[Debian-med-packaging] Please allow relicensing for older versions of two single files from PHYLIP

[Andreas Tille]

Hi Joe,

On Wed, Feb 26, 2014 at 02:52:26AM -0800, Joe Felsenstein wrote:
> Thanks, particularly for copying my emails to the
> author of Seaview.  PHYLIP programs are used
> in many other people's products, and I do not
> have time to notify them of issues like this.

Well, this is part of our goal to connect people using Free Software in
biology.  We would fail if we would hide this relevant information from
the seaview author.
 
> I don't think the UW licensing people will resist a
> change of license.

Sounds good.

> PHYLIP makes them no
> money anyway and basically they don't care about it.

:-)

> As for helping with the move to Gnu/OS, it will take
> awhile (no, probably not in the next 6 months).

While I'm not in the position to put some pressure on you I would like
to admit that I would really love to see PHYLIP in the next Debian
stable release.  The freeze date is just fixed to 2014-11-05 and to make
it up this this date I think it would be nice to have a free PHYLIP in
July this year.  If you could make it up to this point in time we would
be very thankfull.

> You
> could help best by:
> 
> 1. Finding me an Open Source or Gnu license that
> requires that the original programmer get some
> compensation when the program is sold or access
> to it is sold.  I have raised this issue with Gnu/OS
> enthusiasts and claimed that there is none.  Some
> of them look surprised and say "No, Joe, I'm pretty
> sure there is a Gnu license that does that."  I claim
> that this is a mythical beast, and belief in it shows
> that people have mistaken notions about these
> licenses.  But the variety of G/OS licenses is
> very great, and I may well have overlooked the
> relevant one.

According to the widely accepted Debian Free Software Guidelines
which were used as open source definition as well a license like
this would avoid the first item of

  https://www.debian.org/social_contract#guidelines

called "Free Redistribution".  So no, Joe, there is no such license
which will grant you any such compensation ... otherwise it is non-free
(as the current license).

> 2. Pointing me to a succinct online manual somewhere,
> not to the mechanics of Guthub / Sourceforge /
> Google Code, but to how to keep a project like
> this from having its code polluted by unwanted
> "features" put in by well-meaning but self-absorbed
> idiots. How to keep it from being steered down
> paths requiring great amounts of future work, by people
> who will get that started, and then disappear forever.
> How to keep some kind of control over "look and
> feel".

As far as I know all these hosting platforms are doing some kind of user
management were the project leader can add / refuse new programmers.
Considering the lot of projects which are trying to approach the very
same as you describe above I think it is not a technical but rather a
social problem which you are describing.  Since I learned that you can
not fix a social problem by technical means I think that the answer
above is rather:  Any of the platforms above are similarly good - it
just depends how well you manage your project.

> 3. The next time you hear someone say that if
> people would only make their code Free, that
> it would be so widely used that they would make
> lots of money, point out to them PHYLIP, which
> has been "free beer" for 33 years now, and has
> made me no more than $100 in donations.  No,
> that's not $100 per year.  That's total over 33 years.

Well, as I said previously:  You do not really write Free Software to
get money for it directly.  In your situation as an employee of an
University you get salary from the University.  The point of Open Source
software is that you can get more work done in the same time since you
get contributions from external programmers (which are *not* payed by
the University) which is not money but a money equivalent.  Assume you
might have got contributions in form of patches which took the
programmer 10 hours of work.  This is a money equivalent for your
employer of 10*<salary per hour for a programmer> which is more than
this $100.  In this time of 10h you as the University employee could
have dealt with different problems instead of spending this time on a
problem which was solved by somebody else.  Depending how important /
known a project is this could make quite some money-equivalent.

> As to the great excitement that would accompany
> the announcement that PHYLIP is now Free, keep
> in mind that PHYLIP is not first in the phylogeny
> market, but now about 5th (or maybe 6th), after
> MrBayes, RAxML, Phyml, MEGA, and maybe
> one other, as judged by literature citations.

I admit in Debian Med we do not care about ranking.  MrBayes RaxML and
Phyml are included by us in Debian as well and MEGA is non-free so we
can not do this.  We simply want to support all our users and there is
some share using PHYLIP (and as the initial mail said we also want to
support seaview).

> When
> I tried to get money from the NIH program for
> funding Development and Maintenance of Existing
> Software, they turned me down with the excuse that
> PHYLIP was not State Of The Art (of course, it
> probably gets more use than the programs whose
> development they do fund).

I can assure you that we will not turn you down but regard you equal.
:-)

> Anyway, enough from me for now.

Hope my comments were helpful to you

      Andreas.

-- 
http://fam-tille.de