I just returned a few days ago to the USA after one week in Germany. I visited Göttingen for my keynote at Samba XP (which I already blogged about). Attending Samba XP was an excellent experience, and I thank SerNet for sponsoring my trip there. Since going full-time at Conservancy last year, I have been trying to visit the conferences of each of Conservancy's member projects. It will probably take me years to do this, but given that Samba is one of Conservancy's charter members, it's good that I have finally visited Samba's annual conference. It was even better that they asked me to give a keynote talk at Samba XP.
I must admit that I didn't follow the details many of the talks other than Tridge's Samba 4 Status Report talk and Jeremy's The Death of File Protocols. This time I really mean it! talk. The rest, unsurprisingly, were highly specific and detailed about Samba, and since I haven't been a regular Samba user myself since 1996, I didn't have the background information required to grok the talks fully. But I did see a lot of excited developers, and it was absolutely wonderful to meet the entire Samba Team for the first time after exchanging email with them for so many years.
It's funny to see how different communities tend to standardize around the same kinds of practices with minor tweaks. Having visited a lot of project-specific conferences for Conservancy's members, I'm seeing how each community does their conference, and one key thing all projects have in common is the same final conference session: a panel discussion with all the core developers.
The Samba Team has their own little tweak on this. First, John Terpstra asks all speakers at the conference (which included me this year) to join the Samba Team and stand up in front of the audience. Then, the audience can ask any final questions of all speakers (this year, the attendees had none). Then, the Samba Team stands up in front of the crowd and takes questions.
The Samba tweak on this model is that the Samba Team is not permitted to sit down during the Q&A. This year, it didn't last that long, but it was still rather amusing. I've never seen a developers' panel before where the developers couldn't sit down!
After Samba XP, I headed “back” to Berlin (my flight had landed there on Saturday and I'd taken the Deutsche Bahn ICE train to Göttingen for Samba XP), and arrived just in time to attend LinuxNacht, the LinuxTag annual party. (WARNING: name dropping follows!) It was excellent to see Vincent Untz, Lennart Poettering, Michael Meeks and Stefano Zacchiroli at the party (listed in order I saw them at the party).
The next day I attended Vincent's talk, which was about cross-distribution collaboration. It was a good talk, although, I think Vincent glossed over too much the fact that many distributions (Fedora, Ubuntu, and OpenSUSE, specifically) are controlled by companies and that cross-distribution collaboration has certain complications because of this corporate influence. I talked with Vincent in more detail about this later, and he argued that the developers at the companies in question have a lot of freedom to operate, but I maintain there are subtle (and sometimes, not so subtle) influences that cause problems for cross-distribution collaboration. I also encouraged Vincent to listen to Richard Fontana's talk, Open Source Projects and Corporate Entanglement, that Karen and I released as an episode of the FaiF oggcast.
I also attended Martin Michlmayr's talk on SPDX. I kibitzed more than I should have from the audience, pointing out that while SPDX is a good “first start”, it's a bit of a “too little, too late” attempt to address and prevent the flood of GPL violations that are now all too common. I believe SPDX is a great tool for those who already are generally in compliance, but it isn't very likely to impact the more common violations, wherein the companies just ignore their GPL obligations. A lively debate ensued on this topic. I frankly hope to be proved wrong on this; if SPDX actually ends or reduces GPL violations, I'll be happy to work on something else instead.
On Friday afternoon, I gave my second keynote of the week, which was an updated version of my talk, 12 Years of GPL Compliance: A Historical Perspective. It went well, although I misunderstood and thought I had a full hour slot, but only actually had a 50 minute slot, so I had to rush a bit at the end. I really do hate rushing at the end when speaking primarily to a non-native-English-speaking audience, as I know I'm capable of speaking English way too fast (a problem that I am constantly vigilant about under normal public speaking circumstances).
The talk was nevertheless pretty well received, and afterward, I was surrounded by a gaggle of interested copyleft enthusiasts, who, as always, were asking what more can be done to enforce the GPL. My talks on enforcement always tend to elicit this reaction, since my final slides are a bit depressing with regard to the volume of GPL enforcement that's currently occurring.
Meanwhile, I also decided I should also start putting up my slides from talks in a more accessible fashion. Since I use S5 (although I hope to switch to jQuery S5 RSN), my slides are trivially web-publishable anyway. While I've generally published the source code to my slides, it makes sense to also make compiled, quickly viewable versions of my slides on my website too. Finally, I realized I should also put my upcoming public speaking events on my frontpage and have done so.
After a late lunch on Friday, I saw only the very end of Lennart's talk on systemd, and then I visited for a while with Claudia Rauch, Business Manager of KDE, e.V. in the KDE booth. Claudia kindly helped me practice my German a bit by speaking slowly enough that I could actually parse the words.
I must admit I was pretty frustrated all week that my German is now so poor. I studied German for two years in High School and one semester in college. I even participated in a three-week student exchange trip to a Gymnasium (the German term for college-prep high school) in Munich in 1990. Yet, German speaking skills are just a degraded version of what they once were.
Meanwhile, I did rather like Berlin's Tegel airport (TXL). It's a pretty small airport, but I really like its layout. Because of its small size, each check-in area is attached to a security checkpoint, which is then directly connected to the gate. While this might seem a bit tight, it makes it very easy to check-in, go through security, and then be right at your gate. I can understand why an airport this small would have to be closed (it's slated for closure in 2012), but I am glad that I got a chance to travel to it (and probably again, for the Desktop Summit) before it closes.