All I know in 61% of ColonelJessups posts I end up having to look up a word
m
Walker Proof Digital Album Fellas, leave the tight pants to the ladies. If I can count the coins in your pockets you better use them to call a tailor. Stay thirsty my friends......
@Justacommeman said:
All I know in 61% of ColonelJessups posts I end up having to look up a word
m
Amazing - that's the exact percentage for me! BUT the other 39% of the time, I SHOULD be looking up one or more words. It's just that I'm not even aware of it.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
@streeter said:
Well, Heritage had their finger in the pie at NGC, IIRC.
Didn't seem to bother people.
I'm not clear but didn't Heritage also have an interest in CDN? Some people might have been a little leery despite the proclamation of independence.
My thought was that CU was a natural fit to run NGC and CU products (and possibly others) on a collectible auction service. There's going to be an explosion of authenticated and encapsulated product categories in the next few years. CU & NGC et al, products, if they were separated out from eBay by a newly formed auction service would possibly be a valid business model.
eBay is a jumbled mess of offerings and eBay will not fix the shoddy sellers or counterfeit listings. IMHO, people have dropped eBay because of this and other reasons. I just don't see how eBay can fix itself at this point.
I.E. the new CU buys Great Collections and expands it to gradually broaden the categories. They could give Heritage a serious challenge. The collectible market is very fragmented right now. A well run company with horsepower could position itself well and benefit buyers and sellers.
I used to sell coins by mail, but then I had to buy the Post Office and so I stopped.
There is NO REASON why eBay needs to own a grading service to use the services. eBay could require that all collectibles sold on their website be certified where possible. They would, of course, lose a lot of business given that there are a lot of uncertified collectibles out there. [Side note: Does anyone read comics anymore or just admire the covers through the giant hard plastic case?]
I'm not aware of any ownership of NGC by Heritage at any time. Some of the players have moved through different corporate entities over the years but that's not exactly the same thing. But I reserve the right to be wrong.
The only company I'm aware of that does both is CDN and that bothered me for a while. It still potentially bothers me in a post-JA world. It would be very tempting in hard times to slap a sticker on a coin to increase it's value. I trust JA. I'm reserving judgment on the post-JA CAC - assuming I'm still alive.
Heck, there are a number of forum members who are convinced that the big retailers get preferential grading boosts from the TPG's as it is. Imagine if the retailer were actually the same corporate entity. Maybe we need a poll.
@jmlanzaf said:
eBay could require that all collectibles sold on their website be certified where possible. They would, of course, lose a lot of business given that there are a lot of uncertified collectibles out there.
And, due to the expense of certifying coins, there always will be. As noted above, the overwhelming majority of eBay listings are for uncertified coins and I'd bet, if one were to count them all up, one would find that raw coin collectors outnumber collectors of slabbed coins.
Two comments, and excuse me if either has been mentioned. I read most of the thread but not every word.
Automated scanning/grading is entirely possible from a technological standpoint, and would be more consistent than grading by different experts. Biggest problem is agreeing on weighting of factors. No single set of standards will please the wide range of collector preferences or disagreements among the experts as to how much various defects weigh in the grade. Finding a satisfactory coin for the grade might be easier under the present system than in the more consistently graded universe.
Yes, the TPGs are trying to expand into new areas but there's a BIG problem that hasn't been noted. Coins, stamps and currency are designed to be secure-that is as hard to duplicate as possible. This makes the job of creating fakes much harder than making fakes of a snowboard, action figure or baseball card, for that matter. I know what the mint and BEP does with it's dies and plates but what did Topps do with the photos, artworks, plates for baseball cards? Could you possibly locate some unused 50-75 year old cardboard stock somewhere?
There are many collectibles that are MUCH easier than coins and currency to copy and copy quite well. This is the biggest challenge to the TPGs expanding into other areas. It's gonna happen, but this will be a limiting factor and could end certifying of certain items if undistinguishable fakes are discovered.
Collector and dealer in obsolete currency. Always buying all obsolete bank notes and scrip.
@sellitstore - you raise some very good points. Especially with the fate of discarded assets of some of these collectible areas. Given how well die struck counterfeits are now it would be an issue for sure with other collectibles.
@pmh1nic said:
I was on the forum back in the late 1990’s and early 2000’s. Nothing I’ve seen in my recent return rivals some of the utter nonsense that went on back then. I think the moderators have found a good balance in letting discussions go forward, sometimes two steps backward, among people with widely different perspectives on the hobby, short of profanity and threats, allowing some sarcasm and jabs.
I’d like to see more technology brought into the grading process. It’s often said grading is a combination of science and art. I’d like to see more science involved in the grading process. The failed attempts 20 years ago shouldn’t mean that effort should be tossed aside forever.
Not an AI expert by current standards, but coin grading is not "AI-prone" for much collector coinage. The algorithm for 81-S dollars is not the same as that needed for the 83-O. The software to calculate grade based on the algorithm is secondary to the accumulation of data about textures, strike tendencies, etc. that must be derived and loaded. Never ever going to happen done close enough to "right" Not enough object capture events available to build the necessarily-nuanced libraries for self-teaching.
Computer grading of ASE's and AGE's was doable twenty years ago. Scanning capabilities are far beyond what's required.
