Thursday, October 19, 2006

Let The Real Work Begin

After all the fun of Happy Camper, it's time to start doing some real work. I spent yesterday setting up computers in the core logging room. It's going to be a pretty nice setup. There will be two computers for the sedimentologists to use. One will be set up for using the Corelyzer visualization software. It's a new Mac Pro with 2 30" Apple Cinema Displays hooked up to it. The second computer will be used for my software, PSICAT, and is a MacBook Pro hooked up to a 30" Apple Cinema Display as well. So we've got a Mac Pro, a MacBook Pro, and 3 30" Apple Cinema Displays all sitting on the benches in a room that has been known to leak. What could possibly go wrong?

Beyond setting those up, I was setting up our central servers. I got static IPs for them so I can start circulating them around. They are mostly set up, though I had to do some tweaking to the Plone-based Whiteboard stuff last night before I retired home to do some laundry and drink a beer.

The final area of work is going to be updates and new features added to PSICAT, the core logging software I've written. I haven't done much on it since shortly before I deployed but there were some updates that were in the pipeline. I did a few last night and will finish up the rest this morning. I also have to implement a whole new piece of functionality this weekend. Fortunately, since Eclipse is such an awesome platform, this is pretty trivial to do. I simply have to implement a new plugin with my code, add the plugin to my feature, and push it out to the update site. Then all of the clients will automagically download the new code. The whole update site is going to be very useful for distributing updates to both the on-ice scientists and those that are off-ice, and it is the major reason why I'm not freaking out at the moment. I've got a clear way to distribute new features and bug fixes to my clients, so I can fix the problems on the fly and know that everyone is getting the changes.

A related note on the PSICAT front, Stephan Wahlbrink contacted me about possibly translating PSICAT and some problems he encountered. He is using PSICAT in the German locale, and it turns out that I'm not handling numbers in the correct locale-aware way. I display them properly according to the current locale but I don't parse them properly which leads to some major errors and makes PSICAT pretty unusable for non-English locales. I looked at how I was doing things and it turns out I was using DecimalFormat.format() to display the numbers but Double.parseDouble() actually parse the user's text. Doh! So I've got to go through and update the code to parse numbers from the users in a locale-aware sort of way but store them in the XML files in the standard English decimal format so you can share data between different locales.

1 comment:

roonie said...

Don't break my program!!! :D