GPL compliance

[ubuntu at rio.vg]

Daniel Carrera wrote:
> On Fri, 2006-30-06 at 11:24 -0400, ubuntu at rio.vg wrote:
>> I don't see it as all that fuzzy, and if anything, it's looser, not more
>> restrictive.
> 
> I don't know, IANAL. Far be it for me to say whether it's looser or more
> restrictive. I honestly don't know what constitutes a "customary
> medium". It might be that you are right and on-line distribution is
> enough. And I do agree with you that whether this medium is convenient
> or not to the user is probably not relevant. But I wouldn't bet the farm
> on what constitutes a customary medium.
> 

I'm quite certain that the number of programs downloaded off the
internet per day is quite sufficient to comply with the "customary
medium" clause.

>> The requirement, in simple terms is any one of the following:
>>
>> a) Provide the source code right along with the binaries.
>> b) Provide the source code somewhere else.
>> c) Point to a 3rd party that will provide the source code.
> 
> I hesitate to summarize the GPL so much, given that we are trying to
> figure out a relatively fine point of the license. If I was trying to
> explain the GPL briefly to someone, I might say something similar to
> what you said, but your definition loses the details; and it is the
> details that we are talking about here.
> 
> I disagree with your summaries for (b) and (c) btw. If I wanted to make
> them almost as short, I would say:
> 
> b) Be willing to give the sources for the cost of the media for 3 years.
> c) If you got the binary under (b), then point to the third party that
> gave you that offer.

The three year part of it and the cost are there, but immaterial to our
current discussion.

I'm not following your "c" description.  We're talking about methods of
ditributing the source, yet you're pointing to "b"...

>> Don't try to parse a legal document as if it was a C program.  When it
>> gets to court, standards of reasonableness apply.
> 
> Maybe so, but then it depends on whether a judge considers online
> distribution of sources the same as accompanying the binary with the
> sources under a customary medium (when what you got was a CD). He might
> decide it's the same, but I wouldn't bet on that.

The question isn't "is it the same".  The question is simply "Is the
internet a customary medium?"  Under 3(b), the method of distributing
the binary and source are completely decoupled.  Equivalence is not
required.

>> Someone trying to suggest
>> that the internet is not often used to download programs would be
>> laughed at.
> 
> I'm sure you've seen a few arguments that you thought were technically
> laughable go through in court.

Many laughable arguments are made, but all things considered, few
actually get anywhere.  The media loves to sensationalize when it
appears that absurd arguments are getting traction, but it's actually
quite rare compared to the thousands of court cases going on all the
time, and many, perhaps even most of those sensational cases aren't as
sensational when you get past the headline and look at what's really
going on.

For instance, the much laughed at "McDonald's coffee spilled in a
woman's lap".  The headline is easy to make fun of.  But if you look at
the case particulars, the McDonald's in question was heating it's coffee
to and absolutely absurd degree.  The woman suffered 3rd degree burns.
There was a serious issue there.

Another reason for this misconception is how long cases can be.  A
litigant can make absurd statements in court for quite a long time as
the case goes through, because nothing is decided until the end.  At the
end of the day, the absurd guy virtually always loses.