[sane-devel] Going forwared with SANE (was: Re: [announce] coolscan3 release)

[m. allan noah]

On Dec 17, 2007 1:32 PM, Alessandro Zummo <azummo-lists at towertech.it> wrote:
> On Mon, 17 Dec 2007 19:31:04 +0100
> Julien BLACHE <jb at jblache.org> wrote:
>
> > >  eheh.. just kidding, obviously. in fact I'm pretty satisfied the way
> > >  sane works now.
> >
> > There's a real question of how SANE is going forward from there on.
> >
> > It'd be nice if we could answer it at some point.
>
>  we could, but every answer will be different or, at least, slightly
>  different.
>
>  if we answer, who will take the burden of actually choice
>  a single answer?
>

ahh- yet more deja vu :)

I do not wish to hijack this project in any way. Most of you guys have
far more experience than I in programming heavily-used applications.
Oliver's opinion in particular carries some weight, being that Xsane
is probably the most developed and one of the more used front-ends,
and he has put quite a bit of effort into the current SANE2 draft.

My observations:

1. There is a need for more well-known options controlling certain
hardware (ie- adf)
2. There is a need to expose additional image types to specialized front ends.
3. The SANE2 draft is fairly large
4. The number of developers is limited

Frankly, these changes are quite small, not even a new function call,
but if the hang-up is purely a philosophical one around the word
'standard', then what about SANE1.1? Then every packager/frontend
author would have to compile against the new version, and we might
have some systems that continue to ship both versions (even though
they would be identical for most backends).

allan
-- 
"The truth is an offense, but not a sin"