Archive for August, 2005

Toss in the Moss 2005

Monday, August 29th, 2005

Toss in the Moss 2005 at Forsythe Park

For the second week-end in a row, I played in an Ultimate tournament, but this was the first in about two years that I actually traveled to. Our (former) masters team, THOR, decided to try “one last tournament”. We added some young talent and trekked to Savannah for the popular Toss in the Moss tournament.

I don’t know why they keep holding this tournament in the heat of August, but this year wasn’t so bad as the temps stayed in the 80’s at the cost of some occasional downpours. Forsythe Park, shown in the photo, is the primo location for playing, having great fields and a heart-of-the-city location. Unfortunately, 40-plus teams is way too many for the park, and we ended up playing most of our games at satellite fields.

We finished 2-2 the first day to land in the B division for Sunday, where we lost in the semis to a very fast team from Columbia, South America. We older players appreciated a comment overheard from the college team we beat in the quarterfinals, “they’re just too fast for us,” even if it was directed at our younger players.

We didn’t come away empty-handed. We won the Saturday-morning opening parade by chanting a Viking verse as we marched with plastic swords, hammers and home-made Viking helments.

Masters Hat Tournament

Sunday, August 21st, 2005

Dwayne in action at masters tournamentWith a heat index of 109° this must have been the worst week-end of the year to hold a masters tournament, but that’s where I was from 10:30 to 5:30 on Saturday. Fortunately, we were blessed with an afternoon shower to cool things off for a while. I didn’t feel too wiped out at the end of the day (just hungry), but I am sore today and my skinned knee looks a little worse.

Mark in action at masters tournamentIt was a hat tournament organized by Dave Kaminski with teams assigned semi-randomly, and I ended up a on well-balanced team. We went 3-0 through pool play, and, after some minor re-balancing of teams, we kept rolling through the semifinals only to run out of poise in the finals. All of our suppressed mistakes manifested themselves in the second half of the finals where we didn’t score a single goal.

Between games at masters tournamentOne thing about masters players (33+) is that they come prepared. We had plenty of shade tents, watermelon, first-aid supplies, etc. Though the fields were in excellent condition, we had to supply our own shade, as seen in the sideline photo.

Slip and slide at masters tournamentIt was Rich’s first masters event, so perhaps that accounts for the slip-n-slide. He’s making a diving catch onto the slide in the photo. He even went to the trouble of lugging a cooler full of water out to the fields just for the slip-n-slide. The cooler turned out to be very popular for dunking one’s head in to cool down.

Factoring Realized

Tuesday, August 16th, 2005

I found an GNU library for fast multi-precision arithmetic called gmp and decided to try my graphically inspired factoring algorithm. I’m pretty sure it’s just an inefficient version of Lehman’s Method, which is only an improvement over Fermat’s Method. And to illustrate how much a minor player Lehman’s Method is, consider that the best web page I could find for it consists of a photocopied page from a textbook.

Undeterred by reality, I continued on. After working out some details so that my code could factor 10-digit test numbers, I moved on to the smallest RSA challenge number I could find, RSA-140, which was factored in 1999 with tons of computing horsepower. My algorithm didn’t find any factors, of course.

I scaled back to a 20-digit number which I made by multiplying to 10-digit primes. Success! Gaining confidence, I tried the product of two 20-digit primes. No luck there (after 30 minutes), so back to the drawing board.

And it really is luck I’m counting on, since the algorithm is not rigorous — it just looks for factors in places where it can check lots of candidates at once. The real factors may not be in any of those neighborhoods.

Oh well, maybe I should just plug in RSA-2048 and see if I randomly win the prize…

JSPWiki Migration Page Provider

Sunday, August 14th, 2005

I’ve contributed a migration page provider to JSPWiki. The engine supports a variety of page providers that handle the storage and retrieval of page contents from whatever store you want to use. This provider is really just a proxy for some real provider with the key feature that if the real provider is empty (no pages), the migration provider will populate the real provider with content from yet another provider.

The idea was pulled out of the original JDBC page provider which has a similar feature where it will populate itself from another provider. I made it into a separate provider so I can go the other way if necessary. That is, if my JDBC provider proves to be the source of my local wiki crashes, I can migrate the content to a standard page provider.

There is one main shortcoming: the modification dates get lost due to the design of the page provider API (adding a new page to a provider always gives it the current date).

Out with the Old, In with the New (Car)

Saturday, August 6th, 2005

Old and New CarsAfter 10 good years, I’ve finally parted ways with my Saturn wagon. The photo shows the old Saturn and new Prius as we make the sale in the parking lot of Sam’s.

The supply for Priuses is just catching up with demand around here. You can regularly find them in stock at dealerships, but the dealers are still expecting MSRP for them. We got a little bit of a discount (and delivery) by going through the AAA buying service. Not perfect for buying a car in limited supply, since they’re set-up to go find the car you want right then instead of monitoring supply until a perfect match becomes available.

Gas mileage? So far we’re getting about 50 highway and 35 city. Oh, and I did keep the FORTH GO license plate.

JDBC Providers for JSPWiki 2.2.8

Monday, August 1st, 2005

When I first encountered the creatively named JSPWiki two years ago, I thought it strange that the server relies on the file system for storing its content. JSPWiki is very extensible and I found a simple database page provider plug-in and used it with some modifications to bring it up to the then-current version of JSPWiki.

Later I did a major overhaul of the code and forked it into a generic JDBC Provider. A few folks have contributed comments, but I don’t get the sense of a big demand for SQL-based page stores for JSPWiki.

I’ve just updated the code for the now-current JSPWiki 2.2.8 and provided a jar file for easier installation. I’m hoping it will improve the stability of my local wiki which hangs every few days using the original database provider plug-in.