Scan A Photo, Know the Provenance

I ended up submitted one which has been driving me crazy for a bit because I only know of provenance going back to November 2015(and I have at least 5 years provenance for almost all of my RR coins). If it works out I might well submit my "IIS" sestertius next.

 

Log in or Sign up to hide this ad.

Wonderful results, AJ, congrats!

 

You got a great hit there on the stater and sestertius. Not that is was needed but it is a wonderful extra to have.

Based on that, I tried a few more coins that have no provenance. We shall see.

 

Great results AJ. Thanks for posting those.

 

Wow! That sounds like you have quite a lot of old auction records yet to add. I hope you keep the submissions active until the database is complete, or as near to complete as possible. Do you have any idea when that might be? I understand that there will always be at least a trickle of new data added as copies of small or obscure catalogs are unearthed.

Regarding CoinArchives: do you mean CoinArchivesPro or the public version? Does the public version have the same degree of completeness as the Pro version (going as far back as the cutoff date you mentioned)? I don't have access to the Pro version and don't know if there is a difference in the post-2000 depth of data.

 

I just got an answer on the coin I submitted. No matches found. While this is fine, I know that the coin I submitted was sold at auction on CNG 284, Lot: 66, so one might think it would at least show this information. I will try another later today that I have a partial provenance. I will keep everyone posted.

 

They don't track anything post year 2000

 

Three more that received no hits. I went zero out of four. Used up my free credits.

I will stop there.

 

Looks like he's using a foot pump to get an airbed up..... lovely coin.

 

No hits on the two I submitted either unfortunately.

 

I sent in four of my better coins out of curiosity and got one hit. The new information extended the existing provenance I had for it by just a few more years, but I'm glad to be able to add it.



My picture of the coin, with its new (old) provenance added:


MARK ANTONY
AR Denarius
3.61g, 17.8mm
Patrae (?) mint, autumn 32 - spring 31 BC
Sear RCV Vol. 1 p. 284, 1483 (this coin illustrated); Crawford 544/8
O: ANT AVG above Praetorian galley right, III VIR R P C below.
R: CHORTIVM PRAETORIARVM, Legionary eagle (Aquila) between two standards (signa).
Ex Stack's Sale December 2001, Lot #142
Ex Ponterio & Associates Sale 108 August 2000, Lot #307
Ex Gerhard Hirsch Auction 179, September 1993, Lot #770

 

When I mention "new auctions records" it means not only pictures (most already done) but also estimates, texts and price realised.

 

Uploaded 4 of my coins because I figured why not. I don't expect any matches since my coins aren't really high value but you never know and the credits are free.

 

Took the plunge with the 5 credits to try 5 of many that I know have a long history out of the ground but I can't find anything worthwhile. I let them know of the provenances I already have and will update the thread on what I get back.

 

Wow, Carthago => those are 5 total winners!! (you certainly have several sweet coins)

 

I sent three more with one I know the provenance dating back to JQA. All three came back negative. Hmmmm

 

This is why I think the $5 fee is much too high. You have no idea how effective this search really is. There is no data that I see that says "if the coin is in the database, you will get a hit." Several examples below, of why this might fail.

1) The database could be huge, but if the recognition software is mediocre, you would get misses.

2) If the recognition software is bullet proof, but the database is deficient, you would get misses.

3) If the coin is not from an auction (but from a dealers coinshop for instance), you would get misses.

4) you could have both a deficient database and inefficient recognition software, you would get many misses.

Right now, this is a faith based agreement, that what is being promised, actually works as promised. A few hits do not tell us much unfortunately and the other results reported so far, don't tell us much either.

This is certainly better than nothing, the question is whether it is worth $5 per roll of the dice. I can afford $5 if the thing works, but I hate handing it over for a product that may or may not be good, at what it does. I see too many holes here that need to be plugged before I fund someones beta test.

 

I agree 100% with your observations, and would suggest a change to the business model of this site:

A collector must sign up to use the site, but submissions are free. If the submitted coin gets a hit, the collector is notified along with the amount that would be charged to purchase the resulting provenance. The collector can accept or decline the offer.

Of course, this approach will increase the number of submissions geometrically -- tenfold, probably. Ex-Numis might object since their search engine will be performing lots of searches that either (a) produce no result, or (b) produce a result for which the collector declines to pay. But I think this would be a specious reason NOT to try this business model, since lots of submissions, even if Ex-Numis didn't receive payment for them, would actually provide a good, real-world test for both their algorithms and database.

I also think that Ex-Numis' current business model will ultimately fail for the reason mentioned by 4to2centBC, namely, it's not really worth $5 per submission to take a chance on finding a provenance if there is a 80%+ probability that the search will fail.

 

Then there is some, like me, who really doesn't care about provenance & enjoy the coins they have whether it has anything that can be traced or not.

The few coins I do have that have provenance I think,"ok", and make note of it in my own flips and thats about it. I don't spend time looking at every resource trying to find that exact coin from decades old auctions & $5 is a bit much for people like me who is lucky to have $100 a month for coin purchases in general. I would rather save the $5 to add for more coins, or my many other hobbies.

 

We have numerous pleased customers: private collectors, auction houses, dealers. I am sorry that we did not manage to convince you.

Every submission requests a certain amount of work and we cannot afford to work for free.

I think the lost provenances we found, illustrated on this forum, show the potential of our system. But of course, we will do our best to make it better. An do not forget, some of the coins do not have any provenance at all.