[Info-vax] New VSI Roadmap (yipee!)

[Bill Gunshannon]

In article <md2ssm$kpk$1 at dont-email.me>,
	David Froble <davef at tsoft-inc.com> writes:
> lists at openmailbox.org wrote:
>> On Mon, 02 Mar 2015 12:57:47 -0500
>> David Froble via Info-vax <info-vax at rbnsn.com> wrote:
>> 
>>> lists at openmailbox.org wrote:
>>>>> I can't imagine ever using floating point in a new money handling
>>>>> application these days given all the other options available.
>>>> It was *never* acceptable!
>>>>
>>> Even when there was no other options?
>> 
>> Especially when there was no other option.
>> 
>> Integers with software scaling (i.e. treat all amounts as cents) would have
>> probably been better depending on what calculations and how many
>> calculations were involved and what integer types were available. But if
>> you had no integer type big enough to represent the largest amount as
>> cents that is a real problem and strongly suggests the solution should not
>> be coded on that platform.
> 
> Ok, let me see if I can understand this attitude ..
> 
> If you can't do it the way you consider "proper", then don't do it at 
> all?  Gee, I could have been a man of leisure.  Poor man of leisure.
> 
> Integers, huh?  Let me see, -32768 to 32767, yeah, that's more than 
> enough to show the company's monthly sales.
> 
> I'm wondering about your age, because you don't seem to have much of an 
> idea what type of computing resources was available in the early 1970s.
> 
> I'll say it once more.  We used what we had, and we made it work, and 
> IT'S WORKED WELL FOR OVER 40 YEARS!!!  That's one thing more recent 
> stuff cannot say, and will probably never be able to say.

My favorite response to this kind of stuff is: "Hindsight is always 20/20."

Anecdote time!!

One of my early tasks as an application programmer was to write
a system to emulate the online data entry system for a microcomputer
with the idea that at the end of the day all the data would be
uploaded.  Reasoning being that the serial lines used by the data
entry clerks were a rare and expensive resource.  I got assigned the 
task for one very simple reason.  My peers were all mainframe
programmers, CDC, Honeywell and Univac.  The target microcomputer
was a Terak.  LSI 11/02 with 28K Words of memory (minus the I/O
page of course) and SS/DD 8" floppies.  My peers were unanimous in
the opinion that nothing practical could be done with such limited
resources.  I was already playing with Z80's at home and was quite
experienced at shoehorning programs into small memory spaces.  :-)
I did it all in UCSD-Pascal.
It got to be even more fun when I set up the Terak system on my desk
with 4 8" floppy drives and started doing COBOL under RT-11.

bill

-- 
Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves
billg999 at cs.scranton.edu |  and a sheep voting on what's for dinner.
University of Scranton   |
Scranton, Pennsylvania   |         #include <std.disclaimer.h>