[Taxacom] Data quality in aggregated datasets

Fred Schueler bckcdb at istar.ca
Sun Apr 21 20:54:07 CDT 2013

Quoting Robert Mesibov <mesibov at southcom.com.au>:

> I write directly to data providers with messages like 'Could you  
> please check [records]? It looks like there are errors in the  
> lat/lon's, which probably should be [X]. Many thanks.'

* that's important, but as you say there ought to be support for data  
checking all the way along the process. There are problems even  
upstream of those - I've heard of a case where someone gave thousands  
of collections to a certain museum, with carefully prepared labels  
printed from her detailed database for each species lot, and recently  
got the entered data back - and found that in about 5% of the cases  
they'd used the wrong locality, as if their database structure is so  
complex that they can't directly import records into it from a simple  
database table, and need to block & copy fields from the source by  
hand, and tended to pick up the wrong location, and didn't check any  
labels they'd printed against those that came with the specimens.

In Canada one of our problems is problematic data structures, thrown  
together just anyhow, or inappropriately imported from other  
applications, and then used for decades without becoming at all  
improved. Some of us have recently concluded that if you've formed a  
good impression of a person or institution, their data structure will  
make sense, but if you're worried about them on other criteria, their  
data structure will be inappropriate for the use they or anyone else  
will make of it.

          Frederick W. Schueler & Aleta Karstad
Bishops Mills Natural History Centre - http://pinicola.ca/bmnhc.htm
Mudpuppy Night in Oxford Mills - http://pinicola.ca/mudpup1.htm
Daily Paintings - http://karstaddailypaintings.blogspot.com/
          South Nation Basin Art & Science Book
     RR#2 Bishops Mills, Ontario, Canada K0G 1T0
   on the Smiths Falls Limestone Plain 44* 52'N 75* 42'W
    (613)258-3107 <bckcdb at istar.ca> http://pinicola.ca/

More information about the Taxacom mailing list