EuroBillTracker

Hi,

Let me bring to this thread an idea I have been thinking of for quite a long time:

It is something I find weird, to say the least, and somehow useless.

When we see a list of banknotes entered, or a list of hits, we see a hidden serial number that doesn't say much.

If we take the example of a hypothetic banknote with serial number C01234567891 or CT1234567890,
the information we see is just Cxxxx4567xxx in both cases. And sometimes happens that a user logs 20 sequentials,
and what we see is a column of 20 times

Cxxxx4567xxx
Cxxxx4567xxx
Cxxxx4567xxx
Cxxxx4567xxx
Cxxxx4567xxx
etc.

Why not change the set of hidden characters, so we could see

C012xxxxx89x or CT12xxxxx89x
or even
C012xxxx789x or CT12xxxx789x
or
C0123xxxx89x or CT123xxxx89x?

4 digits plus the checksum hidden are quite a good enough protection.

This notation gives more descriptive information about banknote batch, for example, and when someone logged
a list of sequentials it would appear

CT12xxxx788x
CT12xxxx789x
CT12xxxx790x
CT12xxxx791x
CT12xxxx792x
etc., or
CT123xxxx88x
CT123xxxx89x
CT123xxxx90x
CT123xxxx91x
CT123xxxx92x

or, OK, at least
C012xxxxx88x
C012xxxxx89x
C012xxxxx90x
C012xxxxx91x
C012xxxxx92x
and
CT12xxxxx88x
CT12xxxxx89x
CT12xxxxx90x
CT12xxxxx91x
CT12xxxxx92x

In the same way, printer code could be shown with the complete plate number, like A023xx.
Someone can say that if I give 023 for the printing plate, some geek could find out that

CT12xxxx789x is actually CT123xxx789x or CT12x4xx789x.
Yeah, so what? Still 4 (or 5, if your think that's too daring, although I think it's not) digits missing, one of which not serialized...

So, if after a "Planned Service Interruption", banknotes lists, hits lists and reports suddenly appeared in the new form,
does anybody think that it could cause any kind of disaster?

Just a suggestion.

Visit http://www.wheresgeorge.com/wrapper.php?page=hitstoday and see...
They have no checksum digit at all, neither a printer code to match with for hits, and they hide only 3 digits of each serial,
but show parts of the serial number that gives idea of "batch" (OK, block letter) and sequence.
Here, we would still hide 4 AND the checksum, and still without the sheet grid position.

I'm thinking of one immediate advantage in the "Cxxxx4567xxx" layout: anyone browsing the banknotes recently registered by a specific EBTor can, wih that layout, spot a faker when you see a hit in a series. This (hit in series) is one of the most common reports in the "Discussions about investigations" topic under the hidden Database subforum (accessible only to members of the Database Working Group).

Maybe there's other advantages I don't know of.

I haven't gone into detailed analysis, but I know that layouts like "C012xxxxx89x" allow, in various combinations value/printer_letter/serial_letter, makes it possible or someone in the know to derive the plate (like "C001xx") and/or the cell (like "CxxxA1").

If I'm not wrong, there is a stress on accuracy in EuroBillTracker that isn't present in Where's George. Just an educated guess: maybe that is one of the main reasons for such a different hit ratio between these two platforms. It's arguably much harder in EBT (or at least more discouraged) to register notes that do not exist or you haven't actually seen.

Maybe one of the developers have better explanations. Or maybe the layout system can still be improved, but only those deeply knowledgeable (main developers, probably) would know the answer.

lmviterbo wrote: ↑Thu Jun 18, 2020 1:57 am If I'm not wrong, there is a stress on accuracy in EuroBillTracker that isn't present in Where's George. Just an educated guess: maybe that is one of the main reasons for such a different hit ratio between these two platforms.

Hmm, maybe it plays a small part... This is going off-topic, but I think there are two main reasons that account for Where's George's better stats:

1) many users there "url" their notes - all or most of their bills, that is. It's considered standard procedure, and they find it quite odd that we don't do the same lol.
2) equally as significant is the fact that bills can be entered anonymously (ie. "random" people who find "url'd" bills can go on the site it and enter them without the need of being registered members) - as seen in this bill report, that bill has been registered by a mix of members and non-members of the site (as can be inferred by the presence or absence of either of these two icons - ).

