Method for receiving third party data
Hi all. I am interested in methods for organisations to submit their data to us. I believe Google Maps does this so that store owners can upload a file of their store locations and thought it would be interesting if we did the same.
Am thinking that we need a method where we prove ownership - for example and online upload tool that requires the user to prove ownership of their email address. That way we can be sure that any data that is uploaded has come from a known source.
Any ideas how best to do this? I struggled to find a plugin for Wordpress so may need to host this elsewhere.
RobJN Tue 18 Aug 2020 1:37PM
Yes. Also https://su.openstreetmap.it/ uses a fork of OnOSM. Visulally a few chances but not sure if anything else is different. I think we should consider adding this to the OSM UK website for individual contributions. Want to get in touch with sabas to work out what is different in the forked version?
Jez Nicholson Tue 18 Aug 2020 10:59AM
Hang on....do you mean in bulk? rather than one-by-one? Could be another enhancement.
Andrew Black Tue 18 Aug 2020 6:49PM
Is uploading a spreadsheet etc any different from any other sort of file (or am I misunderstanding)
RobJN Tue 18 Aug 2020 9:11PM
Well the thing is where to upload to? That's the question here. As you say the file type doesn't matter (too much).
Dennis Bauszus Wed 19 Aug 2020 8:06AM
Personally I like the idea to use a public GitHub repository. Anybody could upload files via pull request. Meaning a team member would need to approve. GitHub allows versioning and there would be no incognito uploads. The repo could be hosted with GitHub pages allowing direct access to any files. GitHub issues can be used to comment on data issues. I could even build an interface hosted as GitHub page to allow for easier upload via the GitHub API.
RobJN Wed 19 Aug 2020 1:13PM
So if I understand correctly, they would need to log in to a GitHub user account in order to upload, right? In that case we would get a user name and can reply to them, but how would we know that they work for the company they are uploading data about? For example if it was an upload of Tesco store data, how would we know it is a tesco employee? Is that even important?
Cj Malone Thu 20 Aug 2020 10:27PM
I don't think asking people to sign up and use pull requests is a good idea, it's to much of a burden. Once the data has been submitted to OSMUK it maybe a place to hold and monitor changes, but the collection method should be easier.
Dennis Bauszus Thu 20 Aug 2020 11:19AM
I don't think it is important. Another benefit of using GitHub is that we can assign the appropriate licence to the repository. If you make a commit to the repository you need to agree that the content adheres to the licence. More important than you working for Tesco is that you acknowledge the licence terms and that this can be traced back to you.
Andrew Black Thu 20 Aug 2020 9:01PM
If the data omits a branch, it might mean the place has been closed or it could be a mistake. I would trust it more if it comes from Tesco's rather than a random contributor.
Jez Nicholson · Tue 18 Aug 2020 10:56AM
onosm.org exists to do similar. I believe that it adds a note and not a node.
I would be interested in us enhancing it with 'claim my business' + the FHRS data.