            Questions and Answers for the release of
          version 4.01 of the Vaisnava calendar program

                Written by Markandeya Rishi Dasa



Q: Do I have to change to the new version (4.01)?

A: Yes, you should definitely change to the new version. The town
   file (vcal.tow, containing longitudes and latitudes etc.) that
   was included with the previous release (version 4.00), was
   unfortunately corrupted. Because of this some dates in the
   calendars become wrong. I sincerely apologize for this
   mistake.

   To make sure that users do not by mistake use such an old town
   file, I had to make this new release (version 4.01) of the
   program which refuses to use old town files.

   So if you use version 4.01 you know that the calendars
   produced do not have this problem.




Q: The printout of my calendar produced by the old program
   (version 4.00) shows the right longitude and latitude. Does it
   mean that my calendar is correct after all? Can I use it?

A: No, unfortunately it is not that simple. Even if the calendar
   shows the right longitude and latitude it most probably still
   has this problem.

   The reason is that inside the town file, the coordinates were
   stored in two ways - in the ASCII and in the binary way. The
   ASCII storage was intact and that is what you see printed on
   the calendar. The binary storage was corrupted and that is
   what the calculation of the calendar is based on. (The new
   town file uses only one way, the binary way.)




Q: If I get the new version (4.01), is there then a risk that my
   new proper town file can become corrupted in the same way as
   the one which was released in version 4.00?

A: No. The corruption was not caused by the Vaisnava calendar
   program. It was caused by another program which only I have
   (and which has now been corrected).




Q: If I just look at the print out of a calendar, how do I know
   if it is produced by the new version (4.01)?

A: The version number of the program will be written to the
   calendars. The text "Vcal 4.01" will be found in the top right
   corner of each page of the calendars produced by this version.




Q: Has this problem been there already earlier in previous years
   in the calendars that you mailed out or that could be
   downloaded from Com and the Internet?

A: No. This problem is limited to the calendars based on that
   particular corrupted town file included in the version 4.00
   release.

   All calendars mailed out throughout the years, and all
   calendars that could (and still can) be downloaded from Com
   and from the Internet are based on proper town files and do
   not have this problem.




Q: How often does the town file included in the previous version
   (4.00) produce wrong dates? And what about the period for
   breaking of ekadasi fast?

A: It depends on which town we speak about, but for an "average
   town" approximately 4-5% of the dates will be wrong. Looking
   at only ekadasis this means about 1 ekadasi per year.

   The period for breaking of ekadasi fast is also affected by
   this problem but I have not made a closer investigation of
   that.

   It may be interesting to compare the 1 ekadasi per year figure
   mentioned above to the error rate of the old calendar that
   ISKCON used before the computerized calendar was created. At
   that time we used the Mayapur (actually Calcutta) calendar all
   over the world.

   That Calcutta calendar was calculated with old, not very
   accurate formulas, so it had errors even if you only used it
   in Calcutta, and even more "errors" if you used it in other
   parts of the world, which we did.

   The below text is a quote from the documentation of the
   Vaisnava calendar program. It shows how the calcutta calendar
   had 2 ekadasis wrong in the year 1989. I would say that it was
   a rather typical error rate for that calendar.


       "The computer calendar made for Calcutta has been compared
       to the calendar provided by the Gaudiya Math. There where
       differences but after consulting the Lahiris ephemeris it
       was always clear that the computer calendar was right and
       the Gaudiya Math calendar wrong in case of difference.
       When comparing the ekadasis in the 1989 calendars, such
       errors in the Gaudiya Math calendar were found at 11:th of
       September and 9:th of November. These two ekadasis should
       actually be observed the 12:th of September and 10:th of
       November respectively."




Q: Can a new town file be used together with the old version
   (4.00) of the program?

A: Not really. But, yes, from a very technical point of view,
   because the 4.00 version of the program does not check which
   version of the town file it uses. This is because the system
   of having a version number on the town file was introduced in
   the new version (4.01).

   But instead of the longitude and latitude being written to the
   calendar the word Error and 02 will be written there. "Error"
   is used to indicate that something is wrong, i.e. one should
   not use the old version of the program, but instead change to
   a newer version. "02" is the version number of the town file.

   But technically speaking the dates in such a calendar will be
   correct. There is no reason though to use the old version of
   the program, it only increases the risk for mistakes.



Copyright (c) The Bhaktivedanta Book Trust International, Inc.
1983-1999. All rights reserved.

