Monday May 26, 2008

The Inadvertently Interesting Matt Asay

I've just been enjoying the vociferous complaints on the Register's semi-informed podcast Open Season by celebrity blogger Matt Asay about his "innocuous suggestion" (read: simplistic and ill-advised comment) that non-paying users are detrimental to open source. You can read it here, but before you click, you might want to consider the author's own opinion of his writing ("I almost fell asleep typing it, it's so bland.").

Mr Asay shifts his ground between the first piece and the follow up, here. But essentially, he says this: "free riders" (which he holds to be non-paying, non-coding users of open source software) are a problem for commercial open source companies because they may encourage other paying users not to pay. Or, as Mr Asay puts it, "Why should you be the only sucker paying for what everyone else is using for free, and quite comfortably?".

Now, it seems plausible that there is a situation for which Mr Asay's proposition makes sense: perhaps even his own business. But it's worth bearing in mind that if your customer does not know what they are paying you for, perhaps they are, indeed, a "sucker". Perhaps you have sold them something they don't need. Surely a better strategy is to have paying customers who know and value what they're paying for: be it support, indemnity, warranty, training, etc. There is seldom a free-rider problem with products that have a cost per marginal unit produced.

As for Mr Asay's shrill complaints, he hints that it is destructive behaviour and counter to the interests of open source to disagree with him. Well, maybe. But, open source, even "commercial open source" is a broad church of motivation and business models, and it's very hard to speak for everything in the domain (never mind non-commercial open source). Much less if you're falling asleep while writing.

So, for one thing, the term 'free riders' is a pejorative one to my ears, and implies over-consumption and/or problems in production (and frequently implies that a public good ought to be communally produced). While I doubt that Mr Asay meant to be rude about people who use open source software without paying for it, by using a term which has come from a very precise vocabulary (even if it is more commonly used these days), it might seem as if he did.

Further, Mr Asay's comments are only applicable to projects which monetize directly. Mozilla Firefox does not charge for usage, nor does it exist on donations (although I am sure they are welcome). He appears to miss many elements of the bigger picture: that a large and growing market share leads to its own opportunities. And by regarding "contribution" only in terms of money or code (which he later expands to other activities), Mr Asay misses the point that a large user base makes it easier to grow a user base, by making your software both more accessible, and more attractive to others to develop. He seems to think that non-paying/coding users only offer the benefits that a) they are at least not paying your competitor and b) they provide an "emotional safety net who want to buy with the herd". Charming.

Free access to software can drive growth. It can allow businesses to add capabilities which they would not otherwise have explored. In other words, they wouldn't have paid money to anyone. What's more, rather than having an emotional value, a large "non-contributing" user base can have a genuine economic value.

Mr Asay claims that his argument was entirely uncontroversial, restating it to, "we need to find ways to get more code or cash from communities". Given the rampant success of open source, it's dubious that it needs to do much differently, but all the same, we know what he means. I'd prefer to restate it this way: to build a business in open source, usage of the software must translate into some economic benefit for the producer.

If indeed, Mr Asay had made this highly intuitive point, his indignation at people disagreeing with him would be understandable. But by his own admission, he was sloppy in making this point, and didn't really make any others. And so, those who hold that increasing the benefit for a producer of software does not in all cases mean having a higher proportion of users to pay for access to that software could only disagree with him.

Cheer up Matt, at least they read your blog.

ps. the views expressed here are not necessarily those of my employer




« February 2017