Maybe I'm missing something but why does "Pearl" mean 8480 and "Magical" -6129? If that's a hash, I really dislike this, because A: humans can't do hashing in their heads, and B: get one letter wrong and suddenly you don't have the right address. This has also been my big opposition to what3word.
Addresses need to be resilient. If I make a tiny mistake in my address, there should be enough redundant data to help fix the problem. Postal Codes are great for this, because if you get your code wrong, well, it'll probably be one nearby and the delivery service can figure it out. But with hashes? One letter off and you're doomed.
From having worked on addressing in Nairobi, Kenya for the past two years, I can tell you the precision problem is worse than you think actually. You're making the assumption
* someone knows how to use a map to a very high degree of precision (not true often)
* GPS is accurate enough, and this is a double whammy - on initial assignment of the GPS point (i.e. it's the GPS of their door/gate, not of their living room couch, and it was collected with enough precision) and on usage (i.e. the GPS antenna on a $30 smartphone)
* plenty of other issues
There are a couple startups trying to solve this beyond what3words mentioned in other comments - they'd suffer from many of the same issues I mention above. Some encourage people to put a placard outside their door/gate with the code to make last-mile arrival easier.
But a more successful approach from research I've done is what OkHi (http://okhi.com) is building - use a photo attached to GPS. From n>1000 tests, we found that the photo drastically decreases time over the alternative, even reducing average time to find the location by over 50% in the last 500m for some addresses.
(Full disclosure - I was worked on the addressing problem for the two years in Nairobi at OkHi, so I obviously believe in their solution or wouldn't have dedicated that much time there :)
I am really interested in what you've learned because you obviously know a lot about a problem that's very physical and concreate. What is your approach or what will it be? However, you website is horrible at communicating any of that. I learned 4 billion people don't have addresses. OK great. Now what?
"We want to physically connect the world
with an address system that works.
OkHi is unlocking commerce through enhanced logistics, saving lives through improved emergency services and growing access to finance through better personal identification."
The OkHi approach is use your phone number as your universal key to share your address (with obvious privacy controls in place to control who can access it). The address itself is a GPS point + a photo - you can see an example at http://okhi.co/hq (best viewed on mobile - this is a vanity url, but you could very much imagine it eventually being something like okhi.co/+254/700111222).
The OkHi business model (similar to other companies in the space) is to allow consumers to share these addresses peer-to-peer free and charge per-use businesses who work in logistics or need address data (e.g. food delivery, ecommerce, emergency services, banks, postal services). They collect the base address data that users can choose where they live from through our own on-the-ground data collection or smartly crowdsourced data (surprisingly scalable).
Lots of promising progress, though obviously lots of work left to make it global. While it is easier to build a business controling the full stack, over time, the base addressing data could become open-sourced/ user address data stored in a public open db e.g. blockchain. I think the issues folks bring up of open-source benefits are very much in mind as this sort of system is built.
I feel like this a communication and NLP problem, at least in regards to an address being directions to a given location. Like with a picture, an address such as "Next to Sal's general store" would be enough for most mailmen to find the location 100% of the time. Given it does not stand the test of time, but to my understanding this was the system in Ireland for a long time in small disparate communities. In that particular case the instruction would be something like "$COUNTY, $GRANDMOTHERS_NAME's house" in order to reach her grandson.
I guess it depends on usecase. An Amazon drone can't stop and ask for directions. Interesting problem space though.
You'd think "next to Sal's general store" would work, especially given how prevalent that approach is. In fact, through our research, even for folks working the same delivery area day after day, they consistently fail with these sorts of landmarks. I used to say I lived "next to the vegetable stand on XYZ Rd, across from ABC Apartments" - very visible, clear landmarks which turns out most drivers could never use successfully because they usually called me being lost. You could try NLP on it to then lookup in a db like OkHi's, but these text-based directions definitely proved to be suboptimal when consumed by a human drivers such that companies were actively seeking a better way for their deliveries (and, importantly, were even willing to pay).
When I get instructions like that, I often end up in trouble. Once I find Sal's general store, I find there is not just one or two, but three or four spots that could be intereted as "next to" it.
And if none of them is obviously the thing I am looking for, the list starts to grow "maybe he meant across the road, down the street...".
It isn't a third world or African problem. The capital city of Canada's newest territory (Iqaluit, capital of Nunavut) didn't have any street names until a few years ago. Buildings today are still identified by numbers alone, with a degree of local knowledge needed to find many locations.
It's a politically incorrect term but are you sure it's unreasonable to call Iqaluit "Third World"? I live in Oakland, California and I definitely think of it as a Third World city -- the utter poverty, homelessness with no end in sight, squalid encampments in the dirt in derelict land, the lack of help for anyone falling through the bottom of society. I wasn't born in the US but I find it odd that one of the criteria for somewhere to be considered third world seems to be "far away, not in USA".
> I live in Oakland, California and I definitely think of it as a Third World city
Oakland is in the top 20 US cities by median household income. If it is "Third World", then essentially the whole world (including most of the United States) is "Third World".
> I wasn't born in the US but I find it odd that one of the criteria for somewhere to be considered third world seems to be "far away, not in USA".
The First/Second/Third World categorization originally referred to the US and its allies (First World), the USSR and its allies (Second World), and non-aligned (in the cold war) countries (Third World). The Third World were largely developing countries, the former included most economically advanced countries (though it also included developing countries), and eventually the terminology shifted and
"First World" became synonymous in most uses with "Developed" and "Third World" with "developing" (and no one uses "Second World" at all...)
But, in any case, the labels apply to countries, and the US is pretty clearly not Third World country by either the original definition or the one that is more commonly used now.
That's an opinion I think? Take Brazil for example; the difference between middle class suburbs in Sao Paulo and a remote settlement in an Amazonian state straddle our concepts of First and Third World. In fact the middle classes in the southern cities pretty much don't ever go to the north precisely because they view it as the Third World.
For many people in American cities, their access to healthcare, education, sanitation is not different, or in fact worse, than it would be in cities more conventionally thought of as third world. America's lack of social safety net is causing it to be accurate to refer to parts of America as Third World.
Well, the original form was strictly for countries; if you want to use it to describe something that is neither on the level of analysis nor on the axis of variation of the original form, its not worth even using the terminology (the common modern form still uses the country unit of analysis, but is an economic rather than geopolitical axis.)
> Take Brazil for example; the difference between middle class suburbs in Sao Paulo and a remote settlement in an Amazonian state straddle our concepts of First and Third World.
Not really. Actually, have sharp geographic and class divides is typical of the developing world; having elite areas is not at all out of line of the traditional understanding of a developing (or "Third World") country.
Though a binary developing/developed or First/Third divide is problematic for other reasons -- Brazil (like Argentina, Mexico, and lots of other places) really ought to be viewed in a middle tier; its nearly as far above, by most meaures, a lot of the places more typical of the "Third World" as it is behind the places more typical of the "First World". (Perhaps we ought to resurrect "Second World" for this.)
> For many people in American cities, their access to healthcare, education, sanitation is not different, or in fact worse, than it would be in cities more conventionally thought of as third world.
To the extent one wants to discuss claims that that is the case, it may make sense to describe the particular American cities being like developing (or "Third") world cities with regard to the particular concerns, but that's different than describing them as part of the "Third World" which involves more than just those issues.
Yes, on balance I think you win this debate. You are perhaps a little too strongly influenced by the Cold War origins of this term, when for a couple of decades now the term has been almost exclusively used to describe places with poorly developed infrastructure and institutions. Phrases like "third world conditions" are common and as we agree, a feature of modern developing nations is high spatial variance in such things. So, since the Cold War is history, and since semantics evolve, I think a reasonable case can be made for taking the modern definition of "the Third World" to be the union of spatial areas which the speaker deems to exhibit third world characteristics. In any case, I'm shortly going to go out in the northern section of West Oakland, and there I will encounter scenes that I literally will not be able to distinguish from bad areas in Addis Ababa.
Iqaluit is Canada. It isn't third world, by either the original definition or the new. There is certainly poverty and deprivation there as in much of the north, but it isn't the same as a true "third world" nation. Each and every person in Iqaluit is free to move within Canada however they wish, vote for who they wish and exercise all the rights and freedoms of any other Canadian, including education and health care.
> Each and every person in Iqaluit is free to move within Canada however they wish, vote for who they wish and exercise all the rights and freedoms of any other Canadian
The absence of those abilities is not a defining character of the Third World by any definition I'm familiar with.
Just moved to Nairobi (actually, Tigoni on the way to Limuru, where I have no postal address). Pretty sure I saw your team at iHub last week when I was hosting the Mozilla Science Lab. I'd love to connect - gonna email but also please feel free to reach out to me using details in profile (same goes for anyone else in the area).
Unfortunately, your website is terrible. There is no information about what it is or how it works without watching a video, and I'm not willing to do that.
This is why he's incorporated the avatar - in the example he provides you wouldn't see the "boot" avatar if you had missed a letter, the boot is part of the address and should be communicated verbally when giving the address verbally.
So to check whether the address is valid I need to use a smartphone app. If I cannot do it online and only after someone gives me the address I will know it is broken, but will not know what is the correct address.
Bingo, error detection is functionally distinct from error recovery. In this context, error detection helps prevent some misdeliveries, but can't help ensure additional correct deliveries in the face of error.
By analogy, ECC memory can typically detect two bit errors but can only correct single bit errors. That's because the functionality used to detect two bit errors in ECC memory is insufficient to recover both bits.
The concept of an avatar doesn't translate across cultures. And the visual hash isn't an "avatar" as typically used on the internet. Furthermore, the images may be incomprehensible in some cultures.
Keeping a minimal edit distance between code words in the corpus would enable error detection and correction without the burden of sending another piece of info.
Addresses need to be resilient. If I make a tiny mistake in my address, there should be enough redundant data to help fix the problem. Postal Codes are great for this, because if you get your code wrong, well, it'll probably be one nearby and the delivery service can figure it out. But with hashes? One letter off and you're doomed.