Most of the highlights of my day two involved meetings and discussions with people, so there's not a lot to write about. This year there are lots of people blogging the conference, so I'm sure that there'll be lots of detailed information about interesting sessions.

Cliff Schmidt did a good talk that introduced people to the Apache Incubator. Cliff has a lot of cuts and bruises from his experiences with open sourcing XMLBeans. I've never made it to a Lightning talks session, but i went to the Python talk mostly to see Dan Sugalski get pie on his face. Due to technical difficulties with the preparation of the pie, Dan didn't end up getting pie on his face. But it was interesting to learn how far the Parrot team was able to get on running the piethon benchmarks. I think I'm going to be paying a bit more attention to Parrot, and I hope that we'll see a full set of benchmark results at the end of August.

One other notable event: Anna Ravenscroft and Alex Martelli got married last Saturday and are at OSCON as the first stage of their honeymoon. They met as a result of their involvement of the Python community. I happened to be in the right place at the right time to participate in a toast to their new relationship. It occurred to me that their situation was an indicator of the type of community that has grown up around Python. I wrote a bit about the flavor of the Python community when I was at PyCon earlier this year. For me, this little toast was a great indicator about the quality of the Python community.

Here are a few highlights of my first full day at OSCON.

Tim O'Reilly described a new web services API to the Safari online book service, along with Safari Unversity, which allows professors to build textbooks by grabbing individual chapters and then printing the book on demand.

r0ml's keynote on the Semasiology (science/history of the meanings of words) of Open Source was wide ranging and entertaining, and also addressed a very important issue - what is the meaning of the words "Open Source"? He pointed out that folks in the open source community talk a lot about what open means from the view point of licensing, but demonstrated some examples from accounting and horse breeding that suggest other means of what it is to be open. He performed a similar exercise in the notion of source code, using food (ending in a nice play on free beer) and music as examples. One thing that he said which I particularly liked is that one meaning of source is the originator of something. And that under this meaning, the source of code was the people, not code artifact. Quite thought provoking. I only wish that he hadn't run out of time. Nonetheless, what it means to be open source is very important, and an understanding of what that means affects lots of important discussions.

The other interesting talk that I went to was the IronPython talk, which was updated with the results of running parrot bench on IronPython. Jon Udell and Edd Dumbill have covered the talk in gory detail. I don't have much to add other than a favorable impression of the work that's been done so far. I've had a few interesting hallway conversations about CLR related topics.

Talks that I wish I had seen but didn't due to conflicts: Megan Conklin's power law talk, which got pre-empted by the open sourcing Java panel (This is also that last of these panels that I will go to. They just raise my blood pressure, and if things are still at this point by the next conference I go to, I think the issue will be moot, at least for me). Stefano's talk on Darwinian Software Evolution, which got pre-empted by Dan Gillmor's talk on Open Source Journalism.

Today's Chandler presentation and BOF went pretty well. The demo ran without a hitch, and people that I talked to seem to understand the ideas they were being shown, despite the ugly UI. People seem to be extraordinarily patient with us, and I hope that their patience will turn into active support and contribution once we get to a semi-usable state.

After a reasonably pleasant train ride, I arrived in Portland last night just in time to register and stick my head into the SVK BOF. I'm glad that I rushed it to get there. I'm not sure how to give a one sentence summary of SVK, but one way of describing it is as a meta proxy for version control systems. It allows you to work with a CVS, Subversion or Perforce repository using Subversion level functionality, including atomic commits, etc. A lot of work has been done to make offline performance better than Subversion. That's interesting by itself, because it means that you can work with CVS at a higher level of functionality without switching to Subversion.

The SVK team has also added the ability to do star merging, thereby solving the repeated merge problem, which yields functionality at the level of arch. SVK repositories can also be decentralized, much like arch repositories.

The only downside to SVK is that it's written in Perl (some folks will find that to be an advantage). Also, at the moment, installing SVK is a bit difficult, particularly on Windows, but the SVK team said that when Subversion 1.1 comes out they'll work in making installable binaries for Windows.

It's too bad that the SVK folks were only able to get a BOF, because it seems to me that SVK is going to be highly useful once the install issues settle out.

I also ducked into the big Tuesday night event, mostly to hear Paul Graham speak in person. He was highly entertaining, but there wasn't necessarily a lot of content that was new if you're familiar with his essays. During the train ride down, I was able to catch up on some of the ITConversations recordings that I downloaded, including one by Graham. The one thing that I liked about the train ride is the availability of a power plug right next to me. The thing I didn't like about the train was the amount of motion/sway that is transmitted to the rider.

After that, I ducked out to have dinner with some ASF folks. We went to the Chinese restaurant across from the hotel -- something that I never do because I spent a bunch of time in Chinese restaurants as a kid, and I mostly can't stand to teat in them -- but hanging out was more important than food. Unfortunately, this place closes at 9:30PM, and the staff was very pointed about getting us out of there. Needless to say, I won't be eating there again.

It was too early to go to bed after dinner, so I wandered down to the conference level and sat in a wifi hotspot and talked to people who were around. One person that I talked to was P. J. Cabrera, who is working to bring open source to Puerto Rico via an interesting combination of startup and non-profit. He told me quite a bit about the cultural obstacles to adopting open source in Puerto Rico. Really interesting time. The chance to talk to people from outside the ASF and OSAF biospheres is the biggest reason that I come to OSCON, and it's off to a great start.

Stephanie Booth describes her experience taking collaborative notes at BlogTalk. Reading it over reminded me of our experiences at PyCon this spring. It also got me thinking about OSCON next week. Taking notes at a conference is a lot of work, especially if you are the one hosting the notes documents -- you feel like you need to leave your machine running and have SubEthaEdit loaded at all times. With the advent of SubEthaEdit 2, it requires more coordination because you have to worry about which version of SubEthaEdit people are using. I haven't decided how much note-taking I'll be doing this week, but I'm happy to provide some web space for people who want to take notes. If people want to use the template that we used at PyCon, it's here (credits to Trevor Smith).

A reader wrote in suggesting the MacTruck as an alternative to QuickerTek's Powerbook handle. While the MacTruck looks like the invincible case for a Powerbook, it doesn't solve the problem that I have. I like to wander around in conferences, and being a small guy, I don't like toting a lot of stuff around with me. If Powerbooks had battery lifespans comparable to Centrinos, I could just carry the laptop for most of a conference. As it is, I minimally need to carry the Powerbook, the charger, and a spare battery, which is really too much. Since the MacTruck weighs in at 2.4kg (about 5lbs), it weighs as much as the Powerbook does, it's clearly way too much.

Via Feedster's new links feature I discovered r0ml's blog. Showed up even before Technorati found it.

The folks at MobileBurn have a review of the GN Netcom 6110, which is a BlueTooth headset which you can plug your landline phone into, possibly providing a way to use a single headset with cell phone, iChat, and landline. The review is short but postive. Concerns I have:

1. In order to be effective for my landline use, I need an amplifier (with a mute button), a feature that I have on my wired GN Netcom headset.
2. I'd like to have confirmation that it works w/ my equipment -- Nokia 6600, iChat, and landline -- I don't want to be on the bleeding edge for that.
3. The headset is kind of ugly

Morgen Sagen is working on the parcel system that allows os to build Chandler out of a bunch of small modules. Recently we've started using Twisted to handle networking functionality in Chandler. At our staff meeting last week, Morgen demonstrated a servlet system for Chandler which is based on Twisted's web resource functionality. He's got servlets for a telnet server, a simple photo blog, a Wiki that Grant Bowman is hacking on, and the beginnings of a pyblosxom servlet. When we (re)did the design for the pyblosoxm core engine, I tried to make sure that we could use another storage system besides the traditional file system used by *blosxoms. I think we're about to find out whether we did the design well enough.

If you're curious, Morgen has put up a subversion repository with all of his stuff.

