FLHIG

Family and Local History
Indexing Group

Selected articles from the Newsletter

This article first appeared in SIGGNL 7  (November 1994)

What does the user want? Comments on a feature of MACREX
by  Tony Rydings

Introduction

When designing a complicated index, it is all too easy to forget the essential purpose: what does the user want to find from the index?  Having nearly fallen into this error recently, I thought it might be helpful to others to explain how it happened.

The method of compiling the index to the Shropshire quarter sessions was described in SIGG newsletter no 3, and all that need be repeated here is that the project envisaged a contents list in the order of the original documents, followed by three separate indexes, for persons, places and trades.  This is not the place to argue the case for one versus three indexes, which in any case was a decision of the sponsoring bodies, the Shropshire County Records Office and the Shropshire Family History Society.  The first two are straightforward name indexes, but the last presents special problems.

As originally designed, and carried out in the first five-year cumulation, the trades index, which included descriptions as well as occupations, such as "Baronet", "aged 25 years, deceased" or "vagrant", the entries were to be sub-arranged first by place, and then by surname, so that typical entries might read:

Accountant: Priors Lee, WOOLLEY  QR349  69

Auctioneer: Newport, HOLLAND  QR340  93

This proved to be very difficult to set up and proof-read, so it was agreed that for the second cumulation the "place" field would be omitted, and entries would be of the form:

Auctioneer: EDGE, John  QR371  30

Barber: McGILL, John  QR373  132

Another complication which was noted in the first cumulation, but which was kept in the second, was that if more that one piece of information was recorded in the "trades" field, all would be included in the index, thus:

Farmer/aged 37 yrs: DOWNING, Francis   QR363  453

Miller/farmer/juror: HILES, Richard  QR355  354/053

This occurred frequently in the entries for jurors, whose trades were often given, and it was not unusual for two occupations to be shown.  Double (or more) entries are made, e.g. for the last item above there are entries under miller, farmer and juror.

In checking through the file for a final editing it became evident that entries for the same person might be separated because the data in the "trades" field varied, as in the following extract:

Baker: COLEMERE, Joseph  QR369  44, 218, 221

             EVERALL, Thomas  QR369  25

             GREEN, William QR357  96

             THOMAS, Richard  QR364  152

             WALDRON, John  QR364  186,  QR369  256, 257

             WEAVER, George  QR368  127

Baker/confectioner: THOMAS, Richard   QR364  148

Baker/confectioner/grocer: COLEMERE, Joseph   QR369

Baker/juror: EVERALL, Thomas  QR369  364/094

Here the two entries for COLEMERE, EVERALL and THOMAS are quite a way apart (in the actual file there are even more entries in between), and it seemed to me that it would be helpful to rearrange them so that they came together.  Using the "curly brackets" device of the MACREX program, this is easily achieved.  The last three entries are amended as follows:

Baker{/confectioner}: THOMAS, Richard   QR364  148

Baker{/confectioner/grocer}: COLEMERE, Joseph   QR369

Baker/{juror}: EVERALL, Thomas QR369  364/094

and the order of entries is changed to the following:

Baker: COLEMERE, Joseph   QR369  44, 218, 221

Baker/confectioner/grocer: COLEMERE, Joseph  QR369

EVERALL, Thomas   QR369  25

Baker/juror: EVERALL, Thomas QR369  364/094

GREEN, William   QR357  96

THOMAS, Richard   QR364  152

Baker/confectioner: THOMAS, Richard   QR364  148

WALDRON, John   QR364  186,  QR369  256, 257

WEAVER, George   QR368  127

This is not very easy to read, and when I looked at the part of the trades, etc. file with the entries for jurors, it seemed to me that it would look much tidier if they were lined up with the surnames all in one column, like this:

Juror:

          druggist:   BRADBURY, Charles   QR352  68/045

          /yeoman:  BRADBURY, George QR360 310/083, QR361  193/082

          /mercer:    BRADBURY, Walter   QR372  250/050

                            BRADDOCK, Stephen   QR367  215/08

          /farmer:     BRADELEY, William   QR364  469/048

Although this version has a much improved appearance, it shares with the one above it a feature which may lead to confusion.  The entry for BRADDOCK,  is the trade here Juror or Juror/mercer?  Similarly in the previous example, what about WALDRON and WEAVER?  As can be seen from the last entry on the previous page, the answer is Baker rather than Baker/confectioner, though the opposite might well be expected.

It seems that the only way to make the entries entirely clear is to repeat the main heading whenever there are main headings interspersed with main headings/subheadings.  Thus the first four lines of the "Baker" example would read:

Baker: COLEMERE, Joseph   QR369  44, 218, 221

Baker/confectioner/grocer: COLEMERE, Joseph   QR369

Baker: EVERALL, Thomas   QR369  25

Baker/juror: EVERALL, Thomas QR369  364/094

and the "Juror" example would have to be changed so that all the entries for Juror only would come first, and those with other trades as subheadings would follow, in the alphabetical order of those trades, thus:

Juror:  BRADDOCK, Stephen   QR367  215/08

Juror/druggist:  BRADBURY, Charles   QR352  68/045

Juror/farmer:  BRADELEY, William   QR364  469/048

Juror/mercer:  BRADBURY, Walter   QR372  250/050

Juror/yeoman:  BRADBURY, George   QR360  310/083, QR361  193/082

This change of order of entries under "Juror" is as requested by the Records Office.

It will therefore be seen that MACREX gives one almost complete freedom to arrange entries in any order and layout that is required, but the indexer should decide which arrangement best meets the user's needs.

There are  other problems in a rather unusual index such as that for the Quarter Sessions, and in a future newsletter I will attempt to describe them and possible remedies.

Ç TOP