Re: flexfax: Mail to Fax

[David Woolley <david@djwhome.demon.co.uk>]

> >The sendmail shell script comments contain a "." in the CPfax line

Actually it doesn't in 4.0pl2, although multiple people have reported
problems including the . in the problem report, so I wonder if someone
has written some sort of HOWTO document without testing it.

> >and have a dot missing after the token "fax" in the line that
> >invokes the fax mailer; however people should just use the sendmai.mc
> >macros, which, I believe, get it right.
> 
> One might consider, in such a case, grabbing the CVS to check if this bug 
> still exists, and submitting a patch to patches@hylafax.org. Then again, one 
> might not have the time, in which case thanks for mentioning this.

I'm not on the developer list, so I'm not particularly aware that this
CVS facility exists, and in any case I have already reported the problem
on the list once and possibly twice in response to previous problem reports
(one report may have been a back reference to a previous one).  Depending
on when public CVS was started, it is possible that the problem was reported
before CVS.

Chasing CVS for a problem that is not my problem (I only use aliases, not
the .fax domain) would cost me in phone charges or involve the misuse of
my employer's time.  In fact, I don't really use Hylafax at all, as
the system manager in the office went for Cheyenne, because he has the
usual fear about complexity with Unix.  I just follow the list now
because I feel that I am able to answer questions that others don't 
answer - although the quality of questions seems to be going down and
in recent weeks others do seem to have started answering.

Hylafax is very low on my list of things that I might want to develop
in my own time, and if the choice is between spending development time
and not answering people on the list, it is probably going to be the
latter option that is selected.

> Poppycock. That 4.1beta2 is solid on RedHat linux in no way qualifies it
> as an overall release candidate. When it builds on all of the major UNIX
> distributions and is confirmed solid then we release - not before.

In that case take "beta" out of the name of the Red Hat version and
treat it as a branch stable version.  Calling it beta, if it is really
stable, just re-inforces the impression given by people like Microsoft
that beta quality software is suitable for mass market use.

I have seen cases where people have created RPMs of development versions
of the Squid proxy and then not fielded the resulting support problems
resulting in them being asked on the squid-users mailing list, in spite
of a list charter that forbids such questions; my general impression is
that Red Hat encourages the distribution of unstable software, at least
by default of vetting contributed material.