Monday, November 10, 2008

Defining Freedom Again

Sylvia asks: I'm getting that what you mean is that open source means that anyone should be able take the source code and do anything commercial or non-commercial? Yes? Is there any way to prevent the commercial path? Or is that just a bad idea in your view?

More importantly than whether or not I think it is a bad idea is that open source and free software have always been defined to allow commercial or non-commercial use. It is simply what those words mean. As Mako put it:

To be sure, many programmers and software companies are uncomfortable with the freedoms required by the FSD (Free Software Definition). Programmers are welcome to release applications under a license that prohibits terrorists, fascists, or pacifists from using their software but their software won't be free. There are very good and thoughtfully considered reasons for each freedom in the FSD; there may also be very good and thoughtfully considered reasons for choosing not to use them. Free Software draws a line and leaves the final judgment calls up to the developers applying the licenses and the users using the software.

Not every programmer has to write free software. Not every programmer does. But if coders want to call their project "Free Software" or "Open Source," they must pass the bar set in the FSD and OSD. If a programmer wants their software included in Debian, listed in the Free Software Directory, or supported by SourceForge, Free Software's core freedoms must exist for their users. As a result, few coders write "almost free" software today while, proportionately, many more did two decades ago.

I do think that in most cases non-commercial licensing of educational software is a bad idea on pragmatic grounds in part because the market for educational software is so lousy. I don't sit around gnashing my teeth over the fact that the small number of profitable proprietary educational software companies, like, say, Carnegie Learning, don't open source their software. It would be nice if they did, but I've got other fish to fry.

But in most of the cases of academically produced software I'm thinking of, commercial licensing the software won't be a windfall for anyone, either the person who created it or anyone else. The only thing practically accomplished by the non-commercial license is creating a significant barrier to re-distribution. There are other arguments to be made against non-commercial licensing; I primarily think it is inefficient and wasteful and demonstrably doesn't work. It just creates shelfware.


Universities use public funds to make this stuff - so is it fair that some company can take it, not share their work, and sell it?

The flipside to this, which I think is compelling, not to mention politically expedient, is that educational software companies and publishers pay taxes too. I want the government to fund research and development in these fields, but ultimately you're talking about the government creating products that compete with products made by the private sector. Allowing commercial use and distribution of government funded software mitigates that conflict. I think it is a reasonable compromise. You're not going to make an obscene profit by simply selling copies of software that's already available at no cost, and I do hope vendors have incentive to provide training, support, etc. for free software, at a fair profit. Using a more restrictive free software license like the GPL also helps ensure that changes made by the commercial vendors will be made available to the commons.


Stephen Downes said...

It is important to keep in mind that you are defining 'free software' and not 'freedom' itself.

'Free software' is defined as being composed of four very specific 'freedoms'.

This is important because the definition of 'free software' applied *specifically* to software (I heard Stallman repeat this point ovcer and over again when he spoke at FKFT in Spain last summer).

Things that are not software - things like content, for example - are subject to different conditions in order to become 'free'. And there is *not* a single 'by definition' account of 'free content'.

There is a good argument to be made that in many contexts, free means 'non-commercial'. Those are the cases, specifically, when 'commercial' creates barriers to access, barriers that may be impacted not merely by point of sale but by reconstituting entire environments so as to ensure that the only access is commercial access.

Sylvia said...

I'm finding nuance in these posts and comments, so if this feels like you are repeating themes from the past, I'm sorry about that. I hope you know that I appreciate your expertise and I'm not being a devil's advocate or just stirring up dust. I know I've used the term open source without really knowing what it means. I think a lot of people are in my same situation.

It may be that preventing commercial use is simply an idea that sounds good but is not practical in reality. That's certainly not unheard of.

My experience with government grants (not specifically software) is that the rules are so complex that they strangle future use of any kind. Grants often have a lot of language about sustainability, but then the tangle of rightsholders makes it impossible to publish/distribute at the end of the grant. So I see the point there.

At some point, though, I feel like I'm diving into a symantic argument about the meaning of the term open source that gets in the way of actual use. I care, but I'm not sure I CARE - if you know what I mean.

Stephen, if you can explain your last paragraph with an example, I'd appreciate it.

Tom Hoffman said...


I'm happy to explain this stuff -- and I should note that my frustration in the original post wasn't aimed at you.

The thing is that in each individual case, this doesn't seem like a big deal. In aggregate, however, it means an entire ecosystem is choked off.

Sylvia said...

Tom, I never thought it was aimed at me in personal way, no worries.

I do want to understand it better, so this discussion has been helpful.