Tuesday, March 09, 2010

On project management and the Wikimedia Foundation

I have been reading the blog of Guillaume Paumier, the project manager for the Multimedia usability project. Reading it, I am of two minds.. it is tagged as his personal opinion but on the other hand, this is the opinion that he promotes. The question is should I analyse and critique it or not and, to what extend and where.

As a blogger, Guillaume is the opposite of me. I do not use bold, I use illustrations. Thinking on projects, I prefer the scrum approach for the Wikimedia Foundation because the WMF has a track record of waterfall projects that have not delivered on time. Main advantages of scrum are more visible results and consequently easier communication with our communities.

In Guillaume's view a project manager decides on what to do when. Having a keen interest in scrum, I have a problem with this, for me it is the customer who decides what gets done when. NB Customer as understood by scrum. There is one KEY problem though; there is nothing in our communities, in our organisation that serves the role of customer. The result is that the WMF may be doing well but it is increasingly seen as an ivory tower.

When I query his blog there are two words missing: language and culture. Given that translatewiki.net supports 327 languages, it is unlikely that anyone can truly appreciate and predict the needs for software that is to accommodate all these languages. I am really looking forward how Guillaume intends to deal with linguistic and cultural issues. This is maybe a good subject for his next blogpost.

The Usability Initiative did a good job by promoting the localisation of its software and setting up test environments for languages in several scripts.

Finally I disagree with Guillaume that there are no people in our community who fulfil a role of project manager. They do not have the title but they have taken the burden of that role.

1 comment:

eekim said...

I agree that customers are ultimately the focus of SCRUM. I don't agree that "we" don't have a customer. Every company deals with this challenge. The discipline of software engineering requires segmenting and prioritizing the customers.

So, for example, Wikimedia's "customers" could be segmented into editors and readers, which could be further segmented by experience level, language, country of origin, etc.

Wikimedia Foundation's challenge is to work with the larger MediaWiki community to identify the right customers it should prioritize.

I do agree that Guillaume's posts (which I enjoyed) could perhaps expand on these issues. I'm quite sure he's thought about them.