Javascript JVM

Wednesday 30th April, 2008

It’s amazing what programming gymnastics can be achieved when someone sets their mind to it. JQuery‘s John Resig reports on Orto, which is a JVM written in Javascript that accomplishes multithreading in a single thread using yields.

The User Experience of DVDs

Wednesday 23rd April, 2008

I’m a big film fan. I have a fairly substantial collection of DVDs. Some of them I watch once and then leave to gather dust, some of them I watch obsessively, again and again, like a child that’s just discovered the word “poo” and thinks it’s so hilarious it’s worth repeating until puberty.

Some things about the DVD format I love.

  • You get better quality than VHS
  • Control over audio and subtitles
  • You can fit more on a shelf than with VHS
  • Sometimes (sometimes) the special features are actually worthwhile

But I also have some gripes with DVDs, mostly caused by the way publishers assert what they consider to be their interests as being more important than my User Experience (UX).


The film industry collectively decided at the birth of DVDs to split the world into regions. These regions are individually marketed to, meaning different pricing and release dates. With regular consumer hardware you cannot play DVDs from other regions. This means that if I go on a business trip and see a film I like, I probably wont be able to play it when I get home. This is probably the worst User Experience imaginable.

Country selection

Once I’ve bought a DVD for the correct region, you’d think I’d be able to sit down and enjoy my DVD. Not yet. Quite often the first thing you see is a menu asking you to select a country. Okay, it makes sense for me to select my language. What doesn’t make sense is that for a DVD I bought in the UK (Region 2) I have to navigate my way through pages and pages of countries that aren’t even in Region 2 in order to select my country. It’s nuts.


I’ve got the right region DVD, and selected my country. What often happens next is that I’m confronted by one or more, often noisy, adverts telling me that copyright is theft (not technically true). After the first few hundred times this is quite annoying and I reach for the remote to skip forward… only to find that functionality disabled.

I can understand why publishers might want to disuade young minds from copyright infringement, but once someone has made an active decision to skip past that message preventing them from doing so will just make them angry or frustrated. They will be less willing to listen to your message. This is a poor User Experience.


Some people like trailers, some don’t. Most DVDs let you skip them. My copy of Finding Nemo for example does not. It’s a great film, but the publisher decided to make you watch three or four trailers, with skip and fast forward disabled. This is a poor User Experience.

Scratched or worn out disks

If you were buying a DVD as an object to own outright you’d be able to make backups legally in case you scratched a disc. This is not how it works.

If you were buying a licence to watch the content on the DVD, you’d be able to send it back and get a replacement if you scratched the disc. It doesn’t work like this either.

When you buy a DVD you’re not buying an object that you own outright, and you’re not buying a licence to watch the film, you’re buying the most restrictive aspects of both. You can’t make a backup because the publisher owns the copyright, but if you scratch the disc then you have to buy another one because you bought an object not a licence.

This is a poor User Experience.

How should it work?

It’s not complicated. There should be no regioning (I believe this is the case for Blu-Ray). It should not be possible for publishers to override the regular operation of my player. Fast forward should always work. I should be able to either make backups, or send back scratched disks and pay a minimal charge.

Alternatively I could catch the Cluetrain and give up my addiction to one-way, read-only media.

Research and Development

Monday 21st April, 2008

Well, this post does it. I now officially like Simon Wardley’s ideas and have subscribed to his newsletter.

Change of Perspective

Saturday 19th April, 2008

It’s funny how looking at something from another perspective can change it entirely.

The Standish Group has produced a report suggesting that Open Source software “costs vendors” $60 billion per year. Simon Wardley on the other hand describes the situation differently in his post; FLOSS saves industry billions.

Marathon Man

Tuesday 15th April, 2008

Congratulations to Justgiving’s own Jan Braglewicz who ran the London Marathon in 4 hours and 26 minutes, raising over £1,600 so far for Sense who provide support and assistance to those who are both deaf and blind.

Behavioural economics meets charitable donation

Tuesday 15th April, 2008


As a company, the Giving Group (who run Justgiving in the UK and Firstgiving in the US) has a particular set of incentives. Because we take a transaction fee from peoples donations to charity, it’s in our interests to try and increase the amount donated to charity. This is a good thing for charity because by increasing our profits by a modest £50,000 per year means that we have to generate an extra £1,000,000 per year in charitable donations.

You might ask why charities don’t do this themselves and save some money. The answer to that is in the economy of scale. By building a donation service that thousands of charities can use, we can charge a tiny fraction of the price that charities would have to pay to build one themselves. If we make £10k worth of improvements to the service used by our 3000+ charities we’ve effectively saved them a collective £30m. The trick is this; if charities build their own system, the cost of that is hidden from the donor. You don’t know how much of your donation goes on the salaries of employees or service providers.

The incentive structure and the economy of scale add up to explain why Justgiving has irreversibly changed the way people raise money for charity online.

Behavioural Economics

The study of these incentives is called Economics. It may sound dull but it’s actually the study of human nature and how it’s revealed through commerce. There’s a growing weight of argument behind the field of Behavioural Economics, a field that specialises in studying how our emotions and biases effect our commercial transactions.

Behavioural Economics is particularly relevant to the Giving Group because it’s one way to measure the motivations of people who donate through our services. In other words, it could indicate what proportion of our users are sponsoring their friends because they feel obliged, what proportion are showing off in front of work colleagues, donors’ emotional response to different charities, and much more besides.


There is already research underway into why people give to charity. You can read about some of it at the New York Times. I think we’re in a good place to contribute because the Giving Group has a significant amount of data about peoples donation habits. I think that (without ever intruding on people’s privacy) we should be able to discover what motivates donors who use our system.

An example of this is the new phenomenon of donor fatigue (as discussed in the Independent and the Telegraph). The short version is that people who are repeatedly asked for sponsorship are likely to donate less or not at all. By investigating both the effect, and the currency value, of individuals being repeatedly asked to sponsor friends, we can both assess the scale of the problem, predict future trends, and measure the results of changes we might make to counter any problem.

This research is currently underway, watch this space for interesting developments.

Google Apps ported to Linux/EC2

Tuesday 15th April, 2008

As reported by Techmeme, some absolutely inspired work by Chris Anderson has resulted in a port of Google Apps to both Linux and Amazon’s EC2.

Could we be looking at the start of a standard portable application platform?

OpenID + VRM = Cardspace?

Monday 7th April, 2008

It’s starting to dawn on me that OpenID and VRM combined become very much like a version of Cardspace (but with much more potential in areas Cardspace can’t touch).

I actually think Cardspace seems like a great idea, and quite a compelling experience as far as single sign-on goes. That said, I would rather see a VRM-based solution for the following reasons (based on my still limited knowledge of Cardspace).

  • VRM would support subscription access to data
  • VRM would allow broadcast of data to user-selected service providers
  • Installation of client software means I can only use single sign-on with admin rights
  • Mistrust of an infrastructure initiative pushed by a single company
  • VRM/OpenId would allow anybody to be an identity provider for others (it’d be up to the recipient of the identity evidence to decide whether they trusted the identity provider)

You can read more about Cardspace should check out some of the Microsoft identity bloggers: Don Schmidt, Kim Cameron, and Mike Jones.