[Openicc] Drop size calibration

[Robert Krawitz]

   Date: Mon, 4 Feb 2008 12:59:42 +0100
   From: "edmund ronald" <edmundronald at gmail.com>

   The guys I need to talk with at Xrite are offline this week. I will
   have a chat with them *again* about the open source issues when I
   manage to chase someone down. What I am telling Xrite is that
   supporting Robert's efforts to color manage his drivers is
   commercially useful because they are getting to production quality
   and will be widely used.

Thanks!

   It would be useful if people in the open source community also
   realized that a solution needs to be found for open-source *users*
   who want to use spectros.

I agree, but first things first -- let's figure out how to use them,
and then figure out how to put together tools necessary for a
reasonable workflow.

   It seems that in the past, from what people have told me off-list,
   Xrite were willing to designate a maintainer who would distribute
   binaries, maybe one distribution could be chosen for users who just
   want to get on with their work, and act as a reference
   implementation under which everything *just works* ?

Do you mean one particular Linux distribution?  I don't think that
that's going to fly very well.

   Frankly, I think one could for the benefit of everybody consider
   that the driver of a piece of hardware is a part of the hardware,
   and if the supplier of the hardware is prepared to somehow finance
   or encourage the driver development and affirm they will support it
   then one should not oppose this on purely philosophical grounds.

Proprietary, binary-only drivers (if that's what you're talking about)
have a lot of *practical* problems beyond the purely "philosophical".
The ATI and nVidia proprietary X drivers, for example, are reported to
cause a lot of system instability, but nobody can go in and fix them.
Furthermore, they're subject to kernel and user API changes (what if
upgrading your system breaks the driver?) and are typically released
for only one platform (32-bit x86 is the most common), so if you're
running on some other platform, you're out of luck.

Epson has a Linux driver for many of their printers, PIPS.  This has a
proprietary core.  If I took that attitude, I wouldn't bother with
Gutenprint.  The reality is that I suspect that without Gutenprint
Epson wouldn't have bothered with PIPS.

-- 
Robert Krawitz                                     <rlk at alum.mit.edu>

Tall Clubs International  --  http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail lpf at uunet.uu.net
Project lead for Gutenprint   --    http://gimp-print.sourceforge.net

"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton