Re: Most used words in current buffer

[Nick Dokos <ndokos@gmail.com>]

> Nick Dokos wrote:
>> IIRC, Kernighan & Pike say in the "Unix Programming Environment" that
>> there *was* a `head' program, in addition to the `tail' program. It
>> fell into disuse and disappeared almost immediately after sed became
>> available.
>

Eli Zaretskii writes:

> 'head' is alive and well in GNU Coreutils.  E.g., on a garden-variety
> GNU/Linux system:

Indeed - I knew that but I misspoke - I got buried shortly after I
sent and never got the chance to correct it. What I meant was not that
head is not available today, but that it was available on early Unix
and that (I thought) it went away after sed was added in that same
early Unix (maybe version 6 or possibly earlier). Which obviously
would not explain why it is still around today, but never mind: I was
misremembering things - it turns out that Bob Proulx's "feeble" memory
was better than my apparently non-existent one - see below.

Bob Proulx <bob@proulx.com> writes:
> I think you may be thinking of the 'gres' program.  Again here I am
> not going to look up the reference but instead just reply upon my
> feeble human memory.  But I think you are thinking of the gres
> program, global regular expression substitute, which if that route
> were followed would require a lot of greX programs where X is replaced
> by many specific things and was completely subsumed by 'sed'.
>

Yes, indeed: I had to search a bit in Kernighan and Pike to find it
again, and you are right: they were talking about `gres' and `grep',
not head and tail.

> Also remember that at the time knowledge and daily use of ed (and qed,
> ex, and the others) made using sed very easy.  Lots of shared
> knowledge.  However today that sed may seem arcane to people is just
> that they are no longer familiar with ed.  It no longer has that
> shared learning that made sed so familiar back in the day.
>

Agreed: I knew ed (I actually ported the Kernighan and Plauger ed
clone to a Prime OS machine in the early 1980's, before Prime started
selling its version of Emacs - the line editor they were distributing
was driving me crazy), and could transfer the knowledge to simple sed
invocations. At some point I even learnt the more "advanced" portions
of sed programming, but I never used them enough to retain them.

-- 
Nick

"There are only two hard problems in computer science: cache
invalidation, naming things, and off-by-one errors." -Martin Fowler