[hylafax-devel] BINARY HUNT

[Jay R. Ashworth <jra@baylink.com>]

On Thursday, August 10, 2000, at 5:42:35 PM,
   Mark Wilson (P. Kay Associates Ltd.) wrote:
> Reclassify 4.1beta2 as 4.1 (or 5.0) and advertise it as such on
> hylafax.org; punters are cautious of trying a beta package and deterred by
> the age of the "stable" package. Progress will follow an expanding user
> base.

4.1, certainly.  And, there's a philosophical question here, I think
that with the current collection of patches applied for a CVS extract,
we're probably close enough for a production release of 4.1; certainly
the 3 machines I have 4.1beta from the RPM (that's 1, right?) running
on seem to have less problems than the one that's still running
4.0pl2.

> Publish a tarball of the current CVS as x.ybeta1; adventurous punters will
> find and solve problems in it, or revert to the current beta if it's a
> struggle.

Concur here, too.

> Merge the devel mailing list back into the user list; I'd guess that the
> devel subscribers also follow the user list, but feel disinclined to reply
> to queries on it.

I'll disagree, here.  There are good and sufficient reasons for the
people working on code to have their own sandbox.  I make a personal
point of doing my best to have at least *something* to say to most
inquirers on -users (at least when the mail *gets* to me; I'm back,
again :-).  Many of the -devel subscribers in fact may have dropped
-users, understanding Sam's code *and* having a day job comprise a
draining enough collection of responsibilities.  :-)

> Punters try hylafax but drop it if they submit problems
> and don't get helpful replies. Talented punters may follow the user list
> but lose interest when they don't see informed traffic. This is
> particularly significant as the devel list is not available as a
> searchable archive. I know there's a pang of guilt whenever answerable
> messages are silently deleted, but getting a thank-you when your reply
> solves it for someone is very gratifying.

Noted.  And I suspect searchability is coming for the other list; I
smell movement this week.  Besides which, my *head* is a searchable
archive.  :-)

> Don't fret about publishing stuff that's less than perfect, so long as it
> ain't a dog. There's plenty of folks who'll make an rpm from a viable
> package, or able to quickly hack stuff that works for linux to work on
> their SCO/HP-UX etc boxes. 

I have SCO and RedHat 5 and 6 to test on, here, in addition to an
Indigo (set up as a Cyclone RIP, but running some form of IRIX if I
can get it to boot... and if it has compilers), and an HP9000 with
HockeyPUX 10.2... don't know if *that* has compilers, either.

So, this seems like a good time to make the inquiry explicit:

ATTENTION BINARY BUILDERS: if you're on this list, and you've ever
built binaries of HylaFAX from source on your non-Intel platform (or
for that matter, your Intel but non-RedHat, non-SCO platform), please
chime in now if you think you'll have a few hours in the next 3 weeks
or so to download a tarball, build a binary, and do some quick testing
(ie: does it compile).  If you can build an installable package for
your platform (IE: RPM, custom, installpkg, etc), so much the better.

I've changed the subject line so these are visible.

As Darren correctly notes, we have here in our laps one of the
rockin'est packages in Open Source.  Let's get down wit' our bad
selves, shall we?  I'm sure Mr. Hewes will make an announcement when a
source tarball is extracted, and where it's put.  Darren, Robert, is
there anything else you need from us?

Cheers,
-- jra
-- 
Jay R. Ashworth                                                jra@baylink.com
Member of the Technical Staff     
The Suncoast Freenet
Tampa Bay, Florida     http://baylink.pitas.com                +1 727 804 5015




____________________ HylaFAX(tm) Developers Mailing List ____________________
 To unsub: mail -s unsubscribe hylafax-devel-request@hylafax.org < /dev/null