-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Christoph and all:
Sorry for replying so late, but it's been impossible to focus on this import for the around 10 last days. I've been thinking seriously on your idea of reducing the data set to be imported, deleting beforehand those MSD&JSI segments that are already mapped in OSM, except those that are clearly a big improvement on the OSM data. I commented it with Andrew Inglis from JSI, and they like it. For the JSI data sample file, this has meant reducing the original file from 5,965 segments to 2,874 (therefore a reduction of 51.82%) and from a total length of 12,319 km down to 6,864 km (a reduction of 44,28%). This will surely make the import much faster. Now the estimated time is about 100-150 hours, instead of the 475-600 for the initial proposal. Also, the workflow will be more simple, adding to it that the simplification of ways will be done beforehand, and not by the users (I did it already with JOSM, this time with no problems after allocating more memory). I have incorporated to the import wiki the items discussed in this thread [1], and I only have to modify the workflow wiki [2] accordingly (it has to obviously follow the same criteria of the import wiki). This will take me one or two days. Among the modifications in the import wiki: * I added an specific warning about not changing any ref tag from any existing OSM road (this comes on top of a general warning on respecting existing tags, unless they are clearly wrong), * I added a link to the original sample file and the resulting (transformed) file, * Only surface=unpaved/paved values will be assigned to new MSD&JSI roads, * Data transformation process is more clear now (in 5 separated points), * Experience to join the import is a bit less demanding now, * Added the warning on checking river crossings, and * Specified what to do in case of lack of good imagery. After correcting the workflow wiki, and if no more concerns are raised, we would start the import with a first Tasking Manager project in a week or so. Thank you again for your time, an effort that will make this import of the best quality we can. Cheers, Rafael. [1] https://wiki.openstreetmap.org/w/index.php?title=WikiProject_Tanzania%2FTanzania_MSD_%26_JSI_road_import&diff=1212649&oldid=1206542 [2] http://wiki.openstreetmap.org/wiki/WikiProject_Tanzania/Tanzania_MSD_%26_JSI_road_import_workflow On 22/08/15 13:55, Christoph Hormann wrote: > On Saturday 22 August 2015, Rafael Avila Coya wrote: >> >> You can't imagine how many times and for how long I have thought >> about this idea of removing by software and beforehand the >> MSD&JSI road segments that are already present in OSM. The >> advantage would be very clear: we would save a lot of time and >> effort for the import users. But at the same time, I also spent >> quite a few hours going around the data, and my conclusion is >> that a good number of MSD&JSI are an improvement of their OSM >> counterparts. And I think it would be a pity to miss this >> oportunity of improving many of the existing OSM roads >> geometries. > > I understand that - my reasoning however is that if a road is > significantly worse in OSM as it is it won't be close to the better > data along its whole length so it would not be filtered out > anyway. Based on the data shown i would estimate even with a very > tight threshold you could remove at least about 30% of the ways, > especially among the shorter ones in more urban context which would > already reduce work a lot. And even if you miss a few cases of > useful improvements this would be more than made up by the mappers' > time freed to be spent on new genuine mapping. > > Alternatively you could also make available the full data as gpx in > addition to the reduced selection specifically suggested and > prepared for import. > >> There are for sure areas where most roads are already mapped in >> OSM, specially around big towns and cities. But this dataset will >> grow the total road distance in OSM from the aprox. 55,000 km >> that we have now inside Tanzania to around 115,000 km, so this >> means that around 52% of the MSD&JSI road segments aren't mapped >> in OSM yet. That's the average. > > Note that's road length, not number of ways. Most of it is smaller > long distance roads which are not in OSM yet. These seem to be the > main value of the data for OSM and would not be affected by any > automatic filtering, neither would roads that are only crudely > mapped in OSM right now. > >> [...] My suggestion for the workflow in these areas would be >> something like: "1) For segments still not present in OSM, we >> import them, but we will avoid adding the surface tag (we could >> instruct the mappers to add a fixme="Please, add surface tag"). >> 2) For segments already mapped in OSM, we won't import them >> (ignore them)." > > Yes, that's a good idea. > - -- Twitter: @ravilacoya <http://twitter.com/ravilacoya> *Humanitarian OpenStreetMap Team (HOT)* <http://www.hotosm.org> -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with undefined - http://www.enigmail.net/ iQIcBAEBAgAGBQJV6HYQAAoJEB3niTly2pPQzSoP/RaIrb3r2LhWIy1Q1VUGy/Bu iTT2yKlpjRRixMlPY0/BvDOadRWl5YaZdjhQNgxRk7xOeHVSANPejGPCu/vDrGjM Nw3JmVP74CeF+plS2o1+luMwR+eqLqI3n2rXH6GrcrC1pnGCJoCgAMnVhJiDJWdi A2BwChBUq8epOtVDCgoZa4UD/ySqgFYVHKKg2sHeuqEgrAfZ+wnuSjg4VZ0jWHvA AZs8YsCbfSunfUt5h3/iZ1jT4co3BN8Jf6HuI3tlm+9dgksc1Yo8ly7Z/1yPhhJd +y6CdZo7sACiWPrYNHFFsmKloWFnQ7vaYPPBQ7POrFeffqN2e+VAi456j8hoV3U9 vRbUTcMK8k/n5PCcfqQvH2VsrmzOkNNep+H7nz8gplrDAAfWfvImdQqKiPvqVIxa E+HaUJHe/MtFKdnCCHj4pGZht2R3B40wBW/VA4OWA5r+Uikf9xI1MA8ipZsERFm5 M121NcRupUNjy9zpBPSbobvZvfAyqbBrrPZvs3gn+dD/vrAcr2KOTqsEI6uRyLiM gjCQbLV3M/KwV20Xprm73DTL0jKTlSrGCVI14i+M7nHQA4Oh6ict+GFCSlysBzUX mJnAXkK6iajVJMeZE7VM0WgYy7gkwW7e/gB5ulIe13nAbOjIG/kiZOV1NnvyzfWo tJbyjgyQLA43fdJDi6GA =VYqU -----END PGP SIGNATURE----- _______________________________________________ Imports mailing list [email protected] https://lists.openstreetmap.org/listinfo/imports
