Been a long time.
Lots of stuff has happened since the last time I wrote anything in my journal. Most notable was the reworking of our contract here with NASA, changing our group's contract from "you just write this web app as a subcontractor" to "let's out the web services guys and the web interface guys in the same room and see what happens" on a brand new contract.
Which means brand-new machines - specifically a dual 867MHz G4. And OS X. Whee! Initially I partitioned the disk into half for OS X and half for Linux, but between no support for the ATA-100 bus in the new G4s and the, well, niceness of OS X, I backed everything up and devoted it all to OS X.
(Yeah, I could have gotten a Powerbook - the 17" one wasn't out then, though - but I really didn't want to have to deal with worrying about where it was, did I remember to lock it up, etc. With a 15-kilo desktop machine and a big honking monitor, nobody's gonna walk off with it.)
I've been diddling about with trying to settle on maintenance philosophy for the machine. Pure Fink didn't work out for me, and neither did pure configure/make/install. I've settled on a mixed strategy:
Seems to work so far.
Been delving into Web services; our big search application is transitioning to a new server base that's all Web services, so we're getting to work with XML, etc.
There may be at least a session proposal in what I've learned so far on really seriously using Web services. The thing that bugs me about XML is that so much of the knowledge of the structure of the XML ends up embedded in your program. I've come up with a way to externalize a lot of it (using Text::Template to build the XML I send, and XML::XPath and external config files to handle the XML I get. plus an infrastructure to manage multi-step XML transactions).
I'm having fun, and you can't ask for much better than that.
Oh yeah, and the debugger's coming along. Slowly, in between all this other stuff.