Re: Fuzzy machine learning framework v1.2

["Yannick Duchêne (Hibou57)" <yannick_duchene@yahoo.fr>]

Le Sat, 02 Jun 2012 12:52:02 +0200, Georg Bauhaus  
<rm.dash-bauhaus@futureapps.de> a écrit:
>
> Valuations of either type of licenses, MIT/BSD or GPL, assume
> fraudulent entrepreneurship.
Hu?

> There is enough evidence that fraud
> in business is a reasonable assumption.

The following if under conditions I did not erroneously understood your  
words. If the opposite, fix me.

That's the king of assumption (which indeed often come with the GPL  
culture), which make some people look at this license with amusement.

Smalltalk: Some business are fraudulent, not all. Further more, the more  
little your business is, the more you can't be fraudulent. To be  
fraudulent, you have to be able to keep your customers captive, even if  
they don't enjoy you; the kind of thing a small business is unlikely to be  
able to.


> Let product P be a work combined from Foreign_Source and Our_Source.
> P is to be conveyed. In what follows, "theft" should also stand for
> a violation of terms and conditions.
>
> (1) Type MIT/BSD licenses enable closing the source of P and are used
> to prevent theft of privately owned goods, namely theft of Our_Source.
>
> (2) Type GPL licenses require keeping the source of P open and are used
> to prevent theft of privately owned goods, namely that of Foreign_Source.
>
> Type MIT/BSD emphasizes technical prevention of theft,
> Type GPL emphasizes legal prevention of theft.

On the average, I understand the same too, that's indeed a part of their  
respective properties.

> The difference is one of perspective (licensor - licensee).
>
>
> Proof of (2) [GPL prevents theft of Foreign_Source]:
>
> Law abiding entrepreneurs will not choose Type GPL Foreign_Source
> (unless they can get special permission) if P should be conveyed
> without source. Since they don't use Type GPL software, they don't
> steal it.  This shows that the prevention of theft of Foreign_Source
> is effective IFF parties are law abiding.

Yes

> A mere desire to include Foreign_Source under different terms and
> conditions is not sufficient to warrant criticizing the licensors
> for *their* *choice* just because you don't like it.

Yes, there's no excuse for that, but this leave enough room to discuss  
licenses properties.

By the way, an opinion I forget to tell before: one of the property I  
don't like with the GPL, is that it's tricky to me. A permissive license  
may lack some precision, as long as it does not have too much  
consequences. A restrictive license, has heavy consequences, and because  
it has heavy consequences, it should be very precise and explicit.  
Unfortunately, although its text is already rather long compared to  
others, it is not to me. If I find a good starting point with another  
message from this thread, I will make it an opportunity to ask for the  
questions (lack of precision in the GPL) I have in mind. May be someone  
here will have enough knowledge and experiences to suggest me answers.

> Last but not least, I find it telling that the thread has little
> to say in favor of licenses that will allow closed source, but that
> cost money.

As it happens to froggies from time to time, I could not clearly  
understand this sentence (cheese and apologizes).

-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University