It’s been about two week since I went to FOWA2007 in London, and I’ve been meaning to write it up for a while. It was an… interesting experience, and I’ve been trying to find a way to frame why that is. Speeding on a train back from the provinces to London seemed a good way to concentrate my thoughts.

Last year’s FOWA was great: a one-day conference for

A straight-up announcement post here: I’m going to be at Future of Web Apps in Kensington for the next two days. I’m reasonably identifiable (redhead, sideburns, goatee) and I’ll probably have a neck-badge that says “Developer” on it.

I think it’ll be even better than last year’s, in part because it’s spread over two days, and there are more gaps for socialising. So if you don’t see me in the hall, grab me outside it. I won’t be around on the Tuesday night due to prior commitments (The Long Blondes at the Astoria, hurrah) but will definitely be up for much in-pub ranting on the Wednesday. So: do say hello if you’re going to be there, and maybe we can bat some ideas around.

d.Construct 2006 was well over a month ago, now, but I’ve simply been so busy since then that I just haven’t had a moment to write up my experiences.

d.construct is a “grass-roots” web conference in Brighton, run by the nice chaps from Clearleft. It only lasts a day, and, at £75+VAT, it’s insanely good value. I don’t want that last fact to go unnoticed. It was also a great lineup of speakers, including Thomas Vanderwal, Jeremy Keith, Simon Willison, Paul Hammond, Jeff Barr, and Jeff Veen. £88 just to hear all of them speak is, any way you look at it, a very good deal. And then, of course, there’s the all the networking and the chat and the the pub, with four-hundred web-types who descended on Brighton, which is often the highlight of any conference.

When I filled in my feedback form, I realised I didn’t have time to write what I really wanted to say. So I put down the improvements that were the easiest to fix (and, it seems, the most universally agreed upon): more legroom, more power sockets.

Beyond that, what I had to say wasn’t so appropriate for a feedback form, as I’m not quite show how to improve upon my criticisms. But I still think it’s worth putting them down in writing.

At the heart of d.construct is a very good event. I enjoyed this year’s event a lot. But I think there are some areas that could perhaps be improved – or at least addressed – in future years.

First of all, the format. d.construct is much like the Carson Systems’ Future of Web Apps events – it’s a series of speakers talking, one at a time, one after another, in a big auditorium. It’s not stranded or streamed. At the same time… it’s a bit intimidating in that people tend not to want to break out of it. The moment you introduce two or more strands, attendees begin to realise that perhaps they don’t want to go to either talk, and so they decamp to the back room to mingle and chat. And d.construct had a great “back room”, with power, refreshments, and tons of space. As it was… everyone piled into the hall for every session, regardless of whether or not it interested them. Quite often, I saw a number of people ignoring the talk to concentrate on their iPhoto session, and then dumping their latest pictures onto Flickr. I’m sure they could have done that “out back”, or, more likely, found people to talk about things they were interested in. I cut Aral Balkan’s Flash talk – because, whilst I’m a client-side developer, I have zero interest in Flash – and was disappointed to only find about six people hanging around out back, and one or two in the pub around the corner. The chat I had during the session I skipped was great.

I also found the angle a little curious. d.construct, whilst pitching itself as a “web application and Web 2.0 conference”, is very much a web conference coming from the front end. I was disappointed when Jeremy came over very apologetic that he was even showing code at all during his presentation – and it wasn’t really very complex at all. In the end, no matter how many wireframes and PSDs are drawn, websites only exist in code, and I get frustrated when people have to shy away from that kind of expression.

Similarly, I was a bit frustrated that he managed to talk about REST without mentioning HTTP verbs (GET/POST/PUT/DELETE), as they’re as important to the RESTian concept as the URL structure. But I can understand – given what appeared to be the conference’s target audience – why this was the case. It also meant that Jeff Barr got a slightly raw deal – he was going first, and he was easily the most “corporate” of the speakers, but I thought his talk was a great balance of explaining some of the awesome work Amazon are currently doing, and demonstrating how real users have made use of it. But, if you don’t get the importance of S3 or especially EC2 (which is, in some ways, revolutionary), it just sounds like corporate buzzwordiness.

And I was also frustrated by this because the more general-purpose or front-end talks – Thomas Vanderwal’s IA session on tagging (a pretty complete history and explanation of tagging) and Jeff Veen’s phenomenal closing session – were both reasonably high-level (in terms of what they’re discussing) and well received. And for good reason – Veen was absolute dynamite.

Maybe it’s because Clearleft themselves are, primarily, a front-end consultancy, but I don’t want to trivialise things that far or make such bold assumptions. It might just be that it’s the best common ground over which to bring people in the UK together. But I think there’s interesting things about development to be expressed to general audiences, especially given that buzzword-of-the-moment AJAX is all about the point where the front- and back-ends join. And I’m concerned that whilst Web 2.0 (however you understand that phrase) advocates a more holistic, interconnected web, the design and build process is becoming ever-less holistic.

But I don’t want to be a complete downer. Like I said, it was a great-value conference, at which I met many interesting people, and the speakers were all excellent. Maybe next year it will move to multiple tracks; maybe it’ll broaden its scope. Either way, I’m still going to go; it’s a great mixer of an event, and it’s nice to go to something that’s not in London for once. My congratulations to Richard, Andy, and Jeremy on its success; I hope my comments aren’t taken too hard nor too negatively. And I hope you can see why I didn’t quite have time or space to condense them on the feedback form.

Update: One thing I forgot to mention, that’s surely worth a big plus point, is that the “female quota” at this conference was very high. I’m sure that sounds awfully patronising, but given the amount of coverage of “why women won’t go to conferences” (see Mike Kuniavsky on this), it was interesting to note a percentage well into the double figures – I’d say about 20%+, at least. Whether this is because of the front-end focus – or, rather, the less-threatening, less-technical focus – I don’t know, and again, don’t want to trivialise. But bonus marks for this, for sure.)

Exciting news of the day: I’m going to be speaking at the Reboot conference in Denmark this summer.

Looking forward to it lots, though obviously I need to start working on the talk soon. Still, following ETech, I don’t think I’m going to let myself get quite so stressed.

What follows is the rough pitch I outlined in an email (written, as ever, in conference-abstract-ese); final version may vary, obviously, but I think it conveys the gist of what I want to discuss:

“Telling stories – what social software can learn from Homer, Dickens, and Marvel Comics”
or: “Social software as serial narrative”

Social software is playful, and much ludic analysis has been made of it. But what of narrative analysis? After all, we use this software to tell the grand serial narrative of our lives – cataloguing them via Flickr, journalling them (in whatever form) via our blogs. And then consider the wealth of parallel narratives many people have – a delicious account, a Flickr account, multiple blogs, LiveJournals, MySpace accounts, some contradictory, some anonymous, some fictional, some fact. This isn’t a bug, it’s a feature; we should encourage parallel storytelling, encourage the formation of personas, and make the interaction between these different platforms as complete as it needs to be to support this.

We should design our software around these narrative impulses. In ten, twenty, thirty years time, for good or ill, we will want to look back at the stories we told – for they are part of our greater story. So as well as encouraging parallel narrative, we need to consider how best to support the long ongoing narrative that we weave.

So: let’s look at how, over history, serial narrative has been told, distributed, and retrospectively altered, and see what it tells us about the tools we build to tell our stories.

(this may change at some point in the future; I’m still revising this stuff but thought I may as well put it out there).

Etech06 is going really, really well for me (so far): lots of things emerging in my head, at the least. One thing that’s coming out of quite a bit is a discussion of feedback loops.

Feedback loops are really, really important. Amy Jo Kim touched on feedback as an essential part of ludic design – without feedback, play isn’t satisfying (and play is what all early adopters are doing all the time). Feeedback is what generates challenge/reward structures in games. Feedback loops are how communities emerge – I do something, you do something back; it’s the implicit social structures Amy Jo mentioned. Derek Powazek is currently talking about the “new community” – and mentions MeasureMap.

And, of course, MeasureMap is all about feedback – I can see when I made posts, and when comments came; I can track popularity. I’m no longer sending blindly into the ether; I’m sending and tracking response.

And once you track response, you can write so much better; you can design so much better; you can act on the feedback and you get a loop. And that loop’s really important – it’s what keep things going. If there’s not a feedback loop, things tail off, fall away.

That’s why Google bought MeasureMap: they had Blogger already. People can post to the web; they can broadcast into the ether. Once you give them MeasureMap, they become successful, effective publishers. When people say “what’s the point of blogging?”, they say it as an outsider – they just think it’s publishing. They don’t know about the stat-tracking, the refining. Once you have a feedback loop – once you can see the influence (or lack of it) that you have… that’s when it all clicks. That’s when you get placed into a significance grid – your posts get located in space, time, relevance, authority, etcetera.

Play is about feedback; games are about feedback; publishing is about feedback. A lot of the stuff this morning about attention: it’s all about invisible, natural feedback – tracking eyes-on-screens. We can track hits; now we need to track attention. And when we can measure it, we can value it, and we can price it.

That’s the attention economy. Placing everything into this feedback loop of value.

If you’re a business guy, you see how you can price, monetize, and securitize that feedback. If you’re a player or a hacker, you see what loops you can join together in a mash-up, or where you can generate new feedback, what new variables you can track. And that’s the “connective tissue” that Derek’s talking about.