I’ve just spent a few days trying to implement a specific feature in my current AIR project. Having tried several approaches and workarounds with varying degrees of nearly-but-not-quite-ness, I’ve had to admit defeat for now and have moved to a different implementation that I’m less happy with, but still allows basic functionality. At the same time Kieran has recently started work on his first Flex project and it’s been interesting to see how an experienced developer, when faced with a feature or issue he can’t easily implement or find a solution for, has quickly fallen back to a ‘next best’ solution, and moved onto getting the application functional and usable.

Time pressures obviously play a part here – the amount of time you have available before the project has to go out the door is going to impact on how much you are able to include desirable features over essential ones, but I think what I’ve picked up from this is that as soon as you encounter an issue, if you can’t predict reasonably accurately how long it’s going to take you to solve it, you need to set limits on how long you spend trying to make it work and at same time identify what is achievable if you fail to get it done within that limit.

If all this sounds like common sense, it is of course – but I found that in this instance I got too focused on a specific feature, rather than on getting the application functional enough that users can start using it sooner rather than later.

Writing about web page http://arstechnica.com/news.ars/post/20080526-the-promise-of-google-apps-includes-a-shrinking-it-staff.html

Interesting article and discussion on Ars Technica about the increasing use of Google Apps and outsourcing of student email to GMail by universities. The article is discussing the specific impact this is having on IT staff numbers in some universities, but I’m more interested in the trend towards using applications ‘in the cloud’ and moving off the desktop, particularly in higher education.

In the discussion, quite a few people see migration of utility services such as email (particularly student email) as a ‘no-brainer’ decision; letting Google’s massive and well-resourced infrastructure handle scalability and uptime does on the face of it seem a fair trade-off for a bit of advertising and loss of branding, but as Googles provides more desktop-style applications, it is pushing the merits (and cost-benefits) of such applications to institutions in what seems to be a targeted campaign. As some people rightly point out, Google isn’t an exclusive provider of such services and others note there are many other complex issues to consider other than infrastructure and support costs, but on the face of it using free applications like some of those mentioned below seems like a valid move.

Anyway, I digress; how many students at Warwick have tried or regularly use online software like Google Apps (or any internet/browser-based application, like Buzzword, Zoho, Photoshop Express etc.) in preference to MS Office or even free applications like OpenOffice on the desktop? What/how/why do you use it? What would you say is missing from these applications (in terms of specific functionality versus a desktop application, or a whole package). I’m interested!

Writing about web page http://labs.adobe.com/technologies/flashplayer10/

A pre-release of Flash Player 10 is now out on Adobe Labs. Astro has a number of new features; P2P support, Speex voice codec (no more having to use ASAO!), dynamic streaming support and RTMFP, 3D, improved hardware acceleration and more sophisticated text control being the ones I’m most looking forward to.

The full feature list can be found here

I just had to encode a couple of videos from AVI to H.264 format for playing via a Flash-based video player and noticed that the player was having to download the whole file before playing it. I checked out this Adobe Tech article on H.264 encoding and here’s why:

One important thing about playing an H.264 video file as progressive download is that the moov atom needs to be located at the beginning of the file, or else the entire file will have to be downloaded before it begins playing. The moov atom is a part of the file that holds index information for the whole file.

I was using SUPER to encode to H.264, and by default the moov atom gets placed at the file end. Fortunately, Renaun Erickson at Adobe has kindly created a small AIR application (a port of a C++ application) that will take a standard H.264 file and relocate the moov atom to the front of the file, and then progressive download works properly.

Edit- thanks to Marshall’s comment, if you’re using SUPER there’s an option to select ‘Streamable H.264’ – do that and the moov atom will be put at the beginning of the file.