You have these problems with the data quality e.g. in SCO as well. But if you there follow the coins into their collection of origin, then quite often the error comes from the collection of the ANS itself.
I also think that we often don't have errors of cataloging (if a bronze coin is said to be of silver ) but errors of data import:
http://numismatics.org/pco/id/cpe.1_1.357
http://numismatics.org/pco/results?q=material_facet%3A%22Silber%22&start=260 (see the first couple of coins)
But what the ANS organized during the last years to digitize and to bring together different collections is really tremendous and a great gain for all collectors .
And because the budgets of all the institutions involved is limited, these defects arise and cannot be cured immediatedly .
Regards
Altamura
Thanks for your input. Not sure what's going on entirely. The first page link you have shows dozens identical images of a single bronze coin with a single British Museum
catalog number, each listing with a different
weight, and it is classified as a 'silver decadrachm'. Maybe they haven't photographed all the specimens and use the single image as a place-holder. It's hard to interpret such nonsense on a site that obviously took a great deal of effort to build. The IT is beautiful, the content... sometimes not so much.
Here's another PCO page that's fun:
http://numismatics.org/pco/id/cpe.1_2.B290This page refers to 24 specimens of
Lorber CPE B290, and has 21 coin photos and listings. Out of the 21 coins shown, it looks like one actually *is* a specimen of B290. Can you find it? In this case it's the catalogers who appear to have erred - at a rate exceeding 95%. The catalogued
types of most of them don't even correspond to the site's namesake - the CPE book itself.
You may also get a smile out of the nonsensical '
average'
die axis calculation.
Die axes aren't scalar values (like
weights) so you can't just
average them like ordinary numbers; they are directions (they are vectors) so 3h + 9h does not
average to 6h, those '
average' to (
null). It's not a hard calculation to
average some vectors, but someone has to be both numismatically *and* mathematically aware of what
die axes mean. And the folks who make these sites clearly are not, so what's even the point of the ridiculous calculated values?
These sites are released with social media blasts and accompanying praise and fanfare and the folks who are responsible for them are pretty sensitive about their problems. Beautiful, indeed they are. Reliable research
tools, not. They are nevertheless useful for what they can actually do well. They give pointers that may lead to useful data. For example, the photos and listings on the 'B290' page are great - and do include one that I actually was looking for.
The technical achievement is (mostly) brilliant and we can
hope the content will catch up some day. An enormous amount of
ANS's
money (read: grant funding) and time have been expended to develop these very complex sites and we can
hope a small fraction of that fortune will some day make the data on the sites as useful as possible. The sponsors' priorities aren't clear and we could be in for a long wait. New sites (e.g. the recent Antigonid Coins Online) seem to appear faster than data on existing ones is being updated.
PtolemAE