Re: [Talk-us] Disney (was Re: access=destination vs access=private)
Anthony writes: > On Fri, Sep 16, 2011 at 8:33 AM, Phil! Gold wrote: >> The US doesn't seem to have the strict legal categories for rights-of-way >> that the UK does > > I'm not sure what you mean by that, as I'm not familiar with UK law. > But the US definitely has a concept of "public right of way" vs. > "private property". The access=destination key/value was clearly > meant for the former. There is still a note in the source of the wiki > which explains what access=destination means in the US - "no thru > traffic" or "local traffic only" (USA). That is completely different > from "customers only". I think Anthony has it exactly right here. The basic problem is that access= is mostly about legal rights of access, but it's been blurred to be also about permission to access land that one has no legal right to demand. For almost all parking lots that are private property, people have no right of access at all. So we could take the existing tags, where =customer is perhasp not existing, and have a hierarchy: access=yes access=destination access=permissive (no legal right, but not objected to) *access=customers access=private (no right, permission not granted to the public) access=no (hard to tell what this means - doesn't make sense) or we could have access=yes (public right of way, which may be a 'private way') access=destination (legal right of way if reasonable/proper to get where you are going, or only way, depending on local law) access=private (no right of access, but permission might or might not be granted) and also access_permission=permissive access_permission=customer access_permission=private But, I don't think access vs access_permission is helpful in practice, even if it is semantically clearer. I do think it's broken to misappropriate access=destination to mean access=customers (even though I've been doing it also for parking lots). The disney employee discussion points out that while access_permission=customer is a relatively straightforward concept, access_permission=private conveys only If you don't have some special agreement, you can't go here. but doesn't encode the set of people that have permission. I think it's madness to put that in the map, and people should perhaps have a side database of what they've been granted permission to do, or gasp just look at the map and figure it out. The real question is what routers should do. Probably the best that can be done is to put a high cost on access=destination and even higher on access_permission=private and note this on the results. Typically you'd only be routed over those when going someplace where you have permission. pgp38coDIrmx1.pgp Description: PGP signature ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Brainstorm: What should a US map of OSM data look like
On 09/16/2011 03:37 PM, Ian Dees wrote: I'd like to figure out a way to start working collaboratively on getting a style started. Has anyone tried to collaboratively work on a style file before? Is there anyone interested in working with on it with me? I was thinking about firing up a publicly-accessible Tilemill instance on a representative set of US data and going from there. I'd be happy to help working on it - especially since I'm already doing some of the things that were mentioned on TopOSM. Tilemill is nice for trying out new ideas, though it might be some work converting existing styles to carto. Git might work well, too. -- Lars Ahlzen l...@ahlzen.com ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
[Talk-us] Small Place Names (was Re: Brainstorm: What should a US map of OSM data look like)
* Val Kartchner [2011-09-16 10:49 -0600]: > I think that this is also the time to add a "development" level to the > "place" key. There are so many place names that aren't hamlets but are > developments (like subdivisions) that have names. This could include > the name of an apartment complex. As others have mentioned, simple apartment complexes and similar things can be adequately represented by a landuse= polygon with a name= tag. For more complex things, like a major subdivision that includes a variety of landuses (residential, retail, recreation, etc.), there was discussion on the tagging@ list somewhat recently about a place=neighbourhood tag[0] whose semantics would fit fairly well. [0] http://wiki.openstreetmap.org/wiki/Proposed_features/place%3Dneighbourhood -- ...computer contrarian of the first order... / http://aperiodic.net/phil/ PGP: 026A27F2 print: D200 5BDB FC4B B24A 9248 9F7A 4322 2D22 026A 27F2 --- -- "It's today!" said Piglet. "My favorite day," said Pooh. --- -- ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Brainstorm: What should a US map of OSM data look like