I’m not an AI expert but if I had a dollar for every time I’ve heard it can’t be done I be the owner of a 1804 Silver Dollar. We’re approaching the point of facial recognition of billions of people. We can jam billions of transistors in a 2”x2” microprocessor. We have SMT assembly equipment that has the programming and capacity to accurately place over 50,000 components the size is a grain of pepper in an hour. If the human eye can make the distinction there is optics and software that can match or exceed it. PCGS seriously looked at computer grading in the 1990’s. I’m not sure what happened that caused it not to materialize (capability or marketability) but a heck of a lot has happened technology wise in the last 30 years.
How many foci are needed for facial recognition? Did you look at the Compugrade scan protocol? No matter.
From time immemorial (or at least post WWII), grading has part of the adventure. My generation (Daltry stuttering in the background) had an opportunity in the 80's unlike any other. If you were a devoted auction geek or a major bourse-table presence, the sheer volume of truly rare and often wonderful coins was augmented by acres and acres of very choice type and cascades of gold. .
Never again will those 50-75 million coins already graded be available for reference.
Call me a romantic, but I'd rather trust my sense of eye-appeal for plus/minus 0.2 grading points than I would a computer-derived valuation. The subjective nature of your own or anyone else's particular "yummy factor" is beyond calculation. Otherwise, calculably graded coin ETFs would be quite suitable as a meme stock for RobinHood
I don't know the answer to the first question and no to the second but whatever the protocol was 20 years ago probably doesn't have a lot of bearing on the capabilites today given advances in technology and software since then (you remember the Macintosh and floppy drive). That said this might give an indication of where the technology is compared to human eyesight...
"The GaussianFace algorithm developed in 2014 by researchers at The Chinese University of Hong Kong achieved facial identification scores of 98.52% compared with the 97.53% achieved by humans. An excellent rating, despite weaknesses regarding memory capacity required and calculation times."
Keep in mind that this was almost ten years ago and I'd imagine that this capability has improved significantly since then while not much improvement in human grading performance has occurred.
Regarding the 50~75 million coins, you don't need 50 or 75 million as a reference. What you would start with are the best examples of the coins available and use that as the benchmark. Humans would be involved in establishing algorithms for features we subjectively consider attractive. Of course the programs will be consistent in determining attractiveness while humans will tend to vary. The buyer will be the final arbiter on what the find attractive.
1) 50-75 million of earlier best of the best will not be available. You get to scan nothing graded before. Not swill, but the cream has been skimmed off the top. The best of the rest are less than the best..
2) Glad you've read a few articles. "Of course the programs will be consistent in determining attractiveness".
I cannot contrive an epithet sufficient to describe my contempt for this concept on several levels.
The only accurate register for eye-appeal is how one's viscera are effected. Computers have no soul. Useful on facts, useless on beauty.
YMMV
Parenthetically, Before I entered coins, I had 20+ high-level data-base and system network engineers working for me. I have a pal whose work on for the Defense Department using AI for target identification has kept him very up-to-date. $300K a year up-to-date. He laughs at what I know, but is occasionally surprised by an insight. And I had a five-minute conversation with Isaac Asimov!
I know nothing, and less every day . Buh-bye
edited to add: I actually do know one sure thing. No one's going to fund it. Huge expenditure for limited minor incremental reward. As I repeat (once an 'effin again), you are 100% "on target" for ASE's, AGE's and Moderns.
"People sleep peaceably in their beds at night only because rough men stand ready to do violence on their behalf." - Geo. Orwell
@Justacommeman said:
All I know in 61% of ColonelJessups posts I end up having to look up a word
m
Amazing - that's the exact percentage for me! BUT the other 39% of the time, I SHOULD be looking up one or more words. It's just that I'm not even aware of it.
Ironically, I have to look up words for 68% of my posts. But "prolixity" is not one of them.
"People sleep peaceably in their beds at night only because rough men stand ready to do violence on their behalf." - Geo. Orwell
He's easier to understand than trying to
listen to, and understand, Nugget-eese !
Retired Collector & Dealer in Major Mint Error Coins & Currency since the 1960's.Co-Author of Whitman's "100 Greatest U.S. Mint Error Coins", and the Error Coin Encyclopedia, Vols., III & IV. Retired Authenticator for Major Mint Errors for PCGS. A 49+-Year PNG Member...A full numismatist since 1972, retired in 2022
I'm wondering if and when Heritage might be bought out. Seems like a good time to sell to me.
I also consider the business model more compelling, higher prices are a lot more correlated to revenue increases. You can also sell the same items repeatedly and it provides value to the customer. Only so many times you can grade the same collectible over and over again.
I suggest when it's convenient to take the time to read that thread. Lot of good info, links and comment from members who are not on the forum much or if at all.
I only go back to coins to 1962. Sometimes my memory needs to be refreshed. I really don't know much, that's obvious.
@FredWeinberg said:
He's easier to understand than trying to
listen to, and understand, Nugget-eese !
While I've actively resisted joining Facebook for at least a decade, Legend numismatist Greg Cohen convinced me to join, specifically to participate in a special interest group whose main project is trolling Paul Nugget. Unsurprisingly, Kevin has a scatological profile name but, regrettably, is much less effective in print than in person . Scholarships are available
"People sleep peaceably in their beds at night only because rough men stand ready to do violence on their behalf." - Geo. Orwell
@streeter said:
I suggest when it's convenient to take the time to read that thread. Lot of good info, links and comment from members who are not on the forum much or if at all.
