Is the baseline definition then just a convex hull of the geometry? That's not hard to calculate.
Stephen Mather Geographic Information Systems (GIS) Manager (216) 635-3243 [email protected] clevelandmetroparks.com -----Original Message----- From: Michael Krämer [mailto:[email protected]] Sent: Tuesday, May 15, 2012 3:54 PM To: [email protected] Subject: Re: [Imports] Boundary of Tonga Hi, as my first reply didn't go to the list but directly to Frederik here a second try: Am 15.05.2012 16:50, schrieb Frederik Ramm: > I noticed that of all 193 UN member states, there's only one for which > we don't have international borders in our database (instead, a node > tagged place=country). That country is Tonga in the South Pacific. Yes, there's quite a bit left to map in the South Pacific. > I have taken our existing Tonga coastline and computed a 12nm buffer > around that (after having read on the CIA world factbook that they have > a 12nm territorial waters claim), and prepared the following file with a > boundary relation that we could use to replace the country node: Well, this is probably the best we can get for the time being. In a perfect world we wouldn't use the coastline but the baseline. The baseline definiton ([1], 5+6) closes bays by straight lines which makes a difference given the shape of the islands. But manually deriving the baseline using Bing imagery would be a rather complex task. So unless there's a decent way of doing this by some fancy software package I think there's no way of getting the basline. So I think using the coastline is fine. But I would suggest to add a note to the data somehow along the lines of "only approximate, base on coastline". > http://www.remote.org/frederik/tmp/proposed_tonga_boundary.osm After looking at the data I think you missed some islands that have quite an impact to the border. I did some checks with Wikipedia (also adding a few names) and found the follwing ones (with Way ID) ?Ata (22383999) Toku (22390574) Fonualei (22390195) So if you don't mind, I would ask you to add these. > I'm not 100% sure if my reasoning is correct; there is the case of the > Minerva reef which lies about 200km to the SW of Tonga and I read that > Tonga lays claim to that - but, as far as I understand, only EEC claim > and not territorial, so I guess it is correct to not include it in the > country boundary. I agree. From what I understood the reef is submerged so I don't think it's possible to establish a territorial claim. But I'm also not completely sure about this. > I'm not normally in favour of doing imports but seeing that Tonga is > really the "odd one out" in the list of countries, maybe it is worth the > effort to fix that. I think that an import makes perfectly sense for boundaries, especially maritime borders. They hardly can be mapped on the ground. Michael [1] http://www.un.org/depts/los/LEGISLATIONANDTREATIES/PDFFILES/TON_1989_Act.pdf _______________________________________________ Imports mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports _______________________________________________ Imports mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports
