Date: Fri, 12 Dec 2025 20:08:37 -0600
From: "G. Branden Robinson" <[email protected]>
Message-ID: <20251213020837.5rgje3ljltequl4u@illithid>
| Reading and study are acts constitutive of neither plagiarism nor
| copyright infringement.
No, but I also maintain a shell - a competitor to bash if you like,
and were I ever to read the bash code, and later it should turn out
that some code in the shell I maintain looks just like some code that
was in bash, earlier, then it would be hard to defend that, wouldn't
it? But as long as I can truthfully say that I have never looked at
the bash code, that's not going to be an issue.
| Who or what is constraining your freedom to study Bash's source code?
The above (and me).
| I observe that at least one readline clone has been promulgated under a
| BSD license.
editline, aka libedit. yes.
| Nobody is better situated than Chet to tell us if and how
| the FSF attempted to shut down any such effort.
There's no problem duplicating a design - it would need to be patented
to prevent that. I can implement features I hear described (or read from
the man page) that bash has implemented too (as bash has done from ksh).
I have (and had) nothing to do with the editline design or coding - and
also no evidence to suggest that it was copied (other than by features)
from readline's code. Unless there's some evidence that it was copied,
there's nothing anyone can do - whether or not they want to.
Further I would expect that lots of people who slap a GPL licence on
stuff they create would never consider actually taking action against
someone, no matter what they did with the code, most developers of open
source code (I believe) actually believe in sharing openly without
restriction. But as long as the GPL imposes the restrictions it does
(which get worse in each iteration) there's always the possibility that
someone might decide to enforce it.
The same with the groff and clones you mention - unless there is something
in there that's been patented, there's nothing anyone can do to prevent
someone else from implementing a clone. What they can't do however is
just copy the code, slap a new name on it, and call it theirs.
It certainly would be possible for someone to work in both camps, but in
that situation it is very difficult to avoid actual copying, even if it
is entirely subconsciously done. Something seems to be a good idea, that
you don't realise was because you saw it a month (or year) ago in someone
else's code, and had forgotten - but your subconscious remembers what it was
like, and regenerates almost the exact same thing. That's still copying.
If you have ever seen it, and you produce something very similar, then
you are facing potential trouble - avoiding seeing it is the cleanest
way to avoid that.
| So, what gives? Do you have a well-founded reason for refusing to study
| Bash's source code, or have you been drinking from the same pitcher of
| GPL-hysterical Flavor-Aid as Charles Hannum?
I'm not sure what his opinions are, never asked, but I object to any licensing
which limits free use - which the GPL certainly does. Attempting to force
access to sources (however noble) is limiting freedom, not advancing it.
kre
ps: I can only reply via the list, as gmail refuses to accept mail from
me (from my MTA) - it an absurd attempt to control how mail is supposed to
work, for no rarional reason I can see, except promoting (close to)
monopoly control over e-mail. Apologies to everyone else, who is not
interested in this, which has nothing whatever to do with bash bugs.
