Re: 3.0.5 ?

[Carlo de Falco]

2009/4/7 Jaroslav Hajek <address@hidden>:
> On Tue, Apr 7, 2009 at 11:15 AM, Carlo de Falco <address@hidden> wrote:
>> 2009/4/7 Jaroslav Hajek <address@hidden>:
>>
>>> Well the fix is available in the archive, so anyone can download it to
>>> patch the sources.
>>
>> yes, but the info on the web site seems to indicate it's "safe" to use
>> the stable release
>> as it is been tested for long time and the developers consider it more
>> "reliable" than the development version..
>
> "Safe"? I hope not. What site you're referring to? Use of Octave is
> always at your own risk, there's no guarantee of any sort. The main
> Octave site says the stable release is "well tested", which it is

You are right, the website only states that the stable release is
"well tested and recommended for most users".
 I actually got my impression from comparison with the description of
the development snapshots, but I was mistaken.
I think the descriptions are actually accurate as they stand.

> unfortunately, here a bug was created in the very last stage of the
> testing and fixing. Or maybe it didn't get enough testing, but I think
> that Octave is supposed to be tested by its users.

and indeed it was a user who reported the bug just few days after the release.
Also of note is the time that intercurred between the bug report and the fix.
This was just a couple of hours.. and on a Sunday!
I bet that technical support of the competion would never be able to
provide this level of responsiveness!
I guess this should be listed as one of the pro's in those
presentations comparing Octave to its proprietary counterpart..


>> Personally I like the idea of just re-releasing 3.0.4 if this does not
>> mess-up the archives.
>
> I dunno whether it's acceptable for GNU/Linux distributions.

well, 3.0.5 is just fine then...
>
>> IIRC this sort of quick bug-fix release has been done before, does
>> anyone rememer what was the procedure in previous cases?
>
> Yes, an advice would help here.
>

c.