I have just finished reading this long debate ("Debate Round 1"). I think both participants are to be commended.

My comment has to do with the fact that both argue from the premise that remediation has to involve massive changes to application programs.

Hoffmeister starts out with mathematical modeling based on metrics from the work of Capers Jones, and the whole debate is permeated with these metrics.

I admire Capers Jones and his work on metrics. I have one of his books in my personal library.

But I also think that if program encapsulation had been generally adopted, we could pretty well dispense with Jones and his metrics for the purposes of Y2K. I have been espousing both data encapsulation and program encapsulation for over two and a half years at this point.

(Yes, I know that with program encapsulation you still have to adjust hard-coded dates and also the date information you get from the operating system. You can't leave the application programs completely alone.)

Regarding possible responses to this post, what I will respond to with considerable vigor will be batshit comments like "Oh, that would just be postponing the problem".

-- Peter Errington (petere@ricochet.net), August 17, 1999

Answers

Peter,

You need to remember that the programs being remediated were written in an environment that said "Structured Programming" was NICE but not really up to our standards. Also, they are remediating programs written by wizzards who tried to save every byte of core, because many of them had grown up with the "MONSTER" 360-30's with a couple of kilobytes of core.. (In ANY ONE of the 3 machines sitting in my dining room, I have more computing power than there was in the WHOLE of Revere Copper and Brass when I broke in on computers in 1968. And I do mean MORE, more main memory, more DASD errrr hard drive space, more mountable memory, read only and I/O. The only thing I DON'T have is a faster printer than the chain printer).

Chuck

-- Chuck, a night driver (rienzoo@en.com), August 17, 1999.

Chuck -

This is true, however I would not quibble over the speed of the venerable IBM 1403 printer vs. say..a LaserJet 4. Changing the ink cartridge or clearing a paper jam on the LaserJet is "lots" easier than a 1403! (and no "print check" light to tape down with duct tape either!) And just where would one put the carriage control tape? LOL

-- JCL Jockey (WeThrive@OnStress.com), August 18, 1999.

To Chuck:

My use of "program encapsulation" refers to what a lot of people think of as "time-shifting." The latter is probably the more familiar term and I suspect I should have used it in my post. Anyway, the advantage of the approach is that you don't have to worry about the miserable complexities of the application programs. (Also, in modifying an old program, there is what we used to call the "green mold" problem. It's a known fact that the passage of time makes modifying old programs more problematical.)

-- Peter Errington (petere@ricochet.net), August 18, 1999.