Hacker News new | comments | ask | show | jobs | submit login
German town votes ‘No’ to street names (dw.com)
173 points by theBashShell 13 days ago | hide | past | web | favorite | 289 comments





None of the discussion I've seen here so far has raised what I think to be a much more important point: there should be an extra level of indirection between me and my physical mailing address. In other words, there should be a DNS for physical mailing addresses. That way my publicly visible "host name" wouldn't reveal where I actually live (makes it harder for burglars to find me) and if I move I only have to update my real physical location in one place (my "name server") rather than with everyone who wants to send me mail. Also, publicly-facing logical addresses (the analogue to "host names") could be mnemonic, and one person could have multiple aliases. So you could, for example, send physical mail to (say) "Lisper@HN".

This way no one except delivery people would care what these addresses resolved to. They could be lat/long, traditional street addresses, X-codes, whatever.


Congratulations, you have figured out what a P.O. Box is. A bit out of fashion (at least in NL) but was the standard for exactly this. Or people who were expecting to move around quite a bit.

My Dad used to have one (around the divorce of my mom), My Uncle had one, as he was abroad a lot and moved house a couple of times, and wanted a dutch address. etc etc.

As for paper email, aren't there also services that open and scan your mail for you?

Not exactly "DNS for physical mail" as you described but quite handy!

As for the obfuscation part, if the delivery people know how to resolve... so will many others. ;)


No. A PO box is a physical location, it just happens not to be my home. If I move non-locally I have to change PO boxes.

In the UK a PO Box is also an address, which you can use in different ways

Mail addressed to the PO Box -> You collect from post office

Mail addressed to the PO Box -> They deliver to street address (which only they know)

Mail addressed to street address -> You collect from post office

The most annoying thing I found was that they would only let you have a PO Box in a town if you had a street address there, which I did not.


That is interesting, in the scenario that they deliver to your secret street address, are retailers able to give shipping time estimates? Or is that less of a thing in the UK due to distances being shorter?

One the one hand I guess it wouldn't make any difference, since the PO Box must be in the same town as the street address.

On the other hand, the PO Box is provided by Royal Mail. Another delivery company (there are many who handle parcels, RM only deliver letters and small packages) would not necessarily have access to it, and certainly not the associated address.


There are mail services that do this - it's like a PO box, but not provided by the postal service and they will forward your mail to anywhere. These services are quite popular with RV folks and world sailors. Some will even scan your mail for you.

That's still not name indirection, this is more like a 302, because your physical mail goes first to the company and then gets moved around.

But the person who sent the mail doesn't know where it ends up. These services are exactly like what the parent was talking about -- someone sends a letter to a virtual address at the mail aggregator service, and the service then sends the letter on to your physical address.

A 302 is like a letter that's been marked return to sender with a forwarding address. The original sender would then know the actual physical address.


True, 302 was the wrong analogy. But the good point here is that there are three kinds of indirection, not two.

The problem with forwarding is that the mail has to make two physical hops so it takes twice as long and requires twice the resources. This gets particularly annoying if I set the service up in, say, California and then move to New York.

Ideally the US Postal Service could offer a service like this, then it wouldn't require any extra steps.

Yes. Exactly. Except that I would want a private entity to manage the database in the same way that name servers are done now because I want the private carriers to be able to access that info. And I would like the protocol to be open so that I could manage my own entry(s) if I chose to.

Are you sure it works this way? This seems like a very solvable problem.

Over the holidays I researched all major providers of virtual addresses and can confirm that this is how it works.

First your mail is routed to the provider’s local address or central warehouse, then they manually forward it to you. Some providers (incl the popular Earth Class Mail) operate as a hub-and-spoke network: all mail goes to their centralised distribution center, they process it and send it to you. This means if I buy something from a few blocks away, it will first have to Midwest then back to me.


Would you be willing to share your research? It just so happens I’m looking for such a service currently. Email in profile.

How else could it possibly work given current constraints?

> Over the holidays I researched all major providers of virtual addresses

I’m okay with latency and eventual consistency. I just need an analog->digital funnel for paper mail when I’m thousands of miles from home.


This is also my use-case: I'm often away, but cannot miss snail-mail from gov/lawyers/clients. I'm based in NYC and was evaluating this market.

The main discovery was how much of the "virtual address" industry is a nice wrapper around two independent service providers I call "processors" and "sellers".

"Processors" are the people who actually handle your mail. Often they are existing businesses like serviced offices, co-working places, pack&ship outlets etc. These small businesses have a fixed address and personnel, so their marginal cost of handling envelopes is much lower than having to sustain an office dedicated to processing mail. These processors are combined into a whitelisted network and this is how you get many addresses everywhere.

"Sellers" are the endless firms that sell access to networks of processors at arbitrary prices. There are exceptions (e.g. EarthClassMail) who manage their own processing, but most companies I found on the internet are just billing fronts with an app. I found this out because I kept seeing the same "virtual addresses" all the time, so I inspected those buildings on Google Maps or in person.

My selection criteria had three parts: 1. Deliverability (can I send both letters and packages, and will they arrive?) 2. Longevity (will this address exist tomorrow?) and 3. Security (who might be looking through my documents?). Based on this criteria, I have almost no regard for who the "seller" is, and it's all entirely about the "processor". By reviewing addresses given, I found out that sometimes the "virtual address" belongs to a shipping/mailing outlet. These people are dedicated to mail, and tend to be very stable - I prefer them over services offices/co-working places that have other priorities. I also found some mailing stores that were actually UPS stores, and I see this as reducing the risk further.

So for ~$30/mo, all my snail mail goes to a UPS store (managed by the seller service at https://www.anytimemailbox.com). When new item arrives, UPS people send a photo of the outside and I decide whether to trash or forward the item (or pick it up myself). This address looks like a regular apartment number (e.g. "123 Main St #45") and I can use it everywhere (unlike a PO box). In fact, it is listed as my primary address with the phone company, the bank, state driving license, insurance company, government records, and all business registrations. I try to be paperless, but now I am free to travel or move as much as I want knowing I'll never miss anything important from the old school of correspondence.


Thank you so much for sharing your research.

That's like those MITM corporate proxies.

Yuck.

Where I live you can have a "Virtual Address" with the national postal service, which is essentially what you're talking about. You can direct it to various houses or you can direct it to 24/7 access postal locker at any place you like.

You can change your virtual address itself or keep the same one for however long you like, and just change the final destination that it resolves to. The postal service knows where your current postal destination is set and third party freight carriers able to look it up from an API (someone has to deliver it eventually after all). Your destination can be changed in your online account, and if you want a particular package delivered somewhere different you can quickly create a new virtual address and use that instead.


Where do you live? And why isn't this the norm everywhere ?

There are mail forwarding services. They have a bunch of PO boxes, give you an address consisting of the PO box address and a made-up prefix eg Flat one million. They have a web interface which you use to view received mail, trash the mail, forward the mail to a specified address, select forwarding option eg 1st class mail, 2nd class. When you move, add your new forwarding address.

PO boxes are not necessarily a physical location.

That's news to me. Where can I get a PO box that is not bound to a physical location?

I think the main benefit to this isn't obfuscation, nearly as it is management. Instead of storing fully resolved addresses, "pointers" to the addresses exist, which are much easier to manage.

makes it harder for burglars to find me

How so? Or rather, what exactly do you mean here? Burglars don't want to find you personally. They're scoping out a property an determining if no one is home. They aren't looking though the yellow pages for potential victims remotely; they're driving around the neighborhood trying to get physical access.


This was my first thought.

Put another way... If I'm the type of person whom criminals will target by name, address obfuscation isn't what I need. Instead, I need a comprehensive security system (likely including body guards). Think heads of state and other famous people.

Joe Developer isn't being targeted by name. He just gets unlucky because he left his MacBook within sight of the window and didn't lock the back door.


I write a blog. One of the most popular kinds of articles I used to post was travelogues, until I realized that I was pretty much advertising to the entire world that I was away from home. That wouldn't be so bad except for the fact that it is also trivial to find out where my home is.

So you think that burglars are:

  1. Searching through travelogues
  2. Waiting for one where a person is actively posting
  3. Assume that the blogger is still traveling
  4. Researching where that person lives
  5. Hoping they happen to be close by
  6. Hoping no one else lives there/is house sitting/is renting while you're out
Rather than

1. Driving around looking for a place with a bunch of mail overflowing from the mail box and packages piled on the doorstep or some high value target visible through a window, regardless of knowing what the address is of the house

The good news is you're worried about nothing. However, adding a layer of abstraction to an address won't help curb burglaries.


You are overlooking that burglars often have interests outside of burglary.

The scenario where his blog could get him burgled is not burglars who set out to identify targets by reading travelogues. It is a burglar who happens to read the blog because he finds it interesting and happens to recognize that he is being presented with a burglary opportunity.


I always hesitate before posting travel updates on social media, and make sure it's only visible to friends. Probably paranoid, but it makes me feel better to not live with the added risk of entire the world being able to know my house is empty.

I think an enterprising hacker might get it into their heads that there could be profit to be made by systematically searching blogs for travelogues and inferring that their authors might be away, and then selling that information to interested local parties.

A layer of abstraction to your address would make it easier to stop or redirect all deliveries while you're out. This would completely prevent burglaries based on piled up packages and mail.

This is largely a solved problem. Have your mail held (admittedly you can only do this for up to 30 days in the US). And just don't order anything else that will potentially arrive while you're away.

At least some of the major package carriers in the US offer package-holds.

Assuming that putting a hold on the layer of abstraction is no harder than putting a hold on your mail, what you're describing is harder than it would be with the abstraction layer. It's also less effective, as it doesn't handle packages sent by other people, unexpected delays in shipping, or unplanned trips.

So is a friend or neighbour who stops by once a day to feed the plants, water the cats, and empty the mailbox.

Why not just post the travelogue after you return?

For the same reasons that newspapers don't publish week-old news.

News outlets do publish several days old "news" though.

I think the reader feels like it matters but it doesn't. If someone is "writing from $place" it seems more immediate, more authentic, more real? When really it wouldn't matter for most situations.


My travelogues were not just reviews of places, they were often timely pieces about current events in those places. And my trips often last for a very long time.

Because that reduces the narcissistic supply

> That way my publicly visible "host name" wouldn't reveal where I actually live (makes it harder for burglars to find me) and if I move I only have to update my real physical location in one place (my "name server") rather than with everyone who wants to send me mail.

This part is done by mailbox companies like EarthClassMail.com. I've used them for years, and it's wonderfully convenient to have the same postal address despite moving all over the country. Plus, if you want, they'll open specific envelopes, deposit checks for you, scan stuff, etc. I love that because as a traveling consultant, I don't wanna wait until I get home to have checks go into my bank account.

Drawbacks: there's a latency and an expense involved with forwarding packages. I tend to not use it for packages, just for regular postal mail. Anytime someone asks for an address, they get my EarthClassMail one, because I probably don't want whatever they're mailing me anyway.



I don't think it would be advisable to deliver mail from point A to B directly, physically. It would be too inefficient since mail has to be sorted and is often sent to intermediate facilities to do this.

Forwarding is worse not because it prevents A -> B directly but because it forces all mail to additionally pass through the location of the forwarding address.

Well, forwarding really isn't any different because standard, non-forwarded mail often goes to addresses where the recipient isn't present either. Forwarding just requires it. So its an extra stop (which is worse), but the concept is the same regardless of whether mail is forwarded or not.

If I move from California to New York I don't want mail sent to me from New York to have to go to California and back.

You couldnt paste a single sentence?

That had nothing to do with it. I didn't want to bifurcate the same topic into two branches of the discussion tree.

How much would you pay for this? Would you also accept a small delay to receiving mail?

With enough capital it would be fairly straightforward to provide a forwarding service for this. You could also have spam filters or reject mail from specific senders, provide scans of the outside of the letters, specific delivery dates etc.

Given how we are ordering more and more things online nowadays this could be a very exciting opportunity.


I don't want forwarding. If I move from California to New York I want mail sent to me from New York not to have to traverse the country twice to get to me.

That would only work if all post-services would work together. Otherwise it would still be relayed to your redirection-service first, where it receives a new address-label and gets send out again.

I think those services exist in some countries, but not from postal services themself, so delivery-time is still rather slow.


> That would only work if all post-services would work together.

That's the idea: the redirection would be handled by the carriers, not some third party. In place of the current system of (a) scan address; (b) route package; (c) deliver package the process would be (a) scan recipient ID; (b) translate ID to address; (c) route package; (d) deliver package. The address lookup should be completely automated and introduce no discernable additional latency in the mail delivery. It could even be implemented as a special case of the address normalization step carriers already perform.


What I mean is, in many countries there all are third-parties, because there is no monopoly on postal services anymore. So they must cooperate in some way and share the redirection-database. And this means implementing this would be very very complicate and take a long time, because companies are not very eager for cooperation until some strong reason exists. And this feature, as nice as it would be to have, is not really a strong reason for cooperation.

In the first place there is not even enough data to know whether the huge investment would be justified.


Don't these private carriers already share address databases, or outsource address lookup to third parties with incentives to find the right location for any valid address? Adding a level of indirection to that existing system wouldn't take very much effort.

Or do you have one address for UPS deliveries and a different address for FedEx deliveries? (Substitute local carrier names as appropriate.) While that's more work—and pointlessly complex IMHO—you'd just need to set up one ID per carrier and give the correct ID, as opposed to the carrier-specific address.


I don't think they share any information, or maintain address-databases. Usually there are official geodata (maps, postal codes) and customers normally don't report back their own data to the government. Anyway, logistic of postal sercice is highly optimized and well defined, and changing anything is not as simple as hacking some script somewhere.

If they can't easily implement something as simple as a mapping from IDs to addresses, that seems less like "highly optimized and well defined" and more like "inflexible, cobbled-together mess". I don't doubt that what they're doing works well for the situations they've encountered in the past, but situations change and they need to be able to adapt.

Anyway, a shortcut to implementing indirection for these less-organized postal services you're describing would be to look up the recipient's ID when the parcel first enters the system and affix a label with the full address. That way the sender doesn't need to know the address and everything else about the delivery can stay the same.


Highly optimized and well defined solutions are by definition inflexible. Because they are the result of removing everthing which is not relevant for the task. If you are a company handling thousand of letters per minute, then you don't have any reason to be flexible, because the task does not change between the iterations. I mean we are talking here about industrial level of optimization, not some dude in his small office stamping his three letters a day. Companies on such level optimize everything from highest to lowest levels, because any optimization can make a difference of millions of whatever currency they use.

In that scale, nothing is simple, because everything can become expensive on the high run.


> Highly optimized and well defined solutions are by definition inflexible. Because they are the result of removing everthing which is not relevant for the task.

Only if you don't consider flexibility part of the task, which is rather short-sighted in today's world. Maybe postal services could afford to be inflexible once, but those days are long past.


This would be abstracted from the user, and you could have fulfilment in both locations. Moving state is also kind of an edge case for most people. It would make more sense to pilot it in a smaller country first, the US is huge and sparsely populated.

> That way my publicly visible "host name" wouldn't reveal where I actually live (makes it harder for burglars to find me)

Someone has access to this database. And they're selling the data for a price inversely proportional to the exclusivity of their access. Increasingly it seems that if someone is selling data on the open market, this information is increasingly available to bad actors such as burglars.

So it just seems like a very bad form of security by obfuscation. There are definitely things to be said about generally decoupling your physical location from your mailing address if you frequently move, however.


> So it just seems like a very bad form of security by obfuscation

Yes, but that's still an improvement over the status quo where anyone can trivially find my home address with a Google search.