I only go back to coins to 1962. Sometimes my memory needs to be refreshed. I really don't know much, that's obvious.
Actually, both NGC and Heritage are privately owned and Ivy and Halperin appear to have ownership stakes in both.
@FredWeinberg said:
He's easier to understand than trying to
listen to, and understand, Nugget-eese !
While I've actively resisted joining Facebook for at least a decade, Legend numismatist Greg Cohen convinced me to join, specifically to participate in a special interest group whose main project is trolling Paul Nugget. Unsurprisingly, Kevin has a scatological profile name but, regrettably, is much less effective in print than in person . Scholarships are available
@pmh1nic said:
I was on the forum back in the late 1990’s and early 2000’s. Nothing I’ve seen in my recent return rivals some of the utter nonsense that went on back then. I think the moderators have found a good balance in letting discussions go forward, sometimes two steps backward, among people with widely different perspectives on the hobby, short of profanity and threats, allowing some sarcasm and jabs.
I’d like to see more technology brought into the grading process. It’s often said grading is a combination of science and art. I’d like to see more science involved in the grading process. The failed attempts 20 years ago shouldn’t mean that effort should be tossed aside forever.
Not an AI expert by current standards, but coin grading is not "AI-prone" for much collector coinage. The algorithm for 81-S dollars is not the same as that needed for the 83-O. The software to calculate grade based on the algorithm is secondary to the accumulation of data about textures, strike tendencies, etc. that must be derived and loaded. Never ever going to happen done close enough to "right" Not enough object capture events available to build the necessarily-nuanced libraries for self-teaching.
Computer grading of ASE's and AGE's was doable twenty years ago. Scanning capabilities are far beyond what's required.
I’m not an AI expert but if I had a dollar for every time I’ve heard it can’t be done I be the owner of a 1804 Silver Dollar. We’re approaching the point of facial recognition of billions of people. We can jam billions of transistors in a 2”x2” microprocessor. We have SMT assembly equipment that has the programming and capacity to accurately place over 50,000 components the size is a grain of pepper in an hour. If the human eye can make the distinction there is optics and software that can match or exceed it. PCGS seriously looked at computer grading in the 1990’s. I’m not sure what happened that caused it not to materialize (capability or marketability) but a heck of a lot has happened technology wise in the last 30 years.
How many foci are needed for facial recognition? Did you look at the Compugrade scan protocol? No matter.
From time immemorial (or at least post WWII), grading has part of the adventure. My generation (Daltry stuttering in the background) had an opportunity in the 80's unlike any other. If you were a devoted auction geek or a major bourse-table presence, the sheer volume of truly rare and often wonderful coins was augmented by acres and acres of very choice type and cascades of gold. .
Never again will those 50-75 million coins already graded be available for reference.
Call me a romantic, but I'd rather trust my sense of eye-appeal for plus/minus 0.2 grading points than I would a computer-derived valuation. The subjective nature of your own or anyone else's particular "yummy factor" is beyond calculation. Otherwise, calculably graded coin ETFs would be quite suitable as a meme stock for RobinHood
I don't know the answer to the first question and no to the second but whatever the protocol was 20 years ago probably doesn't have a lot of bearing on the capabilites today given advances in technology and software since then (you remember the Macintosh and floppy drive). That said this might give an indication of where the technology is compared to human eyesight...
"The GaussianFace algorithm developed in 2014 by researchers at The Chinese University of Hong Kong achieved facial identification scores of 98.52% compared with the 97.53% achieved by humans. An excellent rating, despite weaknesses regarding memory capacity required and calculation times."
Keep in mind that this was almost ten years ago and I'd imagine that this capability has improved significantly since then while not much improvement in human grading performance has occurred.
Regarding the 50~75 million coins, you don't need 50 or 75 million as a reference. What you would start with are the best examples of the coins available and use that as the benchmark. Humans would be involved in establishing algorithms for features we subjectively consider attractive. Of course the programs will be consistent in determining attractiveness while humans will tend to vary. The buyer will be the final arbiter on what the find attractive.
1) 50-75 million of earlier best of the best will not be available. You get to scan nothing graded before. Not swill, but the cream has been skimmed off the top. The best of the rest are less than the best..
2) Glad you've read a few articles. "Of course the programs will be consistent in determining attractiveness".
I cannot contrive an epithet sufficient to describe my contempt for this concept on several levels.
The only accurate register for eye-appeal is how one's viscera are effected. Computers have no soul. Useful on facts, useless on beauty.
YMMV
Parenthetically, Before I entered coins, I had 20+ high-level data-base and system network engineers working for me. I have a pal whose work on for the Defense Department using AI for target identification has kept him very up-to-date. $300K a year up-to-date. He laughs at what I know, but is occasionally surprised by an insight. And I had a five-minute conversation with Isaac Asimov!
I know nothing, and less every day . Buh-bye
edited to add: I actually do know one sure thing. No one's going to fund it. Huge expenditure for limited minor incremental reward. As I repeat (once an 'effin again), you are 100% "on target" for ASE's, AGE's and Moderns.
@pmh1nic said:
I was on the forum back in the late 1990’s and early 2000’s. Nothing I’ve seen in my recent return rivals some of the utter nonsense that went on back then. I think the moderators have found a good balance in letting discussions go forward, sometimes two steps backward, among people with widely different perspectives on the hobby, short of profanity and threats, allowing some sarcasm and jabs.
