🚀 go-pugleaf

In article <1991Jul16.161324.249@zia.aoc.nrao.edu> cflatter@zia.aoc.nrao.edu (Chris Flatters) writes:
>The extracts from patent number 4,956,809 posted by Mark Costlow give
>the filing date as December 29, 1988.  Why must any proof of prior art
>be from mid-1982 or earlier?  Isn't it sufficient that it is prior to
>the filing date?  I am assuming that neither Mark nor RMS made a typo.

He left out that part.  The most recent filing is actually a continuation
of two prior filings (whatever that means):

			Related U.S. Application Data
	
	Continuation of Ser. No. 934,341, Nov. 24, 1986, abandoned,
	which is a continuation-in-part of Ser. No. 391,223, Jun.
	23, 1982, abandoned.

It is interesting that two Sun publications appear in the "References
Cited" section, though I haven't been able to find any specific mention
of these in the text.

	Sun Technical Report, "The Sun Network File System: Design,
	Implementation and Experience," pp. 1-16, copyright 1986 by
	Sun Microsystems, Inc.

	Sun Technical Report, "Open Network Computing-Technical
	Overview," pp. 1-32, copyright 1987 by Sun Microsystems, Inc.

The wierdest thing is that, while the claim certainly covers any use
of a canonical byte order to make files machine independent, the
particular canonical byte order they chose to use in their examples
is neither little-endian, nor big-endian, but rather the byte order
used by Unix compilers on the PDP-11, with 16 bit words having
the low order byte first and 32-bit longs having the high order
16 bit word first.

Dennis Ferguson