All entries for Thursday 04 October 2007
October 04, 2007
Well, the show’s now over and having been to more sessions than I can count, had a bunch of interesting conversations and seen some impressive applications, it’s time to try and wrap up and summarise what I’ve learned or concluded. It’s important to remember, I guess, that this is a vendor show. It’s not like ETech, where nobody’s trying to sell anything; the objective of this show is to persuade people to use Adobe technologies in preference to other tools. So one needs at least a pinch of the cynicism most elegantly expressed by, was it Paxman?, when he famously said “What I have to ask myself all the time, is, Why is this lying bastard lying to me?”. I don’t think anyone lied to me as such, but inconvenient questions (AIR on Linux, anyone?) were somewhat glossed over.
My previous entries have mostly been “live” from a session, so they’ve been short and factual about what I was seeing at the time; now it’s time to reflect a bit more. There are several topics I’ve been thinking about, some based on what I wanted to learn about before I set off, some arising from the sessions I saw and the emphasis that Adobe themselves and other presenters were putting on different products, technologies or applications.
A no-brainer. I thought before I even came to the conference, and I feel even more strongly now, that video (and audio) playback through the Flash player is the right choice for us to make. At the first day’s keynote, Kevin Lynch mentioned the statistic that over 70% of web video is now in Flash (FLV) format (driven largely by Youtube and the other video sharing sites, presumably). The move to support H.264 video just makes it all the more obvious that this is the best way to deliver video compared with Windows (WMV) or Mac (MOV) formats. We may even be approaching the holy grail of one format to rule them all – H264 video playing back in the browser, on consumer devices such as iPods or set-top boxes, on mobile phones, everywhere. There would still be questions and challenges relating to multiple encodings for different bandwidths and resolutions, but it’s a lot easier to contemplate multiple encodings all in the same format than multiple file formats.
Here the news is mixed. I saw some outstandingly rich and attractive applications, and I have invites to a couple of the ones I considered to be amongst the best – Buzzword and SlideRocket – which I look forward to demoing to people when I get back into the office. But, in completely unsurprising news, there were plenty of other applications which were clunky or slow or hard to figure out. Why should Flex be any different, right? And there are behavioural quirks with even the best of the apps; keyboard support is unpredictable, with even apparently basic things such as arrow up and down to scroll a block of text working in some places and some apps, but not others. Ctrl-Plus to make your text bigger (something my aging eyes are coming to depend on more and more) comes for free in HTML pages, but doesn’t work in Flex apps. None of these things are insurmountable, I expect, and I do think that if you were planning to build an app which you wanted to look and feel as much as possible like a desktop app, Flex would certainly be a tool to consider. But I didn’t come away with the feeling that there would be a big win to be had by converting parts of our existing apps into Flex-based interfaces, or doing any of our near-term planned work (video recording / playback / conferencing excepted) in Flex.
Having spent some time time talking to developers working on relatively largescale projects, I’ve come away reassured that there are plenty of people developing in Flex in ways that we would recognise, using Subversion or something similar, writing unit tests and build scripts, sharing development across a team. Should we decide one day that we wanted to make something in Flex, I see no reason why we shouldn’t be able to develop applications in exactly the same way as we currently do in Java.
A big theme of the conference was the usefulness of disconnected working; the idea that by using an appropriately designed AIR app, you could have your data both on the internet for ubiquitous access, but also on your desktop for when you aren’t connected. It’s not just Adobe who are pushing this idea, either; Google Gears and Firefox 3 both do similar things. But what I’m not sure about is how useful this would be for Warwick. The example that was invariably given in the sessions which talked about this is that of a company with a salesforce, who need to able to take their presentations and spreadsheets and whatever out on the road with them so that even when internet access is unavailable or unreliable, they still have everything they need to annoy people with. But are we like that? I would guess that 95% of our people are connected 95% of the time. And when they’re not connected, how much of the data we can make available would be useful to them? I’m not saying that we shouldn’t make data available offline where it’s quick and easy for us to do so (and we do already, I guess, with Pages-to-go). But I don’t know whether it’s as big a win for us as it might be for other organisations.
A theme which came out quite strongly across a number of sessions was the importance of your application playing nice with other peoples’ applications. Plenty of people talked about how it would be possible to, say, send your data to everyone on your SalesForce contacts list, or get notifications on your Google home page or whatever. It’s not an especially Flash-focussed question, but I do think it’s an important issue which we need to keep in mind. We do a little bit of it already, with SiteBuilder news and calendars being viewable within Google. But looking beyond just web-dev and thinking about all the tools and data which IT Services manages, there’s a long way to go: students can’t get their timetables into Google or on to their phone; staff can’t expose their Exchange calendars to anyone except other staff, and so on. It’ll be interesting to see whether we get pushed by increasing demand from staff and students for more interoperabiliity, or whether we’re enough of a closed eco-system that this won’t happen to us as much as it seems to be happening in the wider world.
I did like what I saw of AIR, the tool to let you convert Flash or HTML applications into Windows or Mac desktop applications. It seems fairly well thought out and although the lack of Linux for now is a bit of a bummer, I do think that it will arrive within twelve months. I think desktop applications to let you interact with our web applications in new ways – upload and download files, write content and then publish it, manage your sites / pages / users with a richer UI than the web one, perhaps – is potentially fertile ground for us to look into (can you look into the ground? Wouldn’t you just see the surface?).
Productivity apps on the internet
A big theme throughout the conference and elsewhere is the idea of moving productivity applications on to the internet. It’s clear that several large companies – Adobe, Google and others – are putting pretty big bets on the idea that having both your application and its data stored and running not on your PC but on a server on the internet, is Going To Be Big. One presentation asserted that this is a change comparable with the way that PCs killed typewriters and then Windows killed DOS; soon the idea that your apps and documents live just on one PC will seem archaic, is the argument. It’s an interesting idea, and there seem to be four distinct benefits being asserted:-
- Ubiquitous access for you to your data. No more not being able to continue working on your draft because it’s on your work PC when you’re at home, or vice versa.
- Easier publication. No longer will publication involve spawning many copies for distribution by email; just open up the permissions, send out an invite and away you go; you’ve published but there’s still a single canonical instance, and you can revoke publication later if you want to.
- More and better collaboration. If you want to work with several people on a document, the model of sending it round by email is clumsy and inefficient. Better to let everyone work on a single copy.
- Easier application upgrades. If the application is on somebody else’s server then upgrades can happen frequently and easily, without the end user needing to download or install or indeed do anything at all.
Against this are concerns about what happens when you’re not connected (but how big a deal is this going to be in the coming years?) and questions of privacy and security; would you be happy with all your documents being on Adobe’s or Google’s servers? Would your employer? But one could see that companies who want to offer this kind of service might be able to find ways around these problems, perhaps at a policy level by committing strongly to privacy and security, or perhaps technically by introducing ways for users to encrypt their data or choose where to store it (nothing to say that the application provider also has to be the storage provider), or even, as Google does with its search appliances, by giving you your own private instance of the application, running in your machine room, but managed by the supplier. But whatever happens, this does seem to be the coming thing; one has only to look at all the people already in this space, or pushing to get into it – Google, Adobe, Microsoft, plus a host of smaller players such as Zoho, Zimbra, Buzzword (until recently), SlideRocket, etc. – to see that there’s a lot of time and effort being invested into this space.
One last thing: if there’s one thing that came across more strongly than anything else throughout the whole conference, it’s that black is the new UI colour of choice. Every damn thing I saw used a black background, with optional accents in graphite or charcoal or carbon. Black, it would seem, is not just the new white, it’s the new any-colour-you-care-to-name.