On Fri, Sep 16, 2011 at 3:37 PM, Ian Dees wrote: > > > On Mon, Sep 12, 2011 at 6:06 PM, Richard Weait wrote: >> >> Let's keep this more "brainstormy" and less "discuss-y and debate-y" >> for now, okay? >> >> Trim and add your ideas to the list for your reply. We'll come back >> around for discussion later. ;-) >> >> A few of us were just asking on irc what a US-style tile theme would look >> like? >> >> - distinguish divided highways >> - distinguish toll roads >> - deconflict trunk / tree colors >> - Render shields at lower resolutions >> - De-emphasize railways at lower zooms. >> - Label motorways with both name and number where both are tagged >> - Fewer road colors > > Richard, thanks for starting this conversation! It was good to hear > everyone's ideas. > I'd like to figure out a way to start working collaboratively on getting a > style started. Has anyone tried to collaboratively work on a style file > before? Is there anyone interested in working with on it with me? I was > thinking about firing up a publicly-accessible Tilemill instance on a > representative set of US data and going from there. > Any other, better ideas? Some of the ideas will be simpler to implement than others. That might mean that those ideas will be simpler to endlessly bikeshed as well. ;-) For example, we could change road colors to address a few of the ideas above. But what should the final color combination be? I predict that there will be endless discussion on that. If one of us comes up with a Super-Contextual-render-a-pub-at-z4-if-it-is-the-only-pub-in-the-hemishpere, then I expect it will get used without question. Also, I think we might want to prioritize some of these and see who has the initiative and knowledge to tackle which ideas. And I'd like to add "- show surface = values in some way, so I'm not surprised by gravel roads." There is a similar thread on talk-ca@ with similar requests from that portion of the community. ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Brainstorm: What should a US map of OSM data look like
On Mon, Sep 12, 2011 at 6:06 PM, Richard Weait wrote: > Let's keep this more "brainstormy" and less "discuss-y and debate-y" > for now, okay? > > Trim and add your ideas to the list for your reply. We'll come back > around for discussion later. ;-) > > A few of us were just asking on irc what a US-style tile theme would look > like? > > - distinguish divided highways > - distinguish toll roads > - deconflict trunk / tree colors > - Render shields at lower resolutions > - De-emphasize railways at lower zooms. > - Label motorways with both name and number where both are tagged > - Fewer road colors Richard, thanks for starting this conversation! It was good to hear everyone's ideas. I'd like to figure out a way to start working collaboratively on getting a style started. Has anyone tried to collaboratively work on a style file before? Is there anyone interested in working with on it with me? I was thinking about firing up a publicly-accessible Tilemill instance on a representative set of US data and going from there. Any other, better ideas? ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Brainstorm: What should a US map of OSM data look like
Yeah, I typically use a landuse=residential/name=* polygon for apartment complexes. When I search for my address in nominatim it correctly says that it is a child element of the surrounding apartment complex. But I guess the original point might still stand. It may not be appropriate to tag an entire new subdivision as landuse=residential if it includes shopping or business areas along with housing that are all part of a named area within a city. Not sure place=development is the best tag for that but as Richard says, that isn't really part of the rendering brainstorm this thread is supposed to be. Toby On Fri, Sep 16, 2011 at 12:43 PM, Richard Weait wrote: > On Fri, Sep 16, 2011 at 12:49 PM, Val Kartchner wrote: > >> I think that this is also the time to add a "development" level to the >> "place" key. There are so many place names that aren't hamlets but are >> developments (like subdivisions) that have names. This could include >> the name of an apartment complex. > > That's a discussion that might be ideal for the tagging@ list. For a > named apartment complex, I'd expect the buildings or site relation to > be named. I wouldn't expect them to be a place. By comparison, I > also wouldn't expect a large university or business campus to be a > place either while they might be named buildings or site relations. > > Looking at our US map rendering brainstorm, how would you restate the > issues from your suggestions above? > > Earlier suggestions include the one you quoted from Brad, which might > be "context-sensitive rendering; show smaller towns if they are the > only towns around". > > ___ > Talk-us mailing list > Talk-us@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk-us > ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Brainstorm: What should a US map of OSM data look like
On Fri, Sep 16, 2011 at 12:49 PM, Val Kartchner wrote: > I think that this is also the time to add a "development" level to the > "place" key. There are so many place names that aren't hamlets but are > developments (like subdivisions) that have names. This could include > the name of an apartment complex. That's a discussion that might be ideal for the tagging@ list. For a named apartment complex, I'd expect the buildings or site relation to be named. I wouldn't expect them to be a place. By comparison, I also wouldn't expect a large university or business campus to be a place either while they might be named buildings or site relations. Looking at our US map rendering brainstorm, how would you restate the issues from your suggestions above? Earlier suggestions include the one you quoted from Brad, which might be "context-sensitive rendering; show smaller towns if they are the only towns around". ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Brainstorm: What should a US map of OSM data look like
