Re: [OSM-legal-talk] public transport routing and OSM-ODbL
On Thu, Jul 8, 2010 at 11:14 AM, Richard Fairhurst wrote: > > Liz wrote: > > Anything this contrived and complex that the potential users can't sort > > it out fails the usability test. > > There are only three possible data licences that aren't complex: > > 1. You may do anything you like with the data. (=PD) > > 2. You may do anything you like with the data. We ask you to be nice and > credit us, and to release any data you mix up with it. (=PD + Science > Commons-like community norms) > > 3. This data is for your own personal use only. Anything else, you have to > ask us. Sign on the dotted line to consent to this contract, and we'll let > you access the data. (=proprietary) > > Anything else has to be complex in order to apply across wildly different > jurisdictions. There ain't no Berne Convention for data and there is > remarkably little case law, especially relating to a database with so many > authors. You simply cannot write an open data licence which is legally > enforceable the world over without some complexity. It's not ODbL's fault - > it's the inevitable result of the OSM community not managing to agree to 1 > or 2. There are levels of complexity, though. CC-BY-SA would be much more straightforward here, as there is no requirement to release anything you don't want to release. If I were advising that public transport operator I'd recommend they fork off a CC-BY-SA version of the database. ___ legal-talk mailing list legal-t...@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-talk] Why quality is more important than routing speed
On Mon, Jul 5, 2010 at 6:13 AM, M∡rtin Koppenhoefer wrote: > 2010/7/5 Anthony : > > > Okay. How do you use speed limit tags when only 8% of the roads are > tagged > > with them? > > Actually it doesn't matter at all, how many percents of the planet are > tagged with a certain tag, it is sufficient that the tags are present > in the area you are interested in ;-). Quite true. But to use an anecdote, I've recently returned from road trip where I was relying heavily on Android's navigation software, and I've found that Google tends to greatly overestimate the actual speed of the less traveled roads I decided to travel on. I'm not sure if it's because they don't have the speed limits for these roads at all and are using an overly high estimate, because they're not adequately factoring in traffic lights, or something else. I'd say this is a great opportunity for OSM, actually. The ability for me to easily correct silly routes (and get instant gratification on those corrections) is something I'm looking forward to. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Why quality is more important than routing speed
I wasn't whining, nor was I suggesting that the project has no potential. In fact, I think there are lots of good answers to the question "how do you use speed limit tags when only 8% of the roads are tagged with them?" But I don't think "think longer-term" is one of them. On Mon, Jul 5, 2010 at 5:24 AM, Lambertus wrote: > Did it prevent me from mapping when I discovered back in 2007 that the map > was almost blank? Did I think that the project had no potential, could not > become anywhere near suitable for routing? No, on the contrary. And look at > where we're now, only three years later! > > Reasoning in half empty instead of half full won't lead us anywhere. Stop > whining, be constructive. > > > 2007On 2010-07-05 01:44, Anthony wrote: > >> On Sun, Jul 4, 2010 at 6:33 PM, Roy Wallace > <mailto:waldo000...@gmail.com>> wrote: >> >>On Sun, Jul 4, 2010 at 10:34 PM, Anthony ><mailto:o...@inbox.org>> wrote: >> > >> > How do you use speed limit tags when >> > only 5% of the roads are tagged with them? >> >>Think longer-term. >> >> >> Okay. How do you use speed limit tags when only 8% of the roads are >> tagged with them? >> >> >> >> ___ >> talk mailing list >> talk@openstreetmap.org >> http://lists.openstreetmap.org/listinfo/talk >> > > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Why quality is more important than routing speed
On Sun, Jul 4, 2010 at 6:33 PM, Roy Wallace wrote: > On Sun, Jul 4, 2010 at 10:34 PM, Anthony wrote: > > > > How do you use speed limit tags when > > only 5% of the roads are tagged with them? > > Think longer-term. > Okay. How do you use speed limit tags when only 8% of the roads are tagged with them? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Why quality is more important than routing speed