I would be more worried about the smart burglars than the dumb ones.

I like your proposal. I was just thinking about this problem recently too. In an era where people are becoming more privacy conscious online, they can make reasoned decisions about how much of their data they shield from megacorp$ that are eager to vacuum it all up. But if you want to partake in the convenience of online commerce, how do you effectively shield yourself from providing your home address to dozens/hundreds of companies?

I wonder if there would be a market for the service you're describing today. It seems you could achieve it in the current postal system. You sign up and are assigned a unique identifier, which you use as your "Name", then ship to a local address maintained by that service, who then forward it on.

All you're doing in this case is just centralizing the trust from each individual online business to the reshipping service. Is that a reasonable first step?


> Is that a reasonable first step?

Sounds good to me.

You know who is well positioned to do this? Netflix.


Is Netflix the kind of company we want doing this though?

I'd imagine it being some kind of non-profit organization that is funded by a combination of donations + fees for the reshipping service.

Put differently, if we want to centralize all this sensitive information (people's residential address), would you want an organization beholden to shareholders and profit margins operating it?


> Is Netflix the kind of company we want doing this though?

Better them than Amazon, Google or FB. (Oh, the horror!)

But I like the non-profit idea better.


Why Netflix?

Because they already have a deal with the post office where you can send something to your nearest Netflix distribution center without having to specify where it is. At the moment the only thing you can send there is Netflix DVDs, but fixing that is just a negotiation away.

It seems Netflix solves that by not requiring the USPS to actually deliver those DVDs, they just get them to a hub and Netflix workers fetch them from there (and occasionally simply check them and re-label them for deliver to another customer), but that seems a bit too complex and expensive for regular mail and packages.

What I think is interesting about this comment is it appears to be a novel enough idea that many of the replies quickly dismiss it without really seeming to grok it.

People are dismissing it because it already exists in multiple countries. In the US for example, you can get a PO Box and have the USPS forward that mail to your home address.

There are also companies that will do this for you with variations on the service (a registered agent is one example, the company Mailboxes Etc also offers this service), which some are dismissing based on "yeah but then a third party is handling your mail". But if your country's post office won't do it then the only alternative is to have a third party handling your mail.

This service exists, and people are dismissing the idea because it is not novel.


What you are describing is not the same as what I am suggesting.

The only difference is that you have to send mail to a normal post office address and not a "hostname". Other than that, virtual PO Boxes exist and fulfill your basic requirement of "no one knows my home address", even if some of the behind-the-scenes implementation details aren't exactly the same (latency etc). We're not discussing something new, we're discussing low-level implementation details of an existing idea.

I'm not saying your dream is bad, which is why I didn't respond directly to you. What I'm saying is I don't think it's a novel/interesting idea for a startup to work on as I believe was being implied by the person I replied to. Even if it became a big business (which I doubt since even UPS/Fedex/Mailbox ETC don't have that many postal customers for their own services) it's just too easy for the national postal service to compete with if it was successful.

I think it is a great idea for the postal service to do, but it would be very risky for a startup to do, even by modern startup standards. The existing postal system is a marvel of efficiency already.


> The only difference is that you have to send mail to a normal post office address

Yes, but that's a big difference.

> I don't think it's a novel/interesting idea for a startup to work on

I ever said that, and neither did whiddershins. Not all novel ideas are good ideas for startups.

The internet itself was a very bad idea for a startup when it started out (much too capital intensive). That's why it was done by the government and not by a startup.


Thank you for noticing.

This. I relocated a couple of times recently, and am about to relocate again. I too contemplated about a similar service. Personally I'm not too worried about the privacy aspect as much as about the pain of notifying every single service provider when I change my address.

It does open up some interesting use cases. E.g., when I go on vacation then hold off delivery, if it's from Amazon then deliver it to such and such a post box etc. I'm not sure about the market size though.


But physical addresses are physical address, they specify a place, not a person. We'd have to set up a paralell system, and expect the government to be able to fund and handle both... Yeah, they can't do either for the current system...

DNS isn't run by the government. Why would this have to be?

Here in the US it’s incredibly easy to find where anybody lives if they are registered to vote. You can plunk my name into the Idaho voter website and it will spit my address out, no fuss.

I find much more crazy that your party allegiance comes out, and that those records can be harvested by apps to harass your friends/family/employees etc.

Huh, they don't have secret voting in Idaho? Isn't that federally required?

Your vote is secret, but your registered party affiliation is not.

Do you have to register an affiliation? What is the scope of that affiliation? Who holds the register? What is its stated purpose? Are you allowed to be in more than one party? Does it bind your vote in some way? It all sounds very strange.

Now in the UK large party donors are publicly listed. And parties have memberships in private.

I guess I should do some research!


You can register as “unaffiliated”, but depending on your state what options you have for voting in primaries depends on your affiliation.

Some states only allow you to vote in the primary for your affiliated party, others let you pick one or the other regardless, some will only let you vote in one parties if you are affiliated but the other has no such restriction. The system is a giant clusterfuck of rules, but everyone just kind of works with it for now.

Your affiliation is part of your voter registration, so it’s held by your Secretary of State’s office.


Thanks for that. The more I learn about USA the more tenuous the idea that it embodies democracy in a serviceable sense.

This.

For the past 10 months I've been travelling constantly staying at hotels, AirBnb's etc and in a lot of cases not having the permanent physical address is very inconvenient. I've tried multiple solutions but none of them work well. Yes, there are companies that offer "virtual mailboxes", but I can't ship any packages to them.


Earthclassmail.Com is a virtual mail box.

You still need an easy way to tell people how to get to your physical address to visit you.

Sure, but then it doesn't matter what format you give that information out in. You can just give them driving directions.

Congratulations! This is xkcd #783.

There's no need to remove addresses from places that already have them, and a system like [Plus codes](https://plus.codes/) could identify physical locations without an address.


A Google maps link would likely save them time rather than telling them the actual address.

But what if they want to use Waze or Bing Maps or Apple Maps or...?

Yeah, using a Geo URI[1] would be a better option. And helpfully, Android already supports them.

[1] https://en.wikipedia.org/wiki/Geo_URI_scheme


Have we finally reached the point where people no longer know what a phone book is? :D

I know what a phone book is. But what are the odds an 18 year old has seen or used one?

Seen, fairly low. Used, that's another story.

I saw one of those taped to the door of a restaurant. I was kinda like, huh what the heck would that be useful for now?

Obligatory reference to the Fundamental theorem of software engineering

https://en.wikipedia.org/wiki/Fundamental_theorem_of_softwar...


Good for them -- it's their town after all!

My grandmother's house (in Australia) was assigned a house number (in the 70s?) but she refused to use it, continuing to give her address as the house name and insisting that we grandkids address letters to her by name. As far as I know they always arrived. There was no number on the house or letterbox. I went by a few months ago (somebody else lives there now) and there's a number on the letterbox.

My in-laws live in a village in Germany (a little south of the town in this article) with only two streets, referred to as "am Berge" and "im Dorf" (on the hill and in the village) and though the house numbers are not unified as in this article's case, there is no actual street sign -- but by "name" it's pretty obvious which is which.


> My grandmother's house (in Australia) was assigned a house number (in the 70s?) but she refused to use it, continuing to give her address as the house name and insisting that we grandkids address letters to her by name. As far as I know they always arrived.

-My parents once received a postcard addressed simply to their first names - period. No last name. No street. No town. No postcode. No nothing.

It goes to show that:

a) Norway is, as such things go, a pretty small country with at least one postal clerk bored enough at work to do a little bit of investigating, and:

b) Having a central database in which every Norwegian is required by law to be registered with the address of his or her residence does have its benefits for the less privacy-conscious. (This was pre-internet everywhere, and they were not listed in online yellow pages or similar - so someone must have made a query to the folkeregisteret database along the lines of 'Is there a $male_name which lives with $female_name anywhere in the country?)


