So here's my live blogging during FOWA London 2007.
I'm late for the first session but I find a space near the back and listen to Mike Arrington. Geez I feel like asking a question about TechCrunch UK. Specially with him going on about the fact he was born here or something. Mike Arrington also talks a lot about Adobe Apollo and mentions his own conference. Suggests that there big differences in culture between UK and America . Considers the fact that Asia might be the place to go for in the future. On removing friction, he considers the difference between Yahoo! and Gmail. Someone hits Mike about his love for Apollo, thank goodness I was about to puke.
Edwin Aoki brings it back to communities, starting off with Tim Berners Lee's vision for the internet. Considers the community features in the leading applications like ebay, amazon, etc. Are Webring and Dmoz the mashups of Web 1.0? Edwin certainly thinks so. Proposes some of the trends to be Disaggreagation and Syndication. Mentions community going mobile through helio phone for example. Secondlife blur the virtual and online worlds. Thinks about our responsibility as technologies not only with security/privicy, spam, social effects but also decentralisation and trust models. AOL now supports OpenID. Applications need to be accessible otherwise we won't ever bridge the digital divide. Then talks about Balance.Power but ease of use, social benefits and commercial interests, Offline and online interactions.
Tara Hunt just gave a fantastic presentation with tons of information which shes going to stick on Slideshare.net soon. The best part was her analysis of Flickr, Threadless, Twitter, BarCamp and another. Well worth looking through.
Simon Wardley gave a good fun presentation about ducks and Zimki which is going open source soon. I love Last.fm, but their presentation certainly sound a little dull for my liking. Next time guys, a little spark would be cool.
Werner Vogels, although very hard to understand because of his accent but made a fantatstic case for Amazon S3 storage. Showing pictures of a datacentre failure and the percentage of hard drive failures after a few years was scary enough. However he made solid points about scaling and uptime. Then they showed off a few figures pre-S3 and post-S3. Elastic compute cloud also recieved simular treatment with Powerset.com and a couple others. Mechanical Turk also came up but only at the end. And Steve from Openstreetmap.org, rightly pointed out that the Turk doesn't work in the UK. Vogels points out that its down to labor laws here. Which I think speaks volumes about the States.
Brandley Horowitz brought the afternoon back into reality with a talk about users. He showed the good and bad sides of user-generated content and then talked about the organic process of interestingness – or turning users into editors. He pointed out it was retroactive and less susceptible to gaming or agendas. Turning users into taggers, highlighted the fact that its quick and brings the level of entry right down. He also pointed out the machine tags (or RDF triples) then quickly moved on to zone tagging which came from yahoo? In the same vein about lower the barrier to entry, he pointed at flickr clustering. Turning his head to the concept of neighboring he also mentioned mybloglog.com. At long last he talks about Yahoo Pipes. RSS is mashup for the masses? Horowitz, makes a good comparison about pipelines as sampling vs synthesizing.
Then we were into the sponsored talks which included QuotationsBook.com, Soocial.com and BT. Soocial.com, left me thinking whats the difference between it and Plaxo.com (bar the cool style). BT showed off BTconnect.com, which I assume was also showed off at BarCampLondon2 but I missed it. BTconnect.com I have to say looks pretty interested although its not live yet (see screenshot).
The last session of the day was by Kevin Rose of Digg.com. He ran through a lot of digg's past and announced that OpenID was coming soon and that a Flashtoolkit will be also be available soon. The toolkit would allow people to take a subset of digg and show it somewhere else. For example allow a bunch of friends to show others their own number ones, etc. He also talked about exporting attention, I suggested that the digg team should consider apml.org (should have mentioned attention rdf too).
The first day had its ups and downs. The biggest down was certainly the wireless which didn't work because of some broadband connection problems. This wouldn't be so bad but the speakers were also on the same connection, so some of the demos couldn't be shown live. Ryan Carson was rightly so, pissed off about it all and made his feelings known. On the upside, Tara Hunt, Werner Vogels, Bradley Horowitz and Kevin Rose were very good and gave me lots to think about. I'm sure Day 2 will be good too. Hey and free parking underground for Scooters was a huge bonus for myself.
