So, I’ve decided to improve details on state boundaries in Peninsular Malaysia (except the Johor-Singapore boundary). District boundaries for Kedah, Penang, Kelantan and Pahang would be imported as well. Selangor and Melaka would get the PBT boundaries.
Wiki page is up, so the last thing is to be scrutinised further, if any, in the imports mailing list.
It’s almost ready.
@angys: I don’t think I want to import electoral boundary data. Things can change by the next general election (could be 4 or 5 years).
Sunday, March 10 update: That’s Penang and Kedah, completed. Perlis and Perak now with improved boundary. The next move would be some clean up across Kelantan first, then proceed with import.
Thank you very much for your update :). And I had noticed that Kuantan-Kemaman boundary very not accurate as Jabor (which is belongs to Terengganu), located in the Pahang side.
I would suggest you to do a Terengganu District update with Pahang and Kelantan boundary, at lease make it more accurate, thank you. (Same situation at Genting Highlands, actually most of all buildings are in Pahang):rolleyes:
I’ll try to complete this import from up north to down south. After Kelantan, it would be Pahang and then maybe Melaka. Then go up north again (Negeri Sembilan first) before trying my best to deal with Selangor('s local council boundaries)…
Yeah, wish me luck though.
Hopefully there’s a way getting open and free/libre data for (sub)districts of Perak, Terengganu, Johor, Selangor, Sabah and Sarawak.
I admit that “life has got in my way” dedicated for OpenStreetMap in these few recent months. I’ll try to complete bit by bit, but then again, please just remind me to complete this import. Can’t promise any ETAs though.
Ok, I believe that all Malaysian OSMers will looking forward for your update. Life is more important but since OSM is a public project, someone (like us) will need to sacrifice our time to this. Anyway, you can just do this when you are free too…
Sorry to disturb, but I still dont see any progress till February 2020, hope that you can spend your precious time to finish these tasks, your works are appreciated.
Hello, can you complete the import within this month? I do believe that you can spend some hours to finish this since outdoor activities are not allowed. TQ
Hi, I don’t connect to a stable, wired connection. There’s a risk of so many upload conflicts, if I happened to stop halfway during the import… and it’s not going to be resolved neatly due to a bad connection, then I really am going to have a good time.
No promises, but if I’m brave enough, I’ll try giving it another go.
(Isn’t it lovely. Weak connection, and I’ve got to refresh few more times to send this reply.)
I sent an email to them and will update as info becomes available.
How about that Sinar Project? Where’s the detail of its copyright/legality to use in OSM?
Seeing a boundary like this sourced from Sinar Project looks like bullshit questionable boundary data to me.
Yeah, it’s not counted as an open[1] data, unless the uploader shows any proof saying “yes, no problem uploading a copy of that into OpenStreetMap…” Best if there’s a copy of any e-mail exchange between the involved parties.
Strange, considering boundaries are uploaded as closed, unbroken ways rather than split ways connected together to form boundary relations.
Hooray, deeper copyright rabbit hole!
[1]open here means you can view, manipulate and do almost everything including distribution or re-distribution of said data
It is a rough sketch based on satelite image features (with hints provided by the portal) and there is no detailed tracing of the boundary. I don’t see a copyright issue there as the boundary is rough
If I were you, I’d refer to the Data Working Group. It does sound not so pleasant, but what’s the difference between that, and “opening Google Maps from a browser tab and an OSM editor; and grab what’s displayed on the browser tab”?
according to OSM Inspector (https://tools.geofabrik.de/osmi/) it show that some boundaries have error, and I found out that some of them have admin_centre included in the relation, but if the admin_centre is a way, then it shows an error message saying that the role should be inner. so I think that we should use a node or relation as admin_centre, so that it won’t get confused. The standard OpenStreetMap layer will render a border along the way even if there is a way which is not tagged as outer or inner. The same goes for subarea role.