That is rather amazing.

There was one case I remember reading about 15-20 years ago where someone in the UK dropped a letter into a mailbox with just a name, street number (IIRC), and street and it managed to get to the recipient in Pennsylvania, US. Someone from the post office guessed that a bunch of misreadings happened to get it close enough.

I guess a creepy possibility is if they have a database of every letter sent they can search that to often figure out the likely recipient (at least in the case I am thinking about where they were in regular contact).


This is a textbook case of tyranny of the majority. The people who aren't inconvenienced by the current status quo vote to preserve status quo. The few people who are inconvenienced aren't numerous enough to change the status quo. In the end nothing is done to solve the problem.

I'm not sure "a group of people has an inconvenience that doesn't apply to the a majority of people, we should change it for everyone" is much better. It happens and I support it when it's specific kind of issues that's worth protecting the minority even if costly (discrimination, fondamental freedoms,...) but for things like that...

It's the same principle as when some want to filter the entire internet to protect a small minority from reaching content that could shock them (often porn). It feels less like tyranny of the majority and more like democracy. Being in the minority in a vote should not be confused with being abused by the majority like you do, because that dilutes the recognition for cases where it really happens.


It's not like people will suffer when their house suddenly gets a proper address.

On the other hand, people already suffer because they have problems with their deliveries, visitors etc.


  It's not like people will suffer
  when their house suddenly gets a 
  proper address.
The 'suffering' is the loss of their cultural distinctiveness.

Like if you told the French they should all start speaking english to better interact with the international business community, or if you told programmers they should start wearing dress shirts so they look like professionals.


It's not even remotely drastic as that. It's a street addressing system. This whole situation seems like a proxy war for something else.

Why do you find it hard to believe that people could become attached to and invested in their town's unusual street numbering system?

People become deeply attached to all kinds of arbitrary things, e.g. the particular shade of red used by their favorite sports team, or the exact shape of the cookies their family has baked for Christmas every year.


I agree that it's not remotely as drastic as telling the French they have to speak English, but it seems fairly comparable to telling programmers they have to wear dress shirts.

> Like if you told the French they should all start speaking english to better interact with the international business community, or if you told programmers they should start wearing dress shirts so they look like professionals.

Both of those are also true btw.


They may be, but is it worth it or necessary?

For the French, I don't know, but you could argue that it would improve the French economy I suppose.

However, for programmers, I don't see the benefit: programmers do not have any trouble these days finding jobs, and in fact are in high demand regardless of how they dress. Dressing more "professionally" doesn't seem like it would actually help them any, so why bother?


I guess those people could continue to use their old addressing scheme, however sooner or later their mail may no longer be delivered to them.

People didn't suffer either with opt-out internet filters schemes that we saw in some anglo saxon countries, they can just opt-out ! On the other hand, people already suffer because they have kids getting access to porn, etc ...

The problem is not always the direct change, but the infrastructure and public mindset change. Add a filter, and now adding things to it is much easier. Germany for better or worse is very much privacy minded, even more so that us neighbors (and compared to the US or asian countries we're very privacy minded). They basically have no google maps coverage because of it, for exemple. So a change in that direction will impact them.


> Germany for better or worse is very much privacy minded

Are you kidding me? 6 German police officers were taken out of their job because they have threatened an ethnic Turk, attorney. They have threatened her new born baby by his name while even none of her neighborhoods know the babie's name. And how did they found that the police did it? Apparently few hours ago the police officers have queried her and her babie's name from their database.

So, while Germans think that they mind their privacy they have no idea that they live in none private country. Germany is not really different from other European countries in term of privacy. Who cares about your shopping habits? Will someone going to point you a gun for shopping or for not using DNT header in your web browser or for using a credit card? Police have all citizens' addresses, names, phone numbers, relatives. And police have guns to point at citizens and arrest them just because. Can you hide your name and address from German police? No.

http://www.spiegel.de/panorama/justiz/frankfurt-seda-basay-y...

https://www.hukukihaber.net/dunyadan/turk-avukata-tehditleri...


Yes they did.

> It feels less like tyranny of the majority and more like democracy.

Those are the same concept. One is just supposed to be good while the other is bad.


From the article it doesn’t sound like much of a problem to me.

And the world is becoming increasingly homogenized so I’m delighted to see some people pushing back.


Pushing back is fine if you have a good reason to do so, and the proposed improvement really isn't an improvement at all.

I fail to see how this is the case here: visitors to the town are getting lost because their address scheme makes no sense at all except to the locals. There's a good reason every other city and town in the whole country (presumably) has adopted the building-number / street-name convention (and in most other countries as well).


You assume there is a problem.

Good for the 60% of the 69% of voters that showed up and voted against street names. Less good for 40% of 69% voters that showed up and voted for street names. And probably 'meh' for 31% of voters that didn't care enough to show up.

Personally I really hope something like https://plus.codes/ will catch up. People should be easily able to provide address without knowing or caring about local arrangements.


> Personally I really hope something like https://plus.codes/ will catch up.

Who needs some proprietary system (there’s also a so-called “three word” competing system). Just use lat/long which are universal.

> People should be easily able to provide address without knowing or caring about local arrangements.

What can be more “local” than an address and why should it be subject to outside specification?

If you consider lat/long user-unfriendly then just stuff them into URIs of some sort — URLs would work — and keep this em in a sort of DNS. Then people could name their addresses whatever they want.


What can be more “local” than an address and why should it be subject to outside specification?

Surely its the exact opposite. Locals don't need an address, they know what you mean when you say "the yellow house behind where the pharmacy used to be". "Outsiders" are the target users for addresses and thus it should be optimized for their understanding and use cases.


  Just use lat/long which are universal.
I'm no supporter of plus codes or w3w, lat/long are no panacea. In particular, (a) they're quite long, which is inconvenient for writing on envelopes or entering into sat nav systems; (b) there's no way to detect swapped or miskeyed digits, which depending on the digit can cause arbitrarily large errors; and (c) they're difficult for humans to parse compared to city and country names.

Also, the idea of expecting average people to memorize them is utterly ridiculous.

> Who needs some proprietary system (there’s also a so-called “three word” competing system). Just use lat/long which are universal. >

Plus Codes are open source and there are no licensing fees. They also compare different location encoding systems like Lat-Long etc.

https://github.com/google/open-location-code/wiki/Evaluation...


Due to the urban canyon effect with limited visibility to overhead navigation satellites, lat/long doesn't necessarily have enough precision for reliable delivery. Are you supposed to deliver the package here or to the building next door? With street address numbers it's easy to verify but with only lat/long it's hard to be certain.

The Japanese Quasi-Zenith Satellite System (QZSS) is designed specifically to work around this problem, but that level of precision isn't available everywhere yet.


> What can be more “local” than an address and why should it be subject to outside specification?

This is a pretty strange statement. Ad address, in my interpretation, is an identifier that can be given to anyone without local contextual knowledge to find a place of interest. You can tell me in my hometown that the cafe is opposite of the hardware store and I'll find it precisely because I know where the hardware store is. In other towns/places, as an outsider, I would need an address to be able to find something.


The issue with lat/long isn't just that it's inconvenient, but it doesn't map to high density structures. A resident of 432 Park Avenue, the tallest residential building in the world with 96 stories, would share a lat/long with 95 other people.

The same applies to just "432 Park Avenue" as well.

Simple fix. Just include altitude :)

You can do exactly what is done now...

LatxLng Apt 23C


Assuming the building is bound on all sides by roads then delivery drivers will need to know on which of those roads to gain access. Why not name the roads, then you can say the road name.

Also, when sorting mail looking up lat-long for every item is tedious, so we could name the address by the town/city name that it is local too.

Now lat-long is getting to be redundant ...


Plus codes are just a well-specified algorithm for encoding lat/long. They are intended to be used alongside a hyper-local identifier, such as a door number (or colour or other symbol).