Hello
I copy here a mail sent to support@eurobilltracker.com
I have connexion pb to eurobilltracker since yesterday

User : EDIT

Result

'There is no EDIT account or your password is incorrect.
Please note that you should use your email address and not your username when you log in.

If you have forgotten your password, click here to set a new password. You will receive an email containing a link to a page where you can set your new password.'



I tried this morning link forgotten password no mail recieved since then 

I receive other mail

I change my account adresse last year because i can no longer access to EDIT

I could connect every day with new adress

But not since yesterday
Don't understand
Thank you for help


Envoyé depuis mon smartphone Samsung Galaxy.

lmviterbo wrote: ↑Thu Jun 18, 2020 1:57 am I'm thinking of one immediate advantage in the "Cxxxx4567xxx" layout: anyone browsing the banknotes recently registered by a specific EBTor can, wih that layout, spot a faker when you see a hit in a series. This (hit in series) is one of the most common reports in the "Discussions about investigations" topic under the hidden Database subforum (accessible only to members of the Database Working Group).

Maybe there's other advantages I don't know of.

I haven't gone into detailed analysis, but I know that layouts like "C012xxxxx89x" allow, in various combinations value/printer_letter/serial_letter, makes it possible or someone in the know to derive the plate (like "C001xx") and/or the cell (like "CxxxA1").

Hi again.

Sorry I have taken so long to reply.

I do not agree on what you write about fake serial hits. As far as I can imagine, opportunity of faking hits exists same way if you have Cxxxx4567xxx as if you have CT12xxxxx89x. A total of 6 digits, one of which checksum, are hidden in both cases.

Also, in my post, I wrote that a skilled user could figure out if one of the hidden digits could be CT12x4xxx89x given printing plate C023 if published. Well: my mistake, this is not true. This middle 4 could be figured out if the printing grid was published, or at least one position, let's say C023Dx or C023x4, but not just for the printing plate number, since plate number 023 actually prints ALL the serial numbers from, for example CS47xxxxxxxx to CT69xxxxxxxx.

And, given the C, you have the checksum, but:

1st: even showing only Cxxxxxxxxxxx, you ALSO give the checksum value, and
2nd: still need to guess a further 5 digits that can literally go from 00000 to 99999, so one out of 10,000 possibilities.

lmviterbo wrote: ↑Thu Jun 18, 2020 1:57 am If I'm not wrong, there is a stress on accuracy in EuroBillTracker that isn't present in Where's George. Just an educated guess: maybe that is one of the main reasons for such a different hit ratio between these two platforms. It's arguably much harder in EBT (or at least more discouraged) to register notes that do not exist or you haven't actually seen.

Maybe one of the developers have better explanations. Or maybe the layout system can still be improved, but only those deeply knowledgeable (main developers, probably) would know the answer.

As for the accuracy or not in both EBT and WsG, my thoughts go to such a different ratio of hits, as you point. But not in your guess: mine is that they have so many hits because they move billions of one-dollar notes. Go to the daily hit list and you'll see that, actually, WsG has a lower hit ratio compared to EBT. The exercise is simple: count only the hits made out of banknotes faced 5, 10, 20, 50 and 100 dollars. In this very moment, there are 430 banknotes with one hit: only 46 are 5$ or above face value. The other 384 are ONEs (or 370 ONEs and 14 TWOs, anyway).

Then, also, as Montgomery Burns writes,

1st: they url (most of) their notes,
2nd: there's lots of anonymous entries,
and 3rd: they ENCOURAGE stamping banknotes, and if you read user comments on hitted bills, you'll see quite often "I saw the stamps/website writings/etc. and I felt curious". And they really DO STAMP: I published quite a few photos in the english, spanish and portuguese forums showing how wild some "georgers" are stamping their banknotes...

lmviterbo wrote: ↑Thu Jun 18, 2020 1:57 am I'm thinking of one immediate advantage in the "Cxxxx4567xxx" layout: anyone browsing the banknotes recently registered by a specific EBTor can, wih that layout, spot a faker when you see a hit in a series. This (hit in series) is one of the most common reports in the "Discussions about investigations" topic under the hidden Database subforum (accessible only to members of the Database Working Group).

