FOWA Day 2

The second day at FOWA brought some interesting speakers, more suitable in my opinion to the many web people present at the event. This day looked much better on paper as far as content and detail were concerned and I was keen to hear some of those speakers, there was also some nice suprises in store later in the day.

The morning kicked of with Mark Anders from Adobe systems, where he explained to us what Apollo was and what it provided. I had also had a chance to speak to Andy on day 1 in the foyer and so had a idea what to expect. Personally I can't see us making the choice to go down the Apollo route for a number of reasons, primarily because it is proprietary and we really do not want to get stuck into a single supplier for our platform, we have experience of that sort of thing in the past and it gets ugly if things go wrong. I don't mean to demean the technology however, it has some great killer features :

1) It is based on Actionscript (V3) which is basically very similar to the new javascript 2.0 (ECMAscript). This is a massive advantage and we all seem to agree that JS2.0 is likely to be the NBL
2) You can create desktop based webapps sharing much of the development features of web apps. these also look really smart due to Apollo's flash heritage.
3) Mixed markup and script based construction, this is good for those already familiar with web development.

On the minus side however would stress:

1) It requires you to download the apollo runtime, some say that this isn't an issue because it is in the Flash 9 plugin, true but thats not for all platforms and certainly not all devices. We have to work to minimum lowest common denominator technologies such as HTM/XHTML and javascript/ajax in order to ensure our projects operate on the target platforms.
2) There is a debate about the invisibility of content to search engines and lack of transparency of its workings after deployment (deployment being binary etc..)
3) It is proprietary to Adobe it means vendor lock in, and I personally don't trust any single vendor to defend our interests. On the positive side however Adobe as part of this has donated their amazing JIT compiler to Mozilla (tamarin) with will certainly help future javascript performance in Firefox.

I also suggested that Apollo was likely to be competing for mind space with the widget platforms such as Google's gadgets, Microsoft's gadgets and the rest, suprisingly Andy seemed to have a more traditional view of what Apollo was and that it wasn't really competing with those things, personally I think all of these things are heady for an almighty fight of our attention,. but we will see..

In addition to the Apollo pitch in the speaker line up we got a surprise 10 minute presentation from Zimki. The presentation itself was brilliant and very entertaining, Simon Wardley is a natural and his tag cloud is somewhat unusual (He has a thing for ducks!). The technology itself however is an interesting take and I later spoke to them in the Foyer. First of all it uses javascript front and back (Next Big Lanuguage anyone?), they actually use spider monkey to script the back-end. Persistence is build into the framework, but heres the new twist, they operate the infrastructure in an amazonian manner. not only do you buy into the framework but also the infrastructure. You get charged based on bandwitdh and cycles consumed, I say charged but actually they deduct from your allowance or credits which to begin with can be free. Anyhow their take was more interesting to Folknology than Adobe's but I worry if Zimki can pick up the critical mass when they are up against so many big players (as well as the resistance of many web devs to backend javascript). I was also concerned about lock in and Steve Purkis assured me that they were open sourcing the technology as well as opening the market place for others to provide competitive infrastructure, thus it is well worth checking these guys out. It is also worth noting that that connection thing kicked in again, as it turns out Zimki uses TrimPath's jasvascript templating engine, remember TrimPath were one of the first javascript based spreadsheeting developer Steve Yen (NumSum anyone), small world huh..

On the Identity front Simon Willison spoke about OpenId, as usual he did a great job of explaining identity and the solution as usual. Simon has done so much for identity he really deserves recognition, especially around OpenId, it is particularly appropriate at this time given recent announcements of support from major providers ; Microsoft, Yahoo, AOL and even Digg announced support at FOWA. At last the digital identity divide and chaos is finding its common ground in OpenId and Simon must take a great deal of the credit. We are now just waiting for Google and the rest to join in with this sea change.

