🚀 go-pugleaf

Something valuable is getting lost here.  I thought I understood the
FSF philosophy: information in general (and software in particular)
should be freely redistributable, to encourage the cumulation of
innovation.

But what does this have to do with proper crediting of one's ideas?

Referencing sources has several purposes:

1. Giving credit.

   This is elementary courtesy.  It is also an encouragement to
   publish ideas -- especially valuable if you don't pay authors.
   (Many GNU files record their authors' names.)

2. Permitting debate.

   The sources of an idea are useful for discussing it.  In the Gawk
   case, it would be useful to compare it (and its documentation) to
   other versions, whether they are "free" or not.  e.g. "Gawk is
   better than Awk (REF) because..."

3. Tracking correctness

   If you know where code or algorithms come from, you can find
   improvements and bug fixes.  For instance, the algorithm used in
   search.c for fixed-string search is the Boyer-Moore algorithm.  The
   original published version of this algorithm contained a bug.  In a
   very incomplete prototype I wrote (Dick King wrote the production
   version), I had the following comment:

   /* Boyer-Moore search (CACM 20:10:762) as implemented by Knuth
      (SIAM J. Comp. 6:2:340ff) and corrected by Mehlhorn in 
      G de V Smit (Software--P&E 12:57 '82) */

   I notice that the 18.55 source doesn't even mention Boyer and Moore
   (although it uses the mnemonic BM_tab).  (In fact, it uses a
   better, streamlined version of Boyer-Moore.)

The traditional way to distribute information is to publish it in a
book or journal.  Perhaps electronic publishing will someday displace
paper publishing.  But today, publishing with Prentice-Hall or in CACM
or Software--Practice and Experience disseminates information far
better than including it in a GNU source file.  Indeed, it's a pity
that more of what FSF does isn't published -- I'm sure there are a lot
of things people could learn from FSF software.

And publishing a copyrighted book or paper does NOT preclude others'
using the ideas; in fact it encourages it.

So benefit from published information, and give credit where credit is
due!

	-s