In Reply to: Software product quality (or lack thereof) posted by Tony Lauck on December 16, 2009 at 09:05:50:
The laws are there, at least in the UK. Sadly they are not applied wrt to IT. Nevertheless, it should be possible to get a refund on software that does not fit the purpose for which it is supplied by a retailer.
When I started scientific computing, it was practice to test each subroutine properly before integrating them. Even after integration, it was practice to hand calaculate representatitive samples to ensure that the answers were correct.
I remember spending a week to 10 days plotting out hand calculated results to verify computer programs. Sadly, few seem to do it these days and I know of PhD failures because finite element computations were not checked aginstof other methods thru estimation or from lack of a common sense assessment of the results.
In audio, two of things that should be avoided with fast signals are curve fitting or sample estimation (very difficult to do accurately with random waveforms) and triggering on the basis of the rise or fall of a near square wave (dac operation). Unfortunately both have been adopted as facts of life. Hence the difficulties with digital audio reproduction and the cost of truly high end systems.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- RE: Software product quality (or lack thereof) - fmak 09:27:34 12/16/09 (0)