27 Aug
2012
27 Aug
'12
6:40 a.m.
On 8/26/2012 12:47 PM, Fred Cisin wrote:
Microsoft then proceeded with an unauthorized port of CP/M to 8086/8088
(QDOS from SCP).
Can you cite some sources for this position? Sources I have note
that
Patterson wrote a new OS to go with the Seattle Computer Products 8086
kit and used CP/Ms API to aid in porting apps, but the internals were
significantly different. If true, I'm not sure I'd call that a "port".
If so, then Linux is a "port" of BSD or ATT UNIX.
Somebody else might think that the most important
"change" was "Long
Filenames", even though it was NOT implemented, merely a system of using
up FPDEs to store long NICKNAMES, such that FILENA~1 filename could be
linked to "Filname of file that contains the content and data of all of
the important stuff", and getting "DISK FULL" messages on an almost empty
disk because the root directory had been filled up with file nicknames,
instead of files.
Dunno, I'd call it "implemented". In fact, I think it well defines the
contemporary "kludge". It was (and is) a masterful use of an existing
structure to add a significant piece of functionality. I had to
"implement" it in the FAT library used by sd2iec, and I was very impressed.
Not that it would ever be used on a floppy disk, but FAT32 did remove
the 512 FPDE limit in the root dir.