Xref: utzoo comp.sys.apple:5045 comp.sys.ibm.pc:13939 comp.windows.misc:422
Path: utzoo!utgpu!water!watmath!clyde!att-cb!att-ih!pacbell!ames!mailrus!umix!uunet!pwcmrd!skipnyc!atpal!tneff
From: tneff@atpal.UUCP (Tom Neff)
Newsgroups: comp.sys.apple,comp.sys.ibm.pc,comp.windows.misc
Subject: Device Drivers and Future OS's (was: Windows for the Blind)
Summary: Those are good examples too, but mine will have a bigger base
Keywords: Microsoft, HP New Wave, Xerox, Lawsuit
Message-ID: <122@atpal.UUCP>
Date: 31 Mar 88 22:47:16 GMT
References: <292@unicom.UUCP> <2112@phoenix.Princeton.EDU> <819@bucket.UUCP> <117@atpal.UUCP> <621@mcrware.UUCP>
Reply-To: tneff@atpal.UUCP (Tom Neff)
Organization: Rational Technologies, Inc.
Lines: 25

In article <621@mcrware.UUCP> jejones@mcrware.UUCP (James Jones) writes:
>In article <117@atpal.UUCP>, tneff@atpal.UUCP (Tom Neff) writes:
>> What characterizes the new generation
>> of OS's is not their graphical nature per se, but the use of hardware
>> independent device *drivers* to control all communication with the user.
>
>Ah, yes...the "new" generation.  Like OS-9, which has supported dynamic linking
>and dynamically loadable device drivers on the 6809 since about 1980, and on
>the 68xxx since about 1984?

You're welcome to those fine examples of sophisticated precursor
systems.  Providing there was fairly strong enforcement of the output
pathway using device drivers (rather than making an end run direct to
the hardware as so many Apple II and PC applications do), then a handicapped
driver could have been written for those systems. (May have been, for all I
know.)

What makes the difference in the case of OS/2 is the presumptive size of the
installed base in a couple of years.  Up till now, the most popular PC
on the market has been rife with nonstandardized I/O handling, making it
hard to justify developing anything but screen "readers."  I think that
will change before long.  TMN 
-- 

Tom Neff