due to contention

[Antoon Smulders]

Hi all,

 

In the year 2000 there was a discussion on the list about a new “feature” (read bug) of SPSS: while saving a datafile one gets a message saying that “due to contention” the file was saved under a different name (something like file_1, file_2 file_3 etc).  The following link will bring you to the discussion.

 

http://www.listserv.uga.edu/cgi-bin/wa?A2=ind0010&L=spssx-l&D=0&P=7150

 

There was a notion from SPSS that you should never save a changed file over an older version. Though I do not agree with this moralistic notion I have another problem:  When this message appears on my screen, my original file, with the original filename  is no longer on the disk!

Can someone explain the rationality behind that?

 

Antoon Smulders

 

 

[SPSS Support]

Hi Antoon,

 

I don't ever recall anyone at SPSS saying that you should never save a changed file over an older version. The message comes when the program "thinks" that the same file is open in two applications or by two sessions at the same time, and there is a danger of overwriting changes made by one when saving from the other. It is most definitely a bug if your original file is somehow erased. In order to try to diagnose the problem, we'll need more specific information about your version and platform and exactly what you're doing.

 

David Nichols 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Antoon Smulders
Sent: Thursday, July 23, 2009 9:28 AM
To: [hidden email]
Subject: [SPSSX-L] due to contention

Hi all,

 

In the year 2000 there was a discussion on the list about a new “feature” (read bug) of SPSS: while saving a datafile one gets a message saying that “due to contention” the file was saved under a different name (something like file_1, file_2 file_3 etc).  The following link will bring you to the discussion.

 

http://www.listserv.uga.edu/cgi-bin/wa?A2=ind0010&L=spssx-l&D=0&P=7150

 

There was a notion from SPSS that you should never save a changed file over an older version. Though I do not agree with this moralistic notion I have another problem:  When this message appears on my screen, my original file, with the original filename  is no longer on the disk!

Can someone explain the rationality behind that?

 

Antoon Smulders

 

 

[Art Kendall]

Although I have never heard anyone at SPSS say you should never save a changed file over an older version, I have often said on this and other lists that it is a poor programming practice. 
A good rule of thumb is to do things in such a way that you can always go back and alter your process to do what you want it to do (as opposed to what you told the computer to do).
In many situations it is a major requirement to have a complete "audit trail".
If disk storage space is a major consideration, you can always save the original input file and the syntax that makes any transformations. You can save blocks of syntax in separate files and comment out ones that are not needed while you are drafting a part of your overall procedure.
something like this untested syntax.
get file 'd:\project\original data.sav'.
*insert 'd:\project\transforms1.sps'.
*insert 'd:\project\transforms2.sps'.
*insert 'd:\project\transforms3.sps'.
insert 'd:\project\transforms4.sps'.



Art Kendall
Social Research Consultants

SPSS Support wrote:

Hi Antoon,

 

I don't ever recall anyone at SPSS saying that you should never save a changed file over an older version. The message comes when the program "thinks" that the same file is open in two applications or by two sessions at the same time, and there is a danger of overwriting changes made by one when saving from the other. It is most definitely a bug if your original file is somehow erased. In order to try to diagnose the problem, we'll need more specific information about your version and platform and exactly what you're doing.

 

David Nichols 

---

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Thursday, July 23, 2009 9:28 AM
To: [hidden email]
Subject: [SPSSX-L] due to contention

Hi all,

 

In the year 2000 there was a discussion on the list about a new “feature” (read bug) of SPSS: while saving a datafile one gets a message saying that “due to contention” the file was saved under a different name (something like file_1, file_2 file_3 etc).  The following link will bring you to the discussion.

 

http://www.listserv.uga.edu/cgi-bin/wa?A2=ind0010&L=spssx-l&D=0&P=7150

 

There was a notion from SPSS that you should never save a changed file over an older version. Though I do not agree with this moralistic notion I have another problem:  When this message appears on my screen, my original file, with the original filename  is no longer on the disk!

Can someone explain the rationality behind that?

 

Antoon Smulders

 

 

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Antoon Smulders]

 

To SPSS-support: I use version 17.02 of Pasw statistics under Vista in a Microsoft Small Bussiness Server environment.

It happens rather often when I try to save a data file. Caching the data before saving does not (always) help.

To Art Kendall as well as and SPSS support: this (saving a data file), occurs mainly in very early stages of a project. Often we use scanning software (Teleform) to get the data or we get Excel-files that were manually typed in. In both cases we have to validate the data and correct the typing or scanning errors. In this stage I don’t see very much necessity to keep track of all the (often numerous) changes.

To Art Kendall: Things get different when the first analyses are done. From that time on, I agree, it is very important to keep your original data files. As fact, I  seldom save the data file after this stage: I save the syntax files.

 

Sorry if I accused SPSS falsely for the advise to never save a changed file over an older version. It was a long time ago and a link in the discussion on the list is no longer valid.

 

Antoon Smulders

 

 

Van: SPSSX(r) Discussion [mailto:[hidden email]] Namens Art Kendall
Verzonden: zondag 26 juli 2009 14:57
Aan: [hidden email]
Onderwerp: Re: due to contention

 

Although I have never heard anyone at SPSS say you should never save a changed file over an older version, I have often said on this and other lists that it is a poor programming practice. 
A good rule of thumb is to do things in such a way that you can always go back and alter your process to do what you want it to do (as opposed to what you told the computer to do).
In many situations it is a major requirement to have a complete "audit trail".
If disk storage space is a major consideration, you can always save the original input file and the syntax that makes any transformations. You can save blocks of syntax in separate files and comment out ones that are not needed while you are drafting a part of your overall procedure.
something like this untested syntax.
get file 'd:\project\original data.sav'.
*insert 'd:\project\transforms1.sps'.
*insert 'd:\project\transforms2.sps'.
*insert 'd:\project\transforms3.sps'.
insert 'd:\project\transforms4.sps'.



Art Kendall
Social Research Consultants

SPSS Support wrote:

Hi Antoon,

 

I don't ever recall anyone at SPSS saying that you should never save a changed file over an older version. The message comes when the program "thinks" that the same file is open in two applications or by two sessions at the same time, and there is a danger of overwriting changes made by one when saving from the other. It is most definitely a bug if your original file is somehow erased. In order to try to diagnose the problem, we'll need more specific information about your version and platform and exactly what you're doing.

 

David Nichols 

---

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Antoon Smulders
Sent: Thursday, July 23, 2009 9:28 AM
To: [hidden email]
Subject: [SPSSX-L] due to contention

Hi all,

 

In the year 2000 there was a discussion on the list about a new “feature” (read bug) of SPSS: while saving a datafile one gets a message saying that “due to contention” the file was saved under a different name (something like file_1, file_2 file_3 etc).  The following link will bring you to the discussion.

 

http://www.listserv.uga.edu/cgi-bin/wa?A2=ind0010&L=spssx-l&D=0&P=7150

 

There was a notion from SPSS that you should never save a changed file over an older version. Though I do not agree with this moralistic notion I have another problem:  When this message appears on my screen, my original file, with the original filename  is no longer on the disk!

Can someone explain the rationality behind that?

 

Antoon Smulders

 

 

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD