Matthew Hutchinson

Multi-touch interface, video demonstration (click to watch)

Microsoft’s Podtech Research Lab has also been experimenting with a different type of interface based around shape/motion tracking. It works with two hands as well, but without any tactile feedback for the user to ‘press against’. To be honest it looks like it would be very odd to use, as Jeff Han mentions, 3d manipulation of objects without tactile feedback is actually less intuitive and a step away from 2d multi-touch interfaces.

Microsoft's Podtech Research Lab (click to watch)

Some of you may know that I work for the BBC, on a little (£150m) e-learning site, BBC Jam. It was with great frustration last week, that I learned of the BBC Trust’s decision to suspend the service, pending the outcome of a public value test – Here lies the official press release describing just that.

Along with some harsh conditions placed on the production of this service (brought about by the same complaints that have caused this suspension), the site has seen its share of challenges. But the quality of e-learning content over the last year has been outstanding, and a great deal of it (more than 80%) has still to see the light of day.

There is a story worth reporting here. Years ago £150m was awarded to the BBC for the production of Jam. The government also granted £530m in e-learning credits (over 4 years) to bolster the e-learning education industry here in the UK. Along with these grants, a set of conditions was imposed on BBC Jam, including terms stating the BBC could only cover 50% of the UK curriculum. Some questions worth asking here;

• Where did the £530m go? How is the spending of this money by schools regulated?
• Why is e-learning content in the UK still so droll? (in comparison with the richer interactive (flash-based) offerings from the BBC)
• How can some companies claim they have lost business/revenue since BBC Jam had (until now) only launched about 15% of it’s content with only a small advertising campaign?
• Why does most of the UK’s educational software market revolve around the adoption of VLE’s in schools (and locking them into a platform) – rather than content and great learning experiences?
• Should the suspension of a free e-learning service for kids ever be celebrated? and who does this really benefit?

Reaction across the web has proved BBC Jam has a lot of support. Even on sites claiming this suspension is a good idea. commentaries have quickly shot down the authors post. There is some more good discussion on the matter here and here.

Here comes the disclaimer: the views expressed above do not necessarily correspond with those of the BBC.

I have managed to turn my work at gravatar caching into a little plugin for Mephisto. Its available to install via subversion like so:

ruby script/plugin install http://svn.hiddenloop.com/public/plugins/mephisto_gravatar_cache

You can browse the source code and the README file, which describes how to install and configure it. Although the plugin will work out of the box with it’s default settings.

Given the nature (and newness) of the Gravatar service, its (highly) likely that they may change their usage instructions again. I originally came across some gravatar caching code from a Daniel Haran – it worked with the old gravatar API, which is no longer available.

This new plugin works in a similar way (wget’ing the images) – however – to check an email has a gravatar at all, a request is made to gravatar.com (with a default no-gravatar image). If a 301 (redirect) is found on this request – it is assumed that the user has no image to fetch. The plugin also includes a rake task (for use in a cron job), image expiry time and other configurable class settings for caching options.

While this plugin is a Mephisto Gravatar Cache plugin – with a little modification it could work for any Rails application. The rake task ‘cache_gravatars’ would need to be modified, and a call to cache_gravatar made on any email addresses.

I have tried to comment my code as best as possible, to encourage further development. Please report all bugs and/or submit patches to me at this post. Your comments, criticisms and thoughts are welcome.

• What is Gravatar? – find out here
• More Mephisto Plugins
• Original Gravatar Caching work by Daniel Haran – the older gravatar API

—Update – Curtis at millarian.com has recently played around with this plugin and fixed some bugs. I’ll be sure to update the repository soon and add some more tests. This was my first plugin release and something I coded rather hastily.

Maybe I am a bit late to the party, but OpenID seems to have made a surging comeback in the last month or so, not least in the Rails community. Its something I am looking to implement on a few projects of my own, and luckily (with Rails) I won’t have to re-invent the wheel. Here are some useful links on the matter:

• OpenID Rails Plugin – using the Ruby-OpenID gem
• Mephisto and OpenID support
• Collection of links at Ruby Inside
• PeepCode – will have a screencast on the matter soon.

Last year during @media2006, Mr. DHH questioned whether login/signup (and managing multiple usernames/passwords across the web) was such a big problem after all. With modern browsers capable of remembering passwords and client side tools to take care of the matter. He does seem to have changed his tune on the matter now. You can see how 37 Signal’s new HighRise application handles it (with a small unobtrusive link in the login box corner).

For some more discussion on the good/bad aspects of OpenID, we have:

• Simon Willison, explains OpenID – Video Interview at Vitamin
• Positive thoughts at ‘We Break Stuff’
• How to best merge OpenID with regular username/password authentication
• Matt Pelletier – OpenID, Rails, and Identity 2.0 – summary of RailsConf talk

One awkward part of OpenID at the moment, is the use of a URL to login with; to the average user this has to be a bit confusing. With that, and few major sites offering OpenID support, it may be a while before it takes off across the web.

Heard of PRojects IN Controlled Environments aka. Prince ? If you’re in the UK you’ve probably seen it, asked for on job descriptions, listed in expense reports and generally talked up among those in the project management industry. Outside of Europe Prince2 holds little clout. Most people have never heard of it, (including Mr. Joel on Software)

Prince2 is essentially a project management framework, that claims it can be applied to (or modified to fit) any project. It is great theory – but that’s pretty much all it is. In reality it’s rarely the right methodology to choose for a software project.

The problem is some large (often bureaucratic) companies place a strong emphasis on having a Prince2 qualification, even in the software industry. Year on year, it seems to make a regular appearance on training expense reports.

To make things worse, whatever good intentions Prince2 had to begin with, have faded away. What’s left now is a money making training industry (with loyal repeat customers). Having to re-take the same training and sit the exam every year or so, just to keep the accreditation. Speaking to those who have taken the course, it doesn’t seem to offer anything more than ‘how to pass the exam’ – and is scant on the underlying principles of project management using the framework (including how you might want to tailor it).

Simply put, Prince2 is too rigid to fit with the complexity involved in managing software projects, especially those with small teams. The amount of work you would have to do, moulding Prince2 into a more agile solution, would offset any benefit in using the framework.

If you’re still not convinced, I’d suggest have a good read through ‘Why Software Projects Fail’ , ‘Getting Real’ and ‘The Agile Manifesto’

Heads up, I know which one I prefer in the Agile Manifesto vs. Prince2 debate.