_old_ mainframe SPSS V9 usage stats program

[F. J. Kelley]

long, long ago ...
in the 1980's as SPSS developed from v7 to v8 to v9, it had a different syntax than it has now or even had in the 1990's.  Not drastically different, but different enough.  When SPSS v9 became SPSS V10, the name of the software itself was changed to  SPSSx and instead of V10 it was V1, and a new age began.  The current version of SPSS is (I believe) SPSS V15.  But a lot of programs were written in the older form.  Notably, the Inter-University Consortium for Political and Social Research (ICPSR) provided code in the old form.  And now, in the fullness of time, it has come to me.......
OK, the LOTR legend stuff is kinda fun, and I used the old SPSS and SPSSx and then SPSS again, but now I need some help.  Specifically, the old mainframe V9 version of SPSS required a punched deck of yearly stats (yes, punched cards) - the "usage statistics".  I still have the old load module and had a request for help from someone who had only the old program to read this data.
but ...
the usage stats need to be generated, so I need a copy of the program that produced the "H90 Usage Report" ... I know this one is kinda off the wall, but does anyone have a copy of this?  This is a program running on an IBM mainframe OS is z/OS (looks like MVS to me).  I know I once had this pgm, but I'm not on the mf very often these days, and it was maybe 10 years ago when i last ran it, so I'm not certain where it is.
anyway, thanks for any help.
-- Joe Kelley

[Richard Ristow]

At 07:19 PM 4/3/2007, F. J. Kelley wrote:

>long, long ago ...

Right. On a mainframe far away ...

>A lot of programs were written in the older form [of SPSS].  I need
>some help.  Specifically, the old mainframe V9 version of SPSS
>required a punched deck of yearly stats (yes, punched cards) - the
>"usage statistics".  I still have the old load module and had a
>request for help from someone who had only the old program to read
>this data.

Let me get this straight: You have the load module for the OS/360
version of SPSS (not SPSSX, nor SPSS PC+, nor SPSS for Windows) release
9, and you need to run it?

That is, you're not trying to convert old SPSS code to run in current
SPSS. (Regarding that, see recent thread "I need some help converting
data from the 1980's", Sun, 25 Mar 2007 <19:55:00 -0400>, ff.)

I'll leave intact your request:

>The old V9 version of SPSS required a punched deck of yearly stats
>(yes, punched cards) - the "usage statistics".

Hmmmmm. The systems people did the loading and that kind of thing, but
I expect it was a card-image disk file, rather than a card deck. But
that doesn't help you much.

>I need a copy of the program that produced the "H90 Usage Report" This
>is a program running on an IBM mainframe OS is z/OS (looks like MVS to
>me).  I know I once had this pgm, but I'm not on the mf very often
>these days, and it was maybe 10 years ago when i last ran it, so I'm
>not certain where it is.

The other thing: is there a particular reason to bring up old SPSS to
run the old code, rather than converting the code for a newer SPSS?

Sounds like fun, either way,
Richard



>--
>No virus found in this incoming message.
>Checked by AVG Free Edition.
>Version: 7.5.446 / Virus Database: 268.18.25/744 - Release Date:
>4/3/2007 5:32 AM

[Peck, Jon]

If someone extracts the data in the same format (but hopefully not on those things that were full of holes), and you can find the program source, you might be able to mostly convert it to modern SPSS syntax using
SPSS PC+ syntax convertor.sps
which is still shipped with SPSS.

The /PC+ program syntax was mostly adapted from the older (and smaller) mainframe system, and the purpose of this script was to migrate that syntax into the style of SPSS-X.

I haven't run this script for years, so I wouldn't make any guarantees, but it's there.

-Jon Peck


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of F. J. Kelley
Sent: Tuesday, April 03, 2007 6:19 PM
To: [hidden email]
Subject: [SPSSX-L] _old_ mainframe SPSS V9 usage stats program

long, long ago ...
in the 1980's as SPSS developed from v7 to v8 to v9, it had a different syntax than it has now or even had in the 1990's.  Not drastically different, but different enough.  When SPSS v9 became SPSS V10, the name of the software itself was changed to  SPSSx and instead of V10 it was V1, and a new age began.  The current version of SPSS is (I believe) SPSS V15.  But a lot of programs were written in the older form.  Notably, the Inter-University Consortium for Political and Social Research (ICPSR) provided code in the old form.  And now, in the fullness of time, it has come to me.......
OK, the LOTR legend stuff is kinda fun, and I used the old SPSS and SPSSx and then SPSS again, but now I need some help.  Specifically, the old mainframe V9 version of SPSS required a punched deck of yearly stats (yes, punched cards) - the "usage statistics".  I still have the old load module and had a request for help from someone who had only the old program to read this data.
but ...
the usage stats need to be generated, so I need a copy of the program that produced the "H90 Usage Report" ... I know this one is kinda off the wall, but does anyone have a copy of this?  This is a program running on an IBM mainframe OS is z/OS (looks like MVS to me).  I know I once had this pgm, but I'm not on the mf very often these days, and it was maybe 10 years ago when i last ran it, so I'm not certain where it is.
anyway, thanks for any help.
-- Joe Kelley