I'd love to use lat/long. I would expect that on the phone with GPS I should be able to paste my current location (after reviewing it and possibly adjusting it on a map) anywhere .. mail, text, message in any app, any text field. And whoever views the message or document with such pasted location should be able to see where it is, navigate there.

But regrettably that's not the case. So I'm hoping that SOMETHING will catch on eventually.


Yeah and guess how good it works offline. You don't even know which country a place is in if you only know coordinates and have no access to the smartphone. Do a typo and there's no redundancy or error correction. You'll end up in some ocean then.

can't you already do this with share intents on android? how many messaging apps do you know of that don't provide an endpoint for share location from google maps?

All of the routing services I have tried work perfectly well Hilgermissen.

For instance, https://planner.myrouteonline.com/route-planner/1752/#, works perfectly in Hilgermissen and has no trouble finding the houses and finding the optimal route.

There is in fact not a problem to be solved, except perhaps between the ears of the people managing delivery companies.


> What can be more “local” than an address and why should it be subject to outside specification?

Your age is even more local. I'd say even personal. Still all others could use objective sense of what a year is and how many of those ago you were born. That's why we need outside specification for local things. Because local doesn't mean isolated from the rest of humanity.


The same reason phone numbers are ? I would have thought that was some what obvious.

Not many people know that the x.400 email standard allowed for physical delivery options in the spec


It's adorable and precious and very "silicon valley startup" to reinvent the coordinate system and then proudly remind you that it's free!

> Personally I really hope something like https://plus.codes/ will catch up. People should be easily able to provide address without knowing or caring about local arrangements.

It would already be a better solution indeed. But ultimately I would prefer to have a kind of unique identifier for my person that is not based on my location nor my name, like for instance A23C-B442-9F72. It could then be associated to a first name, last name, and geographical address in a database. Delivery services can then look-up the geographical address and your name at delivery time to find you or your mailbox.

This would avoid so much troubles for stuff like:

- Moving: I can change my address in one place and everything will be delivered to the new place, no needs for thousands of calls and letters to companies to update every subscription.

- Changing name: Whenever someone is changing his name for any reason (weeding, gender change, etc.) it should only be modified in one place.

- Disposable address: It should be possible to get disposable identifiers that can only be read by delivery services for hiding your real identifier/address from suspicious website

Of course this system is not perfect and probably has drawbacks, but it would be already a step in the good direction for me. Implementation details are left open for discussion (Centralized vs distributed database(s), checksum for spelling mistakes, financial repercussions, etc.)


maybe this is already obvious to you, but a disposable address only hides your location to a certain degree. an untrustworthy vendor probably couldn't figure out your exact address, but I bet they could narrow down your city/town by requesting shipping quotes from many different locations and triangulating via price differences.

>People should be easily able to provide address without knowing or caring about local arrangements.

That's up to the locals. They have the right to insist you should care first about local arrangements before interacting with them.


What if I'm a local but I don't care about other locals and their arrangements and I just want to get my damn package delivered from people who care about local arrangements even less than I do?

Campaign harder for the next referendum, I guess, put up with it, or move.

I could understand it if the existing system were broken, but it doesn't seem to be. From what I can read, it seems to work, presumably people do get their mail and packages.


> Campaign harder for the next referendum, I guess, put up with it, or move.

Or work around local arrangements using lat/long or plus codes or whatever naming scheme delivery companies will choose to use for your town...

Democracy (or any form of government) wouldn't last long if you couldn't work around regulations.


> They have the right to insist...

People have a right to share true information legitimately acquired without the constraints of governments or mobs in other jurisdictions.

Locals can't prevent foreigners from giving directions to their library or the best restaurant. Those are just facts of the world. Visitors are welcome to go discover a town then leave and tell their friends about it.

Then an addressing system is just a shorthand for that locational information. It's just a hash function or a lookup table. No one has a right to prevent others from using data structures to organize their thoughts.

All rights granted to some impose negative obligations on others. Those implied negative obligations should not conflict with other rights. Rights should promote the healthy functioning of society. Also, rights should promote freedom more than restriction.

A right to prevent others from discussing addresses without permission of locals conflicts with other rights, doesn't promote the healthy functioning of society, and creates far more restrictions than freedoms. I don't think it's a good right.


Plus codes work as long as your location doesn't change.

Happens not so often. But it does happen that the coordinates of a building don't stay fixed: https://en.wikipedia.org/wiki/Structure_relocation


What if you move? Your address changes. Why would it be any other way?

If you want something attached to your identity not your location, that's what names are for. Just don't be surprised people can't find where you are if you give them your name instead of location.


I would imagine your classic postal address (ZIP / postal code) would also change if that happened since all western methods for addressing houses are based on location.

I say "western methods" because there are some exceptions to this rule is countries with high numbers of nomadic tribes.


You don't have to resort to countries with nomadic tribes.

House numbering logic wildly differs between countries. In Tokyo, houses are numbered in the order in which they are constructed. I would expect a moved house therefore to keep its number.

London (or even the UK as a whole?) also has a rather random assignment of house numbers even though the natives will claim its completely logical.


> House numbering logic wildly differs between countries. In Tokyo, houses are numbered in the order in which they are constructed.

Isn't that the case for most countries?

> I would expect a moved house therefore to keep its number.

Do you have any reason for this expectation? I ask because your comment here sounds very speculative judging by the way it has been written.

> London (or even the UK as a whole?) also has a rather random assignment of house numbers even though the natives will claim its completely logical.

It's only random where houses haven't been built sequentially down the road. Much like the Tokyo example you gave.

In any case, you're addressing the point about house numbers where as I was discussing postal codes. The two differ. House numbers are an exact designation for your property; be it a house, flat, or even mobile home on a plot of land. Post codes (zip codes, etc) are a region of land which would typically serve a cluster of homes. While you code move a property and decide to keep the same house name (though it would be massively confusing if you kept the same house number if you moved it to another street - so I suspect some authority or other would step in and say you cannot do that), you cannot keep the same post code.

To use a comparison, it's a bit like moving your house out of London and into Tokyo but still putting London, England on your addresses and expecting post to get delivered to the right place. While you can move your house, you cannot move London nor England with it. And the same is true for postal / zip codes - in fact more so because you can in theory put garbage down as your address just so long as you get the house number/name and the post code right.


>> House numbering logic wildly differs between countries. In Tokyo, houses are numbered in the order in which they are constructed.

>Isn't that the case for most countries?

I don't think so.. as far as I can (subjectively) tell, it's more common to set up a logical system. Another poster mentioned the Finnish system, where the number reflects the physical distance from the start of the road (123 = 1230 meters). Another common scheme is to use even numbers on one side of the road, and odd numbers on the other side, and numbering the houses sequentially, leaving gaps for not-yet built houses. In my street, for example, there's a large gap in the numbering because the next building is nearly a hundred meters away. On the occasional 'two buildings squeezed into one spot' just add A and B (and C and so on) as needed. And around here the numbering always start from the most southern part of the road or street (and if you ask "what about roads going exactly east-west", then a) I don't know, and b) It's really hard to find a roading that goes exactly east-west. I suspect, however, that there's a system for that as well - "if in doubt, start from east" or something).


> Another common scheme is to use even numbers on one side of the road, and odd numbers on the other side, and numbering the houses sequentially, leaving gaps for not-yet built houses. In my street, for example, there's a large gap in the numbering because the next building is nearly a hundred meters away. On the occasional 'two buildings squeezed into one spot' just add A and B (and C and so on) as needed.

That's what we do in the UK as well (barring the gaps in the numbering) but for the most part it still ends up following the logic of "houses numbered in the order in which they are constructed" because you either need to know that there should be a gap (in which case the estimated build date is close enough to the rest of the street's build that you might as well consider it one lump development) or else you end up with letter suffixes or unordered numbering (which is what often ends up in the UK as roads get further developed decades after they were first built).

