Re: What so special about PostgreSQL and other RDBMS?

Noons <wizofoz2k_at_yahoo.com.au.nospam> wrote in message news:<40ac9f95$0$1586$afc38c87_at_news.optusnet.com.au>...

> Me too, after 30 years. And guess what: the answer is simple.
> Do NOT upgrade, unless you absolutely have to. I know of many customers
> still running the SAME hardware and software they were running 10 years ago.
>
> If it works and does the job, don't touch it. If it doesn't, then find
> someone who can fix it. If it's too old and no one stocks it anymore,
> then buy a new one. What's the problem with that? It's what the entire
> economy is based on...

The problem is the applications software is not a small interchangeable thing. I know of a company that is running on some old VMS stuff that was heavily customized. Their HQ on another continent was supposed to roll out SAP, but that didn't work for them, too difficult to customize to their business model or something. So now it's even more years later and they are starting all over again, and perhaps having issues finding the rare obsolete skills to keep the old stuff going. So "if it works" depends on definitions, and "buy a new one" is not a matter of running down to Frye's when they have an ad for a $400 computer limited to supplies on hand (dang, didn't see that one till Saturday afternoon -with 17" color monitor, color Canon printer, buncha USB ports... and no way to tell beforehand if RHAS would work).

>
> > When I am able to offer my customers an alternative to
> > this revenue drain, I am happy to do so.
>
> Here is the alternative: if it does the job, do NOT upgrade. Simple.

A balance must be maintained between the wheel and the dead hamster. Simply not upgrading is a false economy, since it loses the effect of getting major enhancements with the development costs amortized among many other customers. Eventually you are in a deep hole and have to spend big capital bucks to get out of it. That amortization argument also works against open source, because you have too much fragmentation and duplication of effort (like all the former linux variants, for just one example). Bleeding edgeism is obviously a mistake for many places, too.

jg

--
@home.com is bogus.
http://www.zapcom.net/~rci/rcigall.htm