At 05:25 PM 1/4/99 -0700, you wrote: 1,000,000 records (a small sample) x 2 bytes x 3 dates (birthdate, initial trx, most recent trx) = 6,000,000 bytes. That was a lot 20 years ago. Keep in mind, also, that reading in a record required enough workspace to hold the record, and this may have been a limiting factor as well. Not to mention bandwidth issues -- we may not think much of downloading a 6MB file now, but imagine doing that 20 years ago with a 300bps modem. Today it is. We keep hoping that one of these days Microbloat will get it right... 8^) But seriously, 20+ years ago, it wasn't so much that it was expected that it would be replaced as it was the idea that if they can get the software out the door now, they may still be around to fix the problem later, whereas if they hold off, they may not be around. It was also a self-perpetuating problem. No point in building a 4-digit date if your source data (internal or external) is 2-digits. In fact, a 4-digit program might fail if given 2-digit data. There are lots of reasons, many of them valid (at the time). --------------------------------------------------------------------- O- Uncle Roger "There is pleasure pure in being mad roger(a)sinasohn.com that none but madmen know." Roger Louis Sinasohn & Associates San Francisco, California http://www.sinasohn.com/