I’d like to see more technology brought into the grading process. It’s often said grading is a combination of science and art. I’d like to see more science involved in the grading process. The failed attempts 20 years ago shouldn’t mean that effort should be tossed aside forever.
Not an AI expert by current standards, but coin grading is not "AI-prone" for much collector coinage. The algorithm for 81-S dollars is not the same as that needed for the 83-O. The software to calculate grade based on the algorithm is secondary to the accumulation of data about textures, strike tendencies, etc. that must be derived and loaded. Never ever going to happen done close enough to "right" Not enough object capture events available to build the necessarily-nuanced libraries for self-teaching.
Computer grading of ASE's and AGE's was doable twenty years ago. Scanning capabilities are far beyond what's required.
I’m not an AI expert but if I had a dollar for every time I’ve heard it can’t be done I be the owner of a 1804 Silver Dollar. We’re approaching the point of facial recognition of billions of people. We can jam billions of transistors in a 2”x2” microprocessor. We have SMT assembly equipment that has the programming and capacity to accurately place over 50,000 components the size is a grain of pepper in an hour. If the human eye can make the distinction there is optics and software that can match or exceed it. PCGS seriously looked at computer grading in the 1990’s. I’m not sure what happened that caused it not to materialize (capability or marketability) but a heck of a lot has happened technology wise in the last 30 years.
How many foci are needed for facial recognition? Did you look at the Compugrade scan protocol? No matter.
From time immemorial (or at least post WWII), grading has part of the adventure. My generation (Daltry stuttering in the background) had an opportunity in the 80's unlike any other. If you were a devoted auction geek or a major bourse-table presence, the sheer volume of truly rare and often wonderful coins was augmented by acres and acres of very choice type and cascades of gold. .
Never again will those 50-75 million coins already graded be available for reference.
Call me a romantic, but I'd rather trust my sense of eye-appeal for plus/minus 0.2 grading points than I would a computer-derived valuation. The subjective nature of your own or anyone else's particular "yummy factor" is beyond calculation. Otherwise, calculably graded coin ETFs would be quite suitable as a meme stock for RobinHood
I don't know the answer to the first question and no to the second but whatever the protocol was 20 years ago probably doesn't have a lot of bearing on the capabilites today given advances in technology and software since then (you remember the Macintosh and floppy drive). That said this might give an indication of where the technology is compared to human eyesight...
"The GaussianFace algorithm developed in 2014 by researchers at The Chinese University of Hong Kong achieved facial identification scores of 98.52% compared with the 97.53% achieved by humans. An excellent rating, despite weaknesses regarding memory capacity required and calculation times."
Keep in mind that this was almost ten years ago and I'd imagine that this capability has improved significantly since then while not much improvement in human grading performance has occurred.
Regarding the 50~75 million coins, you don't need 50 or 75 million as a reference. What you would start with are the best examples of the coins available and use that as the benchmark. Humans would be involved in establishing algorithms for features we subjectively consider attractive. Of course the programs will be consistent in determining attractiveness while humans will tend to vary. The buyer will be the final arbiter on what the find attractive.
1) 50-75 million of earlier best of the best will not be available. You get to scan nothing graded before. Not swill, but the cream has been skimmed off the top. The best of the rest are less than the best..
2) Glad you've read a few articles. "Of course the programs will be consistent in determining attractiveness".
I cannot contrive an epithet sufficient to describe my contempt for this concept on several levels.
The only accurate register for eye-appeal is how one's viscera are effected. Computers have no soul. Useful on facts, useless on beauty.
YMMV
Parenthetically, Before I entered coins, I had 20+ high-level data-base and system network engineers working for me. I have a pal whose work on for the Defense Department using AI for target identification has kept him very up-to-date. $300K a year up-to-date. He laughs at what I know, but is occasionally surprised by an insight. And I had a five-minute conversation with Isaac Asimov!
I know nothing, and less every day . Buh-bye
edited to add: I actually do know one sure thing. No one's going to fund it. Huge expenditure for limited minor incremental reward. As I repeat (once an 'effin again), you are 100% "on target" for ASE's, AGE's and Moderns.
Again, you don’t need 50-75 million best of the best. You need one example of a coin that is the best example of each type as minted. That becomes your baseline for judging all other coins.
I’ve read a few articles and have 30+ years in electronic component manufacturing, components from the size of your fist to others the size of a grain of pepper. I’m also familiar with the optics you use to populate PCB’s at extremely fine pitch with these components. I’m pretty sure those optics can identify any flaw in a coin a grader can see at 10x magnification with easy.
Kudos to your friend and you for your system engineering work, although I’m not sure what bearing that has on the discussion of computer grading of coins. As mentioned eight years ago computers rivaled the human capacity with respect to facial recognition. You’re probably familiar with Moore’s law. Computers have gotten better at the task, humans have not.
As for who will fund it, if we can get Elon interested in numismatics maybe he would fund it. I’m sure it would cost less than one SpaceX rocket launch. There are consumer level facial recognition systems that can be had for $2000 (I looked it up). I’m sure much more sophisticated systems, far surpassing what was available in the early 1990’s, likely costing less given everything electronics drops in price while increasing in capability, would be available if there was really the will to do it.
