I'm not in a position to understand the relevance or quality of the data, but I do have some concerns about your method of confluence based on the proposed import page:
> Add tag noaa:lnam derived from the S-57 LNAM field. This tag is the unique > global identifier of this feature. (and similar tags) It's been my experience in working with various imports that this method has problems. Firstly, users modify tags in ways that don't seem rational. They remove tags, they edit values, they edit keys. Therefore you can't rely on those tags to exist, or to stay in tact. And because your process relies on these tags, additional data made by users themselves won't be discovered by your script, which means that it will make duplicates, and that's not good. Touching on that conflation further, are you saying in this proposal that there is no data of this type other than by NOAA? Data that's only available by a single source always concerns me, as it means that there's no way for a regular user to be able to contribute. If that is the case, then I think a much better way to handle this would be as a data mixin, combining OSM data with NOAA data, rather than to import the data directly in OSM. I'm also wondering if this import would represent data that changes frequently (more than once every six months). If so, then I strongly suggest that this not be imported into OSM's core, but used in OpenSeaMap (or something similar) alongside OSM, much like TopOSM uses topological data alongside OSM. If you need help with this, we can help you. If the data can be contributed by other sources than NOAA, then your process needs to have a way to conflate that data with yours, and not create duplicates. Your argument that there's no current data makes sense for now, but I think it should be addressed, especially as my understanding from your page is that you'll be uploading on a regular basis, so it may happen in the future. Thirdly, about those tags, in my experience, while everyone things that they're useful, it actually seems to dissuade editing. It's a shame, and there's no "logical reason", but lots of users suffer from the "It's official data, we shouldn't touch it" syndrome, and that's not good for the project. - Serge _______________________________________________ Imports mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports
