Postcode loophole enables fraudsters to hijack eBay parcels
theguardian.com
theguardian.com
They must be spending a steady amount on PR. The news seems perfectly constructed for the obligatory quote from their founder when in this case the answer isn't their system, but just that Royal Mail use the correct identifier: the post-code and the identifying information for the property (building name and/or building number as well as company name if applicable).
I've not seen any articles discussing how awful they are -- The idea of copyrighting and privatising the concept of "position" should be rejected at every opportunity.
Now if he said that the customer service was amazing, it would be legitimate to expect some level of criticism response.
We released WhatFreeWords last week and it was taken down after 4 days.
It is back up now at https://whatfreewords.org/ but who knows for how long.
You can read their DMCA notice to our previous hosting at https://whatfreewords.org/download/dmca-20190917.pdf
Once they've submitted a DMCA notice, the hosting company has to take it down within 2 days unless you file a counter-notice, which has to personally identify you and be signed "on penalty of perjury".
None of the DMCA process has anything to do with whether the claimant has a legitimate copyright claim over the work.
Terms of service violation or something maybe.
As a reminder, there is a similar free solution that tries to solve the same problem as what3words from a different angle:
Google Maps shows these today. And you can search Google or Maps for plusCodes.
> The 11-digit codes [...] are hard to memorise and a single mistyped digit would send a courier to the wrong address.
The plus codes FAQ [1] mentions that they thought checksums weren't important for a couple of reasons:
* Not needed for codes that are copy pasted rather than typed or written
* Mistyped codes will be "obviously wrong" due to referring to an unrelated area
I'm not too convinced by the rationale, as playing around on Google Maps shows a typo of one digit can at least sometimes produce another valid code not too far away. And even if typoed codes were always obviously wrong, it's not clear that a computer would recognize incorrect but valid codes as well as a human.
[1] https://github.com/google/open-location-code/blob/master/FAQ...
what3words is so problematic and really isn't the solution to this problem.
Google “PR agency London”. Email the top five asking for a price-per-mention, and get offers back in the region of £500 a pop. Sign a contract. Talk about how you were featured on the BBC.
It says nothing about product, traction or anything else. It only proves journalists and editors can be or already have been bought.
It might also just be familiarity from the writer, i.e. journalist writing article about inexact nature of post codes -> remembers the article he/she wrote or read about what3words -> figures that's a good alternative so contacts the company for comment.
UK postcodes are more precise that in any other country I'm familiar with and are a terrific system [1]: Postcodes are precise to about the street level vs. a whole town or district in most countries. This allows websites that need your address to work it out fully just by asking postcode and house number in most cases.
[1] https://en.wikipedia.org/wiki/Postcodes_in_the_United_Kingdo...
I really like the idea of having a system that is that accurate. Although as someone else pointed out, in the UK your postcode and house number/name are almost always enough to uniquely identify your property.
they are a routing code for the postal service
the number and postcode uniquely identify a mailbox
Adding four digits on the end for a "standardized address" means that a specific block and, often, a subset of numbers on that block, can be identified. For example, ZIP code 90210 covers the bulk of Beverly Hills, California but 90210-5317 narrows it down to just the odd numbers of the 200 block of South Beverly Drive or, according to the USPS database, a maximum possibility of 23 delivery addresses (because not all of the numbers in that block block are assigned).
Piss off, it shows that a postcode without a house identifier is a shitty way for Royal Mail to guarantee delivery is all.
Case is, you can send a letter - no way of tracking that, but if you send it recorded delivery, that's tracked.
So issue here is not the Royal Mail, more case of Ebay etc allowing people to send things using the cheapest untrackable way and allowing such abuse to transpire.
But then every system is abusable - you can send a recorded delivery item to a person, log into tracking and see a copy of that person's signature which you could then abuse in nefarious ways. But many systems open to such abuse. Send a cheque to somebody - they cash it. Then track via your bank and get the bank account details of the recipient. Heck send cheque via recorded delivery and and get the targets signature and bank account details, all legal like. Is that a fault in the systems and even if it is - how do you fix that!
Well, Ebay could insist that all postings are done recorded delivery - would solve this instance.
In the UK, a full postcode e.g B40 4XJ (made up) can be shared by more one than home or building. So the homes numbered 1-20 Acacia Avenue all share the same full postcode: B40 4XJ
Janet lives at 7 Acacia Avenue. She sells her iPhone via eBay for £200 to Patrick living elsewhere. She posts a parcel containing the iPhone to Patrick.
Patrick, via eBay, contacts Janet and says the iPhone is faulty and wants a refund.
Through eBay, Patrick generates a return label (shipping label) which lists Janet's address: 7 Acacia Avenue, B40 4XJ. However, Patrick, alters the address on the label to state: 19 Acacia Avenue, B40 4XJ.
He can do this, of course, because the label is generated online, and using some graphics editing program, he can alter the label before he prints it.
Patrick posts an empty box to 19 Acacia Avenue and keeps the iPhone. Patrick now has the iPhone and a refund from eBay for the £200 he paid for the iPhone.
