Hello!
I would like to know if I have a business/POI that doesn’t have a website but a Facebook or Instagram page, can I use them as a source to add/update that POI?
Thank you!
Yes, that’s fine!
Have seen the Facebook address listed as website, but the more approved version is Key:contact:facebook - OpenStreetMap Wiki.
1 Like
Yeah I’ve done that. Be careful that the data is up to date. You can tell by looking at how recent the posts are. Just do this manually, don’t write a bot which adds loads 
1 Like
Individual facts aren’t eligible for copyright, which is why copying one set of contact info and opening hours from an independent website is not a problem. However, databases of facts do enjoy legal protection (OSM’s own license partly relies on this), and I find it at least plausible that this protection applies to Facebook’s collection of contact information and opening hours for businesses. You may personally only copy an insubstantial amount of facts from that database, but if thousands of mappers use Facebook as a source, they will quickly end up copying a substantial chunk of Facebook’s data.
Because OSM is attempting to be “whiter than white” in terms of copyright, and we try to only use data sources which are 100% guaranteed to be totally legal, I would therefore not consider Facebook an acceptable source.
I routinely consult a shop’s Facebook page, if for no other reason than to add that facebook or contact:facebook tag, so that the end user can see the information for themselves.
These days, as Google and other search engines become less reliable, the Facebook listing can sometimes be the only way to discover the shop’s homepage, other than maybe visiting the shop in person. However, beware that the homepage links on Facebook pages frequently conflate unrelated shops of the same name, so you have to double-check that the site’s details make sense.
For a shop that has at some point maintained an active presence on Facebook or Instagram, a sudden and prolonged interruption in the page’s feed is a good signal that the shop has closed, changed its name, or otherwise undergone a material change. If you see this, you should try to corroborate the closure with other sources, just to make sure it isn’t a case of the owner going on a social media diet.
Otherwise, I don’t have any qualms about using Facebook to verify the continued existence of a place that OSM already knows about. If copyright law is unable to protect the copying of facts, it certainly cannot protect the inference of facts that were not there to copy in the first place.
It’s analogous to checking a shop’s independently hosted website to see if it’s still online:
The difference is that we probably wouldn’t be allowed to scrape Facebook to find all the pages that haven’t been updated in a long time.
Isn’t Facebook making much of this information available under the ODbL as part of the Overture Maps dataset? If so, there would still be a difference between copying the data from Overture versus Facebook directly, but at least there would be a more formal method for bringing in the data. This doesn’t address Overture’s quality issues at all, but if OSM already has a particular POI, then we already have more confidence in the POI’s existence than Overture does.