[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Desert island test (was: Questions about liblouis)

On Friday 29 February 2008 05:29:19 am Josselin Mouette wrote:
> Le mercredi 27 février 2008 à 18:13 -0800, Sean Kellogg a écrit :
> > And not grounded in the specific language of the DFSG but rather a shared
> > aspiration of what the document "ought" to say. I have never seen an
> > attempt to tie the three tests to specific points and thus it is
> > impossible to debate and discuss the test themselves... it has become
> > assumed knowledge.
> The tests are not meant to be extra guidelines, nor reformulations
> specific guidelines, nor assumed knowledge. As the name says, they are
> tests: extreme situations in which you can test how the DFSG (all of
> them) apply to a piece of software in a specific case.

There is a saying in the legal profession... bad facts lead to bad decisions. 
As a developer, this was hard for me to swallow. Any judicial decision should 
be able to handle any set of facts and universality should rain supreme. But 
the more case law I read the more I understood. When a ruling is first handed 
down for a novel issue it sets the baseline by which all subsequent cases 
will be judged. If the baseline is established for the worst case scenario, 
you get a decision that is so hardline, so reactionary, that the decision is 
simply unfair for those whose facts aren't as egregious.

We see this all the time with U.S. Supreme Court cases... an issue 
organization will select the most sympathetic plaintiffs from a group of 
potential plaintiffs and use their situation as the means to persue a 
judicial remedy. As an example, you don't fight for p2p rights using Pirate 
Bay as your plaintiff, you use Project Gutenberg.

The problem is the tests try to give us a judicial decision using the worst 
possible facts and then say: this is the baseline by which all subsequent 
decisions must be made. I, for one, am thankful the world beyond debian has 
rejected reactionary rule making of that nature. I don't have a lot of hope 
for debian though, as it's an organization made up of developers... insert 
obligatory quote about hammers and nails here. But, everynow and then, I 
figure it's worth making a little noise.


Sean Kellogg
e: skellogg@gmail.com

Reply to: