Re: Model Code for the OSD

2003-01-19 Thread John Cowan
David Johnson scripsit:

 My opinion is that deliberately obfuscated source code should be decoupled 
 from documentation. The quality and state of documentation is very 
 subjective, and should not be a part of the OSD.

IMHO that was meant to exclude people from publishing their source code
after it had been fed through an obfuscator, which not only strips
comments, but changes identifiers from meaningful names to barely
distinguishable strings of gibberish.

-- 
John Cowan [EMAIL PROTECTED] http://www.reutershealth.com
I amar prestar aen, han mathon ne nen,http://www.ccil.org/~cowan
han mathon ne chae, a han noston ne 'wilith.  --Galadriel, _LOTR:FOTR_
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3



RE: Model Code for the OSD

2003-01-19 Thread Lawrence E. Rosen
  My opinion is that deliberately obfuscated source code should be 
  decoupled
  from documentation. The quality and state of documentation is very 
  subjective, and should not be a part of the OSD.
 
 IMHO that was meant to exclude people from publishing their 
 source code after it had been fed through an obfuscator, 
 which not only strips comments, but changes identifiers from 
 meaningful names to barely distinguishable strings of gibberish.

Precisely!  You are not required to create high-quality documentation,
but if you do you should make it available along with the source code in
the preferred form ... for making modifications.  (OSL §3.)  I heard a
story about one company (Tivo was the company identified, but I can't
vouch for the accuracy of the report) that is deliberately keeping its
published documentation to a minimum in their Linux contributions
because they want to retain a proprietary competitive advantage.  I
think that's not playing fair.  It seems to me that's honoring the
letter, but not the spirit, of the GPL.

/Larry Rosen

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



Re: Model Code for the OSD

2003-01-19 Thread Bjorn Reese
David Johnson wrote:

 My opinion is that deliberately obfuscated source code should be decoupled
 from documentation. The quality and state of documentation is very
 subjective, and should not be a part of the OSD.

I have to agree with David. The documentation quality of the source
code is orthogonal to the availability of source code, and thus has
nothing to do with the OSD.

Trying to establish what documentation quality is, is difficult
in the first place.

Firstly, people differ in intelligence and experience, so what is
obfuscated to one person, may be obvious to another.

Secondly, should the quality be judge on the choice of human
language? For example, if a russian developer releases source
code with comments in Russian, can I claim that he is
deliberately obfuscating the source code? Can the russian
developer claim that all source code with English comments
are obfuscated to him?

Thirdly, the source code may implement algorithms or domain
knowledge that is inherently difficult to understand, and which
would require a book-sized explanation. Would it be considered
compliant with the OSD to refer to a (commercially available)
book? If not, how does the developer avoid infringing the
copyright of the book author while adhering to the suggested
OSD documentation requirements?

I am sure that there are other concerns as well; the above was
simply off the top of my head. I understand the good intentions
behind the proposal, but I definitely see it as a slippery slope.
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3



OSD Model Code -- Article 1 (Free Distribution)

2003-01-19 Thread Lawrence E. Rosen
Rod,

In your commentary (§1-1) on Article 1 of the OSD (Free Distribution)
you reference several cases on copyright misuse.  That confuses me.  The
copyright misuse doctrine has no application for that article of the
OSD.

Article 1 now reads as follows:

   The license shall not restrict any party from selling or
   giving away the software as a component of an aggregate 
   software distribution containing programs from several
   sources.  The license shall not require a royalty or
   other fee for such sale.

I think this Article really means: 

   The license must permit all licensees to make copies of
   the software without payment of additional royalties to 
   the licensor.  The license cannot restrict licensees 
   from either selling or giving away those copies.

How can this ever be copyright misuse, since this provision imposes no
restrictions whatsoever on downstream licensees?  There is merely a
self-imposed limitation on the licensor's right to collect royalties for
copies made by those licensees, a limitation he voluntarily accepted by
deciding to license his software as open source.

With my rewording, there's also no need for the confusing term
aggregate software distribution.  We only need to rely on the
definition of the term copies in the Copyright Act.  17 USC §101.  

Should we reword the OSD where appropriate to achieve clarity?

/Larry Rosen

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



Re: OSD Model Code -- Article 1 (Free Distribution)

2003-01-19 Thread Mark Shewmaker
On Sun, 2003-01-19 at 14:26, Lawrence E. Rosen wrote:
 
 Article 1 now reads as follows:
 
The license shall not restrict any party from selling or
giving away the software as a component of an aggregate 
software distribution containing programs from several
sources.  The license shall not require a royalty or
other fee for such sale.
 
 I think this Article really means: 
 
The license must permit all licensees to make copies of
the software without payment of additional royalties to 
the licensor.  The license cannot restrict licensees 
from either selling or giving away those copies.

[...]

 Should we reword the OSD where appropriate to achieve clarity?

I'm worried that this rewording could change OSD requirements for the
case of code incorporated into collections for which compilation
copyrights are claimed.

Going back for a moment to the case of the GPL:

| In addition, mere aggregation of another work not based on the Program
| with the Program (or with a work based on the Program) on a volume of
| a storage or distribution medium does not bring the other work under
| the scope of this License.

I have always assumed that if you put together a bunch of GPL programs
on a CD and sold it, that was one thing, but if instead you claimed that
you selected and put the programs together in some artistic or creative
fashion and claimed a compilation copyright on the whole thing, that it
would no longer be a mere aggregation, and that the collection as a
whole would have to be distributed on terms compatible with the GPL.

