DRM - Daniel’s Random Mutterings

Well - it seems that the next Valleyschwag packages (see my previous entry) are finally on their way. I’m not making any predictions as to when they’ll arrive, but knowing the German postal service and taking into account the fact that next Monday is a holiday, I’m not expecting anything to arrive before the middle of next week.

Maybe I’ll post pics when it’s there.

In other news, Amazon finally delivered my copy of “A Guide to the Project Management Body of Knowledge“. An imposing brick of a book. With very few full-page color photographs and cartoons (none, from what I’ve seen so far while skimming through it on the drive to work). I’m not yet sure wether to be thrilled and excited or to be afraid, be very very afraid.

There - that got your attention, didn’t it?

Anyway, I finally managed to finish Paul Graham’s “Hackers and Painters“. After really enjoying the first two or three chapters (see my previous post), I found finishing this book quite a chore and on the whole found it quite a let-down. At the end, I’m not sure what I’ve learned. Except that Paul enjoys LISP. That’s what I found the most painful and booooooringgggg: Advanced LISP fanboy-hood (is that a word? If not, I’ve got copyright).

What totally threw me off was the repeated “Milchmädchenrechnung” (see here for explanation) of developer productivity and the huge gains possible by using LISP. Paul’s argument — condensed in order to make a point — is as follows: Due to syntactic brevity, the average LISP program is five to ten times as short as a Java program solving a similar problem. Hence, LISP programmers are five to ten times times as productive, giving any start-up using LISP an enormous competitive advantage over those using Java. He actually writes: ‘So the more powerful the language, the shorter the program. […] Code size is important, because the time it takes to write a program depends mostly on its length. If your program would be three times as long in another language, it will take three times as long to write […] If [your competitors developing in LISP] spent just three months developing something new, it would be five years before you had it too.’ (p. 192f)

OK. I’ve got a Ph.D., have been in the software business more or less for twenty years. I’ve seen a lot of things come and go. You’d think I’d be a bit blasé about IT stuff. But I’ve still got enough of a fifteen-year-old nerd in me to be immediately thrilled, delighted and ultimately hooked by the idea of Valleyschwag.

Valleyschwag, in a nutshell, is actually paying money for stuff you should get for free: someone’s advertising material. It’s what the brand-conscious teens all do. Why pay a normal amount of money for a normal sweatshirt when you can pay three times as much and be a walking billboard for the company?
Only — these are IT brands. The advertising is about IT stuff and IT related. And being hip IT geeks altogether, the Valleyschwag dudes (and, possibly, dudettes) even say they won’t include stuff from companies considered evil. Now isn’t that something!
I looked at the flickr albums of people unpacking and showcasing what they got in their first schwag “care package”. Stickers, flickr bages, a Technorati t-shirt (some folks even appear to have received an extra pair of oranges, and used them to inspire the biggest flickr comment storm I’ve ever seen). Now, the rational part of my brain said “big deal — they just paid fifteen bucks for company stickers and a t-shirt. Stuff you can pick up at a computer expo for free (or maybe in exchange for a business card — getting you not only the t-shirt, but also an endless supply of advertising junk mail adressed to someone who has almost exactly — but not exactly — the same name as you).”. Yes — sometimes I actually think the parentheses.
The fifteen-year-old geek inside me, who still thinks the ZX Spectrum was a neato system and the Z80 the best processor evah, quickly pounded the rational part of my brain into submission and started a barrage of “Oooooohhhh, I WANT that. Gimme gimme gimme GIMME…!”.
No points for guessing who won. And now the Valleyschwag guys have sent an email letting me know that they’ll ship my first package any day now. I can’t wait! I know it’ll take some time to get to good ol’ Germany. I just hope they air-mail it. And I hope it’s not wrapped in burlap. Who needs natural fibres anyway?

Well — my portal site is almost ready to launch. Roughly two-and-a-half months of part-time effort (a few evenings a week, plus the weekends) from first idea document to implemented software, running on my hosted server, with my own domains, trademarks applied for, etc. Not bad, if I say so myself. Although I have to admit a lot of the development speed is actually due to using the Python-based Web development framework, Django. Being accustomed to Java/J2EE/Tomcat-based development and deployment, I’m amazed at the speed and productivity that this development environment gives me. Python feels quite comfortable, Django fits my way of thinking perfectly. Almost everything is exactly where I would have expected it. The build-test-fix-turnaround is phenomenally fast.

I’m not sure that I would want to use this for every kind of project, though. I’m still not sold on untyped programming languages, for instance.

For one thing, I still miss all the support my Java IDE (IntelliJ IDEA) gives me during the coding — most of which is due to the typed programming language. In IDEA, code completion is very helpful. As is the continuous background checking of the implementation, flagging potential semantic and runtime errors even before I’ve started compiling — sometimes even before I finished typing. The Python IDE in Eclipse just cannot keep up. Since it has no idea what my shiny new variable blaBatz actually is, it cannot reasonably provide me with any code completion suggestions once I write “blaBatz.get” and press CTRL-SPACE (or whatever).

I’ve spent most evenings and weekends for the last weeks busily hacking away and didn’t get much posting to the blog done (never mind much of anything else). Now my new community portal site is almost approaching public beta status. Or, rather, 1.0 status. “Release early and release often” doesn’t mean you have to slap a “beta” label onto everything (I’ll not name any names … {cough}Google{cough}). Anyway - the server is set up and installed, domains have been registered, etc.
I won’t let the cat out of the bag yet, but — as they say in advertising — “Watch this space”.
Oh, and of course the new portal site is