Megan wrote:
quantity in two
four bit hardware registers) to "00". Since RT-11 (I
know - I can't get away from it) has 1973 as the earliest legal date, I
Actually, the lowest date for RT is 1972... that is the beginning of
the 'RT Epoch'... :-)
Jerome Fine replies:
Perhaps you missed the emphasis on "legal" in the "earliest legal
date" with
regard to the DATE command which will not accept 1972 (e.g. 01-Jan-72).
In addition, some of the code in RT-11 in the monitor and elsewhere
rejects a date value with a year value of zero - meaning that 1972 is
considered invalid by that code. On the other hand, DIR has been
coded so as to display a date which includes a year value of zero or
1972 after the bias is added. Which is why I used a pivot value
of 70 (or 2070) so as to be sure to avoid any possible conflict for
the DATE/TIME hardware clock on the third party board. Any idea
how the 11/93 does the 1999 transition with the now Y2K compliant
firmware update?
wrote the
software to assume that any year value less than "70" was after
1999. In addition, back in 1993 or maybe as late as 1994, I patched the
code for the actual RT-11 OS (V5.6 of RT-11) to do the same thing. This
was not the standard that was finally adopted for V5.7 of RT-11 wherein
any date after 1999 was required to specify a four digit year, but it is
very obvious that many other people were already using the concept of a
pivot date long before 1998. I also modified some software for a company
that makes cars (you know those big hunks of metal with four wheels and a
nut in the front seat) - I think that was delivered in 1997 and those
patches also used a pivot date concept. So I doubt that it will take
much to prove that the granting the patent was "inappropriate".
Anyone else have any stories about using a pivot
date some time ago!
Yep...
Back when I was doing some 'private advanced development' of RT-11
post-V5.6, and before Mentec started work on V5.7, I implemented just that
sort of thing (I think it was about 1995 or so, but I'll have to check my
files). I had, for example, the DATE command modified so that it would
accept both 2-digit and 4-digit year specifications. If the year
specified was 72 to 99, it was assumed to be 1972 to 1999. If it was 00
to 71, it was assumed to be 2000 to 2071. Years 2072 to 2099 (the limit
handled by the RT-11 date format) would *have* to be specified in the
4-digit form.
When I started to modify RT-11 (many years before Mentec as well),
I adopted essentially the same procedure that you suggest above. That
was done about two years before you did V5.6 of RT-11. I also extended
the DATE command to accept all four digit years from 1973 to 2099,
essentially the same as the value output by IND using a four digit year
for which I adopted the symbol DATEXT. In IND, I also added
TIMEXT which provided the time to the nearest 10 milliseconds.
It might also be helpful to note that while it is possible to allow
"00" to "71" to represent "2000" to "2071" in the
DATE command,
that is not possible in utilities such as DIR and PIP where there
might be confusion with the "day of the month". Since DIR and PIP are
forced to reject vales between "01" and "31" as a year, perhaps you
also, as I chose, decided to reject "00" to "71" completely and
required all years after "99" to have 4 digits.
I also did V5.4G of RT-11 for most of the utilities (DIR, PIP, LINK,
LIBR, MACRO and BUP) along with the monitor (RMONFB and
KMOVLY) and the tape device drivers. The spooler and things
like RTMON.REL did not seem to be worthwhile, so they were
postponed. Note that Y2K compliance for V5.4G will stay as a
paid for offering until Mentec places versions subsequent to V5.3
on the hobby list (as opposed to Compaq who has now placed
most if not all of VMS on the hobby shelf) or at least for a few
more years or until there seems to be sufficient hobby interest.
Originally, I expected to offer V5.3 Y2K patches this summer,
but as my time was taken up with commercial work, that did not
happen. And since I received less than 5 inquiries about that
possibility, it seemed not worth while in any case. I am still
interested in doing Y2K patches for V5.3, but it will not be
possible to complete them by the end of 1999. However, if
anyone is really interested, SHOUT!
Sincerely yours,
Jerome Fine
RT-11/TSX-PLUS User/Addict
Present RT-11 Developer (unofficial of course - that means that
anything I do is usually ignored by almost everyone else)
