[Info-vax] Meditech in the news

[George Cornelius]

plugh <jchimene at gmail.com> wrote:
> On Wednesday, January 5, 2022 at 2:36:01 AM UTC-7, George Cornelius wrote:
>> plugh <jchi... at gmail.com> wrote: 
>> > kemain... at gmail.com wrote: 
>> >> [...]
>> > That's nice. A longer search while "flogging my memory" 
>> > for lived experience results in 
>> > https://www.academickids.com/encyclopedia/index.php/MUMPS
>> It's not Mumps, an extension of it developed by the creator 
>> of the Mumps language. 
> 
> It's not MUMPS? maybe M? Maybe Mumps? maybe MUmphSsss?

Actually, I was on the Mumps standards committee when it first accepted M
as an alternative name to Mumps.  Along with Mark Berryman.

> Jesus God, talk about damning with faint praise.

My faint praise was reserved for Meditech.  Never touched the thing -
I suppose I should have since my organization eventually expanded to
include quite a few health care entities that were using it, at least
until we developed a new standard organization wide - but by then
I was a VMS system manager only and not a programmer, Mumps or
otherwise.

The person I did talk to about it described it as "weird".  Not a
programmer but someone on the team which had chosen the lab system
we were using at the time - running DSM - so I presume he was
describing Meditech as being an organization odd in that it tended
to hold its cards close to the chest.

Anyway, their extensions may have been a big improvement.  But it
was not Mumps.

Now Mumps suffers from an image problem. None other than Edsgar
Dijksta described it as a "child's disease".  Not being a native
English speaker, he could have just been referring to what it
was named for, but I don't think the quote would have stuck if
he didn't have deeper concerns regarding the language.

I don't go anywhere near that far.  Its developer was an MIT engineering
student, after all.  And he built it as an extension of RAND Corporation's
JOSS, which I have quite a bit of familiarity with having cowritten an
interpreter for it back in the late 60's.  I was impressed with what
he had morphed the language into.

And I point out to people that it is one of the few languages with a
mod function that operates in a way that is mathematically correct
for negative numbers - the remainder r from what mathematicians call
the division algorithm (no, not long division itself).

The problem with Mumps was that you were stuck without a concept
equivalent to a RECORD declaration.  So you tended to use delimited
data.  Or even multiple levels of delimiter, which you eventually
saw showing up in HL7 and ASTM standards for medical data exchange.

Yes, I know, you can do this

 SET ^PA(patno,"FNAME")="John"
 SET ^PA(patno,"LNAME")="Smith"
 SET ^PA(patno,"DOB")="1/7/1917" ...

but this has its drawbacks as well.

Oh, and as a VMS guy I was eventually stuck back at VMS 7.3-2 because
Intersystems refused to port to newer versions or to Itanium, and by
that time they owned Digital's DSM as well as their own (M/VX or ISM).

Somewhere I have licenses but no system to run it on.

When Tom McIntyre and his father were associated with DECUServe
we ran DSM but it's been quite some time since we have even
had it there.

> I appreciate the DECusEerve addr, but really? It's Mumps, errr M.... errr...
> Those of us who actually write software for a living know MuMPS when they smell it. One of the founders of the friggin' company actually helped develop the language.
> I had a contract at Kaiser to help maintain their lab interface written in
> 
> wait for
 
> it

 
> MUMPS!!!!

Thanks for the support. Too bad Intersystems withdrew theirs for VMS.

George
 
> I just love cov