For Approval: NASA Open Source Agreement Version 1.1

[Richard Schilling]

O.K. - I'm going to rant.  Nothing personal here Bryan, but NASA put 
some good time into this so I'm going to pick apart your resonse as a 
non-NASA person.  Specifically, you state some views that are 
misguided.  So, I post my response because so many times on this list 
people try to play "armchair lawyer" and pick apart a license.  It's 
not appropriate and degrades the value of the discussion.  Lawyers most 
likely write NASA licenses and that's something to respect.  When you 
read NASA's license, you may very well be reading the best license that 
money can buy.

I hope that opensource.org applies some sensible legal criteria when 
reviewing submissions - especially this one.  All other discussions are 
just hot air.

see below for some of my own hot air ...

On 2004.02.12 12:21 Lawrence E. Rosen wrote:

[snip]

> 
> >          i.  NASA legal counsel requires that all NASA
> > releases of software
> > include indemnification of the U.S. Government from any third party
> > liability arising from use or distribution of the software.  See
> 4.B.
> 
> All open source licenses include very broad disclaimers of liability.
> Does
> the government also require indemnification? Why? I'm sorry, but I
> can't
> conceive of any open source distributor or contributor to NASA

This is a liability issue.  Why should the government be responsible 
for the misuse of the software they write?  Why should anyone expose 
themselves to third party lawsuits when all they want to do is provide 
some benefit to the public?  Anyone who doesn't understand this part of 
the license has no business using the associated software in the first 
place.

> software who
> would dare to indemnify the US government or any of its agencies.
> You're the
> one with the deep pockets and with a Congress that can legislate any
> protections you need against tort, contract or intellectual property
> liability. Why should we assume those risks for you? :-) Please stick
> with a
> disclaimer of liability; every open source license has one.

You have it backwards.  Why should NASA accept the risk of you using 
their product?  That would put my (however small) tax contribution to 
NASA at risk!  Not O.K.

NASA obviously spent a lot of time putting that license together, and 
those who have even an ounce of a clue about licensing recognize the 
importance of this wording.

Better question: Why should the NASA license stoop to the lower level 
of other open source licenses that are just pseudo legal terms written 
up by a bunch of non-legal people?

The onus to provide a "suitable-for-US-courts" license is on the people 
who wrote the original licenses.  Specifically, the folks who wrote the 
existing licenses endorced by opensource.org have the responsibility to 
show their writing is up to par with what NASA puts out - not the other 
way around.  Is NASA perfect? no, but at least they're open to 
discussing the license in public on this list.  And they'll address 
*legitimate* licensing issues as they come up.

"Big brother" ranting just doesn't belong in a discussion about 
licenses.  I'm glad NASA's got big pockets, and I'm glad they're 
protecting it with their licenses so they can worry about space 
exploration.  
> 
> >          ii.  Federal Statute mandates that the U.S.
> > Government can only be
> > held subject to United States federal law.  See 5.C.
> 
> The Open Software License (OSL, 
> www.opensource.org/licenses/osl-2.0.php)
> establishes jurisdiction, venue and governing law for a US-based
> licensor
> (such as a US government agency) in the United States. See OSL § 11.
> [Foreign readers of this email should take comfort that the OSL is not
> US-centric; jurisdiction under the OSL lies wherever the licensor
> "resides"
> or "conducts its primary business."]

In any license, specifying a venue is up to the writer.  Take it or 
don't use the license.  Or how about asking the folks at NASA, "Would 
you also consider <<venue>> as a suitable venue in your license?"  That 
would at least be POLITE.  Chances are the presently stated venue has 
the best chance of making sure NASA can fend off frivilous and silly 
lawsuits.  Just a guess. . .

> 
> >          iii.  NASA policy requires an effort to accurately
> > track usage of
> > released software for documentation and benefits
> > realized?purposes.  See 3.F.
> 
> Such provisions are not allowed in an open source license. Reporting
> requirements are viewed as unreasonable limitations on the rights of
> licensees to do anything they want internally with open source

Biggest problem of all here - who in all of creation has the authority 
on blessing open source licenses?  I don't even think that half of the 
existing opensource.org endorsed licenses are capable of achieving 
their self-stated goals.  Asking NASA to adopt license language that 
might not serve its intended purpose is just flat out wierd.

I maintain that an open source license has certain characteristics and 
achieves some well defined goals - the primary one being quick, open 
distribution or source code and documentation to the end user without 
charging a license fee.  It is entirely unappropriate to specify what 
belongs and doesn't belong in an open source license.  Either the 
license achieves it's own goals or it doesn't.

The NASA license strives to deal with legal issues, not moral "Open 
Source will save your grandmother and the free world" issues.  The 
beauty is, the NASA license may do that someday :-)


> software
> (e.g., make copies, derivative works, etc.). On the other hand,
> because of
> "reciprocity" (see my further comment below) you'll be able to see
> improvements to NASA software that are distributed by others, and
> benefit
> from them. That will be measurable.

I don't want some hack trying to "improve" NASA software.  There's no 
benefit in having untrained engineers who don't understand NASA trying 
to improve NASA products.  It's like having the boy scouts try to 
improve the Mars rover.  NASA owens the bragging rights on their 
software, and I think it's important to keep it that way.  
> 
> >          iv. Federeal Statutes and NASA regulations requires
> > a prohibition
> > in NASA contracts against representations by others that may
> > be deemed to
> > be an endorsement by NASA.  See 3.E.
> 
> Various licenses do that. See, for example, OSL sections 4 [Exclusions
> from
> License Grant] and 6 [Attribution Rights].

Obviously, NASA's well paid lawyers disagree with what's there.  My 
money's on the NASA lawyers.

> 
> >          v.  Because it is important that each of the aforementioned
> 
> > clauses be a part of each open source agreement relating to
> > NASA released
> > software, the proposed agreement must mandate that distribution and
> > redistribution of the software be done under the aegis of
> > NOSA (mandatory
> > domination similar to GPL).  See 3.A.
> 
> I call this feature "reciprocity." Lots of licenses do that, including
> the

you and anyone else with a similar opinion are most likely not trained 
to call that anything.  *I'm* not even trained to call it anything and 
I've spent a lot of time studying law.  I won't even try.  Spend a few 
years doing nothing but government licensing and arguing cases then try 
labeling it.


> GPL, MPL and CPL; similarly, the OSL contains a reciprocity condition
> that
> requires any licensee who distributes derivative works of OSL-licensed
> software to do so under the OSL. You don't need to write a new license
> to
> obtain that license feature.
> 
> I'm sorry if I've misunderstood your license by relying on your
> summary of
> it rather than the license itself. I intend to read through the
> license
> itself soon, but I wanted to get these general questions on the table
> first.

I think you not only misunderstood the license, but you may only have 
succeeded in bringing out the armchair lawyers in force to debate these 
misguided points.  It just doesn't do anyone any good.


Richard Schilling
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3