I think the key point is to consider what would be the alternative. If
each manufacturer had it's own interface language, then driver
development would be much more difficult, and would require significant
coordination between the printer manufacturer and whoever writes the
driver (which needs to be integrated with the OS). A standard language
gives the OS vendor and the printer manufacturer a reference point. PS
fills this niche quite well (I suspect that some would argue that PCL
also does this job well).
In a world with N printer manufacturers, and M OS vendors, a common
language means you only need M+N solutions, instead of M*N solutions.
Back in the "old days", I wrote a graphics driver for an Anadex DP9501
printer. If I wanted to change to a different printer, I would have
needed another manual (to describe the different set of escape codes,
resolution, graphic data formatting, etc.), and I might even have to
re-architect the driver (e.g. does the printer use polar coordiantes
instead of cartesian, or is it vector vs. bit-mapped, or ???). By
contrast, PS can describe a complex image in a device independent
manner.
The "flies in the ointment" with PS are that different printer features
(like tray source selection, duplex, etc.) can require specialized
code. However, this is a small item, compared to the general case of
describing a page. Of course, these issues must be solved for any
solution.
Hope this helped,
Paul
BTW - No aspersions were received
