees, Lance is easy to root for. Who doesn't love a winner? But that is exactly what drives me crazy about Yankee fans. I mean, how hard is it to be a Yankees fan? Where is the struggle? Where is the commitment that can bring a fan to highest of highs and the lowest of lows? I hate Yankees fans because they don't seem like real fans. I suspect that if and when the Yankees ever actually do fall apart, a lot of fair-weather fans are going to leave in droves.

Speaking of commitment, how is it the Mariners are one of the worst teams in baseball and still continue to draw well? I think the fans are excited to see the M's admit their mistakes (ahem, Rich Aurillia) and doing something about it (ahem, Bucky Jacobsen). It will be a miracle if this team breaks .500 this season, but at least they're fun to watch now.

Team Spirit

2004-07-16T08:27:00.000-07:00

We're facing another critical turning point for the company I work at. We've been given a big chunk of work and in the typical Microsoft way we have given it a "codename", a slogan, a logo and, of course, everyone got t-shirts. Today was "Team Spirit" day at work where we were encouraged to wear our new t-shirts. While I may not be a Vogue-reading fashionista, I can tell you that I won't be wearing a black t-shirt with my company's logo on it unless I'm washing to car, walking the dog or mowing the lawn. Is this high-school all over again and we're going to have a lunchtime pep-rally?

It reminds of a time I worked elsewhere and our department head wanted everyone to dress up for a Halloween party. Those who didn't were branded "party-poopers" and would be required to wear USC colors the following Monday (she was a USC grad and all pumped for the weekend's homecoming game). Fortunately (for her) I didn't need to march into HR with a complaint because USC got it's butt kicked and she quietly dropped the matter.

All of this got me to thinking about what we choose to wear to project to the world our affiliations. I have no problem wearing a hat with a Mariners logo on it. I want people to know that I'm an M's fan. Heck, I even wear a hat for the Java programming language--how geeky is that? So why then, do I get so wrapped around the axle with corporate-mandated dress-up days? Part of it is that I don't let what I do for a living become my defining characteristic. More accurately, I would characterize myself as a software developer way before I would identify myself as the employee for a particular company. What does bother me is that the you see people revert to high-school behavior with in-group and out-group dynamics over silly things such as t-shirt wearing. Honestly, aren't we all professionals? Am I seriously expected to work harder because I have the same t-shirt on as my co-workers? Perhaps more offensively, is it expected that I won't work as hard if I don't have lots of rah-rah trinkets around to sustain my enthusiasm?

Don't get me wrong. I'm as eager to see the places I work at succeed as much as the next guy. But there are lines I will not cross when it comes to identification. Perhaps I'm just a crank standing on my soapbox shaking my fist at the world, but I'd rather see the company save the $10 it spent on the t-shirt and instead get us some decent coffee around here.

The Stars At Night Are Big And Bright...

2004-07-15T08:39:00.000-07:00

Well kids another edition of the mid-summer class, this All-Star game, has come to a close. I saw an article in the Seattle Times this morning talking about the new lows the All-Star game hit in terms of ratings. Well, "duh".

No one, I mean no one, is buying this idea that the league that wins the All-Star game gets homefield advantage in the World Series. Not only is it a pathetic attempt to inject some importance into the All-Star game, it takes away the excitement of teams vying for home-field advantage towards the end of the season. In football this is absolutely huge. It's true that the evidence of a homefield advantage is less demonstrable in baseball, it doesn't take away from the emotional appeal of it.

This notion of awarding home-field advantage has always seemed like Bud Selig was trying to make up for the never-ending tied-up All-Star game of a few years ago where, like Caesar, he ruled from the stands that game would end. If you want to prevent never-ending tied-up games, just make a rule that caps it. What upset people about was not necessarily that it was a boring never-ending tied-up game, it was that the ruling to end it was abrupt and seemingly arbitrary.

The slogan for last year's All-Star marketing campaign was "This Year It Counts"--this year's slogan should have been "This Year It's As Dumb As It Was Last Year".

Weekend Tidbits

2004-07-11T20:06:00.000-07:00

Well the day has finally come...my daughter Audrey nearly made it an entire year before getting her first cold. As it can be expected it has quickly passed to my wife and I and now we're doing an all-family sickness thing. Considering how sick this child is, she's been in a pretty good mood.

It got me thinking about raising a child. I finally get what people mean when they say that you learn as much or more from your child than you teach them. Yes you learn about changing diapers and dealing with young emotions, but I think what you really have is a little laboratory for the human psyche living in your own home. When Audrey sits on the floor and reaches out for something he she can't quite get I can see her determining if she can scooch towards the object or if she will just abandon it and handle the disappointment. These sorts of skills are the nascent beginnings of life-long skills that she will need to nurture. It's fascinating.

Baseball

Great day of days, Magglio Ordonez and Laynce Nix have come off of the DL! I've managed to float in the 4/5/6 positions with all of my injuries so perhaps there's cause for optimism when I get some decent players back in the lineup every day.

The M's released Rich Aurillia outright this weekend. It's pretty clear at this point that GM Bill Bavasi has determined that the limp-along-with-old-guys strategy is pretty-much DOA. I said it after the Freddy Garcia trade and I'll say it again--Bavasi seems to be learning from his mistakes. We'll see.

Tour de France

So far Lance Armstrong has avoided the nearly-daily pileups at the end of each stage. The weather in the Tour so far has been nothing short of lousy and when wet streets are combined with cobblestones, tight turns and funneled finishes wrecks are sure to happen. Both Lance and Tyler Hamilton have been vocal about the dangerous finishes the Tour organizers have put together this year. It's clear they have tried to keep the Tour exciting, but probably to the detriment of the riders' health.

Monday is a rest/travel day and Tuesday we pick it up in the Pyrenees. Stages in these mountains and later, in the Alps, are where we can expect the fireworks between Lance, Jan Ullrich, Tyler, Iban Mayo and other GC contenders.

Mid-Season Evaluation

2004-07-08T08:40:00.000-07:00

Well we're very near the mid-point of the season and it's time to reflect on what the Kids have done right, and what has not been so good. Since the first category is smaller we'll start with that. Laynce Nix was a studly pickup for me, until he got injured. He's part of the Rangers' new crop of talented young hitters and I was most pleased to get one of them. Hopefully his little arm injury in rehab is minor and he'll be up after the All-Star Break. I was similarly pleased with picking up Shingo Takatsu who replaced Billy Koch as the closer in Chicago. I think the White Sox are going to be a team to watch come September.