On Tue, 2011-09-13 at 11:10 -0500, Brad Neuhauser wrote: > Towns appear at zoom level 9 in Mapnik, which seems pretty decent to > me. There are tagged "towns" in SW Kansas that show up, but some > "villages" probably need retagging to "towns" in the N and W. The > Place page recommends tagging county seats as towns regardless of > population: http://wiki.openstreetmap.org/wiki/Place I started > "upgrading" some county seats in NW MN to "town", which helps fill > things in, I think: http://osm.org/go/WprNC4-- > > > Brad I think that this is also the time to add a "development" level to the "place" key. There are so many place names that aren't hamlets but are developments (like subdivisions) that have names. This could include the name of an apartment complex. I suggest this because there are apartments next to each other that have names. The lowest level that these place names can be tagged with is "hamlet". But how many hamlets per square mile would be reasonable? - Val - ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] OSM US Chapter elections and
On Fri, Sep 16, 2011 at 9:54 AM, Kate Chapman wrote: > Sorry the issue was we did not have quorum and did not announce the > meeting. So it could not count as the annual meeting. The > Treasurer's report is finished (a big part of these sorts of things) > and we can have an annual meeting. The bylaws say a quorum "shall be determined by the Board of Directors prior to any such meeting" or at least 10. What number has the Board chosen? > Anyway, the plan was to have an online annual meeting, which is > possible according to our bylaws. Is this planned to happen this month or next? Considering the Board's term should have expired by now I think it's appropriate for the Board to make a formal request for nominations ASAP, and get on with an online meeting+election within a month or so. -Josh ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] OSM US Chapter elections and
Sorry the issue was we did not have quorum and did not announce the meeting. So it could not count as the annual meeting. The Treasurer's report is finished (a big part of these sorts of things) and we can have an annual meeting. Anyway, the plan was to have an online annual meeting, which is possible according to our bylaws. As for SotM-US that is dependent on people willing to organize. Though the event was smaller than SotM it still was a major undertaking. Thea Clay did the majority of the work while she was working for Cloudmade. It would require a larger volunteer effort to do it this year. Though we could certainly start a bidding process to look for possible venues. Best, -Kate On Thu, Sep 15, 2011 at 7:26 PM, Ian Dees wrote: > On Thu, Sep 15, 2011 at 8:20 PM, Richard Weait wrote: >> >> On Thu, Sep 15, 2011 at 7:05 PM, Serge Wroclawski >> wrote: >> > Last year it was State of the Map US. I assumed this year that it >> > would happen at the same time as OSMF has its elections... >> >> The US chapter was scheduled to meet Saturday at SotM at 12:30. Did >> anything happen there? >> http://stateofthemap.org/schedules/saturday-september-10th/ > > I'm fairly certain that wasn't there on Friday when I typed up the Saturday > schedule that ended up on the wiki. Either way, there was a HOT meeting at > that time outside and I don't think a OSM US meeting happened. > >> >> Perhaps a few members will offer to help the US local chapter team >> with organizing things. >> >> On Thu, Sep 15, 2011 at 6:55 PM, Gregory Arenius >> wrote: >> > Another barrage of questions >> [ ... ] >> > I think it was surveymonkey run by >> > some independent (Apache?) observers. Is it possible to do this again? >> > If >> > so, what needs to be done? >> >> As for last year, Mike Collinson and I served as independent observers >> / vote counters. The vote was handled through surveymonkey last year. >> We just used plain-old-email for the Foundation election just passed. >> I think I prefer email since it removes the reliance on an external >> service. > > ___ > Talk-us mailing list > Talk-us@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk-us > > ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Disney (was Re: access=destination vs access=private)
On Fri, Sep 16, 2011 at 8:33 AM, Phil! Gold wrote: > The US doesn't seem to have the strict legal categories for rights-of-way > that the UK does I'm not sure what you mean by that, as I'm not familiar with UK law. But the US definitely has a concept of "public right of way" vs. "private property". The access=destination key/value was clearly meant for the former. There is still a note in the source of the wiki which explains what access=destination means in the US - "no thru traffic" or "local traffic only" (USA). That is completely different from "customers only". ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] OSM US Chapter elections and
