Re: Oracle: Naked King in database land?

Following up on Mark D Powell, 05 Apr 2003:

>
> The problem is that in the short-term it is cheaper to connect the new
> with the old than to convert the old technology to modern technology.
> Unfortunately this usually results in high maintenance costs due in
> part to very difficult modification efforts. But management only
> looks at the short-term numbers.

Nothing wrong with that strategy if it saves $$$. I doubt it does, in the mediumn and long term. But management seems to be judged nowadays on how fast they can make a quick buck and "progress their careers" as a result.

IOW, the "hit and run" approach to management. Which basically means the bastard responsible for the medium to long term mess will NEVER be pinned down, just promoted.

Oh well, who cares? It's their money, they can squander it however they feel. We can see the results now in the stock market, of all the squandering that went on in the 90s. And there is more to come...

>
> I believe that we are going to give it a try. At one time we ran MVS
> under VM and until we reconfigured VM to let MVS take over its own IO
> the setup sucked.

VM was quite a nifty OS, in fact. It could take over by itself or like you say, let portions of the other OS's take over. Quite intelligent.

I dunno how well that will work with Linux I/O, given its Unix background. The I/O system for Unix is character oriented, with byte-addressing and relying on an in-built large cache for efficiency. The VM I/O is very much like other mainframe I/O: VSAM-based. The two shall never meet. But maybe IBM found a way around it? Let us know how you folks go with that, OK?

> As long as you have your health you are not old, because your mind
> appears to still be in working order (not that my opinion matters on a
> subject like that though you did make the comment about age).

what's a "mind"? :D

-- 
Cheers
Nuno Souto
wizofoz2k_at_yahoo.com.au.nospam