Thomas Weber wrote:
> On Fri, Jul 20, 2012 at 07:12:56PM +0200, Philip Nienhuis wrote:
>> Thanks Thomas for an enlightening post. Indeed, there are more
>> perspectives and interests than just Octave users, developers etc. But I
>> have ambiguous feelings on this subject.
>
> <snipped rest of mail>
>
> I didn't want to change your decision[1].

My decision?
Not sure what you mean, but I didn't build the MinGW installer; the 
credits should go to Nitzan Arani.

>                As every non-trivial decision,
> building an installer and chosing its content is a compromise. As long
> as people are aware of what they are actually chosing, that's okay with
> me.

Nitzan made an installer that allows about anything, including building 
.oct files and binary packages, etc.
Soon after Tatsuro compiled Octave 3.3.91 for MinGW I put together a 
minimal but complete setup for a self-contained working win32 Octave 
environment for myself. That allows me to tell from a comparison with 
Nitzan's current binary that the overhead in the latter really isn't so 
very big.
To be able to run Octave and compile a few procs, a lot of MinGW stuff 
is needed; that is unavoidable for an environment like win32 that is 
somewhat alien to *nix-based SW like Octave.

To compare: Michael's MSVC version is much smaller (on disk 225 MB vs 
950 MB for MinGW). But to compile s/th with the MSVC version, Visual C++ 
has to be installed as well, and the minimum size of that would be 2.2+ 
GB; of course probably 2.15 GB of which is just M$ bloat.

> [1] Given that I'm uploading a 100+ MB package with only debug symbols
> for Octave with every Debian package of it, that would be hypocritical -
> summed over all architectures and mirrors, I'm probably generating 100
> GB traffic with each upload myself (10 architectures, ~100 mirrors).

Traffic and storage is only one side of the picture.
I once read somewhere that a trivial Google search costs as much natural 
resources as an average household uses in several weeks (or longer). 
Whether that holds exactly true or not, the message is that there's 
really a lot of waste in contemporary use of IT, and especially that 
most of that waste is not at all obvious. In fact, this is something 
that holds for many socio-economical processes these days.

So in a way I was pleasantly surprised by your post, in the sense that 
scarcity of resources is at least being considered (BTW I'm not so much 
of a greenie).

Philip

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Octave-dev mailing list
Octave-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to