Re: Throughput and storage requirements questions for modern vs old database processing apps

In article <3c2b8efa.43148684_at_news.cybercity.dk>, Lars Gregersen <lg_at_2-control.dk> writes
>On Wed, 19 Dec 2001 21:58:17 +0000, "Anthony W. Youngman"
><thewolery_at_nospam.demon.co.uk> wrote:
>
>>In article <3bffa6cb.8398576_at_news.tcd.ie>, Paul Linehan
>><linehanp_at_tcd.ie> writes
>>>Then you could look at the Open Source databases MySQL (not free if
>[snip]
>>It's easy to learn, easy to use, easy to program ... the only study I
>>know of concluded that multivalue costs roughly half of what SQL does,
>>and if a company knows what it's got and isn't seduced by buzzwords and
>>consultants, they value it very highly.
>
>This sounds interesting. Do you have the reference for this study?
>
Can't give you much detail, unfortunately. The uni that conducted it was Witwatersrand in South Africa. And the gist of it was, they studied what percentage of turnover companies spent on their databases. There were two peaks, the main one at 4%, and a minor one at 2% consisting mostly of companies using Pick or its variants.

However, this could be a statistical aberration. One of Pick's major advantages (and also a major drawback) is that it's easy to use. So most people using it aren't trained in comp sci. So there are a lot of badly programmed systems written by non-programmers out there, and there's the usual "hidden cost" of systems that are costed to the user, not the computer dept. Certainly my company suffered that quite badly, but it had the side effect that the system actually did what the user wanted - after all the user had written it!

Cheers,
Wol

-- 
Anthony W. Youngman - wol at thewolery dot demon dot co dot uk
Witches are curious by definition and inquisitive by nature. She moved in. "Let 
me through. I'm a nosey person.", she said, employing both elbows.
Maskerade : (c) 1995 Terry Pratchett