Author Topic: Page for fakes  (Read 1454 times)

0 Members and 1 Guest are viewing this topic.

Offline Jeru

Page for fakes
« on: 2013 May 15, 10:43:08 PM »
SANDAC suggested that CC should have a designated area for fakes.  His exact words were;

"Concerning fake identification and flagging:  Your team scan the market daily and enter NGC cert numbers sightings so you are in an unique position to track fakes and report their appearances and history.  One easy way to identify fakes are the "DELETED" designation in the NGC database.  When the CC team enter NGC cert number, they should check to see if the coin is marked DELETED.  If so, that information should be made available in the CC homepage.  Examples: CC2777, CC13378, CC13391, CC2871.  My database represents about 60% of NGC MCC and I have about 740 DELETED entries, so I estimate there are a total of 1200-1300 DELETED entries for MCC.  If you created a dedicated FAKE area, I would be happy to put all my DELETED entries there.

This is an important issue and the CC's ability to alert buyers in a visible and timely fashion would be a great service to the MCC community."

I think this is a good idea so we, the team, can also learn how to identify the fake coins.
"...but the greatest of these is love." -1 Corinthians 13:13
 

Offline badon

Re: Page for fakes
« Reply #1 on: 2013 May 16, 03:25:20 AM »
Everyone is supposed to verify their specimens after they enter them to be sure they entered the certification number correctly. If that procedure is being done, then that will give the CC team the opportunity to see that a coin has been deleted from NGC's database. At that point, a "Deleted" problem should be entered for the specimen, and a screenshot of the sighting venue and the NGC verification page should probably be made too. If I understand what SANDAC described, then we would need to generate a list of newly entered specimens with the "Deleted" problem. The CC can already do that, and we had a list almost exactly like that on the main page of the site, but it was updated so infrequently, I removed it.

Also, I think what SANDAC is really asking for is a "Fake" type, or something like that. One thing I've experiment with is putting fake types with genuine types, but that creates a situation where images and sightings are mingled with authentic coins, which is kind of messy. Of course, it's easy to make minor adjustments to the site's code to exclude those, but the number of adjustments we would have to make are unlimited. I had another idea that maybe we could create a new Type form field that is "Fake of type", so when a type needs to be entered for a fake coin, it will be easier to separate it from the genuine coins. That idea is half-baked, and may not be a complete solution.

Right now, I can think of 2 distinct kinds of fakes:

1. Fakes of a genuine specimen
2. Fakes of a genuine type

#1 is basically just a copied certification number on a fake slab. Maybe the genuine certification number will match the fake, and maybe it won't. In other words, 1234567-890 could be a genuine panda specimen, but the fake coin in the fake slab with that number could be a fake lunar. The fake needs to be connected to the genuine panda specimen somehow so anyone learning about it will be aware that it exists, and the fake will need to be connected to the genuine lunar type too. I'm not sure yet what the best way is to do that.

Things get more complicated when it's a genuine coin in a fake slab. What if there is a genuine common panda in a fake slab for a rare panda? Is it #2, a fake type? What if the coin and the slab are both genuine, and it's only the certification number that has been altered to make it look like a rare coin? It's a challenge to classify all known coins in a consistent way that makes the best use of the CC's abilities, but fakes and altered coins are much more difficult, and I haven't researched the problem will enough yet to be sure that it can be done. Maybe there's a way to do it, and I just need to spend some time focused on the problem to solve it.

This is a good time to discuss this issue, because the CC is a more mature, proven technology now. There are a lot of interesting ideas that I thought the CC would be able to do, that it is turning out the CC should not do, even if it can, because there are unforeseen technical limitations would be disruptive of the CC's primary goals (recording all known types, specimens, and sightings).

One thing we can try is just creating a fake type in the CC, and just adding everything we know about to that type. Then, when we have a lot of diverse fake information, we can look for patterns in the data that we can design the CC's fake handling abilities around. For example, when I was first designing the CC, I had no idea what it should look like. After accumulating test data for a while, the pattern of types, specimens, and sightings quickly became clear, and now that pattern is the foundation of the CC.

Jeru, if you want, you can create a stand-alone fake type that doesn't belong to any other types. Then, we can start accumulating data there until we decide what the best way to handle it is. We already have some data in the CC, that's kind of a jumbled mess. We can move those to get them out of the way for now. Also, I've been sitting on a bunch of fake data with no ideas yet of what to do with it. I could hand it over to you, and you could just stick it under the fake type, so we'll at least have the data recorded. Then, we can worry about making the data useful later, when we gain experience with entering it, researching it, and using it.
 

Offline Jeru

Re: Page for fakes
« Reply #2 on: 2013 May 16, 03:01:49 PM »
Like this, CCT4091: Fake?
"...but the greatest of these is love." -1 Corinthians 13:13
 

Offline SANDAC

  • Junior Monkey 10+ posts
  • *
  • Posts: 22
Re: Page for fakes
« Reply #3 on: 2013 May 16, 06:32:07 PM »
A fake type is exactly what I have in mind.  Fake is difficult to classify, because of its very nature, so a dedicated master type to compile various types of fakes and then linked into other types would 1., shine a strong light on fakes and, 2., allow entry of known fakes into CC database so when it surfaces in market place it will be flagged immediately.  I'll now enter all my known "DELETED" NGC numbers under CCT4091.  Please feel free to move them around as you see fit.
 

Offline Jeru

Re: Page for fakes
« Reply #4 on: 2013 Jun 05, 04:24:22 AM »
I have a question...

Now that all the fake types of CC has been transferred to CCT4091: Fake... is it okay that the fakes types are still somehow connected to the real types?  Like this, CCT1905: 1984 22 g silver ancient pagoda temple Kaiyuan fake 1 mirror?  I thought it should be so people would know that the real ones have fake versions, but I don't know if that's just okay.
"...but the greatest of these is love." -1 Corinthians 13:13
 

Offline badon

Re: Page for fakes
« Reply #5 on: 2013 Jun 05, 08:59:58 PM »
No, you should remove the fakes from the genuine types. I'm going to add features to handle fakes in a better way, when I finish the project I'm currently working on.
 

Offline Jeru

Re: Page for fakes
« Reply #6 on: 2013 Jun 19, 03:44:01 PM »
So, if NGC helps us determine fakes by giving them deleted grades, how does PCGS help us determine fakes?
"...but the greatest of these is love." -1 Corinthians 13:13
 

Offline badon

Re: Page for fakes
« Reply #7 on: 2013 Jun 19, 05:08:28 PM »
I don't know! NGC isn't consistent in deleting fakes. That seems to be either a new thing they're doing (in the last few years), or something they do in special circumstances that I don't fully know about yet.
 

Offline Jeru

Re: Page for fakes
« Reply #8 on: 2013 Jun 24, 02:52:50 PM »
"I don't know.  I have not studied PCGS in details, so I don't know how they handle fake slab.  I do know that if PCGS made a mistake and put fake coin in PCGS slab, they'll just remove the entry completely as if it never existed." - SANDAC

Thank you for your answers, badon and SANDAC!


"...but the greatest of these is love." -1 Corinthians 13:13
 

Offline Jeru

"...but the greatest of these is love." -1 Corinthians 13:13