[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: mono and moonlight distribution method make me worried.

Quoting "Bradley M. Kuhn" <bkuhn@ebb.org>:

Steve Langasek wrote at 19:58 (EDT) on Sunday:

we don't consider the existence of a software patent claim to be a
sufficient reason to remove software from main.

Well said.  There are so many USA patents, if you tried to remove every
piece of software from main that might be judged to practice the
teachings of some patent, you'd do a *lot* of removing.

Nevertheless, possible patents on Mono and what Novell/Microsoft's
strategy is with regard to releasing this software is something to watch
and be concerned about.

The Mono Project assures us that Microsoft holds patents that cover the ECMA 334/335 technology; they even go so far as to identify one of the inventors[1]. The Mono Project does not state that they practice those patents, but they also do not suggest in any way that those patents are either invalid or that their practice is avoided in the Mono implementation of the ECMA standards.

Instead the Mono Project asserts that a royalty-free, reasonable, and non-discriminatory patent grant has been provided by Microsoft. Both Microsoft and Hewlett-Packard have complied with ECMA requirements[2] in promising to offer their patents covering ECMA 334/335 under RAND terms[3]; and an archived email from the aforementioned inventor is cited by the Mono Project as further promising that the RAND licensing would be offered royalty-free[4].

However, if one examines the details of the patent declaration, Microsoft states that the RAND licensing is available "to any party requesting it". The royalty-free addendum by Jim Miller does not relieve this requirement to actually request the license from Microsoft. In addition, the patent declaration ensures its validity only for the duration of the ECMA standard.

Not to conflate the issues of patent licensing with copyright licensing, but if it is indeed required that the patent indemnity be requested then from a patent license perspective, the Mono implementation should fail Debian Legal's "Desert Island" and "Dissident" tests for DFSG compliance[5] because upstream must be contacted and the licensees identified. Furthermore, the limitation of the validity of the patent declaration to the duration of the ECMA standard should fail the DFSG's "Tentacles of Evil" test.

Now I would certainly agree that patent claims of persons or companies not involved in the development of a Free Software project should not be sufficient cause for exclusion of that software from Debian Main; however, it seems that when it is the Free Software project itself asserting the claims, and relies upon the licensing from the patent holder as justification for practicing those patents, it is appropriate to consider the actual terms of that licensing.

[1] http://www.mono-project.com/FAQ:_Licensing
  The core of the .NET Framework, and what has been patented by Microsoft
  falls under the ECMA/ISO submission. Jim Miller at Microsoft has made a
  statement on the patents covering ISO/ECMA, (he is one of the inventors
  listed in the patent): here[3]
[2] http://www.ecma-international.org/memento/codeofconduct.htm
[3] (PDF) http://www.ecma-international.org/publications/files/ECMA-ST/Ecma%20PATENT/ECMA-334%20&%20335/2001ga-123%20&%202002ga-003.pdf

[4] http://web.archive.org/web/20030424174805/http://mailserver.di.unipi.it/pipermail/dotnet-sscli/msg00218.html
[5] http://en.wikipedia.org/wiki/DFSG#debian-legal_tests_for_DFSG_compliance

Reply to: