Hacker Newsnew | past | comments | ask | show | jobs | submit | ManAboutCouch's commentslogin

One thing that I make use of but don't see too many services providing is some kind of 'match level' - where the geocoder returns a code indicating how confident it is about the quality of it's result. A result of 1 might mean a building level match, while 100 might be street level etc.

IIRC Google's geocoder does something like this, but it's pretty inaccurate, overstating it's match level consistently.

As others have said, geocoding is very hard to do well, but I commend the efforts being made with Nominatim and komoot/photon.


agree, a simple to understand confidence score is critical.

Also agree nominatim and photon are impressive


FWIW I ran millions of addresses through PostGIS geocoders (using both TIGER as well as PAGC's normalizer functions) and found that MOST addresses geocoded with confidence level 0 or 1. 60% were 0 or 1, the other 40% were spread across ratings 2 - 100.

I don't have a great way to characterize the geographic coverage or data quality of the geocoder, but it is clear that it has a data set which must be maintained to support geocoding into the future. Soon I'll have to start figuring out how long my current data is useful, and how long it will be before the next update from the census bureau.

I'm starting to think that it's crazy for so many businesses to need reliable GIS data and have so few sources to go for it. With the right organizational structure, we could be croudsourcing it it daily.

But I digress.


The OSGB in the UK did something similar last year: http://www.ordnancesurvey.co.uk/innovate/developers/minecraf...

Apparently it was built by an intern, it took him two weeks to create the 22 Billion blocks: http://www.ordnancesurvey.co.uk/about/news/2013/minecraft-ma...


Is it 1:1? It doesn't appear 1:1, given how tiny the A/B-roads are. I am just curious, there are not a lot of screenshots (or information on the scale in fact).


> Each blocks represents a ground area of 50 square metres. The raw height data is stored in metres and must be scaled down to fit within the 256 block height limit in Minecraft. A maximum height of 2 500 metres was chosen, which means Ben Nevis, appears just over 128 blocks high. Although this exaggerates the real-world height, it preserves low-lying coastal features such as Bournemouth's cliffs, adding interest to the landscape.

It seems to be a 1:50 scale.


Fortunately for the makers of the Danish version, the highest point in Denmark is a hill 171 meters high, so this particular problem didn't arise.


> The raw height data is stored in metres and must be scaled down to fit within the 256 block height limit in Minecraft.

Well, I'm a cheapskate I play Minetest, not Minecraft. I'm fairly certain that Minetest has a larger height limit, but it can freeze like a bastard. The other day I went for a dig, and I used the compass to work out my position. I was over 500 blocks under the surface, I kept running into lava flows(!).


I'm now curious enough about Minetest after viewing its website that I might want to contribute. I love Minecraft and have played it since alpha, but it's always bothered me that it's built in Java - and that it runs horribly slow once Feed the Beast comes into play.


Not that Java doesn't have overhead, but quite a bit of the slow down for minecraft servers is caused by bad algorithms (N^2 type stuff) and data structures.


> I love Minecraft and have played it since alpha, but it's always bothered me that it's built in Java - and that it runs horribly slow once Feed the Beast comes into play.

I have never played Minecraft, so I can't comment, but Minetest does slow, it has a lot on its plate.

I just signed up with flickr, look for user "mrtucker257". I've uploaded some photos of my upmarket real estate.


Here is a link to NAFV_P's album: https://www.flickr.com/photos/123873955@N07/


Thanks for that, it was my first time on flickr.


This problem is what finally pushed me away from the game once and for all after a couple years. Minecraft itself is pretty heavy and slow given enough player load (sometimes not even that.. forest fires, lava flows and such can also bog everything down) - mods exacerbate the problem to the point of unplayability.

Which is really too bad, because things like Feed The Beast add an insane level of extra functionality and downright awesome things to the world.

Somehow I think it would be a different game if not for Java overhead. I've spent a truly ludicrous amount of time tweaking heap sizes, obscure garbage collection flags and such trying to keep the blasted thing from consuming all of the memory and crashing (and therefore corrupting the state of the world) on a relatively low-popularity server.


The problem with the iframe approach is that Google's geocoding service TOS explicitly states that the output of the geocode operation can only be displayed on a Google map.

Google's geocoding service is underpinned by a lot of third party datasets which would have to be (very expensively) relicensed if Google started passing the results out to maps under the control of other organisations.


Not quite retail, but Amazon's main Data Centre in Dublin is the former main warehouse used by one of Ireland's largest retailers. It's gone from being full of beans and nappies to servers full of kitten gifs.

http://www.datacenterknowledge.com/archives/2011/02/09/amazo...


The 'spat' broke out on the OGC Mailing list, here's the May Archive: http://lists.osgeo.org/pipermail/discuss/2013-May/ There are about 100 messages about it on there.

There's also this Open Letter: http://wiki.osgeo.org/wiki/Geoservices_REST_API#Open_Letter_...

And James Fee's view on the whole thing: http://spatiallyadjusted.com/2013/05/09/esri-and-an-ogc-stan...


Have you seen CycleStreets: http://www.cyclestreets.net/


Yep. It's great. This one will aim at a different market, really.


Interesting. What other market is there though? They do different confidences of cyclists and leisure routes already?


Their MapBox.js library is a plugin for the increasingly popular Leaflet.js mapping library: http://mapbox.com/mapbox.js/api/v1.0.2/

As an aside, MapBox (via funding from the Knight Foundation) is also behind the shiny new iD Editor for OpenStreetMap: http://ideditor.com/


I believe that other transport modes (public transport, on foot) are in the pipeline, but the amount of additional compute power they need is significant

Also Public Transport route and connectivity information is frequently a bit patchy in OSM, but this is steadily improving.


The creation of so called 'Trap Streets' is well known - http://wiki.openstreetmap.org/wiki/Copyright_Easter_Eggs

However putting a non-existent island on a Marine chart used for navigation is a few steps above that. It looks like it's the result of a mistake, a mistake that was copied far and wide.


Yeah, that's the ticket... guess I wasn't thinking about the requirements for using said charts for effective navigation. I was just wondering if this would precipitate infringement claims once the point of origin is identified. As in it's an accidental "Trap Island" but it's caught some rubes nonetheless...


Except that it never appeared on the official australian and french maps. If you use Google Maps instead of an official navigation chart for navigation, you probably don't deserve better...


I'll add the obligatory tip of the hat to www.openstreetmap.org - it's community basis and the ease of use of it's editing tools make this kind of issue very easy for users to fix.

And as a bonus, any edits you make to the data aren't owned and copyrighted by a large corporate, but are made available to everyone via a Creative Commons Licence (soon to be ODbL)


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

Search: