Re: Aspell-en license Once again.
On Mon, 2002-11-04 at 13:54, Henning Makholm wrote:
> Man, you're way out. Some people (not all developers) point out that
> the Database Directive exists. Not a word has been said about it being
> "supreme" in any way. It exists. That is all. It that so har to grasp?
Since we acknowledge that it exists, what should we do about it?
The problem seems to be centered around the DEC word lists. From my
reading of the license, it seems that the original author of the DEC
list make a serious good-faith effort to identify the contributors to
the list. I doubt that tracking down the remaining authors would be
practically possible, given that we are now removed by two additional
layers of incorporation from them.
In addition, it is worth noting that DEC was a US corporation, and
remains US-owned. I also note from Kevin's site that he lives in the
USA. Finally, Debian's main sites are located in the USA. The Database
Directive of the EU, therefore, does not have the impact that, for
example, the USA crypto laws have.
Finally, I don't know how old the Database Directive is, and whether its
terms are retroactive. This may have an effect on the status of the DEC
list if it can be established that it was created before the Database
Directive went into effect.
So, given that we have:
- a good-faith effort to respect rights that may not even have existed
at the time,
- no way of identifying additional authors besides those contacted by
- a significant burden of proof on the part of any claimant, given that
said claimant did not respond at the time of DEC's inquiry in addition
to the ease which alphabetized word lists may be generated,
- significant jurisdictional issues surrounding the applicability of
the Database Directive to DEC, the upstream author, or Debian itself,
- a significant likelihood that we would be joined in legal liability
by Hewlett-Packard, the current owners of DEC's assets and a friendly
ally of both free software and the Debian project,
I vote that we treat the copyright to this list the same way we treat
patents generally: wait for someone to complain before pulling the
list. The situation is analogous; just as we cannot know which patents
we currently infringe upon, given the volume of patents and the volume
of code we distribute, so here we cannot know which copyrights we
infringe upon, due to our disconnection from their original authors.
At any rate, it seems unlikely to me that anyone will be able to hold us
responsible for willful infringement given the circumstances surrounding
any holder's disregard for his/her rights up to this point.