On Wed, 14 Mar 2018 11:42:28 +0100
Didier Kryn <kryn@???> wrote:

> Le 14/03/2018 à 11:29, Florian Zieboll a écrit :
> > Hallo Didier,
> >
> > just to avoid confusion: this was not my point, but Menelaos
> > Maglis's. I just tried to figure out that the basics of printing
> > (like most things in computong) are a quite simple thing: pushing
> > ones and zeroes.
> >
> > Besides that: Of course, in bandwidth-limited environments, it is
> > more effective to send some (K)B of vectors over the network, than
> > a >15MB bitmap (letter@only1200dpi) to print the usually quite
> > empty page 2 of some essay and let the printer do the rendering.
> >
> > I guess that's part of why we are here: To restore the original
> > meaning of "economical":-)
>
>     I think we all agree there.
>
>     Do you remember any of these comics where the driver of a car
> opens the motor to repair, throws away a bunch of parts, and then the
> engine starts again and the guy goes away with the car? Here we are
> with Linux. The BIG piece to remove was systemd, but there are quite
> a few others... follow my eyes.
>
>     There are alternatives to communicating through dbus. If two
> processes are necessary, a socket or a pipe can do it. If more
> structured communication is necessary and you don't need two
> processes (why would you in this case), other famous applications use
> a kind of dynamically linked libraries (plugins).
>

LOL, I've used the kill command from one shellscript to another to tell
the shellscript when to look for the next "thing". And sometimes a
simple FIFO is enough, as described in pages 18-20 of
http://troubleshooters.com/linux/presentations/leap_digitizing/leap_digitizing.pdf

That setup between asyncronous producer and modifier programs is as old
as computers themselves.

SteveT