On Thu, Sep 15, 2011 at 1:12 PM, Josh Doe wrote: > > Has the Technical, Education & Student Outreach or US Tagging working > groups ever met? > > . > The Technical, Education & Student Outreach wg has been coordinating informally on efforts to outreach to colleges. > > -Josh > > ___ > Talk-us mailing list > Talk-us@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk-us > -- Carl Anderson, GISP cander...@spatialfocus.com carl.ander...@vadose.org ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Disney (was Re: access=destination vs access=private)
On Fri, Sep 16, 2011 at 9:17 AM, Anthony wrote: > As I've said on the wiki, I'd rather see > access=restricted plus access:restriction=customers_only, this way we > can give routers general information (that the way is restricted to a > particular category) without having them understand access=customers > and access=employees and access=judges and access=expectant_mothers, > etc. The way I envision it, a router could see access=restricted, access:restriction=X, and would pop up a dialog "your route includes a restricted area, the restriction is X, would you like to include this area in your route?" ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Disney (was Re: access=destination vs access=private)
On Fri, Sep 16, 2011 at 8:33 AM, Phil! Gold wrote: > * Anthony [2011-09-15 22:19 -0400]: >> Incidenally, for some reason Lulu-Ann put the example of customer >> parking lots in the wiki for access=destination, but I'd say this >> completely changes the original intent of access=destination. Only a >> few months ago the wiki said "access=destination The public has right >> of access only if this is the only road to your destination." > > The US doesn't seem to have the strict legal categories for rights-of-way > that the UK does, but it seems to me that customer-only parking lots match > the sense of access=destination very well, in the sense of, "you're only > allowed to use this if it takes you to an adjacent destination." access=destination on a way through a parking lot would mean "the public has a right to use this way if they are parking here" which most certainly is not the same as "the owner gives permission to use this way to customers". And in my experience "customers only" parking lots don't even mean "you're allowed to park here if you are going to an adjacent destination". The destination which "customers" is referring to is quite often *not* adjacent, and quite often there are other adjacent destinations (e.g. private residences) for which parking is not allowed (which is, in fact, the whole purpose of the sign). > What access restrictions do you think would better fit customer-only parking > areas? Areas, or the ways which lead through the areas? If you're talking about the areas, I don't know - it's not something I've ever thought about writing software to utilize. If you're talking about the ways, then access=customers would be the best fit, though in my opinion it's a bit *too* specific. As I've said on the wiki, I'd rather see access=restricted plus access:restriction=customers_only, this way we can give routers general information (that the way is restricted to a particular category) without having them understand access=customers and access=employees and access=judges and access=expectant_mothers, etc. Also, it would be helpful to actually see the area in question. Is the parking *area* restricted to customers only, or is it just *parking* that is restricted to customers only? ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Disney (was Re: access=destination vs access=private)
* Anthony [2011-09-15 22:19 -0400]: > Incidenally, for some reason Lulu-Ann put the example of customer > parking lots in the wiki for access=destination, but I'd say this > completely changes the original intent of access=destination. Only a > few months ago the wiki said "access=destination The public has right > of access only if this is the only road to your destination." The US doesn't seem to have the strict legal categories for rights-of-way that the UK does, but it seems to me that customer-only parking lots match the sense of access=destination very well, in the sense of, "you're only allowed to use this if it takes you to an adjacent destination." What access restrictions do you think would better fit customer-only parking areas? -- ...computer contrarian of the first order... / http://aperiodic.net/phil/ PGP: 026A27F2 print: D200 5BDB FC4B B24A 9248 9F7A 4322 2D22 026A 27F2 --- -- If the USENET discussion touches on homosexuality or Heinlein, Nazis or Hitler are mentioned within three days. -- "Sircar's Corollary to Godwin's Law" --- -- ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
Re: [Talk-us] Disney (was Re: access=destination vs access=private)
On 9/15/2011 10:19 PM, Anthony wrote: Also, I couldn't find any such sign going in the other direction. Even if this were access=destination, it would be a unidirectional access=destination. If you go the other direction you have to either pass through the main gate on World Drive or pass one of these signs at Reams Road and Center Drive. ___ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
