Re: Fast400 Value to iSeries community is less than zero

["Steve Richter" <srichter@xxxxxxxxxxxxx>]

----- Original Message -----
From: <thomas@inorbit.com>

>> To run a real c++ telnet accessed application that is code bloat to some,
>> well abstracted functionality to others will need every bit of the 1000
CPW
>> to service its 15 users.

>If you mean that the full 1000CPW will be required as interactive, this is
only true if the >application is written to do so. It's perfectly possible
to break the interactive presentation l>ayer out and run the rest in batch
as server functions.
>Tom Liotta

Tom,

Changing your appl design as a workaround to IBM's pricing of interactive
CPW is more proof of how CFINT hurts our platform. When you add a layer ( in
this case many layers ) to any software, you increase its complexity. That
is bad. Instead of your "interactive presentation layer" calling a common
pgm to apply the dsply entry to the database, your approach is to fill a
data struct, write to a dtaq, hope the server job is running, wait for a
response from the server. And then you have to code and document the server
job.  Very complex. If you have a good appl design reason to do this, fine.
But if it is done because of IBM's pricing of interactive CPW then, that is
not fine.


James wrote:
> So I mean no offense, but it can't help but come across that way, I
> guess...:  I think that explains why the idea of a 15-user system needing
> 1000CPW seems pretty doggone funny, to some of us "old dogs".

Tom said:
 >Every once in a while, you get it extremely right.


Tom and James,

I am asserting that computer applictions will always? expand to fill the
available dasd and cpu available to them. This is proven by computer
history. With that "fact" established, I then say that IBM should not be
concerned re: losing revenue if they remove CFINT because the new
interactive applications writen to run on the new 3x fast systems will use
3x the cpu of software they replace or enhance.

And this increased usage of the cpu is not wasted or bloated code. It would
bring an excellent language like C++ to our platform. That alone is reason
enough to scrap CFINT.

Here is an interactive feature that increases functionality, but needs high
cpw to run:
Consider a subfile dsply.  What if the user wants to see a few more columns
of information today. Doable, but not realistic on a 50CPW multi user
system.  But if the CPW is available, your pgm could call some api's to
build the display on the fly, maybe send html to a next generation telnet
client.

etc, etc, etc

Steve Richter