Ultimately though, there is no law in the UK about how houses should be numbered so as much as there are conventions, it's up to the housing development to apply them. However might differ in other countries which might have stricter legislation about naming houses?

Regarding the south->north scheme, I'd be interested to know what the streets are like in places that follow that convention. In the UK our roads are like a big bowl of spaghetti with even new builds wiggling around and even looping back on themselves (the road I live on is crescent-shaped). But where you have straighter roads, I could imagine a numbering scheme based on the direction of the road might work pretty well.


UK house numbering generally numbers away from the town centre. 1 is always on the left. Applies to most places I know. For twisty roads, it's the starting point that matters. I suspect this breaks in London, Manchester etc where the towns they were centred on have merged into one urban sprawl.

Cul-de-sacs number clockwise consecutively rather than following one side odd, other even.

Some of the strange situations where it breaks stem from WW2 bomb damage breaking a road in two, or removing a dozen houses etc.

Most councils do regulate street names and numbering, and they seem broadly in agreement. No idea of legislation or origin of the convention, but seems nationwide.

Example - chosen as it was one of the longer ones search popped up: https://www.swale.gov.uk/street-naming-policy/


Awesome. Thank you

Yes, there is. It is the Towns Improvement Clauses Act 1847.

Thank you, I didn't know that existed. Though having a read through that, I can't seem to find anything that dictates the system in which houses are numbered (only that houses should be numbered).

Section 64 and 65 talk about house numbers from a display presentation perspective but only go so far as saying "houses with such numbers as the commissioners approve of" but I cannot locate where it describes what systems are approved by commissioners.

Are you able to guide me to the part which does describe what the "approved" system is? I'm rather curious to read it because what I've seen and read to date has lead to to believe it's really more of an agreed convention than anything legislative so very interested to read what the actual rules are.


What are you referring to for UK house numbers? They’re just sequential by location on the road in most cases.

>I would imagine your classic postal address (ZIP / postal code) would also change if that happened since all western methods for addressing houses are based on location.

Not really, there are plenty of businesses that have big enough buildings that they have a single postal code to themselves.


> Not really, there are plenty of businesses that have big enough buildings that they have a single postal code to themselves.

Indeed but doesn't contradict my point because it's still just a location. The only difference is in your example it's home to a singular business rather than multiple residential buildings.

As an aside, I've worked in some buildings that actually had multiple post codes. This is common with large government buildings such as County / City Halls.


Hm, in Sweden they are shifting Kiruna, a mining town.

If the location changes drastically, which I guess would only matter for a public building anyway, you could just update the adress. I don't see the problem.

To me, it seems like an utter failure of democracy.

Why wouldn't the locals want street names?


> Personally I really hope something like https://plus.codes/ will catch up. People should be easily able to provide address without knowing or caring about local arrangements.

So… they've reinvented japanese block-based addressing except shittier all around, and proprietary?


Perhaps you could explain the Japanese system, and why it's better. Plus codes are open source though

As the mailman had to visit every house, and did it every day, he'd soon learn the idiosyncrasies of selfish people who refuse to label their house.

I used to deliver pizzas. If she'd ordered one, it wouldn't have arrived. "Joes Cottage, Long Lane", is a completely pointless address, it forces the person attempting to find it to investigate on average 50% of the houses on Long Lane. Given that only about 1 in 10 hours had a number visible at night from the car - especially in the rain - this would be pretty much impossible to do.


The mailman problem

I have relatives who live in Donegal, Ireland. They don't even have house names, just their personal name and a townland.

When post sorting & delivery was run in the town, by generational postmen who grew up there, everything was fine. They knew that Patrick R Flaherty was in the white house and Paddy J Flaherty was in the stone one.

Now the deliveries are by rotating agency workers driving a post van over from Donegal Town. It's chaos.


> I have relatives who live in Donegal, Ireland. They don't even have house names, just their personal name and a townland.

Ha, yeah same with my relatives in Mayo. When sending wedding invites, my English (now) wife was so confused by it all. Particularly so when Mr/Mrs <FAMILY_NAME>, <TOWNLAND>, Mayo would be applicable to a lot of people so first names are a must :)


While it's imperfect, Eircode should help with this.

Having just purchased a house in rural Offaly where several houses have the exact same address (I guess the post person was supposed to know everyone's name?) Eircode is really helpful for disambiguation.


Bingo: Eircodes were invented for exactly this problem.

Eircodes are unique per delivery point, whereas 35% of addresses were (pre-Eircode) non-unique.

This caused problems for, e.g.:

(1) Delivery people, except the state postal provider (An Post). As you rightly said, An Post kept a note at the local post office (not centralised) of who lives where. For anyone else (e.g. FedEx) it would be literally impossible to tell where to deliver an item, if the address is non-unique.

(2) Ambulances not able to find the correct house.

(3) More complicated provisioning for, e.g., fibre broadband - how do you avoid incorrectly disconnecting a neighbour, if the neighbour has literally the same address as your customer?

Now, an Eircode is sufficient in itself to uniquely identify the delivery point.

To see this in action, go to Google Maps and put in an Eircode, e.g., "D02 AF30 Ireland".

Or go to https://www.eircode.ie/ and put in an Eircode, and the rest of the address will be shown.

But to go the other way (from address to Eircode) literally requires, for non-unique addresses, the person to click on the correct property on a map.


Of course, Google Maps has a problem in that it gives people the wrong eircode if I put in my mailing address (I currently live in a building with several apartments - haven't moved yet). Perhaps Google maps can't handle multiple postcodes per building?

Quite possibly, yes - or otherwise related to the non-unique mapping from address to Eircode.

The other way round should always work though, i.e., if you give Google Maps the Eircode then it should put the location marker in exactly the right place every time.


you're correct, but if people type my exact address "apt X, # streetname, <city>" it still gives them a very different eircode.

Given that more people are using google maps than finder.eircode.ie we might wind up with Google determining truth here.


The chaos is not caused by the changing drivers, but simply by the unclear adresses...

If that is the official address, it should be on the map, so you should have to investigate exactly one cottage. It's not like street numbers are always simple and sequential either. I have streets that become other streets in the middle. I have streets that end in a T-intersection, then continue from another T-intersection half a km away. Addressing globally has never been anything close to e.g. Manhattan's grid system. As long as that town's official addressing system works the way the locals say it does, deliveries should arrive as normal with no hassle. And that town isn't even unique within the EU, there are plenty of towns on my side of the continent that just number their buildings and don't assign them street addresses.

It's silly. If every village gets to do its own scheme we'd have chaos.

The standard street names + numbers scheme works well, it obviously works better than their "numbers only" scheme, it should be adopted.


Street numbers + numbers is a joke, and doesn't work anywhere. People are only able to know where some street is if they happen to know the place very well, and you can never trust the ordering of numbers.

given that knowledge of local streets is necessary to find a route there, what would be a better system?

The city I live uses a sum of two numbers given by the house position in the two directions (one on the hundreds, other on the unities) into a coarse grid.

There are plenty of cities that have numbered streets, in order.

Some cities have named "mega-blocks" as a third grouping element beyond the street/number and finer grained than the usual neighborhood name, so you have to memorize less stuff.

In short, there are plenty of schemes that give way better result. If you have to look up on a street-name database to find an address, there is little to gain compared with a house-number database.


Ok, that makes sense. When your streets are lain out in a grid, everything is easier. But most municipal road systems are far less tractable. I can't imagine easily navigating my hometown, or any other town in New England based on grid coordinates.

> Good for them -- it's their town after all!

Doesn't mean they're doing what's actually best for themselves.

FTA:

> There are some house numbers that are about 3 kilometers (1.8 miles) apart,

Sounds like a pain to navigate.


> Good for them -- it's their town after all!

Of course, your grandmother was relying on the kindness of the postal service in this case. The postie could simply have decided it to return it as misaddressed.