OK, now for the bad. Injuries, injuries, injuries. Here's my list of guys on the DL: Juan Gonzales, Kevin Brown, Magglio Ordonez, Raul Mondesi and Laynce Nix. I've got three All-Stars sitting on the DL and being replaced by scrubs the likes of Bubba Crosby, Fred McGriff and Ryan Drese. I will admit that my drafting strategy certainly favored older players over young unproven talent which may partially explain the high number of injuries.

A smart GM would begin to adjust and start looking at other ways to alter the roster. In our league there are no free-agent pickup except under limited circumstances. The best way to make moves is make trades, of which I have made none. I'm always a little hesitant to make trades for a number of reasons. I'm a believer in the laws of big numbers. Over time I expect certain players to perform to a certain standard. So just because a player has sucked for a big part of the year doesn't mean he can't pick it up and end finish with totals close to his career average. I absolutely hate it when I trade a guy that has been dogging it all season and then goes nuts for some other guys team.

I'm also wary of making trades because I'm always afraid I'm getting ripped-off. When another team proposes some trade I'm always trying to figure out the angle. What piece of intel does the other guy have that I don't? Admittedly this is pretty paranoid and I would be wise to follow Billy Beane's motto which is that it's a good trade if you get what you need, not if you get equal or better value.

Tour de Lance

2004-07-06T21:27:00.000-07:00

Alright baseball fans, you'll have to suffer through another non-baseball post. Between the Mariners achieving new levels of sucking and the Swing Kids suffering one injury after another, baseball has dropped a bit in my daily world and been replaced by the Tour de France.

This is the third year in a row that I've been absolutely captivated by this 101 year-old event. I'll admit that watching Lance Armstrong was my original impetus for tuning in, but now that I know more of the riders and the stories behind them I would be watching even if Lance wasn't racing. Like any good sport the Tour has electric personalities, intrigue, raw physical prowess and the intellectual gamesmanship of tactics and strategy.

What could be a more compelling story than a guy who comes back from a 30 percent chance of survival from testicular cancer who then goes on to win the hardest physical event in the world five times in a row? Oh but it doesn't end there folks. No one has ever won the Tour six times. Lance is already in very elite company with five wins, but to win a sixth solidify his reputation as one of the all-time greats.

You also have Lance's erstwhile rival, Jan Ullrich. Ullrich is an extremely talented rider who is a by-product of the old East-German sports institutions. He cuts an increasingly sympathetic figure as we learn how many expectations are piled on him and how little he controls his life. Unlike Lance he doesn't love to ride the bike, he's just very very good at it.

You also have Lance's former lieutenant, Roberto Heras, racing for the newly-formed Libery Seguros team. Roberto helped Lance through the Alps and Pyrenees for the last few years and left U.S. Postal to become the star rider for a new team. Can Lance's former right-hand man challenge him for the maillot jaune?

I'm in continual amazement at what these riders put themselves through to ride the Tour. By the second week their bodies are so taxed from the daily physical toll they subsist on liquid diets. I used to think that soccer players were the most physically fit athletes in the world, I changed my mind once I saw the peloton climb surreal inclines like Mt. Ventoux.

You want excitement? If you saw the way Lance recovered last year after the terrible crash he had Luz-Ardiden you witnessed one of the greatest sports moments I have ever observed. Here is a guy who was on the ropes for the entire ride. When he hooked his handlebars on a roadside fan's musette bag it looked like it was all over. Thanks to his old pal Tyler Hamilton (an amazing story himself), the peloton held up and refused to attack in an act of cycling chivalry. Lance rejoined the main group and suddenly the old fire returned. Like a man possessed he attacked Jan Ullrich mercilessly up the hill and one the stage handily--an absolute exhilarating win.

My rambling can't do this event enough justice. If you get cable tune into OLN and take in the rich commentary of Phil Ligget and Paul Sherwin--the Vin Scully and Al Michaels and cycling broadcasting. This is going to be one hell of a Tour.

Nearing The End

2004-07-01T12:11:00.000-07:00

Yesterday there was a very interesting panel discussion on Aspect-Oriented Programming (AOP) titled "Aspect-Oriented Programming: Great New Thing or Great Leap Backwards?". The panel consisted of one big AOP evangelist (Gregor Kiczales) and four other guys on the fence, including James Gosling. One very interesting point that Gregor made that I had not heard about AOP was that it brought better consistency and encapsulation to your source code. He likened AOP to being the "second wave" after OOP which brought better consistency and encapsulation to programs.

He demonstrated this concept with a GUI drawing tool. It had different classes for the various shapes that understood how to render themselves and all of that was fine. The problem was the niggling little detail that when a Shape was updated it was required to update the Display. This caused little display.update() calls to be sprinkled all across the code. Gregor's solution was to use an Aspect to capture this important detail.

I'm not sure if is really just a design problem that is being ignored or not, but my gut-feeling is that the problem hasn't been sliced correctly. I will say that I have used AOP in a very limited capacity for logging and performance metrics for which it worked quite well. For what was termed "read-only" operations, I think AOP is a nice way to glom some non-invasive functionality on a system. Most other participants on the panel expressed uneasiness when major program functionality could be overridden with Aspects.

The Java community has grown up a little bit and the little idiomatic expressions that we all use are the lingua franca of our day-to-day interaction with the code and the Java language. The changes to the language made in Tiger were primarily motivated to capture these in a much terser fashion. I wonder if non-invasive AOP is not another way of trying to achieve this same goal.

Reusability

I went to an interesting presentation put on by Erich Gamma (that's Mr. GOF to you buddy) on plugin development in Eclipse. I've looked at his book on this before, but sitting in the presentation yesterday I was struck by what the Eclipse folks have really done. They've built a framework that, down to its very core is about extensibility. If you think about any creative work, no one really comes up with anything truly original--if they do we usually call them "crazy" and lock them away. Instead innovators build on common ideas. The beauty of what Eclipse has done is that they have codified that concept in the very fabric of the platform.

This does lead me to another question though. Sun has just release it's fancy new Sun Creator Studio which is once again promising "pluggable components" for rapid application development. Do you remember just how wildly successful the Swing component or off-the-shelf EJB component market was? (Hint: it died). I can't help but feel that the promise of object re-use has been oversold. The power of OO is about expression. Expression, theoretically, leads to more maintainable code because humans don't have to spend so many cycles acting like machines to figure out what code does. To paraphrase Kent Beck, I'm more interested in use than re-use.

It strikes me that when we pay lots of attention to re-usable components, we usually guess wrong on how we will want to use those components in the future. This is the proposition that eXtreme Programming puts forth--design for today, refactor for tomorrow. What I think is impressive is that somehow Eclipse seems to have gotten this right. A lot of projects have tried and failed at this kind of super-pluggability, I'm curious as to why Eclipse has succeeded.

JavaOne Scraps

2004-06-30T12:50:00.000-07:00

Here are a couple of quick tidbits between sessions:

PowerBooks Everywhere!

Windows has something like 95% of the laptop market, but I swear to you more than 50% of folks I've seen here at JavaOne have sexy, sleek PowerBooks. Several of the Sun big-wigs have praised Apple's Java integration into OS X and we're now seeing a strange confluence of Mac/Java/Open-Source/Unix types. For me personally, I'm still desperately trying to come up with some rationale for why I "need" my own PowerBook (my wife has one). If you have any suggestions, please send them to me.

JSF & JDO

Continuing my theme of JavaServer Faces, I attended an interesting session about combining JSF with JDO as a more lightweight solution to the typical database-powered web application. I remember looking at JDO before it became a JSR and even then I thought it was pretty nifty.

There seems to be a large number of folks in the Java community that turn up their noses when JDO is mentioned. Certainly for solving the problem of persisting domain objects to a database I think the solution is way more elegant and less intrusive than Entity Beans. You simply bolt the persistence mechanism on top of your objects a la TopLink or similar framework. To me EntityBeans pervert your design because they are primarily EntityBeans before they are your actual domain objects.

Hip Hip...Hooray!

2004-06-30T10:47:00.000-07:00

There are three things that receive consistent applause and cheering here at JavaOne:

Making fun of Microsoft

Announcing the open-sourcing of an API

The simplification of EJBs proposed in version 3.0

I'm not kidding folks, I've seen two presentations that had a blank slide describing the "deployment descriptor of tomorrow" and people in the audience literally clapped and shouted out "Amen!" like I was at some evangelical revival. I will admit that I found myself caught up in the fervor too.

If you've ever had to add a field to an EntityBean and four files and one deployment later you realized you missed one thing, you will be excited about the changes that are being proposed. While I'm still on the fence about the meta-data features that are part J2SE 1.5 (aka "Tiger"), they make a lot of sense in EJB-land.

SWT Talk

By far the most entertaining presentation I've been to is the one put on by IBM about the SWT. It was very funny and gave some nice insights into the guiding principles of the SWT project. This would be a good one for finding the slides after the conference and posting some of the goals and principles on the wall. Junior developers could learn a lot from the experience and wisdom of the SWT guys.

Taming The Tiger

2004-06-29T13:02:00.000-07:00

This morning I attended a session titled "Taming The Tiger: J2SE 1.5" hosted by Joshua Bloch. It was a nice preview of all of the new language changes that are going into version 1.5 J2SE. Mr. Bloch stated that all of the language changes were put in to "make the compiler write boilerplate code, not the developer". In other words, the goal is to reduce common coding idioms to something more terse. All in all, I think this is a good idea. I think expressing things with the minimal amount of fuss and typing is a good thing. This is what I would call "good terseness". Perl fails this test because it's terse, but not obviously expressive. You need to know several dialects and a mental Rosetta Stone to understand what any particular Perl hacker has written.

Here's a quick rundown of the topics covered:

Generics

We've had a lot of fun at work deriding Generics as Sun's attempt to slap some syntactic sugar on top of Java to make it more appealing for VB monkeys than .NET. I am starting to think though that not all sweet things are bad for you--especially in moderation. Essentially Generics add some compile-time type checking to things that used to work with Objects.

With Generics and the new for-each loop you can take code like this:


List players = getPlayers();

for (Iterator it = players.iterator(); it.hasNext(); ) {

    Player player = (Player) it.next();

    player.release();

}

And replace it with this:


List players = getPlayers();

for (Player player : players) {

    player.release();

}

I have to admit that after having worked with Python this kind of intuitive looping is easy on the eyes.

Autoboxing

I remember when I first started learning Java, the dichotomy between primitives and their object-equivalents seemed really clunky for long before I refined my feel for such things. While this feature feels a lot like .NET-chasing it does serve to clarify the code.

Enums

Ever since being exposed to Joshua Blochs Effective Java I've been a fan of his Type-Safe Enumerations. Considering the enthusiasm with which Mr. Block presented this feature, I figured I'd better pay attention. Essentially the Enum feature is a boiled-down way to implement type-safe enumerations. Again, the language changes have been put in to express the enum concept as simply as possible. Of all the new Tiger features, I like this one the best.

Variable Arguments

With the addition of (thank god) printf() the need for variable-length argument lists became apparent. Essentially it's a quick way to express arrays of objects in a quick way. Consider this example:



public String format(Object... args) {

  return "Hey there " + args[0] + "! You are " + args[2] + " years old";

}



System.out.println(format("Duke", 5));

OK, this is a really dumb example, but the cool thing about this is that the call to the format() method didn't require the usual clunkiness of packaging up my arguments into a Object array.

Unfortunately the presentation didn't cover Attributes which looks sort-of XDoclet-like. This one seems to violate my aesthetic principles the most, although when the J2EE guys start talking about eliminating deployment descriptors my queasiness dies down.

Monday Night Goofballs

2004-06-29T09:30:00.000-07:00

Just a few quick notes on the Birds-of-a-Feather(BOF) sessions I went to last night...

More JavaServer Faces

After suffering through some poor Sun developer's halting presentation, the audience had an opportunity to do Q&A with Craig McClanahan, the inventor of Struts and one of the minds behind JSF. The focus of the session was about using Struts and JSF together and where each one fit. Craig admitted that the scopes of both technologies overlapped but that was due to being able to deploy one or the other independently.

He gave perhaps the best description I've heard so far for JSF, especially as it relates to Struts: JSF is a front-controller for the Struts controller. If I've got this right JSF is really about adapting the clunky HTTP request/response model into a more familiar (if you're a GUI developer) event-based model. Of course there's much more to it than that, but I think that boils it down pretty well.