Chris Wilson group program manager for the Internet Explorer team also spoke. As well as enumerating what IE 7 had to offer he also explained why it took so long for Microsoft to do something with the browser since it's IE6 release. They were not after all completely idle, the service pack/ security update involve vast re-writes and security testing of the browser code. Personally I don't entirely accept that excuse, but he did apologise to all at the event that they had taken so long to do some thing for the community. He also indicated with Vigor that they were back and there was going to be a lot more to come. I managed to grab 15 mins or so with Chris later at the lunch break and got to ask a couple of questions.

First I asked if they would be implementing Canvas element in IE8, his reply was 'which canvas standard' which to mean seemed to be a little dig at canvas itself, Microsoft not having been directly involved in its development (see my link chris for the spec it pretty clear to me). So although he did not answer the question directly, he also did not rule out its inclusion in future releases. My next questions 'Was will you implement Javascript v2 into the next or subsequent IE releases?'. As you can guess I didn't exactly get a straight answer. Chris seemed to indicate that Microsoft did not entirely agree with what is currently being proposed by Brendan et al around Javascript 2.0. It therefore appears that as well as Microsoft getting itself back into the browser development game, it may well be pursuing it's old tricks of placing it's own agendas ahead of that of standardisation regarding new features such as canvas & JS. This could mean once again seeing diverging features in browser support across the different providers. At the same time Chris and his team is advocating standards support as an important design goal of IE moving forward, although some of this may also conflict with Microsoft historical 'embed and embrace'. The other thing Chris stressed is that their first principle is that any changes must not 'break the web', I would personally use the term backwards compatible with other versions of IE rather than laying down older IE quirkyness as a 'the web', but again this is just classic case of Microsoft posturing. Finally I asked Chris whether the browser was actually the future of the web, the question puzzled him as expected so I went on to explain; 'Well what I mean is..' and spoke about how many of the surrounding features of a browser were becoming redundant :

1) The address/url bar, I (and many others) rarely use it now with search engines combined with the links embeded in my feedreader/emai etc.., I just click links to find new stuff
2) Bookmarks, with the rise of Delicious and others the internal bookmarking feature of browsers is rapidly becoming redundant. He also indicated that he hadn't really used Delicious explaining his needs to be more private than public, so I suggested he look at Magnolia as an alternative. I also got the feeling maybe they already had things up there sleeve in this department, but he wasn't letting on.
3) Back and forward buttons, these are becoming much less frequently used, with the proliferation of ajax these buttons are often surplus to requirement, in many cases they actually break for ajax web apps.
4) The emergence of widgets and platforms like apollo make the browser and its page model even less relevant moving forward.

Chris pretty much disagreed (not a suprise given his role) and we left it at that. I must admit he has one of the toughest jobs out there, He has to serve the whole web community at the same time as Microsoft's own agendas and do so while not breaking the stuff they have put out in the past, I for one would not want to be in Chris's shoes!

One of the other major guest speakers was Rasmus Lerdorf from yahoo and of PHP fame. Although we at Folknology don't current use PHP, it was PHP that actually brought me back into web development many years ago and as such it was personally interesting to listen to Rasmus share with us the history of PHP's creation.

I had been waiting to hear from the product manager of Google Docs + Spreadsheets Jonathan Rochelle, as you know how much Folknology's interest lie in and around these efforts by google and others. His presentation was timely given the official release on thursday of their premeir web apps package. What suprised me was that they didn't use FOWA as a platform for that announcement, they did of course cover Techmeme for the day, irrespective of where they announced it. Although I didn't get chance to talk Jonathan I will be following up on this in the near future and his comments where very interesting.

Tariq Krim of Netvibes announced their 'Universal widget' (I notice there might be an issue with Sniperoo here) this means in theory one can develop a netvibes widget that will also run in dashboard, and google gadgets which is a really cool move and is likely to make widgets much more appealing from a development point of view. At Folknology we have been working on widget stuff as a way to modularize business components and this is a welcome announcement. It will be interesting to see how their implementation performs and what caveats may be involved, again watch this space..

There were more speakers and other topics, one of my favourites was one of the ten minute spots by Soocial, this is a new take on contact management and syncing another UK startup worth checking out. And the now famous Moo guys also got up on stage to explain how they put together their innovative business cards.

Technorati Tags: ajax, developers, FOWA2007, web2.0