The longer I live the more convincing proofs I see of this truth, that God governs in the affairs of men. And if a sparrow cannot fall to the ground without His notice is it possible for an empire to rise without His aid? Benjamin Franklin
You don't know how to grade. Me and a bunch of guys you've never heard of invented the modern coin grading system. You've spoken nonsense about AI self-teaching by suggesting that if the highest coin graded is available as an exemplar then other reference points are not needed. You're in love with the beauty of your weapons. Coin grading is artisanal. You have no conception of what a database might be required to hold. If the data was entered, you have no idea what evaluatory techniques would be required to manually curate and classify each image.
Elon will maybe throw some chump change at the project? Maybe he'll adapt the '33 $20 as a hood ornament design.
Moore's Law cannot fix Garbage In, Garbage Out, only produce it with greater alacrity.
Kudos to you for your knowledge of scanning techniques, Shame on you for the desolation of your ignorance about both database structure and design and, mostly, coin grading. I'm just a guy who helped started a grading service and made a few PR69 coins. Whereas you superficially appear to me to be someone with the perquisite knowledge to understand Moderns. And naught else. :head slap emoji
Try this one for size. John Dannreuether and I (and quite a few others) knew a lot more about the underlying technical reasons for grading coins than David Hall did. He had the genius to capitalize on it. No one who was around in the 80's would dispute either part of this.
But you're just insulting my knowledge. I'm insulting your insult to my intelligence
"People sleep peaceably in their beds at night only because rough men stand ready to do violence on their behalf." - Geo. Orwell
@ColonelJessup said:
You don't know how to grade. Me and a bunch of guys you've never heard of invented the modern coin grading system. You've spoken nonsense about AI self-teaching by suggesting that if the highest coin graded is available as an exemplar then other reference points are not needed. You're in love with the beauty of your weapons. Coin grading is artisanal. You have no conception of what a database might be required to hold. If the data was entered, you have no idea what evaluatory techniques would be required to manually curate and classify each image.
Elon will maybe throw some chump change at the project? Maybe he'll adapt the '33 $20 as a hood ornament design.
Moore's Law cannot fix Garbage In, Garbage Out, only produce it with greater alacrity.
Kudos to you for your knowledge of scanning techniques, Shame on you for the desolation of your ignorance about both database structure and design and, mostly, coin grading. I'm just a guy who helped started a grading service and made a few PR69 coins. Whereas you superficially appear to me to be someone with the perquisite knowledge to understand Moderns. And naught else. :head slap emoji
Try this one for size. John Dannreuether and I (and quite a few others) knew a lot more about the underlying technical reasons for grading coins than David Hall did. He had the genius to capitalize on it. No one who was around in the 80's would dispute either part of this.
But you're just insulting my knowledge. I'm insulting your insult to my intelligence
You don’t know what I know.
I’ve seen enough discrepancies in grading to say the so-called professional grading leaves a lot to be desired.
You guys twist the English language so it means whatever you’d like it to mean for your convenience and/or profit. To call them grading “standards” is laughable.
Instead of thinking you know it all take some time to Google and read a few articles on AI self learning. You might learn something.
When did I “insult” your knowledge?
What you may have done in the 1990’s regarding computerized grading has little to do with the technology and software available today.
The longer I live the more convincing proofs I see of this truth, that God governs in the affairs of men. And if a sparrow cannot fall to the ground without His notice is it possible for an empire to rise without His aid? Benjamin Franklin
Most of the technical aspects of this discussion are way over my head, but that it would be a "Huge expenditure for limited, minor, incremental reward." Seems inarguable.
@ColonelJessup said:
You don't know how to grade. Me and a bunch of guys you've never heard of invented the modern coin grading system. You've spoken nonsense about AI self-teaching by suggesting that if the highest coin graded is available as an exemplar then other reference points are not needed. You're in love with the beauty of your weapons. Coin grading is artisanal. You have no conception of what a database might be required to hold. If the data was entered, you have no idea what evaluatory techniques would be required to manually curate and classify each image.
Elon will maybe throw some chump change at the project? Maybe he'll adapt the '33 $20 as a hood ornament design.
Moore's Law cannot fix Garbage In, Garbage Out, only produce it with greater alacrity.
Kudos to you for your knowledge of scanning techniques, Shame on you for the desolation of your ignorance about both database structure and design and, mostly, coin grading. I'm just a guy who helped started a grading service and made a few PR69 coins. Whereas you superficially appear to me to be someone with the perquisite knowledge to understand Moderns. And naught else. :head slap emoji
Try this one for size. John Dannreuether and I (and quite a few others) knew a lot more about the underlying technical reasons for grading coins than David Hall did. He had the genius to capitalize on it. No one who was around in the 80's would dispute either part of this.
But you're just insulting my knowledge. I'm insulting your insult to my intelligence
You don’t know what I know.
I’ve seen enough discrepancies in grading to say the so-called professional grading leaves a lot to be desired.
You guys twist the English language so it means whatever you’d like it to mean for your convenience and/or profit. To call them grading “standards” is laughable.
Instead of thinking you know it all take some time to Google and read a few articles on AI self learning. You might learn something.
When did I “insult” your knowledge?
What you may have done in the 1990’s regarding computerized grading has little to do with the technology and software available today.
You joined this forum in 2001. What the heck happened in the last month to make you so obligated? I'm assuming you recently bought a dictionary...or wrote one.
"Standard" refers to a model or norm not an immutable truth. Marilyn Monroe can be a standard of beauty without that meaning that anyone must be an exact clone to be beautiful.
