Re: [tlug] linux in Japanese schools

["Stephen J. Turnbull" <stephen@example.com>]

Curt Sampson writes:

 > For some rason, this sort of thing, to me, feels more like "compelled
 > speech" than "free speech."

No compulsion is involved.  That is, "viral" in the sense of common
cold is incorrect.  "Viral" in the sense of AIDS is the correct
metaphor.  You have to engage in unsafe intercourse to "catch" GPL. ;-)

 > Yet the author of readline is saying, "the price for using my code
 > is that you have to give away, for free, a million lines of your
 > own code that you spent the last ten years writing."

As a matter of fact, you're wrong about readline.  Brian Fox has been
given ample reason to regret his assignment of readline to the FSF; he
agrees with you that that price is *way* too high.[1]  However, he
doesn't have any practical say any more (the GPL does have an escape
clause for authors, but he would have to disentangle his code from the
contributions of a couple score of other important contributors).

 > Maybe I'll come up with a better way of explaining this if I think about
 > it for a while.

I like "royalty in kind".

 > I find it hard to think of this as a "side effect" when it's the only
 > major difference between the GPL and BSD-style licenses. Or is there
 > some other difference I'm missing?

The GPL is compatible with no other license in the sense of conversion
*from* the GPL.  This is much more important than having to submit to
having your code deproprietized occasionally, because this divides the
free community.

Do you realize that it is illegal to copy text from Emacs code into
the Emacs manual or vice-versa without the FSF's permission?  (Of
course, you'll surely get it in that precise case, but there are
nearly as beneficial applications where RMS has explicitly refused to
give permission.)


Footnotes: 
[1]  This is hearsay, actually; my source is L. Peter Deutsch on the
Ghostscript beta testers mailing list.