If anyone should care about the century, astronomers should. Accuracy in pointing a telescope requires absolute knowledge of the year. Two digits just don't work; but shortcuts in writing telescope control software—particularly the venerable FORTH system —may have been taken over the years. Our Y2K review is complete and we are ready for the Year 2000 to begin.
Testing for the Y2K problem at Kitt Peak began in July of 1997 with the verification of the WWV-based IRIG-B time signal hardware and software. Individual telescopes were tested this past spring. During the 1999 Summer Shutdown, a mountain-wide test looked at the telescope and instrument control, all the computers and networks, and the time signal hardware as a system.
No problems were found with the main calculation codes for the Telescope Control Systems, but some changes were needed where date input routines used two digits. We were able to track and guide with no problems with dates set into next year. Most of the instrumentation software did not have a reliance on dates. However, we did have to update FITS keyword routines to handle the Year 2000 changes in the FITS standard that took effect this year. See http://www.cv.nrao.edu/fits/documents/standards/year2000.txt for information on the FITS update. Most of the computers required operating system patches for Year 2000 problems. For the mountain networks, a new Cisco router was installed, as the previous one was not certified for Year 2000. The network router at WIYN needed a new firmware version.
The KPNO telescopes will be closed on 31 December 1999 to allow our staff to celebrate the last year of the millennium, with a T&E night scheduled for 01 January 2000 to ensure that we have correct operation and that visiting astronomers will not have to fly on the first day of the Year 2000.