I just returned today (unfortunately on an overnight flight, which always causes me to mostly lose the next day to sleep problems) from SCALE 8x. I spoke about GPL enforcement efforts, and also was glad to spend all day Saturday and Sunday at the event.
These are my highlights of SCALE 8x:
- Karsten Wade's keynote was particularly good. It's true that some of his talk was the typical messaging we hear from Corporate Open Source PR people (which are usually called “Community Managers”, although Karsten calls himself a “Senior Community Gardener” instead). Nevertheless, I was persuaded that Karsten does seek to educate Red Hat internally to have the right attitude about FLOSS contribution. In particular, he opened with a an illuminating literary analogy (from Chris Grams) about Tom Sawyer manipulating his acquaintances into paying him to do his work. I hadn't seen Chris' article when it was published back in September, and found this (“new to me”) analogy quite compelling. This is precisely the kind of activity that I see happening with problematic copyright assignments. I think the Tom Sawyer analogy fits aptly to that situation, because a contributor first does some work without compensation (the original patch), and then is manipulated even further into giving up something of value (signing away copyrights for nothing in return) for the mere honor of being able to do someone else's work. It was no surprised that after Karsten's keynote, jokes abounded in the SCALE 8x hallways all weekend that we should nickname Canonical's new COO, Matt Asay, the “Tom Sawyer of Open Source”. I am sure Red Hat will be happy that their keynote inspired some anti-Canonical jokes.
- Another Red Hat employee (who is also my good friend and former cow-orker), Richard Fontana, also gave an excellent talk that many missed, as it was scheduled in the very final session slot. Fontana put forward more details about his theory of the “Lex Mercatoria” of FLOSS and how it works in resolving licensing conflicts and incompatibility inside the community. He contrasted it specifically against the kinds of disputes that happen in normal GPL violations, which are primarily perpetrated by those outside the FLOSS world). I agreed with Fontana's conclusions, but his argument seemed to assume that these in-community licensing issues were destabilizing. I asked him about this, pointing out that the community is really good at solving these issues before they destabilize anything. Fontana agreed that they do get easily resolved, and revised his point to say that the main problem is that distribution projects (like Debian and Fedora) hold the majority of responsibility for resolving these issues, and that upstreams need to take more responsibility on this. (BTW, Karsten was also in the audience for Fontana's talk, has written a more detailed blog post about it.) Fontana noted to me after his talk that he thought I wasn't paying attention, as I was using my Android phone a lot during the talk. I was actually dent'ing various points from his talk. I realized when Fontana expressed this concern that perhaps we as speakers have to change our views about what it means when people seem focused on computing devices during a talk. (I probably would have thought the same as Fontana in the situation.) The online conversation during a talk is a useful part of the interaction. Stormy Peters even once suggested before a talk at Linux World that we should have a way to put dents up on the screen as people comment during a talk. I may actually try to find a way to do this next time I give a talk.
- I also
saw Brian
Aker's presentation about
Drizzle, which is
a fork of the MySQL codebase that he began inside Sun and now
maintains further (having left Sun before the Oracle merger
completed). I was impressed to see how much Drizzle has grown in just
a few years, and how big its user base is. (Being a database
developer, Brian thinks user numbers in the tens of thousands
is
just a start
, but there are many FLOSS projects that would be elated even to max out at tens of thousands users. While I admire his goals of larger user bases, I think they've already accomplished a lot.) I talked with Brian for an hour after his talk all about the GPL and the danger of single-copyright-held business models. He's avoided this for Drizzle, and it sounds like none of the consulting companies spouting up around the user community has too much power over the project. (Brian also blogged a summary of some of the points in the discussion we had.) - Because it directly time-conflicted Brian's talk, I missed my friend and colleague's Karen Sandler's talk about trademarks, but I hear it went well. Karen told me not to attend anyway since she said I already knew everything it contained, and that she would have went to Brian's talk too if my talk was against it. She did however make a brief appearance at my talk, so I feel bad my post-talk chat with Brian made it impossible for me to do the same for her talk.
- I spoke extensively with Matt Kraai in the Debian booth. It was great to meet Matt for the first time, as he had previously volunteered on the Free Software Directory project when I was at FSF, and he's also contributed a lot of development effort to BusyBox. It's always strange but great to finally meet someone in person you've occasionally been in touch with for nearly a decade online.
- Don Armstrong was also in the Debian booth. I got to know Don when we served on one of the GPLv3 discussion committees together, and I hadn't been in touch with him regularly since the GPLv3 process ended. He's continuing to do massive amounts of volunteer work for Debian, including being in charge of the bug tracking system! I asked him for some ideas in how to help Debian more, and he immediately mentioned the Debian/GNOME Bug Weekend coming up this weekend. I'm planning to get involved this weekend, and I hope others will too.
- Finally, I had a number of important meetings with lots of people in the FLOSS world, such as Tarus Balog, Michael Dexter, Bob Gobeille, Deb Nicholson, Rob Savoye and Randal Schwartz. Ok, enough name-dropping. (BTW, Tarus has written about his trip as well, and mentioned our ongoing copyright assignment debate. Tarus argues that he can do non-promise copyright assignment in OpenNMS and still avoid the normal Open Core shareware-like outcomes, which he dubs “fauxpen source” for “fake open source”. Time will tell.)
SCALE is really the gold standard of community-run, local FLOSS conferences. It is the inspiration for many of the other regional events such as OLF, SELF, and the like. A major benefit of these regional events is that while they draw speakers from all over the country, the average attendee is a local who usually cannot travel to the better-known events like OSCON.