Tim Bray observed some nasty code that he had to write for his Zeppelin project:

Another routine in the grotty underbelly of Zeppelin does a bunch of file/directory maintenance: create a shadow directory tree if it’s not there and copy a bunch of files selected by extension hither and yon. I’m sure programmers will still be writing this kind of code when I’m a hundred years in the grave. Well, Java is really not the right tool. What would have been a few lines of straightforward Perl turned into half a dozen klunky-looking Java methods. If I’d been able to rely on Jython just being there and in my IDE, I would have come out way ahead. Fortunately, I think we’re going to fix that one.

Tim concluded that the language (Java in this case) was to blame. Much as I love dynamic languages, I don't think that the problem here was exclusively the language. The design of the accompanying libraries is as much to blame as anything else. The Java libraries are full of good object-oriented design - small methods that do one thing, tons of getters and setters. Unfortunately, this style of design tends to overlook actual use cases of the code. So tasks that could be done in a few lines of code end up taking many lines of code. Of course, all of this sort of thing is influenced by the language.

[ via the Gentoo Weekly News ]

The Gentoo project has ported their Portage package manager to the Mac and released Gentoo Mac OS. It will be interesting to see how this interacts with the previously announced MetaPkg project.

I've tried both fink and darwinports, and had trouble with both (mostly when trying to get a decent subversion that didn't also require me to override Apple's Python). I wonder if Gentoo Mac OS will be any better in this respect.

InternetNews.com and Java.net picked up my report on this month's SeaJUG meeting. I was a little surprised that the Java 5 transition would be considered newsworthy, since platform transitions are somewhat common in the computer industry. Java is going to have the transition to Java 5, which is big if you are want to use the genericized or annotated APIs. .NET is going to undergo a huge transition for Longhorn. Apple has just completed their transition from MacOS to Mac OS X, and the desktop Linux folks are talking about a transition to either Java or Mono. Compared to these other transitions the Java 5 transition doesn't seem so bad.

As Mitch announced in his blog, the staff of Creative Commons is sharing office space with OSAF. I'm very pleased to see this happen, because Larry Lessig's writing have had a big influence in shaping my thinking about open source. I think that the notion of a commons is an important idea for understanding the places where open source is relevant and important, as well as for thinking about what happens next.

Here is a recent CNN article about the Apache Software Foundation. It was also picked up by Forbes and the Marin Independent Journal.

The deadline for ApacheCon submissions has been extended to Monday August 2.

I would love to have one of these in time for OSCON, but I can't seem to find anybody who's actually selling them. I hate carrying tons of stuff around conferences. It would also solve the problem of using a Powerbook at a good typing height. The only thing I'm not sure about is how it would work if you need to put the Powerbook into a carrying bag.

The other night Wilhelm also gave me a brief demo of Aleks Totic's pydev, which is ironic because Aleks volunteered at OSAF before I started there. pydev looks promising but it looks to me like it still needs some more work. Aleks is no longer working on it, which is a bit scary but it looks like there are some other developers as well. Perhaps Dana Moore will be talking about pydev at his session next week at OSCON. I mostly use Wing just for the debugger, and they've been really late with the Wing 2.0 beta for OS X. Not only that, but Wing on the Mac requires X11. Here I go, turning back into a Mac snob...

I go to SeaJUG to hang out with people. If the presentation is good, that's just a bonus. Here's an example of why. As we were riding over to the meeting, Wilhelm was telling me about see, which is basically emacsclient/gnuclient for SubEthaEdit. It allows you to invoke SubEthaEdit from the command line and have it open a file in a running SubEthaEdit, or start a new SubEthaEdit if one isn't running. Wilhelm picked this up off some blog that I never heard of. It comes in Objective-C source format.

After last night's SeaJUG, I had a discussion with my friend Bob Coombs, who is a long time NeXTStep programmer. He was explaining some features of Objective-C to me. In particular, categories, which allow you to group up methods (which sound a lot like aspects in some ways), and posing, a way for one object to pose as another. The posing functionality sounds a lot like features that Bob Ippolito and Donovan Preston described as part of pyprotocols. As usual the world is small, especially the world of ideas.

Aside from issues of licensing and market share, Cocoa/NeXTStep is probably the most advanced commercially available system for building desktop applications, so I've been getting more interested in learning how things work. It's kind of like the scene in The Matrix when they are about to extract Neo from the Matrix. Neo touches a mirror, his finger goes silvery, and all of a sudden the silver is crawling up his arm. Next thing he knows, it's taking over his head. Yep, it's kind of ... exactly ... like that.