XP In Schools

I went to a fascinating presentation last night on robotics work done at the University of Utah using Java and Legos. It was a fairly hardware-intensive discussion, but before my eyes completely glazed-over I noticed that the presenter said that he had the students use XP as their development methodology. He specifically cited pair programming and unit-tests.

I didn't have time to stick around to chat with him but two things struck me: I think it's wonderful that there are some people in education that seem to be paying attention to what's going on in the "real world". The second thing is that I wondered how strict test-driven development would work for such a hardware intensive project.

Two Tacos And One Margarita Later

2004-06-28T20:01:00.000-07:00

Let me just start this post with the following cathartic exercise...

I hate you Microsoft. Every piece of software you write is crap! Your OS is crap. Your applications are crap! I am sicking of living in your world of poor-quality solutions!

I've probably spent a total of an hour just trying to get the XP laptop the company sent me to SF with to work on a simple WEP-less 802.11b network. You would think it would not be that hard. Guess again skippy. I'm telling ya, when I am king, Bill Gates will be first up against the wall. Actually Larry Wall will probably up first, then Bill.

Follow-Up To The Morning Session

I forgot to mention the coolest thing I saw this morning which was Project Looking Glass. This is a tres sexy desktop environment that had the entire audience oohing and ahhing. Definitely check it out.

The Futrure Of Persistence

I spent the afternoon in a couple of object-persistence workshops that pretty much beat the hope out of me that anybody really understands the problem. The first workshop was about EJB performance-tuning for CMP entity beans. Perhaps the best statement handed-down by the so-called gurus was "avoid putting business-logic in your entity beans".

But wait. Wouldn't the very class names I would use to model my problem domain be the same as the entity beans I would use for persistence (which also probably match my data model's table names)? If that's the case then the behavior I would want on these domain objects must instead live somewhere else. Sometimes I think this lives under the misnomer of a Service Layer and is usually an indication of poor design.

The scary part of this is that if this is the recommended "best practice", then what good are entity beans for anyway? Who really makes money from storing objects in the database? It's a side-effect of needing to keep object state in some non-volatile medium. It should not be the driving force behind my application. Telling someone you are developing an app using Entity Beans is akin to saying that you develop applications with binary numbers. Who the hell cares?

Java Server Faces

I finished a hands-on session with Java Server Faces before splitting for dinner. This is some pretty cool-looking stuff. Those wankers at Microsoft had a similar event-driven web development idea with the abomination-known-as-dot-NET that I though was intriguing. This could definitely be a new and better way to make web apps. At the very least it was worth picking up a book at the bookstore to investigate it further.

Now that I've had a little dinner it's off to some Birds-of-a-Feather session. Depending on my stamina I may be geeking-out until 1:30 in the morning. We'll see. Thank god for coffee.

Convergence Of The Nerds

2004-06-28T13:36:00.000-07:00

After a harrowing night of travel last night I finally made it to the nerd-fest known as JavaOne. The morning session was essentially a big Java love-fest with Sun running out one technology evangelist after another.

The Morning Session

Some of the presentation was inspiring--you felt like you were in a room with 15,000 other charged-up nerds who were into solving problems as much as you are. Some of the presentation was creepy. All the talk of always-on, ubiquitous connectivity made it easy for me to imagine this technology turning into some sort of corporate/Orwellian state. But then I'm a bit prone to lefty paranoia at times.

Some of the presentation was just dumb. Once again Sun thinks that the drag-and-drop application development approach will succeed. This time they are taking .NET head-on and have come up with some all-powerful IDE to visually connect web apps together like VB. I've always been a little suspicious of these tools because they tend to slice the application in ways that are at odds with how I would like to think of apps. I mean come on, can you really point to a VB or PowerBuilder app you wrote that you were proud of?

J2EE Changes

Sun's Bill Shannon came out and gave a nice preview of J2EE 5.0. When he gave examples of proposed changes to EJB code the crowd cheered. No more 20-line stanzas of boilerplate to be an EJB, to configure an EJB or to locate and use an EJB. Yippee!

Sun seems pretty excited about Tiger and all of its new features. It smells a little bloated to me and I'm not convinced that things like Generics are really any more than empty-calories. In particular the new Attributes features seems to be quite the rage across the various Java SDKs. They strike me as Vatican-approved XDoclets. We'll see.

Hans und Franz

They also brought some archetypal German engineers out from Siemens to show off their new infotainment prototype for BMW. Given the ease with which drivers are distracted today, I'm thinking the last thing they need is a rich network client. The armed forces have spent the last fifteen years working on user-interfaces for jet-fighters in an attempt to provide just-in-time data display to pilots. Studies had shown that in the past fighter-jockey's biggest problem was information overload. Since the Navy spends way more money training fighter pilots than mom and dad spend on junior driving, I have to think that traffic accidents are about to go up. Who knows though. Maybe those Siemens boys will put some proximity-detection features in the car.

Back From The Great Outdoors

2004-06-25T07:51:00.000-07:00

Well I can tell you that all of the things you may have heard about Yosemite are absolutely true. It's a gorgeous little corner of the earth. I can tell you that raccoons are clever little devils. We had one that broke into our room twice in one night in an attempt to pull a box of food out on the deck with him. I can also tell you that I now know that when a bear does defecate in the woods it is, in fact, quite large. Okay, enough nature-talk, let's talk baseball.

I was inspired by my father-in-law to pursue a simulation in an attempt to understand the makeup of a winning roto-team. The idea is to run a simulation of randomly drafting players to fill out 10 AL teams and then use the 2003 totals and see how the teams fell out in the final standings.

Now of course there are all sorts of little details that could hamper this process. This simulation would not account for trades, injuries or other mid-season roster changes. Again, if you run the simulation enough times this should be a wash. It may also be helpful to run the simulation using data for more than just last year. Ideally you would do the simulation for the last three to five years.

So what would you look for in the results of such a simulation? Well, I have some vague notion rolling around in the back of my head that all of the statistical categories are not equal in value and attainability. I've looked into Linear Programming as a possible solution to quantify the weighting between categories, but since we use a ranking system and not simple totals, that approach seems to break down. What I would be looking for is the "winning formula" of performance across the categories. Is locking up the SV category really better than locking up HR? I don't know yet and that's why I want to run this simulation. It may even turn out that there is no statisically significant pattern that emerges from the results. This would at least confirm how much of a crap-shoot roto-ball really is. Enquiring minds want to know.

