To me, "imports" are any bulk transfer of data from an existing data source into OSM. They can be big (10's of 1000's of nodes) or small (10-40 nodes); they can be mechanized, curated (reviewed by eye and hand), or otherwise. I'm not sure a different name will help other than to create a faux taxonomy. If it looks like a duck....
* How will this import improve the community mapping effort? > This first question might best be asked last, as much of the answer to this questions should be contained in answers to questions about the targeted import data and planned import process. Plus, asking this first is fairly aggressive. Someone has taken the time to get involved in OSM, to identify a data set, and to come forward to ask permission. Clearly, they have some idea that this would be beneficial. Listen first. Hear them out. Presume innocence. Last, this is a qualitative, subjective question with widely varying opinions and interpretations across the OSM community. Why derail the review on first question? * Would it perhaps be better to "mix and match" this data at the rendering > stage, rather than adding it to OSM, since this data is unlikely to be > edited by anyone anyway? (An extreme example for this is height contours - > nobody would dream of uploading them into OSM, yet many a novice will > mistakenly think "it has to go into OSM to be on the map, right?") > > * Would it perhaps be better to stick this data into a WMS/WFS/Snapshot > server/... and offer it to the mapper community as an additional data > source instead of importing it outright? > This is a great question. OSM could help inform potential importers prepare for this question by describing what types of data OSM *is* good at - e.g. roads, buildings, addresses, POIs, land *areas*, whatever.... I think that, when confronted with new third-party data sources, many > people have a knee-jerk "let's import that" reaction which experienced > OSMers should counterbalance by asking questions like the above. To improve efficiency and reduce frustration, we should document those questions so that the people planning an import have a chance to prepare answers to the questions before submitting ideas to the committee. I'll be adding many of the questions from this recent thread to the wiki (not that that's the perfect answer... just to record the data outside of email archives...) On Tue, Dec 18, 2012 at 8:21 AM, Serge Wroclawski <emac...@gmail.com> wrote: > Frederik, > > > On Tue, Dec 18, 2012 at 11:16 AM, Frederik Ramm <frede...@remote.org> > wrote: > > > I'm keen on the "import process" including a number of "exit lanes" - it > > should ask questions like: > > questions like: > > We talked about this a bit last night; I agree. Many times an import > isn't useful for one or another reason. It could be license related, > or data quality, or the data doesn't belong in OSM, etc. > > > I would be happy if your working group would embrace this idea and find a > > name that doesn't explicitly say "imports" - what we need is people who > help > > with the responsible use of third party data for the benefit of OSM - > which > > might occasionally mean an import, but many other things as well. > > Name ideas welcome. > > I was hoping that large automated imports (like the tiger expansion) > would be included, but maybe you're saying they're separate problems? > > - Serge > > _______________________________________________ > Talk-us mailing list > Talk-us@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk-us > -- Jeff Meyer Global World History Atlas www.gwhat.org j...@gwhat.org 206-676-2347
_______________________________________________ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
