🚀 go-pugleaf

I recently posted an application and some code that showed how to
achieve draft-quality printing from the Macintosh on most of the world's
printers.  The only thing my code requires is that the printer be able
to generate a new line upon receipt of any of CR, LF, CRLF.  With
embedded control characters in a document, access to a printer's native
fonts is also possible.  Baud rates from 300 to 19200 are supported.
The program and source are free.  I received the following communication
as a result of my posting:

>IM tells how to write applications that do not depend on the printer. In
>particular, if you use the appropriate printer control calls rather than
>sending your own CR/LF etc., it should work on all printers that have drivers
>written according to rules. Remember - the rules are there to make portability
>easy to achieve.

(Signature deleted by me.)

I purchased the Macintosh about two years ago, and I have yet to see
the printer driver that would allow any Macintosh application to 
print on my printer (a Tandy DMP-110).  The problem was in the main
that this printer executed a carriage return and a newline upon
receipt of CR (13, \015, \n, 0x0D, whatever), and another newline
upon receipt of LF.  If I followed the "rules" when I programmed
printing into my application, I would now have a program that prints
on the ImageWriter and LaserWriter, neither of which are owned by me.
I would also have a perfectly good Tandy DMP 110 gathering dust.

I have investigated what it takes to write a printer driver that
follows the rules in Inside Macintosh: rewrite QuickDraw for my 
printer.  I am aware that there exist printer drivers for SOME
third-party printers, would anyone who wrote one care to give 
an estimate of the man-hours involved?

I have read the sections in Inside Macintosh on device-independent
printing, and I must say that the objectives are noble, and the 
concept is admirable.  However, for most of us the only printer
drivers that exist are those released by Apple.  Device-independence
is a fine idea, but it falls flat when only two or three devices
can be addressed by the device-independent code.

Is it just me, or do others share this beef?