Tonight's SeaJUG meeting was the annual panel of JavaOne attendees. One interesting thing that happened during the discussions was the amount of concern raised over the new language features, particularly generics and attributes. There was some feeling that Java 5 was going to create two languages, pre Java 5 and Java 5. The fact that many of the new JSR's are already incorporating generics and attributes will only strengthen that separation. A few people felt that it would be even worse, that due to the timing of the releases, Java developers would actually have to deal with 3 dialects of Java. Takes a bit of wind out of the write once run anywhere sails. Thanks to an unknown SeaJUGer for the title of tonight's post

Another thing that I found interesting was that there was almost no mention of EJB 3, despite the fact that it's much simpler than EJB 2. Most of the time was spent talking about J2ME, JNDC and JDIC, Looking Glass, and generics. A few people lauded the benefits of JNDC -- apparently one of the panelists discovered that some of the JNDC components were Swing extensions that were written when members of the Swing team tried to actually write a Swing application. A few people spent portions of their reports talking about whether or not Sun was understanding the importance of developers, and on their perceptions of Sun's financial health.

Joe Bowbeer and I disagreed over the usefulness and openness of the JCP. Joe has had a good experience with the JCP, while I've mostly had indigestion over JSRs, and the long process of getting the JCP to be more open. It appears that some SeaJUG members have me pegged as Mr. Groovy, a title which should be reserved for James. I haven't actually been very involved with Groovy in the last few months. I haven't made up my mind over whether I'm going to participate in the JSR process, for a variety of reasons. One if them is that since the announcement of the JSR, I feel that there are more people pushing for Groovy to be more like Java, which (at least in my mind) defeats the purpose of having Groovy in the first place. Apparently, I'm not alone in my thinking -- this week eWeek published an interview with James Gosling in which he said

I think they could be a little more outlandish and get a little more interesting.

(thanks to Dion Almaer for the pointer to this - you'll have to read Dion to find the eWeek interview). I'm not really in the mood to fight with a whole bunch of people who have a different outlook on what Groovy ought to be, especially when it's starting to look like there are other people out there who have an outlook closer to mine (more on this another day).

As we got into the car to go over for our ritual after meeting beer and food, I turned to Wilhelm and said "this session made me glad that I'm not doing Java stuff at the moment". He responded that all the hue and cry over generics was reminiscent of the hue and cry when objects got introduced. We talked a bit more, and I said "if you're going to have static typing, then I want the computer to figure out what the types should be". I've done templates in C++ -- I used some of the template metaprogramming techniques that Barton and Nackman first showed. Of course, I'm not a fan of statically typed languages to begin with.

Java programmers are going to have a lot to swallow over the coming years. The good news for them is that C# isn't any better. C# already has annotations (attributes) and is about to add generics, so the two languages are remaining essentially equivalent. But as one person asked "what happened to making it easier"?

Dan Bricklin's latest article is titled Software That Lasts 200 Years, but "Societal Infrastructure Software" was the idea that really resonated with me.

I'm not sure that I agree with a 200 year horizon, especially for a field that is less than 50 years old. However, I think that many of the issues that he describes in the article are important and worth thinking about. Foremost among them is the notion that a certain classes of software are societal infrastructure. While he doesn't use the terminology explicitly, Bricklin, argues that these classes form a commons, and because they form a commons they ought to be conceptualized, designed, developed, and funded differently than other classes of software.

I haven't worked on the kind of software that he classifies as Societal Infrastructure Software, but I think that there is a class of software, call it Personal Infrastructure Software, which plays a similar role for individuals. At the moment, my personal description of that class includes RSS Aggregator, web browser, text editor, e-mail program, address book, calendar, and program development environment. I think that instantiations of these programs form a similar commons, and that we're seeing that come about in various degrees via open source implementations, but there are still issues that need to be deal with. Probably the most important one is data format lock in. Close behind it would be the staying power of the various communities working on various pieces of that commons.