This is a good example of humans' inherent bull-headed conservatism and hostility to change. The desire to keep something the same, simply because that's the way it is, regardless of the downsides. I actually saw someone defending this as the villagers preserving their "cultural distinctiveness" - people believing it's a key part of their identity that they have an awkward error-prone addressing system in their village! It's quite something.

It is a part of that town's identity, and the inconvenience to courier corporations is no good reason why they should give up on that. It's not about individual people's identity, it's about the town's identity. This bit of cultural distinctiveness - no need for scare quotes as that's exactly what it is - makes it of historical interest and attractive to tourists.

" the inconvenience to courier corporations is no good reason why they should give up on that."

This is an inconvenience to themselves - the courier is just providing them with a service that is impaired by the addressing system. It is by definition a good reason why they should replace the error-prone system.

Let me ask you - in your job, if someone suggests a change that will improve how things work, would you be against it because the bad way of doing things is party of the business's identity? It's nonsense, really isn't it.


Jobs are different from towns. The people living here aren’t optimizing for efficiency.

> Let me ask you - in your job, if someone suggests a change that will improve how things work, would you be against it because the bad way of doing things is party of the business's identity? It's nonsense, really isn't it.

So, you're already rewriting everything in Rust, then?


Allowing people to eat more than one or two types of meals is also a massive inconvenience. Society would be much more efficient if we all just got our weekly bag of rice and beans.

Painting and other art forms are also huge inconveniences. They are neither edible nor do they provide shelter.

Definitely don’t forget letting people speak their native languages. The incredible economic friction of language barriers has cost the world billions in duplicated efforts.


> would you be against it because the bad way of doing things is party of the business's identity? It's nonsense, really isn't it.

Complete false equivalence. If you think the identity of a business is anything like as important as someone's home, then I'm just sorry for you.


What inconvenience? There are fewer than 2000 people there. Each address can easily be associated with latitude and longitude in a public database. Such databases already exist for ordinary street addresses, for instance https://www.latlong.net/ which works perfectly for own address in Norway. Why would this not work in Hilgermissen?

Anyway maps.google.com works even better so how does any courier company get to the wrong address?


Ok so the postal workers deliver to a designated central point an you pick up your own mail :-)

I am surprised that Germany of all places doesn't have a centrally planned and enforced addressing scheme like the UK post code or the USA's zip code


For -ahem- historical reasons, Germans have a healthy skepticism to anything which may let the powers that be know more about them than what is strictly necessary; it is not my impression that this is the case here, my point is merely that Germans on average are quite privacy-conscious.

They do have postcodes in Germany, though - the gist of this story was that the lack of street names (identifying houses only by their (not quite sequential) numbers.)) caused delivery services much grief.

Basically not enough granularity on the micro level - they know what town you're in, but finding the house you reside in is cumbersome.


>They do have postcodes in Germany, though - the gist of this story was that the lack of street names (identifying houses only by their (not quite sequential) numbers.)) caused delivery services much grief.

A see a simple solution to the problem: pass a national law that towns can be like this if they want, but if they don't adopt the national addressing system, then delivery services (and companies sending goods that way) are completely free of all liability if anyone in that town complains about packages being late, mis-delivered, or not delivered at all. People in the town won't be able to pursue the delivery company or the merchant for the money they spent on the package which never arrived.

With a law like that in place, the town would probably very quickly change their addressing system.


> finding the house you reside in is cumbersome

But it isn't. Google Maps, and others, find the locations perfectly. There in fact is not a problem that needs to be solved.


It is not cumbersome to deliver a package to one house in this town.

It is cumbersome to deliver packages to, say, a hundred houses in this town, where every day, the houses with a scheduled delivery varies.

If you are to load up your delivery vehicle efficiently, you need to have some idea as to the order the packages will be delivered in; this is made harder by the lackadaisical numbering and lack of street names.


> you need to have some idea as to the order the packages will be delivered in

No you don't. That's what route planning software is for. Trying to do it in your head is a waste of your time when a computer can do it faster and more reliably.


...but presumably route planning software is based around the concept of street names, which is a pretty universal concept? (I have no doubt a computer can solve this issue - but neither do I think logistics companies would create a fuss over this unless it was an actual issue; an educated guess is that this town's approach does not integrate well with whatever route planning software they are using (while not being a large enough market to defend the investment which would need to be made in order to serve it more efficiently - hence, we're stuck at cumbersome...))

The maps know where every house is so the software simply has to route to those coordinates. We know it works because Google Maps and all the others I tried do it already. So all that is needed is to get the coordinates for each address, ask the api for the time between each possible pair of locations plug it in to your favourite travelling salesman solver and start driving. This is hardly a difficult task, especially as pretty much all of it is already done by various companies already.

For instance https://www.mapquest.com/routeplanner. It only covers the US but it doesn't look as though there would be any difficulty extending it to other countries.

And this one, https://planner.myrouteonline.com/route-planner/1752/#, works perfectly in Hilgermissen and has no trouble finding the houses and finding the optimal route.


> For -ahem- historical reasons, Germans have a healthy skepticism to anything which may let the powers that be know more about them than what is strictly necessary

If this was true, Germany surely had abandoned the compulsory resident register (Meldepflicht) introduced by the Nazis.


The data from these local registers is not collected and merged, so there is still enough decentralization in the system to make abuse very difficult.


Or have identity cards and with your religion on to boot!

> Ok so the postal workers deliver to a designated central point an you pick up your own mail :-)

Welcome to a lot of reservations in the US. It always ticked me off when some company would say "we don't ship to PO Boxes" since that's all we had. Even all the 911 stuff didn't change that. I ended up getting folks to ship UPS and putting "House 311 Behind the School" for my address. Probably why UPS is used so much and FedEx seams to hate us to the point of rudeness.


It isn't even an inconvenience to the couriers. Google Maps, etc., find the houses by number perfectly well.

couriers should stop delivering to the town or ask for more money for deliveries. I don't see the problem.

I think this is the only reasonable decision. My sons 43213 and 44123 and my wife 38431 agree with me.


Likely the numbering model follows a sequential scheme which reveals a bit too much (at least some side channels)

The numbers should be wholly random. --- Also the genders (sons - male) and wife (female) reveals too much, should be replaced with gender neutral forms.


Reminds me of the NIST codes in the cryptonomicon

I am trying to like that book. Does it pick up at all after the first 50-100 pages?

Its a slow growing book - depends if you like detailed world building. Though technically he gets the lay out of BP wrong - the bit where they vist the bommbe house

I do, at least in the likes of Dune.

German inefficiency at its best. :)

It's the problem with democracy, it tends to create results some people don't like.

There are some things where we don't allow democracy to work at a local level.

There are things we do allow but I think we should discuss ways too eliminate: like nimbyism and efforts to increase property taxes without fear of a population drain to unscrupulous counties/states that are waiting to pounce. I'd argue that for a resident in California, it doesn't matter when jobs go away whether it goes to China PR or to Texas if the jobs go away because of a more LAX, fast and loose labor laws or just simply corporate welfare. I don't have any answers but there are some things beyond the basics of life and liberty of individuals who belong to a protected class. Is there a good solution to this "race to the bottom" (other than a Central world government)?


so german.

Where I was brought up had no street number, just a road and house name. Was an utter pain having to actually describe to everyone who visited drive along the road (not included in street directories), look for a side road, then look for a particular speed sign, then turn, oh and if you can't find it call from the phone box in the town which had a map of the roads (but not houses) nearby (20% did). Eventually got RAPID numbers which gave a grid co-ordinate and then finally rural addressing where you got a house number which started at 1/2 at one end and then incremented by 2 every 20m along the road. So much easier.

