[Math-atlas-devel] 3.3.15 lurches out the door

[R C. W. <rw...@cs...>]

Guys,

OK, this is the fourth consecutive last developer release, 3.3.15, and this
time I mean it, or at least I might very well do so far all you know . . .

I've fixed (or avoided) all so-far reported errors, and have updated the docs
about as much as I see myself doing.

As before, developers should at least ensure that AtlasCredits gives them
the creds they want/deserve.  Now's the time to scope all the docs, point out
errors, do any testing you are going to do, etc (I already noticed that 
ChangeLog is not as up-to-date as that provided by sourceforge).  If I get
some good vibes from this guy, it's transforming into 3.4 in short order; if
I get an afternoon free to work on it, it might happen any afternoon from
tomorrow on (with the way it has slipped, I sometime think my grandchildren
will finish 3.4, just in time to run on the new 'hammer series :) . . .

I have new PPCG4 (linux) defaults, but don't have access to such a machine
at the moment, so if you do, please weigh in with it.  

For any testing, sanity_test is the min, and the full tester is the max.

On other fronts, I resigned from the atlas-comm mailing list today, in
an attempt to cut down on the 10-50 spams I get a day, so you will want to
send to the sourceforge developer list if you want me to see it.  If you
remain on atlas-comm, you might forward particularly worthy mails to
sourceforge if it is not cross-posted . . .

Cheers,
Clint