> Under the system, users can input these seven-digit codes on online shopping websites, and their addresses will automatically appear on the sites.
Why not just let people mail to that code, and the post office then looks up the actual address? That'd also avoid any issues with leaking personal information.
> Their digital addresses will not change even if their physical addresses change.
Apparently he new system works a bit like DNS: the physical location may change, but the symbolic name stays. The resolution is done at the order time, not at the delivery time.
I suppose it's because the numerous e-commerce sites already support the physical address system. With the post office resolving the 7-character symbolic address to a physical address requires approximately zero changes in their existing systems, it's just an extra API call on the frontend. Support for direct use of the 7-character address would require serious changes.
Also, the 7-character address resolved to a physical address right before the customer's eyes works as an extra sanity check, and should limit the number of orders to a wrong address.
Also there is an existing system of 7 numeral post codes that autopopulate Prefecture, City, District - leaving you to enter Area # (丁目）, Block #, House #.
If they amend this to make it alphanumeric and then autopopulate this last three datums - it fits very neatly into the existing scheme.
> Support for direct use of the 7-character address would require serious changes.
No, it doesn't. You can just format that 7-character address like a valid address, essentially like:
番号０１２３４５６,日本
Fill any extra fields with placeholders or ０s as necessary.
The most you'll have to change is to maybe skip validation for those if you have address validation.
> skip validation
Here's whence real problems would begin.
What? That sort of vague dismissal with likely zero reasoning behind it doesn't really deserve a reply, but here you go anyways:
Many places don't even do address validation for regular addresses, because it's generally dogshit and either too lenient to be useful, or too restrictive to not filter out lots of valid addresses. Even Amazon just checks that the postal code and city match, nothing more. You can put in an entirely made up street name and they'll happily try to ship it.
If you really want to have validation, you can add parity bits to the number, so you can validate it fully client side.
As a business you may learn an address is no good when you're trying to request and print the digital shipping label - that is if you're lucky and the provider does any validation at that point for your in-country shipment (for international shipping there's usually simply none besides checking that they can route it at least to whoever they hand it over to). Worst case you'll get a return.
For these IDs the provider should be able to tell you pretty quickly they can't route the package, unlike when the customer gets their house number in their regular address wrong, which is usually only caught once the deliveryman learns that nobody with that surname lives there. If you're lucky he'll remember you live on number 21, not 12.
In any case as an online retailer you'll already be dealing with lots of returns anyways - invalid addresses is just one of many reasons your shipments don't reach the customer. Anyone who does shipment has a processes to handle these anyways. Using numbers that can be quickly determined to be correct/incorrect should actually improve things.
So again: What are you on about?
This sounds like cross-shipping, where you ship to a distribution centre which then ships to the final destination. My guess is that this is more expensive and slower for a lot of cases (this is typically cheaper for extremely high volumes or cross-continental shipping).
Shops with which I’ve worked will have several different carriers pick up packages during the week. Some pickups are for city-local packages, where others are cross-province packages. Finally, lots of packages are delivered from the shop directly by motorbike. The logistics are quite different for each, and they use different providers depending on situation.
Shipping everything to one location that redistributes would explode in both costs and complexity.
Easier to roll out this way would be my guess?
With your approach, the burden is on the post office to update their handling process.
With the implemented approach, nothing changes about the postal process, and the burden of work is shifted to the sender, who must look up the code for the recipient’s current address.
Maybe they will, but this seems to be separate in that if it resolves the physical address from the code during the checkout process like a DNS call, it isn't reliant on actually shipping with Japan post and even if it does, it now has the physical address, so why not use that.
One thing about the long physical address is that it adds a bit of redundancy. Misspelled or wrong name... still arrives. Misspelled street or city... might still arrive. I'm guessing 1 number off on that code and it's wrong. For the example in the story, a customer inputs the number, three store makes the call to Japan post and renders the physical address to the customer so they can verify before committing.
Surely they won't be so incompetent the specifications won't include a correction code and mandatory checking of it ?
The address would need to be looked up in some cases due to different delivery costs and time estimates.
I'm sure having a publicly available API that shows the region a code is in would be fine as long as the courier honours pricing based on that information.
I imagine things would get complicated if private courier companies start getting involved
Imagine inputting: "IiIIil" but meant to input "IiIIi1"
Oops now my package is going to Joe Blow on the other side of the country.
That’s probably why.
If my password manager has an option to not use similar looking characters when generating a password, it should be trivial to do the same for this system.
25^7 is 6 billion, and there are fewer than 200m people in Japan. That's enough for one character to essentially be error correction.
25^6=244,140,625 - I think I have seen estimates of 100,000,000 addressable entities in the Japan postal address system (a bit half of them being households, the rest companies,public buildings, various organizations and so on )
TFA says "seven-digit combinations of numbers and letters".
Yes absolutely. But my comment was just to confirm what you were saying about 25^7 (with one control digit) i.e. 25^6 would be enough to cover everything that is addressable in the system today. If we increase that to alphabet+ numbers 35^6, it is even more feasible.
Oh, I see, thanks for clarifying.
Given that this is Japan, they probably just mean digits 0-9, which are quite often rendered that way rather than 一 to 九.