On Wed, February 2, 2011 10:40 am, Karen Coyle wrote: > It occurs to me that ISBNs may be getting normalized in batch input > but not when they are keyed by a user. I must admit that I have no > idea if any quality control is done on user input data -- can anyone > speak to that? There are probably very few fields where QC can be > applied, in fact, but in the case of ISBN it might be worth doing. > > kc
While normalization and validation should be done on any number of input fields (dates are absolutely the most commonly abused) it need not necessarily be done at input time. An algorithm to normalize both sides of the equation before comparison should be trivial to implement for simple fields like ISBN, and it would be much more time-effective than trying to go back and fix all the bad data that currently exists in the OL data sets. _______________________________________________ Ol-tech mailing list [email protected] http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech To unsubscribe from this mailing list, send email to [email protected]
