Interop Event, April 16-17, 2007


See [WWW]DeWitt Clinton's announcement.

I'll send around the full details regarding logistics as soon as I know the exact building and rooms, but we have space confirmed for:

The rooms will have public wifi and projectors, and we can definitely fit up to 20. Food, spaceships, and dinosaurs will be provided.

Update by DeWitt on 2007-04-06:

I just confirmed the food arrangements. We'll have a lunch set up for Monday as people arrive, and then a dinner brought in for the group Monday evening. We'll have the space as late as we'd like, so if people want a game of Werewolf...

Tuesday we'll have breakfast in the meeting room, and for lunch I thought it might be nice to reserve a few tables at Charlies, one of our larger cafes. An afternoon snack will be arranged as well.

No plans for Tuesday evening; perhaps someone else wants to plan something for after we finish?

Update by DeWitt on 2007-04-09:

See also [WWW]Directions To Google Campus.

Update by DeWitt on 2007-04-13, copied from an email sent to atom-protocol:

Who's Invited

Anyone with working APP client or server code.

Proposed Agenda

See [WWW]Tim Bray's proposal.

Here's how I would like to run it:

  1. W/W/W/W. Everyone gets a few minutes stand-up time to do the who/ what/why/where of their APP client and/or server implementation. No slides! Since I imagine that some people will drift in late, whenever someone arrives, they do the w/w/w/w as soon as they've put their bag down. Let's try to capture everyone's W/W/W/W essentials somewhere on Sam's wiki.

  2. Interop! We will build a grid with the clients across the top and the servers down the side (because there are more servers than clients), then we'll try to interop all the cells. The grid is at InteropGrid.

  3. Issue-gathering. This is the most important activity. Let's do this on the wiki in real-time, so that people on IRC can play. An APP Interoperability Issue can be raised by anyone during their initial w/w/w/w preso, and I assume that lots will turn up during the grid-filling. I do not suppose that we will achieve 100% consensus on what's an issue and what isn't, nor on how to fix any obvious ones that turn up. That's OK; the important thing is to gather the information. I see two forms of issues: "XXX in the spec has been shown to cause interop problems" and "XXX in the spec is sufficiently badly written that implementors have been shown to draw incompatible conclusions about what it says."

Remember, per IETF rules and common sense, the people at the interop session are *not* empowered to make changes to the protocol. Only the WG, via the mailing list, can do that.


I don't suppose there will be time for a rousing game of werewolf?! Can that be added to the agenda? :) [ByrneReese]


For folks who cannot attend in person, several of us will be hanging out on #atom during the interop.


Please include your name, your affiliation if you feel like it, and identify the software you're bringing, and whether it functions as an client, server, or both.

Note, if you're adding your name after 11:00 AM PDT today (Friday, April 13, 2007) please contact DeWitt via email as well. Thanks!