Re: [ARSCLIST] MIC and cataloguing

[George Brock-Nannestad <pattac@xxxxxxxx>]

From: Patent Tactics, George Brock-Nannestad

Jane, you wrote:

> George:
>
> I'm not sure if you're talking about the mapping of the actual data elements
> (i.e., the fields, such as title, date, contents), or mapping the definitions
> of the actual data *content,* such as a subject descriptor, but allow me to
> address both.  Please see my comments below.
>

----- I thought the expression that you had written ("MIC data element
equivalents") related to the definitions of the *content*, and my comment
really was a warning that if there is a mismatch in the definitions, then
precision in retrieval will be reduced (as we agree). Also, if there is not a
log kept of changes, with sufficient interpretive content, we would end up
with a situation that years from now, someone will say "how on earth did we
get off on this tangent?". In other words, there must be traceability in
revisions.

----- I speak from an experience in a previous life, in which we maintained a
Thesaurus relating to electronics with a view to absorb the explosive
development in the early 1970s. We needed precision in retrieval because we
were attempting to find more precise prior art than the various national
patent offices had used when issuing patents. So, full Boolean search and
range search were very important, as were - something quite unique at the
time - "directional operators" which would allow us to encode the statement
"dog bites postman" in such a way that it would not come out as a false drop
when we searched for "postman bites dog". Oh, had we had interactive data
entry at the time - our productivity would have gone up dramatically!

----- in order to keep responses short and simple I have skipped the rest of
your very informative response - read it at
Fri, 20 May 2005 11:05:45, if you have been tempted by the above.

Kind regards,


George