[F. J. Kelley]

reasonable questions ...
both SPSS v9 and SPSSx v4.01 are on the mf; my thought is to create a portable file he can use in a Windows environment.  There are several hundred lines of variable and value labels that need to be converted to current form.
as for the usage stats, we had stopped using v9 (on the IBM -- we still used it on the CDC Cyber) by the time I became a user servant, so I cannot say how this was handled.  I kept the old version active as the Cyber was going to be deactivated; there were a few researchers still using v9 so I wanted to keep something available for migration assistance.  I used to perform the SPSSx installation on the mf, our ibm systems folks did SAS.
i would not even care about the stats, but running a v9 pgm generates a hard "write to operator" (WTO) message about the stats that must be responded to (I had to call the data center as these days a wto is more commonly an error message).  I have sent a note to SPSS about this, but kinda don't think they will have any handy archive of this.
The load module dataset has some additional members - one of which is almost certainly the usage stats module, but I seem to recall there was a parm string or even command as well.
I'll check the mainframe list, but thought someone over here might still recall the pgm.  the idea is _not_ to get this poor student on the mf running an obsolete version of SPSS (i mean, geeze, he'd have to learn enough about MVS, ISPF, JCL and all that to do anything -- how do you explain DD-stmts to someone who has never encountered them?).  and then there would be the old SPSS syntax.
I'm going to test the other members of the load module, but thought someone might recall how this went.
--Joe

---- Original message ----

>Date: Wed, 4 Apr 2007 00:35:07 -0400
>From: Richard Ristow <[hidden email]>
>Subject: Re: _old_ mainframe SPSS V9 usage stats program
>To: [hidden email]
>
>At 07:19 PM 4/3/2007, F. J. Kelley wrote:
>
>>long, long ago ...
>
>Right. On a mainframe far away ...
>
>>A lot of programs were written in the older form [of SPSS].  I need
>>some help.  Specifically, the old mainframe V9 version of SPSS
>>required a punched deck of yearly stats (yes, punched cards) - the
>>"usage statistics".  I still have the old load module and had a
>>request for help from someone who had only the old program to read
>>this data.
>
>Let me get this straight: You have the load module for the OS/360
>version of SPSS (not SPSSX, nor SPSS PC+, nor SPSS for Windows) release
>9, and you need to run it?
>
>That is, you're not trying to convert old SPSS code to run in current
>SPSS. (Regarding that, see recent thread "I need some help converting
>data from the 1980's", Sun, 25 Mar 2007 <19:55:00 -0400>, ff.)
>
>I'll leave intact your request:
>
>>The old V9 version of SPSS required a punched deck of yearly stats
>>(yes, punched cards) - the "usage statistics".
>
>Hmmmmm. The systems people did the loading and that kind of thing, but
>I expect it was a card-image disk file, rather than a card deck. But
>that doesn't help you much.
>
>>I need a copy of the program that produced the "H90 Usage Report" This
>>is a program running on an IBM mainframe OS is z/OS (looks like MVS to
>>me).  I know I once had this pgm, but I'm not on the mf very often
>>these days, and it was maybe 10 years ago when i last ran it, so I'm
>>not certain where it is.
>
>The other thing: is there a particular reason to bring up old SPSS to
>run the old code, rather than converting the code for a newer SPSS?
>
>Sounds like fun, either way,
>Richard
>
>
>
>>--
>>No virus found in this incoming message.
>>Checked by AVG Free Edition.
>>Version: 7.5.446 / Virus Database: 268.18.25/744 - Release Date:
>>4/3/2007 5:32 AM

[Richard Ristow]

At 08:58 AM 4/4/2007, Peck, Jon wrote:

>If someone extracts the data in the same format (but hopefully not on
>those things that were full of holes)

Now, Jon, those old SPSS versions weren't *that* bad ....

-Ducking quickly,
  Richard