Posts tagged: OSCON

If my scanty coverage of the 2005 O’Reilly Open Source Convention wasn’t enough for you, check out these excellent, thorough post-show wrapups from Andy Oram at O’Reilly and Slashdot (though I recommend setting your score threshold to 5 as always, in this case to boil down a “Ruby On Rails Doesn’t Scale” thrash).

The above were my two favorites from O’Reilly’s general OSCON wrapup page; it also includes links to Technorati, Feedster, Bloglines, and Delicious tag searches if you want to read every last bit of prose posted about the convention. Also, you may find some of the presentation files interesting or useful, especially if you’re an attendee with spotty notes.

It’s possible that there’s still some guy doing laps around the Oregon Convention Center on a Segway with security chasing after him, but as of 1:30PM today OSCON 2005 was officially over.

This was my first time attending the conference, and I had a great time. I learned a lot, had some excellent discussions and unexpected laughs, and got myself fired up again about being part of the open source anti-massacree movement.

I have to admit that I carry big shield of skepticism when I circulate exhibit halls. Luckily a fellow attendee tipped me off to OpenLaszlo, an extremely spiffy system for server-side, declarative generation of Flash content. What this means for somebody like me – someone who, despite a lot of background in visual design, would really prefer to work directly with code – is that very sweet Flash-based interfaces can be constructed via XML. Their XML dialect, LZX, impressed me with its elegance and (for XML) relative lack of verbosity. The generation code is Java, which you can either run live on your server or run offline to generate standalone .swf files. They offer a nifty playground/demo for you to check it out.

Belatedly, here are my notes from Guido van Rossum’s “State of the Python Union” talk this past Wednesday.

Guido discusses (with illustrations) his recent eye problems.

SD Magazine/Jolt Productiviity Award given to Python 2.4. O’Reilly Python book sales are up. Unscientifically extracted Sourceforge stats: about 4000 python projects.

Discussion of PEP 342 and PEP 343 broke records on python-dev (“What about decorators?” somebody asks. “I think this was worse,” Guido says.)

PEP 342: Generator enhancements. Can we make yield do some of the things that it does in Ruby? Yield becomes an expression instead of a statement.

The place to be during the afternoon break today was the Portland Ballroom, where the Artist Currently Known as why the lucky stiff and musical accomplices unleashed a multimedia explosion involving Ruby, cartoon foxes, animation, repeated MPlayer crashes, video artifacting, rocking out, shadow puppets, and network timeouts. I think this is the show he did at FOSCON last night. The quality of the stuff that did work was so high that the stuff that didn’t work wasn’t such a big deal. My favorite bits were the hilarious animated imaginary Ruby Cabal meetings. A good time was had by all and I don’t think that there will ever be another conference presentation where someone says, “You’ll notice we’re using octagonal paper… as seen on Battlestar Galactica.” Check why’s site to see if he comes through on his promise to post those media files.

The PDX Ruby Brigade (“PDX” is the airport code for Portland) is hosting a sort of alternative/extended OSCON called FOSCON. It’s being held at Freegeek.

Several of us in Portland got into OSCON free the last couple of years by volunteering. When we found out that there would be no volunteer opportunities this year we decided to see if any of the Ruby speakers would like to practice their OSCON talks on us local Rubyists - thus FOSCON (or as Lennon calls it: OSCANT) was born.

O’Reilly has been running CodeZoo for a few months now. Today they announced CodeZoo subsites for Python and Ruby. CodeZoo is very slick – you can track changes to a particular app or component via a special RSS feed, for instance. Downloads are fast and simple, even for Sourceforge-hosted projects. And they’ve got this new thing called DOAP (why do you think they call it DOAP?), an XML schema for component information. (Tangent: I’m thinking that DOAP could be a nice standard upon which to build phone-home version checking features.)