Plone Conference 2006 blows away ================================ Cab driver: "So, what brings you to Seattle?" Me: "A software conference." Cab driver: "Ah, of course." Me: "No, it's not Microsoft. In fact, it's quite the opposite. We make software that doesn't cost anything." Cab driver: "Software that doesn't cost anything? Really?" This is how my first day (or shall I say evening?) in Seattle started. Alex Limi later reported in his keynote that he had a similiar conversation in a taxicab, except his driver actually knew what Open Source and SourceForge were. That's almost hard to believe. Fact is, though, Plone's more popular than ever. And so I was not the only one who came to the city that is usually better known for its proprietary software vendor. 360 other people decided to do so, too! There are some statistics about that number that are worth mentioning: * For most of the participants (roughly two thirds), it was the first time being at a Plone Conference. * A surprisingly big group stated that they are in involved in some form of government, about the same number as people from an NGO background. Another large non-business group was education. * About 80% of the participants of the pre-conference trainings said they would not have come if it weren't for the trainings. * Over 50 interested veteran and rookie Plonistas took part in the post-conference sprint. The conclusions I draw from this are these: * Jon Stahl and his team at ONE/Northwest have done an incredibly good job at marketing this conference to people previously outside of Plone and to sectors previously not so well represented in Plone (government, education). * Plone's growing fast and newcomers still participate in the community by coming to conferences and participating in sprints. * Pre-conference trainings make an otherwise boring-sounding conference attractive for newcomers. Of course, I bet most of them realized pretty soon that the conference was not at all boring, but I guess many others are. The pre-conf trainings were the bait, so to speak. The talks --------- There were so many good talks that there were very hard choices to make. Unfortunately, sometimes the best talks of the day were at the same time, making it even harder to choose. Fortunately, however, most of the sessions were video-taped. On Wednesday, Paul Everitt kick-started the conference with his *What makes Plone different and better* presentation. In this for Paul typical "content-free" presentation, he put us in the right mood for the rest of the conference. After that, Tres Seaver talked about GenericSetup, especially Seaver's Law ("Persistence means always having to say you're sorry"). Too bad it was up against Kai Hänninen's PrimaGIS talk which talked about spatial indexing and mapping in Plone and left, as I heard, most of the audience jaw-dropping and sulliver-drooling. For technical people, the afternoon was then dominated by Rocky Burt and his tutorial on *Using Zope 3 Technologies* in Plone. Thursday was started with Alex and Alan's keynote which, as typical for the Plone founder duo, was pretty much just a random collection of slides. Martin Aspeli followed with a tutorial on how to develop content types for Plone 2.5. The afternoon had shorter talks, such as *REST in peace* by Lovely Systems CEO Jodok Batlogg and Zope 3 developer Stephan Richter. It was a short and very energetic talk, despite the fact that it was very technical. The highlight of the conference, Eben Moglen's keynote, was on Friday morning. I unfortunately missed it, being quite hung over from the night before. I barely made it to the conference in time for my talk after lunch. Unfortunately, I was up against the KSS/Azax tutorial by Godefroid Chapelle and Balasz Ree, so not too many people came to my thing, and I ended up missing another talk I was interested in. For me, the conference ended with a workshop on improving Plone's add-on story, lead by Martin Aspeli. There were a lot of fantastic lightning talks at the end of each conference day. Definite highlights were Kevin Teague's *History of Plone* which featured funny facts and quotes of past and present days, Alan Runyan's presentation of Entransit ("It's all transactional, bla bla bla") and *plone.portlets* by Martin Aspeli ("Portlets in Plone 2.5 are pretty nice, so we thought we'd rewrite them from scratch"). The organization ---------------- The organization of the conference was superb. Thanks to Jon Stahl and the rest of the ONE/Northwest team for putting so much time and thought into this. There were only a few things that didn't work out so well. The WiFi network was a definite problem in the beginning, but that's something that never seems to work right at any conference. The projectors could also have been better and bigger. Last year's conference in Vienna definitely had the better equipment there. Other than that, there's little to be critiqued. One innovation this year was the video taping of selected talks. `MrTopf a.k.a. Christian Scholz`_ also streamed the talk whichever he was in via a QuickTime link and into Second Life. I think that was a great idea and it should be continued for the next conferences. Even better, **all** talks should be taped and put on some subsection or subsite of plone.org. .. _MrTopf a.k.a. Christian Scholz: http://comlounge.tv/ploneconf.html The nightlife ------------- Seattle is a relatively quiet town, at least that's how I perceived it. There wasn't much going on in the streets, especially not at night. But if you know the Plone/Zope community at all, you know that that didn't stop us from having fun. Hanging out at one of the fine bars in the close Queen Anne or the distant Ballard districts, or with beer from the liquor stores on the roof or in the conference room (sic!) of the Mediteranean hotel, we managed to definitely keep up with the drinking. The sprint ---------- More than 50 die-hard Plone developers, newbie developers and non-technical (marketing, docs) people attended the sprint -- the biggest sprint I've ever been to, I think. I was anxious to see how the organizers Rocky and Nate would manage such a large crowd. To be honest, I wasn't expecting much. How the heck would you manage 50 people and *still* get something done? I was surprised, positively surprised that is, with how well the sprint went. We split up into several groups (membership, KSS/Azax, multimedia, syndication, etc.), each led by an experienced Plone core developer. It seemed that everyone knew what to work on very quickly so at the end of the first day, each group had a lot of results to report. At the end of the second day, a lot got more got done. And while not everything was finished right there, it seemed like the sprint brought everyone a big step forward, even though it lasted for only two days. One observation of mine (and this is true also for the conference) is how much Plone is now pushing toward Zope 3 technology. While Plone 2.5 contained *some* Zope 3 views, everything that's new in Plone 3.0 is now written using Zope 3. Probably half of the people at the sprints were writing code that actively uses the Zope 3 Component Architecture. Wow! I myself formed the smallest group with just Alec Mitchell. We picked up something I had worked on about half a year ago at the `PyCON 06 sprint`_ to which I was sponsored to go: The *Customize* button for Zope 3 views. After many discussions with Martin Aspeli on local components (plone.portlets is using them heavily) during the previous months, it occured to me that there's a slightly easier and better way to handle a customized template than how I had done it in Dallas. I also finally had an idea how to deal with security for the view class. So Alec and I started out, taking tests and view code from my Dallas prototype and implementing the actual customization code from scratch based on my new ideas. By the end of the first day, we had customization working in unit tests; by the end of the second day, the UI was in place and even security checking worked accordingly. The new code is now in a separate package called `five.customerize`_ which can be installed into any Zope 2.10 instance. It's not done yet, I hope to polish it up during the next few weeks. I would then very much like to find a maintainer from within the Plone or greater CMF community (as they're mostly interested in it and probably want to integrate it tightly into their next releases). .. _PyCON 06 sprint: http://www.z3lab.org/sections/blogs/philipp-weitershausen/2006_03_07_pycon-2006-sprint-recap .. _five.customerize: http://svn.zope.org/five.customerize/