[Foomatic] Driver maintainers?

[Grant Taylor]

Stian Sletner <stian at sletner.com> writes:

> I have an old HP LaserJet IIP which works perfectly, and I'll be
> damned if I'll throw away something fully functional over a small
> problem like this.  The problem at hand is that there's a bug
> *somewhere* in the complex chain of events that constitutes printing
> from Linux nowadays, but I have trouble finding out where.  The
> particular problem is described in this post I sent to the hp-list
> four months ago:

> http://www.linuxprinting.org/pipermail/hp-list/2004q2/005234.html

> It went unanswered just like all previous descriptions of this
> problem by other people, that I could find.  It makes me kind of sad
> to observe that most of the questions and bug reports around
> linuxprinting.org go unanswered, owing no doubt to the inherent
> complexity of this domain, but also, I think, to the somewhat opaque
> and confusing structure of the linuxprinting.org site and other
> resources on *nix printing.  

Hmm.  Opaque and confusing I can maybe fix; what was most confusing?

The problem of questions going unanswered I don't know how to solve.
It is the classical problem of self-help forums; nobody who knows the
answers cares to hang out in a place full of questions.  In our case,
the problem is made worse because we get few repeat visitors.  People
just want to know what driver to use or whatever, and then they're
gone again, never to return.

> Possibly, the only people who can accurately answer these questions
> are the people who wrote the drivers, and in the case of my old
> printer, they are probably long gone, and in the case of newer
> printers, I'm sure they don't have the capacity to follow everything
> up.

Exactly correct.  Most drivers - even the good ones, that work well
and have active developers - are effectively unsupported.  Any
programmer can theoretically figure out any driver and fix problems,
but there is often no reason for anybody to do so.  I occasionally do
driver work as a side effect of consulting gigs, a number of more or
less random drivers have gotten "tune-ups" that way.  I am not the
only one; patches occasionally appear from all over and get rolled
into the distributions.

> Some "rot" of old drivers is to be expected as older devices go out
> of use, I realize that.  But I can't help thinking (with all due
> respect) that there may be something to be gained from a more
> transparent system, maybe with clearly defined maintainers for each
> driver, or some such -- because I'm sure I'm not the first dunce to
> be overwhelmed and give up when faced with this type of problem (I
> think this is my fourth attempt to find a solution to this).

Most newer printers that work (ie assorted HPs and Epson) have active
developers, who do participate in, or at least occasionally monitor,
the forums here.  This isn't to say that the drivers are perfect, or
that there is anything resembling "support" for them.  But at least
the odds of it working are a bit better.

If there's a maintainer web page for a driver, we do link to it.  The
"ljet2p" driver is one of the old unmaintained ones, and probably
points to ghostscript.com or the moral equivalent.  This is of course
not very useful...

> As for my specific problem, I'm going to ask just this before giving
> up again. :) Is it worth looking into?  Is it _probably_ an easy fix
> somewhere?  Are there any chances of getting any assistance, maybe
> from some place I've not thought of looking?  What frustrates me is
> that it seems like such a small bug, but I just have no idea how to
> get at it.

If the runtime alignment options don't address the issue, then it
might be either a imageable area misdefinition in the PPD file (are
you using a PPD-aware application to print?), or more likely it's a
misdefine in the driver source code itself.  A third possibility is
that it is a physical limitation of the mechanism that is correctly
handled in the driver, although I assume you've verified this
possibility against some other driver's output?

-- 
Grant Taylor
Embedded Linux Consultant
http://www.picante.com/