I'm sitting in the computer room at the Radisson RTP after Science Online '09 has wound down, and most of the attendees have left -- though I'm looking forward to dinner with a few fellow stragglers this evening.

Many thanks are due Anton, Bora, David and their various helpers, sponsors and assorted minions for running another wonderful conference. I was happier'n a pig in a puddle with this year's program, as I was able to attend an Open Something (or related) session in almost every slot. There's nothing quite like indulging an obsession with a crowd of like minds, especially when there remains enough diversity of opinion to (mostly) avoid the echo chamber effect. There was only one thing I can point to that wasn't essentially perfect, which is that the web connection, wifi or wire, was flaky and slow quite a lot of the time. That observation must be taken in context, though: although everyone commented, no one complained. It just isn't that sort of gathering.

My session with Björn went well (OK, I can't really judge that -- but I had fun!) -- although it would have gone better if I'd shut up sooner. Having not been to an unconference before, I wasn't strict enough with my introductory blurb and took up time that would have been better spent on the ensuing discussion, which was just terrific. I'll know next time -- and Björn was careful to learn from my mistake, limiting himself to a quick intro for his session with Peter Binfield and obstinately driving the discussion away from echo chamber territory, challenging the participants to come up with new ideas and ways forward. (If you're interested in the Impact Factor question -- that is, metrics and measurement in science -- there's a collaborative bibliography underway in a Google Doc here. I'll make it publicly editable as soon as I figure out how; in the meantime email me if you want an invite to collaborate.)

I definitely prefer the unconference format to a traditional lecture-style conference. When there is a subject that needs more intensive coverage by the speaker(s), the flexible format easily accomodates that -- for instance, John Wilbanks' talk on the semantic web was of necessity about half informal lecture and half rowdy discussion, simply because it's a complex topic about which few of us knew very much. (Before John got through, I mean, since it was an informative and inspiring look at the technology which will probably underpin the next truly radical leap forward in scientific capability.)

As Eva Amsen and Henry Gee both observed, the line between people I know online and people I've met in meatspace is getting very blurry these days. I was nonetheless pleased to meet Eva and Henry f2f for the first time, and also Björn, Peter and John, Cameron Neylon (more like "nylon" than "nay-lon"!), Victor Henning, Martin Fenner and a dozen others to whom I apologize for being too tired to remember you right now! I was of course no less happy to catch up with old friends, repeat offenders like me who were also at the 2007 and 2008 events.

And now it's too late for me to get a nap before dinner, so I think I'll go see if a shower will wake me up instead. More later as I process the many new ideas and insights I collected in the course of two very enjoyable days.

Science Online '09 is less than a week away, and I'm going to be co-moderating an unconference session with Björn Brembs, the theme of which is "Open Access publishing: present and future".

Björn has already put some notes up on the wiki, and there's an interesting contribution from Antony Williams of Chemspider. As both Björn's and Antony's notes make clear, we think the future of Open Access (indeed, all scholarly) publishing will feature prominently the long-overdue death of the Impact Factor. In fact, audience willing, we plan to use some of this session as a sort of preface for Björn's Sunday session with Peter Binfield, which is titled "Reputation, authority and incentives. Or: How to get rid of the Impact Factor".

It's difficult to overstate the extent to which that single figure has come to dominate scholarly and administrative decision making: where to publish, who to fund or promote, which candidate to hire, and so on. It's also difficult to overstate how bad an idea it is to put so much weight on a single journal-level metric derived by undislosed calculations and decisions from a proprietary database.

But that's the future of publishing, about which much more from Björn and Peter. Regarding the past, I thought I would do a five-minute definition-plus-potted-history, cribbed almost entirely from my earlier talk and Peter Suber's timeline.

That leaves us with the present, and in the spirit of an unconference about science online, I thought I'd simply ask the audience: what do you want to know about Open Access?

There are two things I must clarify. Firstly, by audience I mean both online and on the day: if you're there, you can ask in person, but if you're not going to the meatspace conference you are welcome to ask your question here, on the conference wiki, or by email to me, at any time. Secondly, I'm not claiming I'll have the answer ready to hand -- but OA and related Open ideas are pretty much an obession with me my hobby these days, and if you have a question I can't answer I'll be sure to find out and get back to you. (In addition, the conference will be packed with OA experts and I have no hesitation in bothering them for answers!)

So: what do you want to know about Open Access?