ChangeLog patch

[Craig Burley]

>  In message < 199709241901.PAA19059@melange.gnu.ai.mit.edu >you write:
>  > >Here's a simple one...
>  > 
>  > Heh, yeah, I also noticed the 2.7.2.3 release's ChangeLog entry
>  > has a day that disagreed with the date.  (Aug 22 was *not* Thursday.)
>  > 
>  > Somebody's gotta convince HJ to try out these newfangled "computer"
>  > things that do much of the typing automatically.  ;-)
>Don't be too quick to blame this on HJ (or anyone else); it's
>quite possible it's my goof since I usually have to generate
>most of the header myself and just append the submitter's
>name & email address.

I thought HJ took care of the long-needed 2.7.2.3 release -- keep
in mind I have yet to download *any* egcs stuff, as I can barely
get on Internet long enough to read, maybe answer, email, occasionally
(when the Cygnus modem pool is willing to answer ;-) news, hardly
ever any large-scale ftp'ing anymore.  (And the prospects for
fixing this are ever more dim; the Alpha UDB I was planning to
turn into my ISP-access system can't even boot RedHat Linux 4.2
in under 25 minutes, though this is probably a RedHat problem --
it's taken me two days just to get their email support system to
*accept* my email, who knows how long it'll be before I'll get
a useful response, etc. etc.)

HJ did do a decent job of 2.7.2.3 AFAIK; his only mistake was
in assuming 2.7.2.2 had consistent .tar.gz and .diff.gz packages
which, it turns out, was not the case, the result being that
it's difficult to end up with a correct ChangeLog file using
vanilla FSF distributions (unless one accepts fuzz from patch
by going from 2.7.2.1.tar.gz to 2.7.2.3 via the patches).

Making releases is not a job for the quick of finger (like me --
I have a severely-in-need-of-clean-rewrite piece of paper with
all sorts of instructions I must *carefully* follow, which I learned
by making mistakes, which I do about every other g77 release or so
despite all the notes I've taken ;-).

        tq vm, (burley)