Now the trick is to make a computer do as much of this work as possible. My wife has a high-powered statistics package for Windows but these days I'm so tired of crappy Microsoft software I'm more inclined to go out of my way to find an open-source alternative. Surely there is some other Linux nerd out there that has put together a statistics package.

Conference Time!

For the first time in my professional career I will be attending the JavaOne conference in San Francisco next week. I'm super-excited to go and since I'll be all hopped up on coffee and general computer nerdiness, next weeks posts will probably be very Java-oriented.

Out On Vacation

2004-06-18T20:55:00.000-07:00

I'll be off for a couple of days taking in the wildlife at Yosemite, so you won't see any posts until late next week. I've never been there before so it should be a lot of fun. Hopefully Audrey will handle all the travel well.

Dropping Like Flies

2004-06-16T08:06:00.000-07:00

I suppose it was inevitable that Kevin Brown would go down with an injury at some point this season. I knew what his Griffey-esque health record was like when I drafted him so I knew what I was getting into. Fortunately I needed to slot a reliever on my roster so I picked up Shingo Takatsu, who is replacing the uber-flameout Billy Koch.

The injury-bug wasn't done biting my roster yet. My super-sneaky-pick Laynce Nix also went on the DL this week. Man there are no outfielders left in our league. So I picked up Bubba Crosby of the Yankees based almost purely on the name alone.

Slaying Goliath

2004-06-15T20:39:00.000-07:00

I'm not much of a basketball fan, but I figured I'd better tune into game 5 of the NBA Finals just to see if Detroit could really knock off the Lakers. Not only did they win, but they beat the living tar out of those pretty-boys! I have hated the Lakers for about as long as I can remember and maybe it's just schadenfreude, but I sure enjoyed watching them lose. Like the Yankees, they are the big bully on the block and can do whatever they like to secure championships with no regard for the consequences. They have the biggest war-chests so they can easily outspend the competition.

Ah, but now we have a beautiful example of an overt attempt to buy a championship that has failed. Perhaps the universe has some vague boundaries that the Lakers finally pushed too far. I'd like to think that no one team/nation/person/corporation can truly dominate their domain for an extended period of time. Eventually the piper must be paid and the long reign is over.

All of this gives me some hope that the Yankees' attempt to buy a World Series trophy this year will fail. Getting Kevin Brown and Gary Sheffield was bad enough, but getting A-Rod is just too much. As an M's fan that can recall the bitter days when A-Rod left for Texas I can tell you that seeing him spray champagne in October might possibly drive out the love of the game for me....Nah, not really.

Lousy Free Service

2004-06-15T08:27:00.000-07:00

Apparently Yahoo! has decided to give all of us free-email-seeking leeches an upgrade to our accounts. Fancy (meaning buggy) DHTML interfaces coupled with a boost in disk space. Yippee. Sounds great. I can't wait until I can actually get the Yahoo! mail server to actually let me in. WTF?! I guess I may actually have to start using my Comcast address, alexvollmer@comcast.net. What a bunch of goofballs.

Update
OK, I finally got to my inbox after a lot of screwing around, including removing all Yahoo cookies and closing the browser. Now it appears that Yahoo! Mail has managed to circumvent Firefox's link access via keyboard so that I actually have to use the mouse to click links! On top of that, their shortcut keys don't work either. Uh, guys? Did you test this thing or is this another crappy site that only works with that-giant-security-hole-in-your-desktop-known-as-Internet-Explorer?

On top of that the reply/compose page has a font size somewhere around 5pts which makes it practically unreadable. If this is Yahoo's response to GMail, they should be embarassed.

Ahhh...Greener Pastures

2004-06-13T17:52:00.000-07:00

Today I went to only my second M's game of the season--but what a game! It was the first sweep of the season, albeit against the lowly Expos, but hey, I'll take what I can get. It was Audrey's first ballgame and she loved it! She has just learned to clap and enjoys doing it a lot. Anyone who has been to a baseball game knows that there is a lot of clapping going on so she had a very good time.

Despite the M's pathetic showing this year I've still managed to tune in about 75% of their games. Today I forked out a good bit of dough to make to the game live and I enjoyed myself immensely. So as much as I love to gripe about the M's I'm still very fortunate to live in a town with a Major League Baseball team that plays in a beautiful stadium. Sure this season is a basically a write-off, but it's still fun to watch your ballclub win, no matter how they are doing for the season.

While there is a part of me that would really love to see attendance and TV ratings drop precipitously to send a message to the front office, it's a lot more fun to go to a game when there are 43,000 in the stands as opposed to 14,000. So for one day this season, I felt like I was on top of the world. I had a hot dog the size of a surfboard, I sat in plush seats with my beautiful baby daughter and, for once, my M's looked good out there.

The Pages of War

2004-06-11T08:32:00.000-07:00

Okay, I can't seem to shake the WWII theme. One reader asked for my top ten list of WWII books. While I'm trying not to turn this blog into a top ten list site, I'll bite on this one and write down my list. They are, in no particular order:

A Bridge Too Far, Cornelius Ryan
This is the first real piece of WWII military history I ever read. I love the movie too so this book has a special place in my heart.

D-Day, Stephen Ambrose
I've read most of Ambrose' WWII work and will admit it's not the most academic material out there--even he claims to be affected on occasion by his own self-described hero-worship in his books. However, I think his treatment of D-Day is one of the clearest I've read.

The Second World War: Asia and the Pacific, John H. Bradley, Jack W. Dice, Thomas E. Griess
This is a very matter-of-fact recitation of the Pacific War. It's a great overview of events in the PTO and is a great companion with the Atlas (sold separately!)

Knight's Cross, David Fraser
A fascinating biography of Erwin Rommel. The author depicts him as a professional soldier caught in the politics of the madmen he served. One interesting note is that the author claims that Rommel admitted as early as 1942 that he thought the war was already lost. The fact that he carried on is a testament to his adherence to duty--especially when you consider that he eventually fell in with the crowd that tried to kill Hitler.

Delivered From Evil: The Saga of World War II, Robert Leckie
This ambitious book attempts to cover all important events of the war from North Africa to Europe to Asia.

A Genius For War, Carlos D'Este
A great biography of Gen. Patton.

At Dawn We Slept, Gordon Prange
The definitive account of Pearl Harbor and great ammunition in arguments against any would-be revisionists who claim that Roosevelt was goading the Japanese into war.