On Sun, Jul 4, 2010 at 2:20 AM, Frederik Ramm wrote: > Nic, > > > Nic Roets wrote: > >> There is a lot of talk around better algorithms (e.g. contraction >> hierarchies), distributed routing, stress tests etc. So I'm going to >> put in into perspective with a few calculations. >> >> For a 40km journey, Gosmore takes 50ms*. >> > > It's all a question of user experience. Of course if I want to plan a route > it is not a big deal if I have to wait 50ms or even a few seconds. > > However if you have a fast service you can switch over to a more > interactive way of route planning (think Google's UI where you grab a via > point and drag it and get a new route instantly, even while dragging; or > where it computes a number of routes for you initially instead of just one). > This is also an aspect of quality, and allows the user to find a better > route. > That's a great point. Being able to quickly recalculate a route, when going off course (intentionally or unintentionally), or when traffic conditions change, is very important. On the other hand, I'm sure it's possible to cache some extra detail when creating the route which should make this relatively simple. If nothing else, finding a shortest (*) path when you've already got a relatively short (*) path calculated, is much quicker than finding a shortest path from scratch. Really I think the biggest cost factor contributing to quality is going to be keeping up to date with things like traffic, construction, etc. I don't foresee OSM being very useful on that front, especially up-to-the-minute traffic conditions. (*) With "short" defined however desired, not necessarily in kilometers. > Quality is not just how many tags you support, although in a rich > environment like OSM it is of course desirable to use as much information > from the data as possible. > The problem with that is that once you get past the most popular tags you're going to have very spotty coverage. How do you use speed limit tags when only 5% of the roads are tagged with them? Based on the title of this thread I thought Nic was actually going to say pretty much the opposite of what he said: that using high quality OSM data is more important than trying to squeeze a few minutes of time savings using less well-maintained OSM ways. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Changed highway=*_link meaning?!
On Thu, Jun 24, 2010 at 2:21 PM, Lester Caine wrote: > Anthony wrote: >> >> You could always have highway=link. >> > But some links ARE motorway rules and some ARE trunk road so just saying > link does not work. I guess, but now you're using a different definition of *_link. Not "tag-for-higher" nor "tag-for-lower", but "tag-based-on-the-rules". That's excellent if we can come up with some good definitions for what those rules are. > But IMO, it's not a big enough deal to bother changing. >> > There is a lot of good detail already mapped that does not need changing. > Just using as it was intended. Right now the definition of motorway_link is a link road between a motorway and another road. There's no mention of a requirement that the road be subject to "motorway rules". And whether or not the link road is connected to a motorway is something that is inherent in the nodes/ways themselves. So there's no detail which wouldn't be given by highway=link. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Changed highway=*_link meaning?!
On Thu, Jun 24, 2010 at 1:14 PM, Lester Caine wrote: > Ed Avis wrote: > >> Isn't this tagging redundant? If a link road leads from a primary to a >> secondary, or whatever, this can be seen by looking at the tags for the >> two >> roads it connects. In principle there is no need to duplicate the >> information. >> > > But how do you know that a way IS a slip from one road to another, and not > just another road? You could always have highway=link. But IMO, it's not a big enough deal to bother changing. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging for street danger levels
On 6/23/10, Andy Allan wrote: > On Wed, Jun 23, 2010 at 2:49 PM, Anthony wrote: >> highway=path wouldn't work if the way is already tagged with >> highway=secondary. >> I was thinking something like highway=secondary, footway=both (a la >> highway=secondary, cycleway=track). >> The alternative would be to have three different ways. >> Neither solution is particularly nice, though. I'm actually hoping >> someone >> will come up with something better :). > > Without attempting to express much of an interest in tagging > discussions, it would be great if you were to use the word "sidewalk" > to describe the paths-along-the-edge-of-roads. The word footway might > well mean the same thing but is used elsewhere in osm, and will lead > to confusion. I don't get it. Isn't the term footway used in OSM for precisely this sort of thing? It's a way primarily used by pedestrians. Perhaps it's not "designated", but that's only because nothing is "designated" around here. Or do I just not understand what "designated" means? What am I missing? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging for street danger levels
On Wed, Jun 23, 2010 at 9:29 AM, M∡rtin Koppenhoefer wrote: > 2010/6/23 Anthony : > > I figured it would be footway=* (a la cycleway=*), but apparently that > was > > proposed years ago and never adopted > > (http://wiki.openstreetmap.org/wiki/Proposed_features/Footway). > > > do you suggest to set it to abandoned? Features in OSM are (mostly) > not introduced or adopted by voting. Just use it, if you find it > useful, even though there are alternatives (highway=path, > foot=official, bicycle=no) > highway=path wouldn't work if the way is already tagged with highway=secondary. I was thinking something like highway=secondary, footway=both (a la highway=secondary, cycleway=track). The alternative would be to have three different ways. Neither solution is particularly nice, though. I'm actually hoping someone will come up with something better :). ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Changed highway=*_link meaning?!
