The midseventies SHARE survey 63

recent posting in comp.arch about personal computers

my personal computer was 64kbyte 360-30 ... normally the university shutdown the computing center from 8am sat. until 8am monday. I got a key to the machine room and could have it all to myself for 48hrs on the weekend ... but it was a little hard after staying awake for 48hrs to go to monday clbuttes.

anyway ... nearly 40 years later ... i'm typing this on (almost) 4gbyte machine ... in little less than 40 years i've doubled the number of memory address bits on my personal computer from 16 to 32.

the issue back then was price-performance ... the cost of the hardware versis the cost of programmers improving performance ... did it result in a net benefit. the cost of hardware has declined significantly while the cost of programmers has gone up (making it a lot harder to show any net benefit of improved system performance against the cost of people doing the improved performance.

there is also the equation of lost opportunity ... given scarce programming resources (regardless of the cost) is the benefit greater applying them to new feature-function versis improving current performance. back then ... once you got the application running ... it needed to run well.

also, back in the dark ages ... with the relative high hardware costs ... the number of different business processes that could be justified automating was much fewer (and it was easier to justify making the ones that could be justified, run better).

with increasing people costs and declining hardware costs ... it is possible to cost justify automating huge numbers of additional business processes ... making more sense to apply scarce programming resources to automating additional business processes (as opposed to optimizing existing automated business processes).

--