However I've seen the Japanese system and that works only because there are maps everywhere telling where to find each house and every house has a nice white lettering on blue plaque in exactly the same style so it is easy to spot. Also people stick their names on letterbox/house so you can be sure. The only problem is you have to get on the right street and in roughly the right area which makes for complicated directions for the taxi driver, but for post/courier deliveries it works well. [Mind you a Japanese GPS will do addresses, but for some reason a party of foreigners seemed to cause it not to work with some drivers, no problem I brought my own GPS and can give directions!]

The article doesn't say if there are maps available, if they are then fine, otherwise I'd vote for street names. Not sure if having a random house number (German) instead of a road name (my former situation) is better or worse.


OT: Since we like discussing Google Maps vs OSM quality here at HN. This is just one example where OSM has the lead:

- https://www.google.de/maps/@52.844593,9.1547416,15.75z

- https://maps.openrouteservice.org/directions?n1=52.843476&n2...


Osm has the lead in the majority of places. I used a random number generator to generate coordinates (latitude between the southernmost and northernmost point of land, I forgot the numbers, and -180 to +180 degrees longitude). Any point that was not water, I'd zoom in to about street level by default and zoom out until I saw a reasonable amount of features on at least one of the maps. The maps I compared were Google, OpenStreetMap, Bing, and TomTom. I wanted to also compare Apple but their maps are only for customers it seems. To determine the best map, I considered mainly which had more features (a more complete map), but between near-ties I also looked at satellite imagery to determine which one was more accurate.

OpenStreetMap beats everything else.

I still want to write this up in a blog post with visual examples, and I'd like to redo the test (for comparison) with random coordinates biased by population centres (does anyone know how to generate those?). I'm curious if that paints a different picture. But purely geographically, this de facto standard isn't so great if you look beyond the western world.


Interesting! Yep, a lot depends on the distribution you use for your random sample.

But also, any attempt at objective weighting like this is going to pick a single winner, but it's not the case that everyone is better off using the same mapping service, since you mostly want coverage for the places you travel to and that's different for each person.


Right now, that's true: some maps are better in certain regions than others. But I don't think it's a given that "it's not the case that everyone is better off using the same mapping service". Being an OpenStreetMap contributor, I like to think that one day, the map might be complete enough that this is the common basis for all mapping services. The data is free to use, as long as you're willing to give credit and release any additions under the same license.

Hosting is not free and even open source apps like OsmAnd limit the number of downloads unless you buy the app. It's also still a big challenge to present the data nicely and make a nice user interface. There will definitely still be opportunities for competition and profit, even if all data is open. So hopefully, one day, everyone is better off by having a single, complete map available to use for anyone :)


This is actually quite common around here. In Austria you have e.g. Schlitters and Alpbach that also don't have street names and they are even bigger than Hilgermissen.

Related: use caution with Google Maps in Austria. Here e.g. it suggests you to go through a steep mountain road (in current weather it's likely not passable without chains) instead of taking a normal street: http://imgur.com/a/If4aMI7


I remember walking through a nearby village for two hours in the middle of the night with a friend... we were invited to a party, but only knew the house number.

Nowadays, everyone has maps on their phone so it's really not an issue any more.


This has always been the norm for rural places in E.g. Sweden, sometimes even with hundreds of houses.

I worked extra some weekends doing delivery, and in the age before good maps and smartphones, this usually meant driving to one random house in the village and asking where house 123 is.


Ditto in Finland, although we base the numbers on the distance to the start of the road, so number 123 would be 1230 metres from the start of the road. Makes it quite easy to find the way, actually.

Interesting. How does that work for a place with more than 1 road? Say 20 houses in a village right at the intersection between 2 roads (say houses randomly placed in a circle within 1km from the intersection)?

I'm sure there are other variants as well, but buildings on private roads with no name usually get an address based on where they intersect with the closest public road.

If the road is public, it should have at least a number. Almost all public roads have names, and many have two (as some parts of the country are bi-lingual -- it makes address validation quite interesting).


...which makes it pretty obvious what a bad system this is vs. the more common sequential numbering scheme.

My claim to fame throughout my teens were that my house number was 1337.

This is entirely common in Japan, few streets have names. Although, almost the whole country does it, so it’s somewhat more accepted than one town.

https://en.wikipedia.org/wiki/Japanese_addressing_system


Kyoto is the only city I've been to where I found street names..

And yes, it's a bit tricky to navigate. Even for locals, navigation tools or not. It's usually a lot of talking on the phone while (sometimes frantically) driving around, and trying to figure out where the nearest landmark is.

As for myself I usually go around by bicycle, and I'm not online when out of wi-fi range. I can use an offline map, and if I know where to go I can get there. But it's infinitely easier to navigate in my home country because of street names+numbers (particularly when there's some system to the street names - a flower theme in this area, a maritime theme in another area). Because I don't need a satnav, or a map, or anything. I just go there. My wife (Japanese) is still a bit surprised about that. Oh, and the postal service now demands street names everywhere, so the last decade or so has filled in street names wherever there weren't any before.


It's terrible. You need to know the phone number of a place to enter its address into your car navigation.

There is an address system though. It's not just city + house number, like in the city in article.

I remember reading a about a snobby put down, I think it was between officers in the British Army, where someone was made fun of "as his house has a number"...

Ah like the famous Alan Clark quote on Heseltine "his father brought his own furniture" which is of course topped by an even posher tory by "that's a bit rich as his father brought his own castle"

Interesting, I've seen the same thing in Iqaluit, Nunavut, in the mid/far north of Canada. Each of the roads have names but the house numbers aren't re-used on each street. There's only one building 626 in Iqaluit. People don't really know/use the street names, preferring to refer to just the number of the house. Even the taxi drivers don't seem to like using them.

Why not go with GPS coordinates? :)

Anyway, I suspect this is more a vote against change, than a vote against the concept of street names.


My current address has no street name, just a house name and village name. Technically, there is a street name, but no-one uses it because it's the main street through the village.

I really don't see it as a problem: the local postman knows every house, and these days with GPS the postal code gets delivery drivers within a few yards without a problem. In fact, a post code and flat/house number or name is technically enough to identify any property.


Reminds of Bhopal, a city in India. People there are very fond of numbers. You see people getting on a bus and buying a ticket for "aath number" (number 8) or even "saade paanch" (five and half). You are right, thay are names of places. Not feeling well? get youself checked at "baara sau pachas" (twelve hundred and fifty). Yes, that is the name of a public hospital. It took me a while to get used it.

> Those advocating the introduction of street names pointed to postal and courier companies and holiday accommodation guests, who often struggle to find their destinations because consecutive house numbers are not always close to each other.

No, they struggle because their employers are treating them like shit. It's not like they never struggle unless there's no street names and consecutive house numbers. I often take packages for my neighbours when they're not home and I am, and when the couriers express their heartfelt gratitude I kinda feel bad, the whole situation is fucked up. And where I live, people have no issues finding a house, but every second counts, and adds up to make the difference between a shitty day at a shitty job, and a hellish day at a hellish job. It stinks.

So I would vote against street names, too, just on the basis of not giving an inch to pretending street names are the issue. Pay people by the hour for doing their job correctly, and then let it take as long as it takes, then we can talk.


Openstreetmap has two street names though (but not for the "main street").

Another nightmare for programmers.

It’s fine, it’s a nightmare with or without street names.

You already shouldn’t assume streets to have names and also accept arbitrary strings.

If you really need something reliable (i.e. it needs to be accepted by an external authority or you will actually deliver goods) you can ask the user for a validated adress from an official source.


The residents of Hilgermissen are quite likely well adapted to entering 'Hilgermissen' as the street name as well as the place in any form that forces it.


Not much worse than zip codes with letters.

If you're trying to store numbers that might start with a zero using an integer type, you're doing it wrong anyway. A rule of thumb is to ask yourself the question: what does the result of adding two such values mean?

Does that mean that e.g. numeric user ID should also me strings since adding them is meaningless?

Nothing difficult about zip codes as long as you treat them as strings :)

Canada says hi.

More

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: