Cobol - [Semi-OT] IBM's IMS V10 Announcement

Author

[Semi-OT] IBM's IMS V10 Announcement

William M. Klein

2007-10-09, 6:55 pm

(I think it was in this group that someone stated that IBM was
"grand-fathering" - or some such term - its IMS support. That is PART of the
reason that I am sending this note)

Today, IBM announced its latest Version of IMS (both database and transaction
servers). Even for those NOT on IBM zOS, I think you might be interested in
going to:

http://www.ibm.com/vrm/ newsletter_.../WKlein12584487

and look particularly at the section,
"Product positioning"

I don't think that the information (options?) would be universally accepted, but
I think some within this group would find them interesting AND they do represent
IBM's current OFFICIAL "position". They address both the database (hierarchical
vs relational) issues and the transaction (web and non-web-based) issues.

--
Bill Klein
wmklein <at> ix.netcom.com

Pete Dashwood

2007-10-09, 6:55 pm

"William M. Klein" <wmklein@nospam.netcom.com> wrote in message
news:lXNOi.83394$6L.59679@fe03.news.easynews.com...
> (I think it was in this group that someone stated that IBM was
> "grand-fathering" - or some such term - its IMS support. That is PART of
> the reason that I am sending this note)
>
> Today, IBM announced its latest Version of IMS (both database and
> transaction servers). Even for those NOT on IBM zOS, I think you might be
> interested in going to:
>
>
> http://www.ibm.com/vrm/ newsletter_.../WKlein12584487
>
Sorry, Bill, I am unable to see anything at this link. ("Internet Explorer
cannot display the web page" Looks like a 404 but can't be sure)

If it is a Newsletter, could you please forward to me, (or put it on a
server where it CAN be accessed?)? I'd really like to read it.

Cheers,

Pete.

William M. Klein

2007-10-09, 6:55 pm

I don't know if there is something that checks for viewing from within the US or
not. Try the following LONG URL and see if it works:

http://www-01.ibm.com/common/ssi/cg...num=ENUS207-241

That would get you to the official version. I have also (in case this doesn't
work) put a PDF version on my website at:

http://home.comcast.net/~wmklein/IBM/IMS_V10.pdf

--
Bill Klein
wmklein <at> ix.netcom.com
"Pete Dashwood" <dashwood@removethis.enternet.co.nz> wrote in message
news:5n2f4tFfuj22U1@mid.individual.net...
>
>
> "William M. Klein" <wmklein@nospam.netcom.com> wrote in message
> news:lXNOi.83394$6L.59679@fe03.news.easynews.com...
> Sorry, Bill, I am unable to see anything at this link. ("Internet Explorer
> cannot display the web page" Looks like a 404 but can't be sure)
>
> If it is a Newsletter, could you please forward to me, (or put it on a server
> where it CAN be accessed?)? I'd really like to read it.
>
> Cheers,
>
> Pete.
>

Pete Dashwood

2007-10-09, 9:55 pm

"William M. Klein" <wmklein@nospam.netcom.com> wrote in message
news:egUOi.87746$zh6.26538@fe10.news.easynews.com...
>I don't know if there is something that checks for viewing from within the
>US or not. Try the following LONG URL and see if it works:
>
> http://www-01.ibm.com/common/ssi/cg...num=ENUS207-241

Didn't work for me.
>
> That would get you to the official version. I have also (in case this
> doesn't work) put a PDF version on my website at:
>
> http://home.comcast.net/~wmklein/IBM/IMS_V10.pdf
>

That was fine, thanks.

I've always had a soft spot for IMS :-) Used it for years before DB2 became
available.

The main things I got from the document are as follows:

1. IBM is being HUGELY influenced by what is happening in the real world,
and have realised they need to provide bridges to the OO environment. I
think what they are doing is good.

2. Like all sales literature there were statements in the document that made
me smile and laugh :-) The justification for hierarchic access as being best
for "Mission Critical" and where the "query is known in advance" (hard
coded) just made me chuckle. If IBM had abandoned IMS years ago there would
be no such writing. I remember the pain in the arse traversing hierarchical
DBs was and how we used Command codes to "remember" our positioning across
dequeue boundaries. (Command Code 'C' stored the concatenated key of where
you were in the "tree" so you didn't have to re-traverse it on re-entry.)
The fact is that the Relational model can implement a hierarchical
structure, but the hierarchical model cannot easily implement a relational
structure. This is marketing justification for an existing product, not real
data processing theory.

It is even more amusing because it was IBMers who formulated the Relational
theory and showed it to be mathematically pure... Funny how Marketing can
trumpet something when promoting one product, then ignore it when promoting
another... :-)

You could substitute "VSAM" for IMS in the arguments described, and it
wouldn't change the meaning (more efficient when queries are known in
advance, good for saving object structures, etc.)

Nevertheless, they have obviously poured investment into IMS and the results
are impressive.

3. IBM are demonstrating commitment to a web based world and the support
for SOA and Web Services is good. As a co-owner of the SOAP copyright they
obviously realise that they can leverage much with platform independent XML.
Good to see.

Thanks for posting this, Bill. It was interesting and quite nostalgic for
me.

Pete.
--
"I used to write COBOL...now I can do anything."

Sponsored Links