[License-discuss] Open source license chooser choosealicense.com launched.

[Richard Fontana]

On Thu, Aug 15, 2013 at 10:17:47AM -0400, Bradley M. Kuhn wrote:
> Sorry for posting a month late on this thread [I hadn't poked into the
> folder for this list in some time], but I didn't see a consensus and
> wanted to add my $0.02.
> 
> Luis Villa wrote on 16 July:
> > In the long-term, I'd actually like OSI to promote a license chooser
> > of its own. But in the meantime I'm pretty OK with linking to a
> > variety of license choosers.
> 
> Richard Fontana pointed out in his OSCON talk that license choosers
> generally make political statements about views of licenses.  He used
> the GitHub chooser as an example, which subtly pushes people toward
> permissive licenses.
> 
> I was told GitHub's chooser accepts patches, and I was planning at some
> point to try to patch out this bias myself and see if my patch was
> accepted -- but of course any patch I produce is going to have subtle
> copyleft biases -- which I think was Fontana's point.
> 
> (Fontana, do I have that right?)

It is fairly obvious that the GitHub site's presentation of what
licenses it recommends, and why, is politically biased, though I
wouldn't say it "subtly pushes people toward permissive licenses".
Rather, the impression I get is that it was originally designed to
blatantly encourage non-copyleft licenses but threw in the GPL in its
top three recommended licenses as a kind of simple political
concession.

It is certainly true that any patch you make will reflect your own
political biases and maybe this is difficult for anyone to avoid
unless they either struggle to overcome or correct for bias or don't
have any bias to begin with.

Independent of this point, I'm concerned about inaccurate statements
made on the choosealicense.com site (one that we talked about was the
assertion that GPLv3 "restricts use in hardware that forbids software
alterations"). This is a general problem with all efforts I have seen
to summarize or provide simplified explanations of licenses, and it
happens even where the summary is provided by the license steward.

> Therefore, I think OSI should likely avoid license chooser lest OSI
> end up in the quagmire of taking a position in the copyleft/permissive
> debates.

Speaking just for myself, it is difficult for me to imagine any
license chooser or license explanation site that I wouldn't think was
more problematic than useful. Linking to a *wide* variety of license
choosers or summary sites with a very strong caveat emptor statement
might be okay.

 - RF