What of the empty parcel posted to 19 Acacia Avenue? Presumably, the parcel isn't signed for when it is delivered? Or is it? If the parcel requires a signature on delivery, surely the resident of 19 Acacia Avenue will immediately notice the parcel is addressed to someone else?
Please do correct the above if it's incorrect.
It's not that they 'can' be shared, like it's some sort of bug or corner case - they're designed to be shared. A full address is a house number or name plus the postcode.
I assume that the unsuspecting resident of 19 Acacia Avenue, B40 4XJ will sing off receipt for the parcel, then open it and find the box empty, but since they weren't expecting any delivery at all they will do nothing about it, maybe just tell the story as an anecdote next time they go to drinks with friends.
Another thing that enables the fraud is that eBay only compares postal codes, so the delivery to 7 Acacia Avenue and 19 Acacia Avenue is all the same for them.
That said, Royal Mail say anything over £100 should be Signed For so why did eBay let somebody return an iPhone without using Signed For?
This is addressed in the article. They sign for it, see it’s empty, assume error on someone’s part, and dispose of it.
Where does Banana Man fit into all this???
- eBay are supremely uninterested in being an unbiased arbiter of disputes.
This goes back as far as I can remember and likely to the dawn of eBay. People have been scamming each other on there for 15 years at a minimum.
- Royal Mail do not provide proof of postage at a granular level.
Combine that with the above where eBay will probably take a piece of tissue paper as proof of postage and there you are.
It's not just eBay though, it's also the credit card companies. You can't escape this by switching platforms.
E: I guess this is just one of many possible ways to force a carrier to "lose" a parcel, enabling the fraudster to use the same ebay account more times without attracting complaints from sellers.
What? Think about it for a second, what proof would they have?
The trick described in the article just improves upon it slightly by making the seller think that royal mail lost the package. There's no dependence on this trick though, the scam works 100% of the time even if the seller receives an empty box.
Ebay will always believe a "hey my box was empty", no matter your account history. They'll just ban your account at some point.
At first glance the empty box seems pretty meaningless, it certainly won't win a dispute for the seller.
I guess you're right though, now the seller won't even realize that they're getting scammed and won't report the buyer account to ebay.
This is just a cost of doing business. Really sucks for amateur sellers though.
If the scammer is in the UK, there's always small claims court. I'm not even from the UK, and I got a scamming seller from there to refund me by just threatening to make a claim.
Stolen money makes no sense, this kind of trickery would have zero impact on chargebacks.
But it actually works both ways, this scam has happened on sellers and buyers.
TLDR - fraudster orders good, returns empty box with tracking, tracking label is photoshopped to different property in post code. Fraudster knows that tracking only records post code not property within post code. eBay issue refund.
Problem is granularity of tracking nothing to do with post code loophole.
Is the granularity of the tracking really the problem? I believe fraud is the problem, and it's an impossible one to solve.
Even for proofs of postage sometimes (always?) the certificate only lists the recipient's postcode instead of the name and full address.
Since, this particular scam is originating from a very specific ecosystem; in this case, the responsibility ultimately lies with eBay and to a certain extent with PayPal ─ who are blasé about these scams and have zero intent on solving these issues, they are even happy to front-load the system, so it favours certain types of scammers.
https://business.help.royalmail.com/app/answers/detail/a_id/...
Apparently they also use a field "Delivery Point Suffix"
"A postcode is not always sufficient to uniquely identify each delivery point in the UK – for example, in shared buildings.
Some postcodes relate to only one delivery point, while others may cover up to a hundred. The average is 15.
For that reason a Delivery Point Suffix (DPS) has been developed. The DPS is a two character code (a number and a letter) which enables each delivery point to be identified.
The DPS can be created by extracting the postcode together with the house number or name using Royal Mail Postcode Address File (PAF)."
At least if the seller is the one receiving an empty box he can set up procedures to attempt to prove it.
>At least if the seller is the one receiving an empty box he could set up procedures to attempt to prove it.
I definitely agree that would be a better outcome, but I think it is a stretch to blame royal mail for this. If someone really wants their package to get lost in shipping there's not much you can do about it.
First Last
Street # or Building Name if any
18.0000 -7.0000
USA
Basically the “ZIP city, state” line should be replaced by coordinates; 4 decimals are precise to 11 meters.And in less dense areas you might not need the second line either:
First Last
18.0000 -7.0000
Mail is not handled manually anymore so why act like it still is?We can do something similar to your system in the UK now - "10 SW1A 2AA" is a unique address, but we don't as it's less robust. "10 Downing Street, London, SW1A 2AA" makes the intention clearer, both if there is an error when writing ("10 Downng Street, Lendon, W1A 2AA" - we can make a fair assumption, particularly if the name indicates it's the same as other mail going there) or if it is damaged in transit ("10 Downing [Unreadable], London, SW-- 2A-").
I was just complaining about addresses and not this specific handling of ZIP codes by UK’s post.