(That is, I've always assumed that you can't claim that you've put
together a mere aggregation of programs while at the same time
claiming that you've been creative enough in your selection to warrant a
compilation copyright on the whole thing.)

Although the current Article 1 restricts licenses from making claims on
aggregate collections that include covered code, it doesn't seem to make
such restrictions if the collections are more than just a mere
aggregation of covered code.

I'm worried that your new Article 1 might restrict licenses from making
claims on collections that are more than mere aggregations, because even
though those collections for which a compilation copyright is claimed
are necessarily a derivative work (IMHO), programs would likely be
included as whole copies, and your new Article 1 says that the license
can't restrict licensees from selling or giving away copies, with no
exceptions made if in merely making the copy they're also making a
derivative work.

(BTW, article 8 is a moot point.  Even if the license allows the new
compilation to be distributed under the same license, Article 1 seems to
be implying that it can't make any particular requirements about the
compilation as a whole, such as saying it must be distributed under this
license.)

That's my worry about your proposed change.  (I'm not a lawyer here, so
maybe I'm thinking about this all wrong, or maybe I'm confused on some
basic points somewhere, but it seems to me as if your rewording would
really change things.) 

 -Mark Shewmaker
  [EMAIL PROTECTED]
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3



RE: OSD Model Code -- Article 1 (Free Distribution)

2003-01-19 Thread Lawrence E. Rosen
I think you're confusing copies, collective works and derivative
works.  These are each defined in the Copyright Act, 17 USC §101.
There is no such definition for component of an aggregate software
distribution.  Claiming a collective work copyright doesn't mean you've
created a derivative work.  

Here's an example.  If an author exercises great creativity to collect
together into one work the best poems of the last 10 years, he is
creating a collective work and not a derivative work.  The degree of his
creativity isn't a factor unless he exercises no creativity at all
(e.g., simply a list of poems randomly selected), in which case his
collective work copyright might not be valid.  

Here's another example.  If an author exercises great creativity to find
the best open source utilities for distribution on a single CD, he is
creating a collective work and not a derivative work.  He requires
licenses to make copies, not licenses to create derivative works.
That's true no matter how much creativity he put into bringing the
collection together.

/Larry Rosen

 -Original Message-
 From: Mark Shewmaker [mailto:[EMAIL PROTECTED]] 
 Sent: Sunday, January 19, 2003 8:03 PM
 To: [EMAIL PROTECTED]
 Cc: 'Rod Dixon'; [EMAIL PROTECTED]
 Subject: Re: OSD Model Code -- Article 1 (Free Distribution)
 
 
 On Sun, 2003-01-19 at 14:26, Lawrence E. Rosen wrote:
  
  Article 1 now reads as follows:
  
 The license shall not restrict any party from selling or
 giving away the software as a component of an aggregate 
 software distribution containing programs from several
 sources.  The license shall not require a royalty or
 other fee for such sale.
  
  I think this Article really means:
  
 The license must permit all licensees to make copies of
 the software without payment of additional royalties to 
 the licensor.  The license cannot restrict licensees 
 from either selling or giving away those copies.
 
 [...]
 
  Should we reword the OSD where appropriate to achieve clarity?
 
 I'm worried that this rewording could change OSD requirements 
 for the case of code incorporated into collections for which 
 compilation copyrights are claimed.
 
 Going back for a moment to the case of the GPL:
 
 | In addition, mere aggregation of another work not based on 
 the Program 
 | with the Program (or with a work based on the Program) on a 
 volume of 
 | a storage or distribution medium does not bring the other 
 work under 
 | the scope of this License.
 
 I have always assumed that if you put together a bunch of GPL 
 programs on a CD and sold it, that was one thing, but if 
 instead you claimed that you selected and put the programs 
 together in some artistic or creative fashion and claimed a 
 compilation copyright on the whole thing, that it would no 
 longer be a mere aggregation, and that the collection as a 
 whole would have to be distributed on terms compatible with the GPL.
 
 (That is, I've always assumed that you can't claim that 
 you've put together a mere aggregation of programs while at 
 the same time claiming that you've been creative enough in 
 your selection to warrant a compilation copyright on the whole thing.)
 
 Although the current Article 1 restricts licenses from making 
 claims on aggregate collections that include covered code, it 
 doesn't seem to make such restrictions if the collections are 
 more than just a mere aggregation of covered code.
 
 I'm worried that your new Article 1 might restrict licenses 
 from making claims on collections that are more than mere 
 aggregations, because even though those collections for which 
 a compilation copyright is claimed are necessarily a 
 derivative work (IMHO), programs would likely be included as 
 whole copies, and your new Article 1 says that the license 
 can't restrict licensees from selling or giving away copies, 
 with no exceptions made if in merely making the copy they're 
 also making a derivative work.
 
 (BTW, article 8 is a moot point.  Even if the license 
 allows the new compilation to be distributed under the same 
 license, Article 1 seems to be implying that it can't make 
 any particular requirements about the compilation as a whole, 
 such as saying it must be distributed under this
 license.)
 
 That's my worry about your proposed change.  (I'm not a 
 lawyer here, so maybe I'm thinking about this all wrong, or 
 maybe I'm confused on some basic points somewhere, but it 
 seems to me as if your rewording would really change things.) 
 
  -Mark Shewmaker
   [EMAIL PROTECTED]
 

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