When I presented my “Fauxpen Source” post, I had several goals. One was to introduce the term “fauxpen source“, which I think is clever. The second was to firmly state my position on the term “open source” as it is applied to “open source companies”. Finally, I wanted to show that open source qualifies as free software per the Free Software Definition.
I figured it would stir up some debate, but I was actually quite disappointed that while my post was attacked as being “tired” and of limited usefulness, not a single person stepped up to offer a different definition or to debate the points I raised.
I’m a pretty open minded guy, and I would love to see someone take a stab at a definition that takes into account the open core folks in a way that differentiates what they offer from commercial vendors, as well as what companies like OpenNMS offer. Instead the debate wandered all around the subject and never addressed it directly.
Matt Aslett questioned whether or not the “industry” needed a definition at all. I would claim it does, since many governments, including that of the US, are considering mandating or at least exploring requiring the use of more open source software. Isn’t that important? Shouldn’t we define terms? Aslett doesn’t like the OSD to be applied to business, but he didn’t go as far as to offer an alternative.
Some dude over at Information Week expressed the fact that he was bored with the whole thing, and asked “at what point does it cease to even matter?” (I would have commented on his post but it appears that they are unable to get comments working over there). I would claim that if it doesn’t matter, then why do all of these fauxpen source firms emphasize that they are “open source”? Just this week I saw a press release from one of them that started out “Open source network management” – if it isn’t important why lead with it?
The best comment I saw on the whole thing was from David Dennis of Groundwork:
So if a vendor isn’t primarily making money from license sales, but only partially, would they be misusing the term open source?
If a vendor gets 40% of its revenue from license sales, and 60% from support and services, they’re not ‘primarily’ getting money from non-open-source software.
Yay! An actual question that is relevant and is hard to answer. It is easy for me to claim a company is fauxpen source if their business plan is based on software license sales (i.e. the hockey stick revenue graphs they presented to their VCs ). But what if only 1% of their revenue comes from such licenses? What about 10% or 20%? At what point does it cross the line?
I don’t know, but I’d love to discuss it.
Ben recently posted about intent. I think intent does have a lot to do with it. If your intent is to basically force people toward your “enterprise extensions” by purposely crippling your “community edition” then that obviously isn’t in the spirit of open source. However, can I imagine a situation where a complementary add-on in no way blocks the functionality of the main project, so would it be okay to license that as commercial software and still call oneself an open source company?
Unfortunately, as it is not possible to objectively measure “intent” so we can’t use that as part of a definition. Again, the question is not an easy one.
Furthermore, it doesn’t seem like anyone is really interested in talking about this in a rational manner. People seem to prefer making blanket statements and not backing them up. Not to purposely pick on Matt Asay, but in one of his posts today he stated:
Red Hat is an example of “free done right,” following analysis from TechDirt. We’ve moved beyond the business models that insist that every line of software be open source: they couldn’t scale and tended to treat openness as an end in and of itself, rather than as a means to an end.
Today, if you look at the most successful open-source businesses, none of them pass the ideologues’ unrealistic and counterproductive “100-percent freedom” litmus test. Not a single one of them.
He uses Red Hat as an example quite a bit, but when Andres Garcia (and myself) asked “which software made by Red Hat has lines of code that aren’t open source?” there was no reply. This is because, to my knowledge, all of the software that Red Hat actually distributes (versus uses in house) comes with the source. So does OpenNMS, but my guess is that Matt wouldn’t categorize us as a successful open-source business. However, it is hard to argue that Red Hat is not (plus they pass the CentOS Test). But my guess is that Matt will continue to insist that Red Hat is a hybrid vendor since many of his arguments come crashing down if they are not.
I also assume am I one of those “ideologues” he talks about, but that is not the case. I just insist that software called “open source” meets the Open Source Definition – not that all software is open or free. I’m even open to an alternative definition, but so far no one has come up with one. Thus Matt hopes to sway people by using rhetoric, informal fallacies and irrelevant examples – not to enter into a real dialog.
Unfortunately, this is the state of the debate. Heck, it is the internet after all, so I guess I was a bit naive to expect otherwise. I’ll probably take a break from it for awhile, since it’s like the old joke about the pig – you always lose and end up dirty while the pig enjoys it.