Thursday, October 11, 2007

Lending people / ignoring demands

I saw this piece on libraries lending people as well as books on one of my favourite blogs, David Wilcox's 'Designing for Civil Society.'

As techies say, the idea is conceptually finished (i.e. there are no actual working examples). Still, using conversational approaches to achieving many policy goals is - I believe - a hugely under-rated approach.

Could it be that the kind of people who thrive in officialdom and public policy formulation didn't get where they are today because of their conversational talents? In my experience, this is often emphatically not the case.

Elsewhere, changing the subject, I found this article well worth reading. It's about the thoughts of Matt Mullenweg, the lead developer on Wordpress, the open source content management platform.

As demands have diversified, Mullenberg has had to face ...
"...an avalanche of feature requests. Mullenweg advocates a minimalist approach to software quality, and believes that many software products have too many options included to satisfy divergent opinions.

"Lots of software, especially open source software, goes the option route because it makes everyone happy. But it creates terrible software. I think WordPress as it is now has too many options. So in making decisions, I piss people off. Sometimes they take it personally, and that's a difficult thing. But if we continue to be a successful product and to grow, I feel we're making the right decisions.

"There are two main methodologies of open source development. There's the Apache model, which is design by committee - great for things like web servers. Then you have the benevolent dictator model. That's what Ubuntu is doing, with Mark Shuttleworth. Ubuntu is doing amazing things, and I think it's going to change the face of the desktop. That's also WordPress, and ultimately that's what's going to work for consumer applications.""
So. Instead of giving people exactly what they want, he advocates making a standard one-size-fits-all solution that meets most requirements - and doing it in a disciplined and competent manner that doesn't lose sight of the most important objectives: Make it easy to use, make it work well, make it work safely.

This is food for thought for economists, managers, politicians and bureaucrats everywhere, I think?

No comments: