On about 1983 a Chabad shliach, Shlomo Laiken, had the idea to produce a Jewish art calendar for various US cities. Producing such calendar was of course possible, however in those years it was very difficult to compute the precise sunset times in order to be able to show candle-lighting times properly for each city.
Today, one click on a phone app and we have the precise time, even calculated based on our precise GPS location! But in the early 80’s there were no such apps, and computing power was very limited and expensive. The IBM PC was not even on the scene yet!
Shlomo was chatting with a friend, Rabbi Motty Berger (at the time shliach in Ottawa, Canada), and mentioned his challenge. Motty knew that I was in IT, and put Shlomo in touch with me.
I did some research and found out that it was almost impossible to find a program that could do such calculations with enough precision for our needs. The only place I could find was the Canadian Observatory, which needed this precision to accurately align their telescopes.
Upon contacting them, I was told that they would be willing to share their code with me, however it was quite complex and required significant compute power. Since I was working for Litton Systems in Toronto, in charge of powerful computers (for the time), I decided to accept.
Their program required some code changes to produce what we needed for a Jewish calendar… To calculate the precise time offsets for candle lighting and end of Shabbos or Yom tov. I made the changes, including formatting each page to show the city and data in a tabular form that would be easy to understand.
Such calculations are based on geographic coordinates, and in researching this, I discovered that the official coordinates of each city were based on the location of their city hall! To test whether or not this was a problem, I got coordinates of various locations and ran the program to compare calculated times. While the differences weren’t too significant for smaller cities, it turned out that if a large city’s Jewish community was far from city hall, there were times of the year when the calculations could be as much as several minutes off!
Without GPS (which didn’t exist) it was difficult to get accurate coordinates for any specific location. After much discussion, we decided the best we could do was to include a warning that for larger cities one should be very careful not to wait until the end of the 18 minutes to light (I think we suggested assuming 15 minutes to be safe), and should wait another couple of minutes to end Shabbos.
Not knowing whether I’d be able to produce this data easily in the future, I decided to run a full 10 years charts for each of the cities (I think about 10) that Shlomo wanted to cover. As I recall, the program was very complex because it also did many other computations that we were not interested in, so it ran for over an hour for each city! Printouts were on standard fan-fold computer paper. At the end, I delivered a huge carton to printouts with 10 years of data.
While it was fun to be a small part of this new initiative, I imagine that within a few years new, faster programs probably made the printouts obsolete. However, until the GPS satellite system was created, I believe that we helped publicize the possible inaccuracies of “standard” sunrise/sunset times in terms of Jewish observance.