Re: How long is it acceptable to leave *undistributable* files in the kernel package?

[Humberto Massa <humberto.massa@almg.gov.br>]

@ 18/06/2004 09:52 : wrote Raul Miller :

> On Thu, Jun 17, 2004 at 04:41:42PM -0400, Michael Poole wrote:
>  
>
> > This is not the way the law works.  The presumption is not "this work
> > is a derivative work because Raul Miller claims it is."  Humberto has
> > cited reasons why the kernel tarball (or binary images) should be
> > considered a compilation rather than a derivative work.  You have only
> > claimed that they should be a derivative work, without referring to a
> > "why."
> >    
>
> I just took a closer look at your cites
>
> http://digital-law-online.info/lpdi1.0/treatise6.html
> http://www.copyright.gov/circs/circ14.html
>
> the digital-law-online page says: "There is necessarily some overlapping
> between the two".  You quoted this yourself.
>
> The copyright.gov page doesn't really say anything that would make me
> think differently.
>
> In other words, it's still not a dichotomy.  I was too hasty when I
> wrote my message saying that it was.
>
>  
The overlappíng resides in that more often than not, you have to put 
"glue code" to make an anthology. A preface explaining why you tought 
that those were the “Best Short Stories of 2001”; an introduction to 
each story citing facts about the author; and so on. If you take a good 
look at the example of "primary linux patches" I sent you, you can 
consider the part of the kernel that Linus has exclusive rights and the 
anthology as a single part. That's the overlapping.

--
br,M