Hacker News Reader

Comment by jjwiseman | Hacker News Reader

Because I don't think a lot of people know about it, I'll mention pelias here. If you have 16 GB RAM and 600 GB of disk, you can have a free, open source, pre-populated geo database that covers the entire planet.

https://github.com/pelias/pelias

Pelias does geocoding (turns addresses, POIs, and administrative regions into coordinates), reverse geocoding (given coordinates, finds nearby POIs, addresses, neighborhood, city, county, state, country), full-text search, and autocompletion. You can also import CSV to use your own custom data.

I run it on an extra PC, where it powers about 20 twitter bots that turn lat/lon coordinates of aircraft into descriptions like "flying over Silver Lake, Los Angeles, 0.5 miles from Circus Liquor."

There are docker configs for the whole-planet DB as well as various countries or metro areas if you don't need the whole Earth. https://github.com/pelias/docker/tree/master/projects

View on HN

Thanks for the mention John.

Co-maintainer of Pelias here so I'll chime in with some more background info.

Pelias came out of Mapzen which (sadly) shut down about 3 years ago, but we had a great team of people working on it for 4 years before that, and the project has continued to grow since then. We have contributions from my company, Geocode Earth, which I plugged in another thread so I won't repeat, other contributors like Jawg(https://www.jawg.io/), and of course individuals and teams from many other places.

The core architecture uses Elasticsearch, Node.js, SQLite and some Go to build a fully featured, modular geocoder.

Geocoding is both a fascinating and incredibly frustrating challenge so we are always looking for more people reporting bugs and contributing fixes or other improvements. We truly believe a quality geocoder can't be built outside the large bankrolls of Google, etc without growing a strong open-source community around it.

Pelias is great, I ran it + open route service + a custom router all on one server for 30€ a month + a few other non maps related services. It made for a surprisingly coherent maps experience.

Can it be run for a subset of locations on a subset of those memory requirements? For example, just North America?

Yes, for example the north-america docker project says "As a fairly large build, this will require significant resources to complete quickly. It can still run on a personal computer with 8GB+ RAM, but will take a while. It will require 300GB or so of disk space."

https://github.com/pelias/docker/tree/master/projects/north-...

(Building the db typically takes more resources than just running the services using the db.)

It's buy, not build culture where I've been working for, we decomissioned a lot of self-host services for transitioning to managed services.

What are your impressions so far?

Another good option for self hosted geolocation is using Postgresql+PostGIS.

That's not really comparable. It's kind of like saying you can have a self-hosted search engine by using Elasticsearch. The devil is in the data and details.

Reply on news.ycombinator.com