We had this thing at Google. Heated toilet seats. Catered food. All you can drink soda, etc. Very decadent, I see now why Google has done so well at hiring people.

Our facilitator, Jon Stahl, did a great job of driving the sessions. He looked like he was about to pass out the last time I saw him about an hour ago. It must have been exhausting.

The PSPS itself was a series of organization and brainstorming exercises designed to steer people towards a consensus about "what needs to be done" in the future. Jon had everybody move around and put up notes on big pieces of white paper and run around with stickers and mark issues they agreed with, and some other exercises. Virtually no one used computers the entire time (at least relative to the audience, some folks hauled out laptops for brief periods to check email and whatnot).

The first two days were about framing the issues. The last day was largely about attempting to define very specific tasks and to find people willing to do them, or at least to drive other people to do them ("champions"). I think maybe 20 or 30 tasks got created, and the majority of them were claimed by someone.

The audience was relatively balanced, with probably 50% of the people there being integrators (lots of consultants) or internal deployers, and the other 50% being "Plone core" folks (also a good number of consultants in that camp too), who are in charge of actually creating the framework upon which new features might be based.

I think this was a useful exercise. Although I personally enjoyed the first two days of exercises, I probably would have been better off just attending the last day, as I intended to "champion" the things that I care about anyway and I could have just said "yes I'll do that" on the last day and discussed it with folks afterwards.

We are now supposed to create some sort of proposal about creating a zc.buildout that installs Repoze and Plone together. Agendaless is also on the hook to deliver a proposal for a stripped-down Plone theme for use with Deliverance (or another transform engine).

I accidentally hit a nerve today when I suggested as an example of stratifying the things that currently make up the amorphous "Plone core" that we make Archetypes optional. There was lots of strong disagreement on that one from almost everyone (each task was "voted on" by coloring in of dots on a range of "strongly agree" to "strongly disagree"). I suspect, however, that many folks may have misunderstood "make Archetypes optional" as "destroy Archetypes altogether." And perhaps there has been some talk about that in the past that colored people's expectations about what the task actually was.

There really wasn't enough time to clarify the intent at the summit in person but I believe it'd be possible to preserve backwards compatibility but allow for an Archetypes-less Plone deployment (which is more or less what I'm after; I'd actually like more things than Archetypes to be optional). But I did get into some side conversations with folks on the "framework team" which leads me to believe that the idea of stratifying Plone isn't as stillborn as the response to that proposal suggested. As long as we're willing to supply some of the elbow grease, I think it might be acceptable and desirable to the folks to whom it would matter: integrators would basically just not need to be aware of the existence of an Archetypes-less Plone, and the core folks would just create dependencies from "Plone the application" to "the Plone core" and "Archetypes".

As always, it was a lot of fun to hang out with the folks that showed up at the summit. Tip: hide the beer when the Plone folks show up at your offices.

Oh yeah, and I forgot to mention, it was great to see the winner of the GHOP Plone project (who was 14). He rocked.