In reply to Stefan Seefeld <[EMAIL PROTECTED]>
>
>[EMAIL PROTECTED] wrote:
>>
>> that would be the ideal situation, but would require modifying gs
>> itself. i don't have time for that, but it's a more than worthwile
>> goal.
>
>not gs, but a gs driver. That's the main idea behind every modular
>system.
yes, but i think you didn't get my point. as far as i know gs needs to
be compiled with the driver itself, that is, drivers are not
dinamically loadable --- which for me means that gs is not
`sufficiently modular.' this means that either you reinstall gs by
yourself (which only solves your case) or you convince every linux
distribution on earth to compile your driver in. summarizing, you need
to get the driver in the official gs distribution. that's not what
happens with ggi, berlin, for example.
in this respect the modularity of gs is somewhat useless. consider
gnuplot, for example. it has a ggi driver (i wrote it!). can any of
you use gnuplot with the ggi driver in your system? not really unless
you recompile gnuplot by yourselves. and it is not guaranteed that the
linux distributions will compile the drive in when the new version
comes. gnuplot is 'modular,' but the kind of modularity it has is
somewhat useless for end users...
>> i think having gs to get to know ggi is a worthwile goal for the ggi
>> project itself. isn't somebody willing to try?
>
>no, gs shouldn't need to care about ggi as little as gs should need to
>care about X. gs should have a 'driver interface' (much like ggi has
>an (internal yet abstract) visual interface) and one such driver would
>know about ggi visuals...
agreed. but that't not the way it works in practice, since you need
the gs maintainer and the linux distributions to include the X driver
by default. if i write a ggi driver it will be at least months before
non-hacker people can use it normally, and that's not counting for
installations that don't get upgraded often.
>Yes, indeed: big prizes for the first to write a ggi driver for gs ! :)
idem ibidem. and while this guy's at it, let's lobby for dinamically
loadable modules in gs and everywhere else in the world.
--
Cesar Augusto Rorato Crusius __o __o __o __o __o
Stanford University _`\<, _`\<, _`\<, _`\<, _`\<,
e-mail:[EMAIL PROTECTED] (_)/(_) (_)/(_) (_)/(_) (_)/(_) (_)/(_)
www.stanford.edu/~crusius
o _ _ _
__o __o __o __o /\_ _ \\o (_)\__/o (_)
_`\<, _`\<, _`\<, _`\<, _>(_) (_)/<_ \_| \ _|/' \/
(_)/(_) (_)/(_) (_)/(_) (_)/(_) (_) (_) (_) (_)' _\o_
He who sacrifices functionality for ease of use
Loses both and deserves neither
