[Printing-summit] [Desktop_printing] Ghostscript leading edge is now GPL!

Till Kamppeter till.kamppeter at gmail.com
Thu Aug 17 09:21:26 PDT 2006


What about the future of GhostScript? Nothing heard for a long time.

I think we should decide on something now. ESP GhostScript as used by
all distros got somewhat old and has bugs which are fixed in GPL
GhostScript 8.54, for example the colors are off when printing on color
laser printers with Rick Richardson's foo2zjs drivers (Rick, can you
tell some more details about it), GhostScript is segfaulting in certain
cases with IJS drivers (Todd?), ... I think making GhostScript 8.54 the
standard GhostScript as soon as possible is much better that if bug
fixes are backported to ESP GhostScript. But we should get somehow all
drivers into GS 8.54 to avoid the need of every distro doing tons of
patches independently.

It would be best to add everything to upstream GS and to close ESP GS
instead of forking GS 8.54 to ESP GS. WDYT? See also my earlier posting
below.

   Till


Till Kamppeter wrote:
> Great news!
> 
> So I think we (mainly the GhostScript team, Mike Sweet, me) must plan
> now how to make the GhostScript for the future distributions.
> 
> Some ideas:
> 
> - Move all modifications done in ESP GhostScript into the HEAD branch of
>   upstream GhostScript.
> 
> - Drivers added to ESP GHostScript are probably not very difficult to
>   move over, as they are in the addons/ directory of ESP GhostScript
>   (http://svn.easysw.com/public/espgs/trunk/addons/). This directory can
>   be easily excluded for commercial builds.
> 
> - The "cups" backend should go into a cups/ directory which also should
>   be easy to exclude for commercial builds.
> 
> - Later on, the drivers in the addons/ directory (or nearly all built-in
>   drivers of GhostScript) should be decoupled from the core GhostScript,
>   for example by making one or more OpenPrinting vector driver modules
>   carrying these drivers. For that a piece of code needs to be developed
>   which on one end plugs into the "opvp" backend of GhostScript and on
>   the other end offers the internal driver API for GhostScript's
>   built-in drivers. This way (often unmaintained) legacy drivers do not
>   need to be adapted to changes in the internal API of GhostScript. Also
>   if there are third-party drivers with questionable license we will
>   have no problems any more. The new GhostScript developer could perhaps
>   also work on that.
> 
> - Bug fixes done in ESP GhostScript must be checked whether they have to
>   be done also in upstream GhostScript.
> 
> - Open bugs of ESP GhostScript (http://www.cups.org/espgs/str.php)
>   should be checked and merged into upstream GhostScript's Bugzilla.
> 
> - It would be nice if the developers and contributors of ESP GhostScript
>   (Mike Sweet, me, Epson Avasys, ...) get commit access to the CVS or
>   Subversion of upstream GhostScript.
> 
> WDYT?
> 
>    Till
> 
> 
> Bastian, Waldo wrote:
> 
>>Some nice news on the ghostscript front. Ghostscript used to have a
>>staged release process where new releases where first released under the
>>AFPL and only after a year under the GPL. This approach led to a fork by
>>people who needed to get patches in the GPL version. It’s great to see
>>that this has been resolved.
>>
>> 
>>
>>See http://advogato.org/person/raph/diary.html?start=411
>>
>> 
>>
>>Waldo Bastian
>>
>>Linux Client Architect - Client Linux Foundation Technology
>>
>>Channel Platform Solutions Group
>>
>>Intel Corporation - http://www.intel.com/go/linux
>>
>>OSDL DTL Tech Board Chairman
> 
> _______________________________________________
> Desktop_printing mailing list
> Desktop_printing at lists.osdl.org
> https://lists.osdl.org/mailman/listinfo/desktop_printing
> 
> 




More information about the Printing-summit mailing list