Re: Fast400 Value to iSeries community is less than zero -- MIDRANGE-L

On Saturday 10 November 2001 04:08 pm, Nathan M. Andelin wrote:

> I suggest that IBM get rid of its cycle burning governor, and ask a price
> for a 5250 license.  My prediction, however, is that Rochester will close
> the loophole that Tiger Tools discovered, and continue business as normal.

Well, I am sure I agree on the second statement. I also agree that I would
prefer a different solution to whatever problem IBM is solving with the
CFINT governor.

> Features as software.  You seem to be saying that a cycle burning
> interactive governor is good for IBM.  I disagree.  I think artificial,
> contrived stuff like this alienates IBM from its customers and hurts IBM
in
> competitive situations.

Well, I don't know that I'd say I think it is "good for IBM." I am saying
that I believe that IBM has a problem they are solving with it. That
problem appears to be a pricing issue although some have said they feel the
whole purpose of CFINT pricing and control is a form of "punishment" for
IBM customers who like the green screen.

I think that IBM needs a solution for the pricing issue. CFINT may be a
poor solution, but it is one. So I'm saying it is "better for IBM than no
solution." I'm also saying that I really want IBM to have a solution to
this problem, because I really want IBM to continue making iSeries
machines.

> Two points.  5250 workload is not a constant.  Sometimes it's 100%.
> Sometimes it's 0%.  Companies must buy according to peak demand, even
though
> usage between 5:00 p.m. and 7:00 a.m. may average 3%.  A session license
is
> not much different.  Having the service available, when you needed it, is
> what counts.  Secondly, its usually pretty easy to convert POS stuff to
use
> a data queue, socket, or BSC interface.

Well, the POS terminals I'm refering to are actually just simple terminal
emulators. I know it would be easy to write something else, a PC based job
that just accessed the info, but right now there really isn't a demand for
that. The reason I think these work so well is that even when these people
are really busy, the majority of their work isn't interacting with the
system. So response time might go from sub second to several seconds, but
that is still acceptable in their particular use.

I'm not saying your idea wouldn't be much better than the current one.
Maybe it would be the best solution and the amount of complaining would be
far less. I'm pretty sure, though, that no matter what the solution is to
this problem there will be some people crying about it. Others complaining
it is an IBM conspiracy, etc. But if the whole point of it is to give IBM a
different way of reaching the point we are at today, then I question the
need for it.

Before you respond (if you were going to), I get the idea that you perceive
the need for it being to change the perception of how the control is being
done (in other words, IBM looks pretty crappy using CFINT governing). I
just think that there are plenty of people to point out how IBM is crappy
for whatever solution they use.

> Nathan M. Andelin
> www.relational-data.com

--
Chris Rehm
javadisciple@earthlink.net

And thou shalt love the Lord thy God with all thy heart...
...Thou shalt love thy neighbor as thyself. There is none other
commandment greater than these. Mark 12:30-31