Comment to The Post-MARC Era, Part 2, Where the Problems Lie, Part 3 by Roy Tennant, who discusses inconsistency in catalog data.
You mention “… we all need to own this problem and work against the forces of inconsistency …” (concerning the inconsistency in the place of publication information [Place of publication not identified])–JW)
but it seems to me that the current trends in cataloging don’t care much about consistency. Titles can now ALL CAPS if someone wants; additional authors can be traced or not; you can add a subtitle or not. Adding the relator codes are optional. Often, thousands of records are made into a semi-MARC format and dumped into the catalog with almost no controls at all. So, it is tough to say that we need consistency in something like [Place of publication not identified]
when users probably won’t even notice it or care much about it if they do.
I am a #1 advocate in favor of consistency, but it seems that information such as [Place of publication not identified] is a perfect candidate for allowing a measure of inconsistency, especially when compared to the other points I mentioned above.
If we want to turn place of publication into an area for searching or limiting, that may be another factor, but there is such wide variation in the spelling of place names in the place of publication area that getting half-way decent results for place of publication would make the inconsistency found in [Place of publication not identified] look like child’s play.