The open source download heist

Javier Soltero (CEO, Hyperic) has a great post taking the download bean counters to task for…counting downloads. Actually, his qualm is not with those who tout their download counts so much as those who a) inflate their download counts (so that they can flaunt them), b) assume that download counts substitute for community/value/revenues, c) both “a” and “b.”

The correct answer, of course, is “c.”

This means that while [an open source project’s download counts are] somewhat indicative, and mildly amusing, tracking download counts is in no way a real diagnostic of a project’s success. When there’s money involved (as there always is in Commercial Open Source companies), there’s a built-in incentive to use raw download metrics as evidence that one project is ‘taking off’.

Javier then goes on to suggest a few ways that companies inflate their download statistics, disparaging both the outcome and the purpose behind doing it. (Btw, at Alfresco we felt ‘less than’ for a long time because we didn’t have any tricky ways to game the numbers, but now that Javier has shown us how to do it…watch our downloads skyrocket!!! 🙂

He therefore suggests – in the same way that grocery stores do for products with a “$0.00 per ounce” metric – that someone invent a way to measure real utility (which he suggests is best measured with community data points, rather than download numbers) in a project:

This is what the open source community needs. Price per ounce for downloads. A normalized download counter. The first step will be to defraud the numbers as they stand. The second, and more important one, is to start measuring the heartbeat of the community.

I agree. I’d actually suggest that a useful (though hidden to all but those who work for the companies in question) metric would be dollars-per-download. Or, in other words, how much revenue does your company have compared to downloads. It’s not a good way of measuring community, but it is a very good way of measuring how much the buying community values one’s software. I know from some experience that you don’t need to have millions of downloads to have millions of dollars in sales.

In fact, as I wrote last year about JBoss, that company’s downloads remained relatively flat while sales rose dramatically. It wasn’t that downloads mattered more, but that the company figured out how to wring more dollar value out of them.

Isn’t that what companies should be measured on? Money? I don’t think shareholders care very much how many downloads one has if they’re not feeding little Nathan….