Counting changes to paragraphs, figures, tables (Word2000/SR1)

Topic

New Reply

My management wants to impress our client by establishing a quality control process that counts the number of paragraphs, figures, and tables that have changed from one document version to the next. The idea is to show how the number of these types of changes to documents decreases over time through implementation of the process. They want this change-counting method to be applied to all future documents as well as to look back at existing ones and gets counts for those as well.
Using Compare Document seems overly tedious for getting these counts, since it finds more types of changes than we need and it doesn’t list counts by change type. Does anyone know of a method or macro that comes close to doing what my management wants?
Thanks in advance!

Reply | Quote

Replies

This is one tool that does the job. It is relatively inexpensive compared with others. Alternatively, I suspect a macro to do this could be quite lengthy; however, I cannot help you with that. Perhaps one of the moderators can. HTH

http://www.softinterface.com/Compare-File-…le-Programs.HTM%5B/url%5D

Ron M

Reply | Quote

Thanks, Ron for the info on the Diff Doc tool! It’s loaded with features and can probably do what I need it to do.

Reply | Quote

Thanks, Ron for the info on the Diff Doc tool! It’s loaded with features and can probably do what I need it to do.

Reply | Quote

After evaluating DiffDoc and talking to their developers, it turns out that the package can’t count changes the way I need it to. A typical example: A document has five figures in it. If the only change I make is to insert a new figure at the beginning of the document, it will become the new Figure 1. I consider that I have made only one change, even though old Figures 1 through 5 have been renumbered as Figures 2 through 6. DiffDoc counts this situation as 6 total changes (it includes all of the incidental renumberings as changes), and they tell me it can’t be reconfigured to ignore the “incidental” changes. Any quality process I implement using such change counts will be seriously flawed, especially with my large documents. Anyone out there got any suggestions?

Reply | Quote

You could write code to count up various types of objects, but it’s not easy to determine whether the count proves anything… You’d almost have to capture all of the user’s actions, i.e., insert figure, delete 25 characters, insert 12 characters, etc. etc., to get any sense at all.

You could do a binary file compare and look for percentage differences. I don’t know any software that does that off the top of my head, but I’m sure it’s out there.

Reply | Quote

You could write code to count up various types of objects, but it’s not easy to determine whether the count proves anything… You’d almost have to capture all of the user’s actions, i.e., insert figure, delete 25 characters, insert 12 characters, etc. etc., to get any sense at all.

You could do a binary file compare and look for percentage differences. I don’t know any software that does that off the top of my head, but I’m sure it’s out there.

Reply | Quote

Hi George,

I think you’re likely to end up with a very comples beast if you have documents using heading level numbers, fields etc, and you don’t want to count the incidental changes to them too. Even though you might only insert or delete a numbered para, Word tracks all of the consequential changes and tracks them too. But if you do both an insert and a compensating delete, the tracked change in para numbers disappears. Then you also have the problem that a simple thing like re-ordering paras is treated as both a delete and an insert in each case.

Cheers

Cheers,
Paul Edstein
[Fmr MS MVP - Word]

Reply | Quote

Hi George,

I think you’re likely to end up with a very comples beast if you have documents using heading level numbers, fields etc, and you don’t want to count the incidental changes to them too. Even though you might only insert or delete a numbered para, Word tracks all of the consequential changes and tracks them too. But if you do both an insert and a compensating delete, the tracked change in para numbers disappears. Then you also have the problem that a simple thing like re-ordering paras is treated as both a delete and an insert in each case.

Cheers

Cheers,
Paul Edstein
[Fmr MS MVP - Word]

Reply | Quote

After evaluating DiffDoc and talking to their developers, it turns out that the package can’t count changes the way I need it to. A typical example: A document has five figures in it. If the only change I make is to insert a new figure at the beginning of the document, it will become the new Figure 1. I consider that I have made only one change, even though old Figures 1 through 5 have been renumbered as Figures 2 through 6. DiffDoc counts this situation as 6 total changes (it includes all of the incidental renumberings as changes), and they tell me it can’t be reconfigured to ignore the “incidental” changes. Any quality process I implement using such change counts will be seriously flawed, especially with my large documents. Anyone out there got any suggestions?

Reply | Quote

This is one tool that does the job. It is relatively inexpensive compared with others. Alternatively, I suspect a macro to do this could be quite lengthy; however, I cannot help you with that. Perhaps one of the moderators can. HTH

http://www.softinterface.com/Compare-File-…le-Programs.HTM%5B/url%5D

Ron M

Reply | Quote

First of all, if your

Reply | Quote

First of all, if your

Reply | Quote

Could you do something using change tracking, or it’s close relative Compare Documents.

This would give you a single document with all changes marked as revisions, you could then use a Macro with code like…

For Each paraNext in ActiveDocument.Content
If paraNext.Revisions.Count > 0 then iParaCount = iParaCount + 1
Next paraNext

For Each tblNext in ActiveDocument.Tables
If tblNext.Revisions.Count > 0 then iTableCount = iTableCount + 1
[Next tblNext

I’m not sure how you would do this for figures, if you have been consistent about the placement of captions it might be possible to do something.

StuartR

Reply | Quote

I’d like to thank all who responded to my request for the special kind of change counting. Jekyl makes a good point that my management may be overemphasizing the numerical solution and should concentrate on more practical ways to impress the client (a military one, I might add). But I do think that their request for software that reports how 2 documents differ “realistically” (i.e., by ignoring insignificant/incidental changes) is a valid one. In fact the more I think about it the more I’m surprised such a tool doesn’t already exist.
I am going to try Jekyl’s recommendation, DeltaView, next, and maybe CompareRite (or Compare Plug-in?).
Thanks, Stuart, I tried to use your code but since I’m not a VBA expert I got errors. Nor could I tell what the code was designed to do.
Macropod, if I’m reading your “compensating deletions” workaround correctly, it seems to require a lot of effort when dealing with autonumbered figures and tables, and wouldn’t be possible to apply to old documents. Please correct me if I’m wrong about these inferences.
Jscher, I’m still looking into your binary file compare suggestion.

Reply | Quote

I’d like to thank all who responded to my request for the special kind of change counting. Jekyl makes a good point that my management may be overemphasizing the numerical solution and should concentrate on more practical ways to impress the client (a military one, I might add). But I do think that their request for software that reports how 2 documents differ “realistically” (i.e., by ignoring insignificant/incidental changes) is a valid one. In fact the more I think about it the more I’m surprised such a tool doesn’t already exist.
I am going to try Jekyl’s recommendation, DeltaView, next, and maybe CompareRite (or Compare Plug-in?).
Thanks, Stuart, I tried to use your code but since I’m not a VBA expert I got errors. Nor could I tell what the code was designed to do.
Macropod, if I’m reading your “compensating deletions” workaround correctly, it seems to require a lot of effort when dealing with autonumbered figures and tables, and wouldn’t be possible to apply to old documents. Please correct me if I’m wrong about these inferences.
Jscher, I’m still looking into your binary file compare suggestion.

Reply | Quote

Could you do something using change tracking, or it’s close relative Compare Documents.

This would give you a single document with all changes marked as revisions, you could then use a Macro with code like…

For Each paraNext in ActiveDocument.Content
If paraNext.Revisions.Count > 0 then iParaCount = iParaCount + 1
Next paraNext

For Each tblNext in ActiveDocument.Tables
If tblNext.Revisions.Count > 0 then iTableCount = iTableCount + 1
[Next tblNext

I’m not sure how you would do this for figures, if you have been consistent about the placement of captions it might be possible to do something.

StuartR

Reply | Quote