[Of course, you will simply accuse my example of further twisting the meaning of words, so you can save the time required to post that. ]
moving along, I suggest an internal FAQ for customer service. Not only describe what to do when things go right, but also what to do when things go wrong
@ColonelJessup said:
You don't know how to grade. Me and a bunch of guys you've never heard of invented the modern coin grading system. You've spoken nonsense about AI self-teaching by suggesting that if the highest coin graded is available as an exemplar then other reference points are not needed. You're in love with the beauty of your weapons. Coin grading is artisanal. You have no conception of what a database might be required to hold. If the data was entered, you have no idea what evaluatory techniques would be required to manually curate and classify each image.
Elon will maybe throw some chump change at the project? Maybe he'll adapt the '33 $20 as a hood ornament design.
Moore's Law cannot fix Garbage In, Garbage Out, only produce it with greater alacrity.
Kudos to you for your knowledge of scanning techniques, Shame on you for the desolation of your ignorance about both database structure and design and, mostly, coin grading. I'm just a guy who helped started a grading service and made a few PR69 coins. Whereas you superficially appear to me to be someone with the perquisite knowledge to understand Moderns. And naught else. :head slap emoji
Try this one for size. John Dannreuether and I (and quite a few others) knew a lot more about the underlying technical reasons for grading coins than David Hall did. He had the genius to capitalize on it. No one who was around in the 80's would dispute either part of this.
But you're just insulting my knowledge. I'm insulting your insult to my intelligence
You don’t know what I know.
I’ve seen enough discrepancies in grading to say the so-called professional grading leaves a lot to be desired.
You guys twist the English language so it means whatever you’d like it to mean for your convenience and/or profit. To call them grading “standards” is laughable.
Instead of thinking you know it all take some time to Google and read a few articles on AI self learning. You might learn something.
** When did I “insult” your knowledge?**
What you may have done in the 1990’s regarding computerized grading has little to do with the technology and software available today.
I'm insulted that you don't know how well I could almost justify my self-importance . Actually, it's what I've done with coins since the 70's that inform me. And, also, what I've learned about the productions of fools who might convince upper management with power points but not the technical folks who would do a cost analysis on the specs.
I implore anyone paying attention to this gent know that I have full support for his anti-Luddite efforts. I'm thinking he's capable of making much more meaningful contributions to society than this but just wondered into the Forum as a casual visitor and, like Tron, can't escape back to reality.
Hey, Call up John Albanese at CAC. Ask him if Rick Sear knows and he'll say yes. Ask him if I'm difficult. He'll go "Two for two and it sounds like you know him personally"
"People sleep peaceably in their beds at night only because rough men stand ready to do violence on their behalf." - Geo. Orwell
I’ve seen enough discrepancies in grading to say the so-called professional grading leaves a lot to be desired.
You guys twist the English language so it means whatever you’d like it to mean for your convenience and/or profit. To call them grading “standards” is laughable.
In sports they are called "rules." And yes, they also employ humans to make the call. And yes, on rare occasion they are wrong. But, they have the flags and they have the whistles. Are you one of those dads who encourages his football son that it was the referee's fault?
@ColonelJessup said:
You don't know how to grade. Me and a bunch of guys you've never heard of invented the modern coin grading system. You've spoken nonsense about AI self-teaching by suggesting that if the highest coin graded is available as an exemplar then other reference points are not needed. You're in love with the beauty of your weapons. Coin grading is artisanal. You have no conception of what a database might be required to hold. If the data was entered, you have no idea what evaluatory techniques would be required to manually curate and classify each image.
Elon will maybe throw some chump change at the project? Maybe he'll adapt the '33 $20 as a hood ornament design.
Moore's Law cannot fix Garbage In, Garbage Out, only produce it with greater alacrity.
Kudos to you for your knowledge of scanning techniques, Shame on you for the desolation of your ignorance about both database structure and design and, mostly, coin grading. I'm just a guy who helped started a grading service and made a few PR69 coins. Whereas you superficially appear to me to be someone with the perquisite knowledge to understand Moderns. And naught else. :head slap emoji
Try this one for size. John Dannreuether and I (and quite a few others) knew a lot more about the underlying technical reasons for grading coins than David Hall did. He had the genius to capitalize on it. No one who was around in the 80's would dispute either part of this.
But you're just insulting my knowledge. I'm insulting your insult to my intelligence
You don’t know what I know.
I’ve seen enough discrepancies in grading to say the so-called professional grading leaves a lot to be desired.
You guys twist the English language so it means whatever you’d like it to mean for your convenience and/or profit. To call them grading “standards” is laughable.
Instead of thinking you know it all take some time to Google and read a few articles on AI self learning. You might learn something.
When did I “insult” your knowledge?
What you may have done in the 1990’s regarding computerized grading has little to do with the technology and software available today.
You joined this forum in 2001. What the heck happened in the last month to make you so obligated? I'm assuming you recently bought a dictionary...or wrote one.
"Standard" refers to a model or norm not an immutable truth. Marilyn Monroe can be a standard of beauty without that meaning that anyone must be an exact clone to be beautiful.
[Of course, you will simply accuse my example of further twisting the meaning of words, so you can save the time required to post that. ]
And no one puts Rick Sear in a corner!
Her adored father got a nose job right before she did, and I've wanted to murder him ever since
"People sleep peaceably in their beds at night only because rough men stand ready to do violence on their behalf." - Geo. Orwell
Comments
All I know in 61% of ColonelJessups posts I end up having to look up a word
m
Fellas, leave the tight pants to the ladies. If I can count the coins in your pockets you better use them to call a tailor. Stay thirsty my friends......
Amazing - that's the exact percentage for me! BUT the other 39% of the time, I SHOULD be looking up one or more words. It's just that I'm not even aware of it.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
I used to sell coins by mail, but then I had to buy the Post Office and so I stopped.
There is NO REASON why eBay needs to own a grading service to use the services. eBay could require that all collectibles sold on their website be certified where possible. They would, of course, lose a lot of business given that there are a lot of uncertified collectibles out there. [Side note: Does anyone read comics anymore or just admire the covers through the giant hard plastic case?]
I'm not aware of any ownership of NGC by Heritage at any time. Some of the players have moved through different corporate entities over the years but that's not exactly the same thing. But I reserve the right to be wrong.
The only company I'm aware of that does both is CDN and that bothered me for a while. It still potentially bothers me in a post-JA world. It would be very tempting in hard times to slap a sticker on a coin to increase it's value. I trust JA. I'm reserving judgment on the post-JA CAC - assuming I'm still alive.
Heck, there are a number of forum members who are convinced that the big retailers get preferential grading boosts from the TPG's as it is. Imagine if the retailer were actually the same corporate entity. Maybe we need a poll.
And, due to the expense of certifying coins, there always will be. As noted above, the overwhelming majority of eBay listings are for uncertified coins and I'd bet, if one were to count them all up, one would find that raw coin collectors outnumber collectors of slabbed coins.
Two comments, and excuse me if either has been mentioned. I read most of the thread but not every word.
Automated scanning/grading is entirely possible from a technological standpoint, and would be more consistent than grading by different experts. Biggest problem is agreeing on weighting of factors. No single set of standards will please the wide range of collector preferences or disagreements among the experts as to how much various defects weigh in the grade. Finding a satisfactory coin for the grade might be easier under the present system than in the more consistently graded universe.
Yes, the TPGs are trying to expand into new areas but there's a BIG problem that hasn't been noted. Coins, stamps and currency are designed to be secure-that is as hard to duplicate as possible. This makes the job of creating fakes much harder than making fakes of a snowboard, action figure or baseball card, for that matter. I know what the mint and BEP does with it's dies and plates but what did Topps do with the photos, artworks, plates for baseball cards? Could you possibly locate some unused 50-75 year old cardboard stock somewhere?
There are many collectibles that are MUCH easier than coins and currency to copy and copy quite well. This is the biggest challenge to the TPGs expanding into other areas. It's gonna happen, but this will be a limiting factor and could end certifying of certain items if undistinguishable fakes are discovered.
@sellitstore - you raise some very good points. Especially with the fate of discarded assets of some of these collectible areas. Given how well die struck counterfeits are now it would be an issue for sure with other collectibles.
TurtleCat Gold Dollars
1) 50-75 million of earlier best of the best will not be available. You get to scan nothing graded before. Not swill, but the cream has been skimmed off the top. The best of the rest are less than the best..
2) Glad you've read a few articles. "Of course the programs will be consistent in determining attractiveness".
I cannot contrive an epithet sufficient to describe my contempt for this concept on several levels.
The only accurate register for eye-appeal is how one's viscera are effected. Computers have no soul. Useful on facts, useless on beauty.
YMMV
Parenthetically, Before I entered coins, I had 20+ high-level data-base and system network engineers working for me. I have a pal whose work on for the Defense Department using AI for target identification has kept him very up-to-date. $300K a year up-to-date. He laughs at what I know, but is occasionally surprised by an insight. And I had a five-minute conversation with Isaac Asimov!
I know nothing, and less every day
. Buh-bye 
edited to add: I actually do know one sure thing. No one's going to fund it. Huge expenditure for limited minor incremental reward. As I repeat (once an 'effin again), you are 100% "on target" for ASE's, AGE's and Moderns.
Ironically, I have to look up words for 68% of my posts. But "prolixity" is not one of them.
I'm glad I'm not the only one on dictionary dot com after reading his posts.
He's easier to understand than trying to
listen to, and understand, Nugget-eese !
for PCGS. A 49+-Year PNG Member...A full numismatist since 1972, retired in 2022
https://forums.collectors.com/discussion/803809/heritages-owns-30-of-ngc
I'm not sure that resolves the question. CU was publicly traded. What is the ownership structure of Heritage? NGC?
I'm wondering if and when Heritage might be bought out. Seems like a good time to sell to me.
I also consider the business model more compelling, higher prices are a lot more correlated to revenue increases. You can also sell the same items repeatedly and it provides value to the customer. Only so many times you can grade the same collectible over and over again.
I suggest when it's convenient to take the time to read that thread. Lot of good info, links and comment from members who are not on the forum much or if at all.
I only go back to coins to 1962. Sometimes my memory needs to be refreshed. I really don't know much, that's obvious.
While I've actively resisted joining Facebook for at least a decade, Legend numismatist Greg Cohen convinced me to join, specifically to participate in a special interest group whose main project is trolling Paul Nugget. Unsurprisingly, Kevin has a scatological profile name but, regrettably, is much less effective in print than in person
. Scholarships are available 
Actually, both NGC and Heritage are privately owned and Ivy and Halperin appear to have ownership stakes in both.
79% for me:
The Heritage owners have non-voting shares in NGC
Again, you don’t need 50-75 million best of the best. You need one example of a coin that is the best example of each type as minted. That becomes your baseline for judging all other coins.
I’ve read a few articles and have 30+ years in electronic component manufacturing, components from the size of your fist to others the size of a grain of pepper. I’m also familiar with the optics you use to populate PCB’s at extremely fine pitch with these components. I’m pretty sure those optics can identify any flaw in a coin a grader can see at 10x magnification with easy.
Kudos to your friend and you for your system engineering work, although I’m not sure what bearing that has on the discussion of computer grading of coins. As mentioned eight years ago computers rivaled the human capacity with respect to facial recognition. You’re probably familiar with Moore’s law. Computers have gotten better at the task, humans have not.
As for who will fund it, if we can get Elon interested in numismatics maybe he would fund it. I’m sure it would cost less than one SpaceX rocket launch. There are consumer level facial recognition systems that can be had for $2000 (I looked it up). I’m sure much more sophisticated systems, far surpassing what was available in the early 1990’s, likely costing less given everything electronics drops in price while increasing in capability, would be available if there was really the will to do it.
You don't know how to grade. Me and a bunch of guys you've never heard of invented the modern coin grading system. You've spoken nonsense about AI self-teaching by suggesting that if the highest coin graded is available as an exemplar then other reference points are not needed. You're in love with the beauty of your weapons. Coin grading is artisanal. You have no conception of what a database might be required to hold. If the data was entered, you have no idea what evaluatory techniques would be required to manually curate and classify each image.
Elon will maybe throw some chump change at the project? Maybe he'll adapt the '33 $20 as a hood ornament design.
Moore's Law cannot fix Garbage In, Garbage Out, only produce it with greater alacrity.
Kudos to you for your knowledge of scanning techniques, Shame on you for the desolation of your ignorance about both database structure and design and, mostly, coin grading. I'm just a guy who helped started a grading service and made a few PR69 coins. Whereas you superficially appear to me to be someone with the perquisite knowledge to understand Moderns. And naught else. :head slap emoji
Try this one for size. John Dannreuether and I (and quite a few others) knew a lot more about the underlying technical reasons for grading coins than David Hall did. He had the genius to capitalize on it. No one who was around in the 80's would dispute either part of this.
But you're just insulting my knowledge. I'm insulting your insult to my intelligence
You don’t know what I know.
I’ve seen enough discrepancies in grading to say the so-called professional grading leaves a lot to be desired.
You guys twist the English language so it means whatever you’d like it to mean for your convenience and/or profit. To call them grading “standards” is laughable.
Instead of thinking you know it all take some time to Google and read a few articles on AI self learning. You might learn something.
When did I “insult” your knowledge?
What you may have done in the 1990’s regarding computerized grading has little to do with the technology and software available today.
Most of the technical aspects of this discussion are way over my head, but that it would be a "Huge expenditure for limited, minor, incremental reward." Seems inarguable.
AKA Pakasmom
Which makes sense to preserve their objectivity.
You joined this forum in 2001. What the heck happened in the last month to make you so obligated? I'm assuming you recently bought a dictionary...or wrote one.
"Standard" refers to a model or norm not an immutable truth. Marilyn Monroe can be a standard of beauty without that meaning that anyone must be an exact clone to be beautiful.
[Of course, you will simply accuse my example of further twisting the meaning of words, so you can save the time required to post that. ]
And no one puts Rick Sear in a corner!
wow
in before the lock
moving along, I suggest an internal FAQ for customer service. Not only describe what to do when things go right, but also what to do when things go wrong
🙈🙉🙊 Perhaps a respite of levity might ease the increasing tension.
Life Coach & Philosopher W.C.Fields muttered these words: "I spent 50% of my money on alcohol, women, and gambling. The other half I wasted” 🤣😂
Stuart
Collect 18th & 19th Century US Type Coins, Silver Dollars, $20 Gold Double Eagles and World Crowns & Talers with High Eye Appeal
"Luck is what happens when Preparation meets Opportunity"
I'm insulted that you don't know how well I could almost justify my self-importance
. Actually, it's what I've done with coins since the 70's that inform me. And, also, what I've learned about the productions of fools who might convince upper management with power points but not the technical folks who would do a cost analysis on the specs.
I implore anyone paying attention to this gent know that I have full support for his anti-Luddite efforts. I'm thinking he's capable of making much more meaningful contributions to society than this but just wondered into the Forum as a casual visitor and, like Tron, can't escape back to reality.
Hey, Call up John Albanese at CAC. Ask him if Rick Sear knows and he'll say yes. Ask him if I'm difficult. He'll go "Two for two and it sounds like you know him personally"
In sports they are called "rules." And yes, they also employ humans to make the call. And yes, on rare occasion they are wrong. But, they have the flags and they have the whistles. Are you one of those dads who encourages his football son that it was the referee's fault?
Give Me Liberty or Give Me Debt
are you going to go towards video review?
I'm officially starting a rumor the HQ will be moved to Texas.
Jacksonville, FL
Give Me Liberty or Give Me Debt
Her adored father got a nose job right before she did, and I've wanted to murder him ever since
This thread has gotten wildly off-topic now so locking to discourage the continuation of this argument.
Heather Boyd
PCGS Senior Director of Marketing