Its Adobe on board talking about Flex and Apollo. He's revealing the code behind Flex aka mxml. I'm still having a hard time understanding what the real difference between flex and xaml. In places they could have used SVG like syntax, they didn't, where they could have used xpaths they didn't. Thankfully they used CSS at least. Now he's showing off picnic.com which I've seen before. It seems Flashplayer 9 has been redone and now supports ECMAscript for XML (E4X). A guy from Scrybe shows how much faster Actionscript 3 is compared to actionscript 2, it looks like 12x faster across the board. AVM2 source was donated to Mozilla (Tamarin), it looks like it will make its way into Firefox 4.0. Now finally with only 3mins left, Apollo. Maptastic and a ebay application on the desktop, oh wow – i'm so impressed. Not!
Chris Wilison from Microsoft is now on stage, expressing why he felt thing went wrong in 2001. Hacking also became lucrative is one of his slides, but generally he talks about lack of vision of web developers trying to build everything in a their own silo. 2005 saw the rebirth of the lower s semantic web. RSS, Microformats, tagging and other browsers arise. Now the IE7 pitch, but some details on the integrated RSS platform. And back to security… and back to IE7. Wasn't this talk meant to be about the future of the browser? Talk about quirks vs strict mode and running multiple IE's side by side. Microsoft are releasing a virual PC image which has IE6 and IE6 sp2 for free every few months. Now a section about Microsoft Expression designer, ASP.net Ajax and WPF/E. WPF/E is like WPF but is a subset of XAML and works like a Mozilla plugin at 1.1meg.
I'm left wondering if the new york times get it at all after hearing Khoi Vinh talking? There still building lots of stuff internally and not reaching out to the rest of the web – silo building?
Simon Willison did a great job outlining the problem with a non-single sign on eco-system. The slide for Microsoft and Six a Part did get a laugh when asked if you trust these people. The rest of the presentation was pretty fluid (as is Simon when usually talking). Intellengence on the edge is actually a very good model for why OpenID makes sense. Simon then outlined a load of scenarios which could be done now we're using OpenID. Simple examples using OpenID and XFN and hCard microformats. Plus more complex examples using Social Whitelists using OpenIDs. On the whats wrong with OpenID, he mentions phishing, privicy and what happens if your openID provider goes down? He mentions Microsofts cardspace as possible solution, but generally thats a browser issue which openID can not solve. The others can be solved by using adding multiple OpenIDs to a ID and using multiple providers.
I attended the Panel Debate in the Council Chamber. The debate was what could be learned from the Americans when it comes to startups in Europe. It wasn't a very good debate at all (too many panel members I think) and the setting was kind of odd, Anyway some of the highlights or lowlights depending on where you sit included, Mike Arrington deciding it would be best for the UK market if the BBC was shutdown (I caught the whole rant on camera). Another BBC employee jumped in and made the point about the public service test but Mike Arrington was having none of it. His point about the Office was simply stupid. The reason why there is no more seasons of the office is because we know when to stop (Mike should take a note out of that book) and it looks like we may have sold the rights to the American version. Distoration in the market my ass arrington.Later in the debate Chris Messina made the point about define goals and success. Mike being the one sided guy which I now think he might be, said to Chris, what do you mean? Thats pretty simple – How much money does the startup make. Chris I believe didn't say anything but shruge his sholders, knowing this was simply a arguement which couldn't be won with such a one sided guy. Maybe I was wrong but its certainly what I felt. The rest of the debate was pretty un-eventful except Arrington grinning everytime Mike Bucher avoided my question (which wasn't going to be about the BBC actually). I can't help but wonder if Mike is more upset because he didn't get the full details of CBBC world?
I took a few sessions off to catch up with Ben Metcalfe and others after the panel debate.
Rasmus, compared opensource projects to the current crop of web 2.0 sites. System that harness network effects and get better the more people use them in a way that to there own self-interest. He wonders what PHP, any Sourceforge project, Wikipedia and Flickr would be like without partcipitation from its users
Tariq from Netvibes just quickly announced a universal widget API (UWA) for Netvibes and then came back on and announced OpenID support coming soon. Mike Arrington blogged it 30mins before the annoucement.
Moo cards guys talk about the differences between them and there rivals Qoop who launched there cards 2 weeks before Moo.com. Made a good point about marketing. They think about and pay for marketing before the product which is very cool. Moo is very much the company I think of when saying the word boutique. They hand check every single pack and send it all via Royal Mail in the morning.