[sane-devel] SANE2, what do we want ?

[Johannes Meixner]

Hello,

On Mar 28 22:49 Julien BLACHE wrote:
> Till Kamppeter <till.kamppeter at gmail.com> wrote:
> 
> > ago and no one answered). This will help us that it is much easier for 
> > scanner manufacturers to ship drivers with their scanners. They can 
> 
> This is all but a good thing. Currently, binary backends provided by
> the manufacturers are nothing more than a total annoyance.
> 
> They're crappy, non-debuggable, under-documented, badly written, badly
> tested, and available for Linux/i386 only.
> 
> Encouraging binary backends undermines the efforts done by backends
> developers over the past years to obtain specification/documentation
> from the manufacturers.
> 
> Think twice, please. You're actually actively harming both SANE and
> the users by doing so.

Could you please avoid such harsh wording and be gentle
and assume that those who post something are no idiots.

What have broken binary drivers to do with a LSB standard?

Did Till write that the LSB standard applies only for
broken binary drivers?

For example the hpaio driver from HPLIP is free software.
For example the epkowa driver from IScan is free software.

I would appreciate it if there were more free drivers from
manufacturers but manufacturers ask for reliable standards.
If SANE1 was LSB, there would be such a reliable standard.

If SANE1 was LSB, it would not mean that there cannot be SANE2
but it would mean that a nice migration path from SANE1 to SANE2
would have to exist or that SANE2 is backward compatible to SANE1
(software for SANE1 would have to work in a SANE2 environment).


Kind Regards
Johannes Meixner
-- 
SUSE LINUX Products GmbH, Maxfeldstrasse 5, 90409 Nuernberg, Germany
AG Nuernberg, HRB 16746, GF: Markus Rex