I worked from home today partly because I’m battling a cold and want to take it easy. It was actually a very busy day. I finalized the release 4.1 implementation plan with James E. and we got the RFC approved, so hopefully everything will go smoothly on Monday. I also crafted a unit-test strategy for the daylight saving time change. I collaborated with some of the other developers and we came up with a battery of test suites to verify if the DST patches work.
The issue is that for the first time in over twenty years, the date that daylight saving time occurs is different this year. Instead of being on the first Sunday in April, it will now be the second Sunday in March. This change (a law enacted in 2005 and beginning this year) requires patches to the operating systems and java. JDK/JRE 1.5 is already complainant, but anything older than JDK 1.5 needs to be patched. We came up with a series of tests which show when DST occurs this year for the following scenarios for our HPUX environment:
- Java 1.4.1
- Java 1.4.2
- Java 1.5
- C++ Standard Library
- C++ RogueWave Library
- ‘date’ shell command
I wrapped them all up into a script and have it packaged in a tar file that we can deploy on whatever server we need to test the patches on.
I also dealt with a lot of emails and worked with Vasu for a couple of hours getting his Makefile changes incorporated into our master Makefiles which are included by all other makefiles. He is working on a small project to convert our dataservices from using Oracle to now be using DB2. We do a pre-compile step that needs to be modified. Currently we use the proc compiler from Oracle. Going forward we will be using the DB2 precompiler (the C++ equivalent to sqlj). My changes seemed to work so it will mean that there will be minimal changes to the local makefiles for each project being rebuilt.
One thing I always push for is to eliminate duplicate code & redundancy, so instead of having all of the same ’stuff’ in all of the local makefiles (we probably have over 100), we have as much as possible up in some master makefiles that are included and used as templates. It’s something I came up with years ago and we still use it today. I’m actually pretty impressed how well the approach has held up considering that I didn’t know a single thing about makefiles when I did all of this - I just learned as I went.
I stopped working around 6pm and took a shower and got dressed just as Jen got home. We worked together to get the house ready for our ‘Game Night’ tonight. We’re having a bunch of friends over to play Pictionary. I wish I was feeling better, but can’t do anything about that.

(The dining room with all of the food)
We had a great time. Katie, Chris, Lynne, Tremayne, and Tracy came over. Jen put some large sheets of whiteboard-type contact paper on the windows that we used dry-erase markers to do our drawings. We played three games (broken out into two teams each time and mixed-up the teams after every game).

Jen also prepared some horderves and drinks. It was really nice. At the very end we decided to come up with some really difficult things for the opposing teams to draw. My team (Katie, Chris, and I) came up with some things like: “Abstract, Oxymoron, & Personality”. The other team came up with some words like: “Mozart, & Tai-Chi” for us to draw. It was extremely difficult but fun. I’m looking forward to the next ‘game night’ which should be hosted by Clint & Erin.

(Jennifer looking happy)