[Taxacom] HHDB: hemihomonyms

Francisco Welter-Schultes fwelter at gwdg.de
Sun Jan 17 17:20:38 CST 2010


These are good proposals, Pat. Will think about how we can implement such
ideas in the BHL project.

> 2) BHL needs to code their titles (volumes) as botanical, zoological,
> bacteriological, mixed, or not taxonomic.  This is not as big a task
> as it might seem as entire runs of journals are often exclusively
> botanical or zoological.

As far as I can see, this would not be easy. Which specialists would be
paid for the indexing? I don't see any budget anywhere for that. We could
do that in the AnimalBase group. But who would pay us?
Alternative solution: metadata could be added to works by users. I made
likewise proposals concerning other metadata, but the BHL associated
libraries did not like that. They preferred a procedure by which users
would have an opportunity to make suggestions, which is then forwarded to
whom it concerns, so that the library which digitized the work would
modify the metadata.
Possibly they would agree in such a solution for user-created metadata in
fields in which libraries are almost absolutely not skilled.

Another problem in pre-1900 works is that OCR text recgnition is not 100 %
reliable. And if you observe an error rate of only 0.01 % in a work - you
can bet that these errors will concern exactly the taxonomic names.
Try to search for Planorbis duryi in Wetherby 1879, Journal of the
Cincinnati Society of Natural History 2 (2), published in p. 99.
http://library.cincymuseum.org/starweb/journals/servlet.starweb?path=journals/STARArchives.web

English journal, good printing quality, but you will not find it,
incorrectly OCR'd.

> While at it, the works might also be coded
> by discipline: entomology, ornithology, malacology, etc.

If you can explain a librarian the difference between malacology and
entomology... this will be a funny job...

> 4) Finally, the BHL Advanced Search option will need to allow
> taxonomic searches to be restricted by Code, and whether unresolved
> names should be included or excluded.

We would have a budget to implement such a function, AIT in Austria could
be encharged to do that. But I would predict that it would not be perfect.
Which of course, should not be a reason for not trying it out.

Francisco





More information about the Taxacom mailing list