The Second World War, John Keegan
Like all of Keegan's work, this one really uses historical events as a framework to explore other issues. He takes four major campaigns and discusses the strategic concerns and dilemmas faced by each of the major powers. An absolute fascinating read.

Citizen Soldiers, Stephen Ambrose
This is the only Ambrose work I know of where he actually has a theme to express beyond simply re-telling the facts. It's a good treatment of the American G.I. and his ultimate ability to triumph.

Your Link O' The Day

2004-06-08T08:58:00.000-07:00

What are blogs better for that regurgitating links published by someone else? Today's case in point is an essay about abolishing the FCC that I found this morning on Slashdot. I'm not sure I buy every argument in this guy's article--especially the bits about no single corporate entity able to dominate the airwaves--but it's an interesting read nonetheless.

Risk And Reward Pt. II

2004-06-08T08:30:00.000-07:00

I got the following comment on yesterdays post:

...the data has to be looked at in perspective -- in my case, strikeouts are a positive 3 which sounds promising, but the reality is I'm second to last...so, there is only one point to lose downward...

Let me use this as a springboard for some further thought. There are two important points that come out of the risk/reward analysis. One is that we have normalized the risk/reward value across categories so that you can compare between them. So while our guy may not have much to lose in K, he may have more to gain in a different category. Knowing differences like that are important to guide your mid-season drafting and trading strategies.

The second point is that even at the extreme ends of the categories the analysis is still valid. Clearly if you are on top of a category you have nowhere to go but down. The question is how much are you at risk? Unfortunately this line of thought won't give us an absolute answer, only one that is relative to other categories. But if you were tops in several categories this analysis would allow you to compare the relative risk between them. This would obviously work the other direction if you are at the bottom of several categories.

BTW, I'm not picking on our reader, he just served as a good foil for today's post. I couldn't come up with anything better on my ride into work this morning--which was glorious. It was a beautiful morning and I hit 35 mph on the way in. A few more long rides and I'll be ready for the STP.

Risk And Reward

2004-06-06T12:54:00.000-07:00

OK, back to baseball posts--I've been playing Call Of Duty and had WWII on the brain since I started it. Today we're going to revisit an idea I've had about quantifying risks and rewards for the Swing Kids in my roto-league. There are two questions a team needs answers to in order to make mid-season adjustments:

In which categories am I close to teams ahead of me?

In which categories are teams below me close to me?

You can imagine being relatively far from teams ahead of you in, say, RBI, but close to teams below you. You could then say that RBIs are a high-risk category since you are more likely to lose points in this category than gain them. Similary you could be really close to teams ahead of you in K, but far away from teams below you. You could label K as a high-reward category. Then you can imagine other categories being somewhere in-between.

What we're most interested in then, is the spread of totals across a particular category. This spread can best be characterized by Standard Deviation (SD). Put crudely, SD is the average difference between the mean value and all values in the data set. If we look at the category totals and figure out how many teams are within one SD ahead of the Swing Kids, and how many teams are within on SD below the Kids we get some quantitative idea of the potential risks and rewards. Here are the results:


	        R	HR	RBI	E	AVG	OPS	W	SV	K	ERA	WHIP

Teams Above	4	2	2	3	2	3	3	4	4	2	4

Teams Below	2	3	4	3	1	3	2	2	2	2	1

Risk/Reward	2	-1	-2	0	1	0	1	2	2	0	3

The "Risk/Reward" row is simply the "Teams Above" value minus the "Teams Below" value. This has the nice effect of showing high-reward categories as positive numbers, neutral numbers as zeros and high-risk categories as negative numbers. Using this value we can say that HR and RBI are the categories the Kids stand the most to lose. Similary R, SV, K, and WHIP show the biggest room for potential improvement. E, OPS and ERA are neutral categories in terms of risks and rewards.

Now that I have this kind of information I can guide any trade or mid-season draft strategy with some hard data. It looks like I probably need another close on my team (SV, K and ERA) and I'm probably going to be hurt pretty badly by Magglio Ordonez going on the DL (HR, RBI).

In Remembrance

2004-06-06T08:11:00.000-07:00

If you have opened a paper today or turned on your TV, it's impossible to miss the recognition of the 60th anniversary of D-Day, the invasion of France. One can easily dismiss the hubbub as a desperate grab for human-interest storie by news organizations or as opportunities for politicians to ride the coat-tails of the ceremonies to push their agendas.

But let's cut through the hype. The fact of the matter is that D-Day is a big deal. Nothing like it had ever been attempted before, nor has anything like it happened since. It was the ultimate "Coalition Force" ever assembled for a purpose that united much of the world. All attempts at coaltions thereafter have attempted to imitate this, usually without the same success militarily or morally.

Take a moment to reflect on what may have happened if the Allies had not been able to gain that precious toehold on the beaches of Normandy--a toehold that was ultimately leveraged to bring about the downfall of one of the greatest evils the world has ever seen.

Memorial Day Tribute

2004-05-28T08:17:00.000-07:00

This weekend is the official opening of the World War II Veterans Monument in Washington D.C. Considering the impact that war had on our nation, it stuns that we have not recognized it in this fashion until now. However, one could argue that up to this point the World War II generation had a lasting monument of great films. So in honor of all veterans this weekend, here's my top ten list of World War II movies.

Twelve O'Clock High
Like most people my age, my first exposure to Gregory Peck was the film adaptation of Harper Lee's To Kill A Mockingbird. In that film Peck plays the honorable Addicus Finch who is an upstanding citizen and paragon of virtue. In "Twelve O'Clock High" he plays a barely-in-control egomaniac with a disturbing self-destructive streak. It's quite a shock for the Addicus Finch fans.

The Thin Red Line
This film came out not too long after "Saving Private Ryan" and suffered the inevitable comparisons that caused it to get lost in the shuffle. However, having watched the film a few years ago I was able to appreciate just what a masterpiece Terrence Malick produced. The cinematography alone is worth the rental.

Objective Burma
This is the only Errol Flynn movie I have ever seen, but I can tell you that it's a dandy. A wonderful taut "mission" movie where a special group of Marines sets out to take out an important Japanese radio station on the island of Burma. Classic black and white fun for the whole family.

Tora! Tora! Tora!
This film is a faithful, but entertaining, recreation of Gordon Prange's excellent book At Dawn We Slept. One interesting note about this film is that all the scenes of the Americans was shot by an American director and all of the Japanese scenes were shot by a Japanese director. The movie does a good job of showing just how confused the American chain-of-command was on Dec 7 1941.

A Bridge Too Far
Hollywood has churned out several cast-of-thousands WWII films--this is one film that delivers. Based on Cornelius Ryan's book of the same name, the movie deftly moves through scads of historical characters to give the viewer a good impression of the overall picture. I will admit this movie is more enjoyable if you know the general story of Operation Market-Garden or, better yet, if you have read the book, but the production quality alone make this an achievement in combat filmmaking.

Das Boot
This is the definitive submarine movie. Yes The Enemy Below came out first, but Das Boot set the standard for the genre. I saw the director's cut in a theater a few years ago and when the lights came up at intermission the entire audience let out a collective gasp. When we all returned from the lobby to get snacks, the theater smelled sweaty from the tension the audience was feeling along with the German U-Boat crew on the screen.

Stalingrad
This German film is one of the grittiest combat films I have ever seen. Unlike the all-too-slick Enemy At The Gates this movie does an excellent job of showing just how horrible the Battle of Stalingrad really was. It can be a hard film to watch, but it probably has more humanity in the characters and story than any of the other films mentioned here.

Saving Private Ryan
This film re-defined the World War II film and sparked a bit of a mini-craze for WWII films in Hollywood. An outstanding cast, amazing cinematography and excellent sound and effects are only marginally knocked down a notch by Spielberg's heavy hand. Tom Hanks proves why he's won so many Oscars and the supporting cast is also equal to the task.

Band of Brothers
Okay okay, I know that this is a mini-series and not a film, but there is no production about World War II that has equalled what this series has done. An absolutely amazing cast, a wonderful story and the same production crew that put together "Private Ryan" makes this series one of the all-time greats. One of the amazing things about this series is the consistent quality achieved across the episodes despite having a different director for each one. This is an absolute must-watch series.

What's Goin' On?

2004-05-26T21:23:00.000-07:00

Sticking with yesterday's two-part theme, I've got some broadcasting gripes and some fantasy baseball analysis.

Can somebody please tell me when it was okay for sports interviews to be reduced to a recitation of the events for the athlete in the hopes that they will respond with some kind of insight? For example, an interviewer will start with "Ichiro, you hit that homer in seventh and this team has been struggling a lot lately..." and then end with a trailing pause as a signal that the player should jump in and say something. Some interviewers will tack on a weak question at the end--something like "describe your feelings", but the style is the same. I wonder if this type of questioning is what leads to a lot of the canned answers one hears a la Nuke LaLoosh ("we just play 'em one at a time", "I'm just here to help the team"). This is really amateurish and is another reason why I am enjoying watching sports with the sound off.

Roto Ball

Last night I spent a little quality time with the Fantasy Baseball Website and Open Office getting some spreadsheet action going. Let me take a brief detour and point out that extracting the data off of the Yahoo site was a relative breeze thanks to the best damn web browser in the world, Mozilla Firefox. For all you poor people living with Internet Explorer do yourself a favor and download Firefox. It will be like a heavy veil lifted from your eyes.

You may recall that I had three theories explaining my team's current poor position in the standings. Last night I investigated the first one which was that I have a lot players that don't get as much playing time relative to other teams in my league. Below are the numbers:


Team	        Rank	Off     AB	Pitch	IP

                        Ranks           Ranks

Veal Rollups	1	45.5	1980	35.5	331.9

Chin Music	2	28	2199	40	311.8 

Porcupines	3	36	1868	28.5	312.5

Cthulhu	        4	25	1918	33.5	370.6

Sumo	        5	39	1858	16	328

Sandmen	        6	16.5	1888	37	363.7

Angry Guys	7	33	1834	15	369.1

Swing Kids	8	28	1916	23	381.1

Roy	        9	6	1856	32	312.9

Kristi	        10	18	1988	11.5	368

					

Overall Rank/AB Correl	    -0.39				

Total Off Rank/AB Correl     0.06				

Overall Rank/IP Correl	     0.46				

Total Pit Rank/IP Correl    -0.39				

Median AB	             1902				

Median IP	            347.8

You can see that the Swing Kids are in the middle for AB and on the high-side of IP. I tried correlating AB totals to overall rank (-0.39) as well as AB totals to total offensive category points (0.06). For you non-stats people correlation is described as a number between -1 and 0 or 0 and 1. The closer to 1 (or -1) the more correlated the two data sets are. The closer they are to zero, the less correlated they are. These correlation numbers are awfully weak and lead us to conclude that opportunities are not well correlated with success.

I did a similar thing for pitching by correlating IP to overall rank and points for pitching categories and they showed a similar lack of correlation. So what can we conclude? Well, the Swing Kids are certainly getting the opportunities to play, they simply aren't performing well. Now we move on the theory number two, which is to see who is underperforming their current levels and by how much. More to come...

Be Like Hendu

2004-05-26T08:26:00.000-07:00

This year I've decided to take a cue from my grandfather and watch Mariners games with the sound off. Thus far, I have found this to be a much more enjoyable way to watch the game as I find that I pay better attention to actual game and less to what the commentators are saying. Now don't get me wrong, I like Dave Niehaus and I think Rick Rizzs does a fine job, but I can't stand listening to the ex-players. It used to just be Ron Fairly, who acts as Seattle's very own Ralph Kiner ("I forgot to mention something today that bears repeating..."). Then they got Dave Valle who comes off like a frat-boy who can't believe he just got a job in the broadcast booth. I could deal with these guys since they were paired with either Niehaus or Rizzs and their effect on the game was minimal. But then they got Dave Henderson...

I should be fair and say that some ex-players turn out to be fine sportscasters later in life. People may forget just how wooden Harold Reynolds was on Baseball Tonight until last year. After a couple of years on the job, I think Harold does a bang-up job. Similarly, there may be hope for Hendu, but he's got a few things to work on--namely reading off statistics to the viewer as if they couldn't see them for themselves.

