labeling screen shots

[gyaker -at- csc -dot- com]

I?m re-writing a document that was used with our old mainframe-based
application. The new document, more than less, is a users? manual for the new
widows based version of the application. The client wants the new document to
resemble the old one as much as possible (which is completely asinine IMHO, but
that?s besides the point).

 The old document had screen captures of the mainframe screens from a terminal,
 i.e. the old screens are text based. There would be one screen capture on a
 page followed by a few pages of the data elements? names and descriptions. The
 data elements were numbered with a distinctive font set  on the screen shot.
 The screen shot would have an element such as "Last name," and next to those
 words might have been a "1" so the reader of the guide knew that the number
 wasn?t an actual part of the screen. A page later, the #1 data element was
 "last name" with it?s description along side in a numbered table of all data
 elements on the corresponding screen.

 So, is there a better way to do this? Is it enough to have a clear screen shot
 and then in the subsequent pages list the names of the Data Elements, so that
 the user views "last name" on the screen shot, and views "last name" in the
 list of data elements, and that?s enough for him or her  to make the
 connection. Or, do users need numeric clues in addition to the verbal clues,
 like in the original document?

 Thanks

Gil Yaker
gyaker -at- csc -dot- com
Computer Sciences Corporation
Federal Black Lung Project