[Printing-architecture] On the Continued Need for PostScript Workflows

[James Cloos]

I've been trying for months to write a whitepaper on this subject, but
haven't been comfortable with the tone I end up with.  This is getting
more important, though.  So I'll try to be succinct, even if that might
come off as blunt.  Apologies in advance for feather ruffling.

*Any* conversion of PostScript into PDF in the print work flow -- ie, in
the cups filters -- is broken and unacceptable.

PostScript files exist; they do not magically disappear just because PDF
is available.

PostScript printers are still more common than PDF printers.  And they
have *long* lifetimes.  (In the US, printers have a five-year depreci-
ation schedule and one can expect a much longer service life.)

When printing an existing ps file -- including when printing from
programs which can generate ps but not pdf -- the postscript MUST be
modified only by pstops when sending on to a ps printer, or MUST be
rendered directly to a raster format when printing to printers which
require raster input.

Conversions can only cause damage.

Artifex refuses to ensure that gs' pdfwriter generates device indepen-
dent colour when given a source file with device independent colour
(evidently it is a large job for which they do not have a paying
customer).  Cf the relevant WONTFIXes.

And it is not always even possible to convert jobs which use postscript
as a language into equivilent PDF.  The formats are just too different.

It is perfectly OK -- even welcome -- to prefer PDF when the original
file is not already in a page description format or language.

But the only molestation a postscript file should endure is that
specified by the ppd and user-selected options.

-JimC
-- 
James Cloos <cloos at jhcloos.com>         OpenPGP: 1024D/ED7DAEA6