Here's the typical situation: it's the seventh inning and the manager brings in a situational lefty. The familiar Fox Sports graph pops up with a list of statistics for that pitcher. As soon as that graphic pops up Hendu starts with the reading. Now mind you, he doesn't just read them. No no, he's trying to make them interesting by surrounding each stat with some conjunctions and other connective phrases. Hendu would translate this:

IP: 86
BB: 14
K: 12
R: 4
SV: 2
W: 0
L: 3
HR: 2

Into this:

Anderson has gotten a steady amount of work at 86 innings with just 14 walks and only 12 strikeouts. He's only given up 4 runs and gotten 2 saves out of it. Those 3 losses came with just 2 home runs.

Arrgghh! I'm not blind! I can see the numbers! Save your breath and just let me read them without the read-along storybook phraseology! OK, rant over.

Fantasy Baseball Thoughts

I'm languishing in 7th place right now and I only have a gut-feel for why my team isn't doing so well. I'd like to get a more quantitative understanding for just what is going on with the Swing Kids. This is going to require a little spreadsheet time so to avoid any excessive marital strain, I had better know what is I'm going to look at. As I see it, my team is suffering due to any combination of these reasons:

I have a lot of guys that don't play much. Fewer opportunities means less totals in the counting stats.

I have a lot of guys that are performing beneath their historical levels.

The teams ahead of me have guys that are out-performing their historical levels.

I should be able to account for the first reason by looking at the various opportunity counts (AB for hitters, IP for pitchers) and compare them to the other teams in the league. I know that guys on my roster like Eric Byrnes or Lou Merloni are not playing fulltime, but I don't have a good understanding of how much that is hurting my team. The question will be seeing how much opportunity totals correlate with the overall success of each team. Therefore I'll need to get the opportunity totals for each of the teams in my league and compare them to each other and their place in the overall standings.

The second category will require a little more processing. If you read my post about my Draft Day strategy you know that I used a linear-regression analysis for each player. Essentially I looked at their three-year totals and looked at whether they over- or under-performed the expected output of the average major-leaguer. To see which of my guys are underperforming and by how much will require looking at their historical output and figuring out how far below that output they are currently at.

The third category is of the least concern to me right now, mostly due to the fact that I have no control over it (although I have thought of flying to New York and pulling a Tonya Harding on Jorge Posada). But like the analysis of the second item, this one would look at historical output and see who is looking like 2004's Brady Anderson-like overachiever.

Ready...Aim...Fire?

2004-05-24T07:49:00.000-07:00

The Fall of Troy

My wife and I hired a babysitter and went to go see the movie Troylast night. It's been awhile since I had been in a theater but now I am reminded as to why I hate the Loewe's Cineplex chain. I think we sat through nearly thirty minutes of advertisements and previews before we actually got to see the movie we paid money for. Considering that how irritated I was by the beginning of the film, it must have been pretty good to have me leave the theater satisfied. Now mind you. this is not a great piece of cinema, but I'm pretty sure it will beat the snot out of all the other summer movie candidates.

Ah but I have some nits to pick--or maybe more like questions for people who are more informed than me. First of all, it seems strange to me that in films featuring archers (like Troy or Lord of the Rings) there is always some commander directing the troops to "fire". Fire? Isn't that gun terminology? I would venture to guess that the command to "fire" was originally used when guns first appeared. They would literally apply fire to the gun powder to ignite the powder and let loose the bullet. Since guns have been around for so long now this must be some residual notion so ingrained in our collective conscience that we can't imagine calling the action of releasing an arrow from a bow anything but "firing".

Now prepare yourself for some super-nerdy historical nit-picking. Despite the power of CG these days, battles of the ancient Greeks were not carried out in long lines clashing together. The most important contribution of the Greeks to military science was the phalanx, which was nowhere to be seen in the film. Seriously folks, the reason that college teams are named after residents of Sparta is because they were undefeatable in the field. A 10x10 square of quick, well-armed hoplites was something that other armies of the ancient world could not figure out for a long time. Alright...history lesson over.

As for the film, I'd give it three (out of five stars). It's a good summer flick, but probably not one destined for the halls of greatness. My biggest complaint was the overly dramatic music and exceedingly long large-scale battle scenes. This is my beef with the Lord of the Rings movies too--the battles scenes are impressive the first time you see them, but in this case less is more. Now I must note that I was extremely impressed with the fight scene between Achilles and Hector. Great cinematography, great choreography and no music--which was perfect. This film was well-cast and even Brad Pitt gets a thumbs up for knowing when to let a better actor take over when he plays the scene between Achilles and Priam with the amazing Peter O'Toole. Now I'm going to have to check out Mean Girls to figure out what all the buzz is about.

Back To Baseball

Moving from Ancient Greece to baseball, I wrote yesterday, there's talk a-brewing about what the M's are going to do in the face of a season that holds no post-season hopes. The leading article in this morning's Times sports section is a full-blown analysis of all the things Blaine Newnham talked about yesterday. Apparently anyone but guys playing in Tacoma and Ichiro are on the trading block. Now I'm not too sentimental when it comes to ballplayers, but it would be a shame to see Edgar Martinez end his career in somewhere like Cleveland.

Tidbits

2004-05-23T10:14:00.000-07:00

I have no earth-shattering remarks to make today but for once Blaine Newnham wrote a column I agree with in this mornings paper. I have to agree with the strategy he puts forth for the M's: don't go after the big rent-a-player (ahem...Carlos Beltran), but try out some of the PCL-leading talent the M's have in Tacoma. Mr. Newnham gets extra credit for the shrewd suggestion of preparing for the departure of Freddy Garcia this season by looking to get some power-hitting talent in a trade and sending some of our talented AAA arms up to the bigs. Hats off to you Mr. Newnham for a column well done.

One other M's note is that the boys over at USS Mariner are making the first suggestions I've seen that Paul Molitor is not working out as the hitting coach. The old adage that those who can't, teach could be happening in reverse in Seattle: those who can (especially at a Hall of Fame level) can't teach. We'll see. I'm not sure I would put all of the blame for the M's sorry offense on Paulie's shoulders, but I think one has to question the M's new make-the-pitcher-work-strategy. It may be that a lot of these guys are just better aggressive hitters and are starting to think too much.

As for roto-ball, I must be leading the league in no-decision pitchers. I think I've had at least five or six starters play in the last week that have pitched decently only to walk away without a decision. When you add the fact that Gary Sheffield and Juan Gonzalez have been extreme underperformers I can't help but feel hopeful. Additionally Keith Foulke only has nine saves so far because the Red Sox seem to win or lose by wide margins. I gotta think Keith is going to pick it up soon. Hope springs eternal does it not?