Maybe there's other advantages I don't know of.

I haven't gone into detailed analysis, but I know that layouts like "C012xxxxx89x" allow, in various combinations value/printer_letter/serial_letter, makes it possible or someone in the know to derive the plate (like "C001xx") and/or the cell (like "CxxxA1").

If I'm not wrong, there is a stress on accuracy in EuroBillTracker that isn't present in Where's George. Just an educated guess: maybe that is one of the main reasons for such a different hit ratio between these two platforms. It's arguably much harder in EBT (or at least more discouraged) to register notes that do not exist or you haven't actually seen.

Maybe one of the developers have better explanations. Or maybe the layout system can still be improved, but only those deeply knowledgeable (main developers, probably) would know the answer.

Maybe, there are "problems" with hits in EBT other than "serial hits" because of one bill.

Actually, if I log bill CT1234567890, I could also log bill CT1234567891 that I have never had in my hands.
OR; if I hit bill CS0987654321, I could try to log CS0987654322 and "see".
OR; trade bills.

I don't think at all that showing the first and last digits of serialnumber instead of the meaningless central ones would increase the ratio of fake hits in EBT list, even showing printing plate number, since most skilled EBTors KNOW that bill CS09xxxxx32x has been printed out of printing plate C009, and non-skilled users just don't have any further clue.

On the other hand, these are REAL SCANDALS and occur EVERY SINGLE DAY...:
.
.

klapotec wrote: ↑Thu Dec 19, 2019 9:01 pm

uvy wrote: ↑Thu Dec 19, 2019 3:38 pm There are still no F (Oberthur Fiduciaire AD - Bulgaria ) in the Diffusion (Europa series notes)

have a bit more patience, the development team will be somewhat operational in January, and this is near the top of the list

Any progress with bulgarian F notes in the Diffusion ?

I don't know what happens in the EBT site ,
Each time i want to see the maps of hits ( for instance, here, the map of hits during the current year ) , and the link is : https://fr.eurobilltracker.com/hits/?co ... imestats=5 ) , an empty map appears, and also information "This map is currently generated" ,
but, whatever i do , (waiting or refreshing the window) no map appears, never, and you can wait for hours, no map appears ever.
Surely somerthing doesn't work correctly anymore :

https://www.eurobilltracker.com/notes/?id=166733263
https://www.eurobilltracker.com/notes/?id=210579979

Am I seeing things? Isn't that the same bill? Two note reports for the same note.

Montgomery Burns wrote: ↑Sun Mar 28, 2021 1:38 pm https://www.eurobilltracker.com/notes/?id=166733263
https://www.eurobilltracker.com/notes/?id=210579979

Am I seeing things? Isn't that the same bill? Two note reports for the same note.

As is usual for a hit note - one note report for each user who has entered the note.

Really? wow, I never realized. Thanks.

klapotec wrote: ↑Sun Mar 28, 2021 1:40 pm

Montgomery Burns wrote: ↑Sun Mar 28, 2021 1:38 pm https://www.eurobilltracker.com/notes/?id=166733263
https://www.eurobilltracker.com/notes/?id=210579979

Am I seeing things? Isn't that the same bill? Two note reports for the same note.

As is usual for a hit note - one note report for each user who has entered the note.

I do not agree.

The id number refers to the bill, not to the user. That's why it is a hit. In this case, bill has two ids because most probably one of the users made a mistake in the printer code. For this reason, banknote not being identical, it received a new id number. But it became a hit because serialnumber is the same.

Go open the list of hits today and then open hits by both users: bill id will be the same.

No. The ID refers to that user's entry of the note. For hit notes there are multiple entries. Whether the users used the same printer codes or not is irrelevant. I can see why you would think differently, but the database schema is not arranged that way.

avij wrote: ↑Sun Mar 28, 2021 5:24 pm No. The ID refers to that user's entry of the note. For hit notes there are multiple entries. Whether the users used the same printer codes or not is irrelevant. I can see why you would think differently, but the database schema is not arranged that way.

OK, you are the programmer. Nothing further I can say, except I see I had made myself a wrong idea of how it worked.