Is SFLC Shooting Open Source in the Foot?

The academic article by SFLC about ZFS is troubling and may unintentionally shoot free software licensing in the foot.

When I was at Sun (as part of the team that released the Java Programming Language by starting the OpenJDK project) I often heard community concerns about the CDDL license. At the time the big complaint was about the "Choice of Venue" clause.

I got involved because Sun had developed many essential Java libraries and distributed them under CDDL. The community requested a more permissive license and I was able to convince internal project leaders (and Sun's lawyers) to make a licensing change for a handful of these projects. And there was much rejoicing.

Based on my experience in helping Java to become open source I came to appreciate the legal hacks on copyright which make open source possible. It's the free software license which uses copyright to enable sharing (vs. the default of disabling sharing).

Open Source Licenses

And so I have appreciated many of the writings and speeches from SFLC on the mechanisms of software freedom. I was particularly moved by the talks about the "Freedom Box" concept.

That's why this SFLC post on ZFS sounds so off key: if open source works because of free software licenses it seems weird to weaken that foundation by prioritizing the "equity" (or intended spirit) of the license.

Allow me to mention that as I do most of my computing these days on GNU/Linux I miss the super cool features of ZFS from Solaris. I did try an early version of btrfs and was quite disappointed (but that's another story).

In this happy case the source code for ZFS is available, but what about the future, when we aren't so lucky and someone asserts in court that the "you know, the software license was really about the spirit of sharing and that means we are allowed to use it -- and not be held to the pesky details as written in the license".

A lawyer I respect called this out: "Equity" has no place in US law. The point is that for lawyers software licenses work because they have clear, written rules to guarantee the spirit is upheld; but spirit doesn't work in front of a judge -- clear rules do.

Free and open source software has made so much progress in all facets of life why on earth would we second guess the licensing tools that made it possible? And why would SFLC try to shift the spotlight (and in this case the legal burden) to "a good-faith belief that the conduct falls within the equity of the license". Especially given the earlier comment which clearly states "[the combination] is inconsistent with the literal meaning of GPLv2 section 2(b)."


The entire raison d'ĂȘtre for open source software licenses was so that developers (and users) would have clarity and wouldn't have to ask permission to use the software!!!

As stated elsewhere (and like I did with those Java libraries) the easy solution is to have the ZFS copyright holder (now Oracle) reclicense (or dual license) the code under a compatible license (permissive or copyleft). If OpenSolaris was still a thing I might understand some hesitancy, but why not liberate ZFS now?

So we have to wonder what could possibly be motivating this odd "spirit of the license" position on the part of SFLC? Fortunately charities that enjoy non-profit status are required to make public filings of their income in something called a "Form 990". The latest SFLC 990 I could find shows SFLC getting 78% (or just over $5 million) from "non public support" (see page 14).

A number with "two commas" would even be interesting to for-profit companies. Just whom is making these "donations" and what exactly do they get in return? Apparently I'm not the only one wondering about this question.

On one hand it's important to know if SFLC as a non-profit is, indeed, acting in the public interest (as the IRS requires). Yet the even bigger issue here is would "asking for a consensus about the spirit" trump the written copyright license and set a scary precedent for open source software in general?