Re: mxe-octave stable-octave

[Thorsten Liebig]

Am 02.01.2014 21:41, schrieb John W. Eaton:
> On 01/02/2014 02:26 PM, Thorsten Liebig wrote:
>> Thanks for the update, I have build the stable version without any
>> problems.
>> But I have not had the time to test it on some windows machine.
>> If anybody is interested, you can download it here:
>> http://www.openems.de/download/octave/octave-3.8.0.exe
>>
>> Btw: The mk-dist script should create a more suitable name?
>> octave-installer.exe is a bit too generic for my taste...
>> And the name for the portable zip version
>> ("stable-octave-2014-01-02-12-02.zip") is even more stupid?
>> Maybe just "octave-3.8.0.zip" ?
>>
>> Will there be a official windows build for download on octave.org soon?
>> IMO it is a big mistake that there is still nothing to download for all
>> the potential new Octave users ...
>
> Please take a second and think before saying things like "more stupid"
> and "big mistake".  There's really no need to tell other people who
> are working on Octave that their choices are "stupid" or "big mistakes".
I agree, and I apologize! It sounded much harsher than I meant it...
>
> Now, about the merits of your complaints.  Sure, the installer and zip
> files could have a better name.  I chose to use a timestamp because I
> was often making more than one in a day for the same version of Octave
> and I needed an easy way to distinguish.  Given that requirement, does
> it still seem like a stupid choice?
It makes perfectly sense for the dev/default version or tree and is
absolutely not stupid.
But for the stable version I think there could be a more useful naming
convention?
And I really mean that only as a suggestion ;)
>
> Yes, we would like for there to be an "official" binary for Windows
> available from the Octave web site.  But it's not ready yet.  The last
> time I checked, the installer didn't create a desktop shortcut or
> include any compiled packages.  I think those things need to be done
> before we distribute it.  Would it cause more trouble to delay
> distributing an installer, or to distribute one that doesn't really
> work very well or provide features that nearly everyone would expect?
I think we agree that distributing something not working is even worse
than not having one.
What I just meant is, that in my own opinion I consider it a problem,
that we don't have a perfectly working Windows installer yet.
Because the announcement (hopefully) created some attention, but a lot
of users will be disappointed that they cannot just try it out right away...
 
>
> jwe
>
>