On Wed, Jun 23, 2010 at 9:16 AM, James Livingston wrote: > I, and from what I see in use where I live quite a few other too, have > always used xxx_link tags to join a highway=xxx with a higher one, because > we think what was documented on the wiki (xxx_link joins highway=xxx with a > lower one) is silly. > So what's a motorway_link? I've always assumed motorway_link joined a motorway with something less than a motorway. Perhaps this is jurisdiction-specific, though. Where I've lived, the offramp of a highway is under the jurisdiction of the highway (offramp of a state highway is policed by state troopers, not county police). So an offramp of a motorway would be, I'd assume, a motorway_link (if not motorway), not a primary_link. In fact, if I hadn't seen that there was a such thing as motorway_link, I'd probably tag them as motorway. If this is something which is opposite in other jurisdictions, that definitely would be a consideration we'd need to address. Of course, highway=* tags are already jurisdiction-specific. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Changed highway=*_link meaning?!
> > > Having a wiki is great but the 'anyone can edit' model is not good for > pages > > that are meant to be authoritative, > > Luckily we don't have authoritative pages in OSM. > I don't know about luckily, but yeah. For data to be maximally useful, it needs to be well-defined. Instead of trying to create authoritative wiki pages, we must make it > clear to everyone that these pages are *not* authoritative. The wiki reflects life, which reflects the wiki, which reflects life, which reflect the wiki... ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging for street danger levels
On Tue, Jun 22, 2010 at 11:25 PM, Stephen Hope wrote: > On 23 June 2010 00:14, Paul Houle wrote: > > > >I'd like to see some tagging that tells cyclists not to ride on > > sidewalks, for instance: as a pedestrian I've been involved in > > accidents where cyclists were ~illegally~ riding on a sidewalk and > > Surely that is just highway=foot, bicycle = no? I figured it would be footway=* (a la cycleway=*), but apparently that was proposed years ago and never adopted ( http://wiki.openstreetmap.org/wiki/Proposed_features/Footway). ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Changed highway=*_link meaning?!
On Tue, Jun 22, 2010 at 6:08 PM, David Paleino wrote: > Hello people, > does someone know the reasoning behind: > > > http://wiki.openstreetmap.org/w/index.php?title=Template:Map_Features:highway&diff=490719&oldid=485601 > > ? > http://www.mail-archive.com/tagg...@openstreetmap.org/msg02653.html > I can't find any discussion about those tags on t...@. And no, I haven't > checked whether it's been discussed in the tagging mailing list, but > changes of > this importance should be given ample diffusion, and announced everywhere > (and > you can't expect everyone to be subscribed to the tagging mailing list). > And > talk@ is one of the minimum requirements, IMHO. Agreed. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging for street danger levels
On Mon, Jun 21, 2010 at 2:40 AM, Cartinus wrote: > On Monday 21 June 2010 01:21:19 Roy Wallace wrote: > > On Sun, Jun 20, 2010 at 6:17 AM, Anthony wrote: > > >Personally I don't mind if they add some sort of subjective hazard level > > > tag as well as these objective tags, but I think the objective tags > will > > > be much more useful in the long term. > > > > +1. Please map the cause of the hazard, instead of (or at least as > > well as) a vague, subjective meta-description of a conglomeration of > > factors. If you are having trouble tagging any of these factors, e.g. > > traffic flow, let's discuss and fix that instead. > > So we are going to retag all > highway=primary|secondary|tertiary|unclassified > to highway=road and then tag the number of lanes and their width and > surface > in stead? > That would probably piss people off. Would be fine with me, though. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging for street danger levels
On Sat, Jun 19, 2010 at 1:51 PM, Toby Murray wrote: > They want to produce a map of the city that highlights the dangerous > roads to avoid in order to show how they act as barriers and make it > very difficult to move around town on a bicycle. Sounds like it's in both their interest and ours for them to include information on *why* a particular road is considered dangerous. Hopefully some patterns will arise, and we can create tags to map that rather than a generic "hazard_level=8". I could see > using numbers (0=safe path or bike lane, 1=residential/low traffic > road, 2=bigger roads with higher speed traffic, 3=avoid at all costs) > or some other string based identifier if that is deemed easier to > understand. According to http://wiki.openstreetmap.org/wiki/Cycleway bike lanes should be mapped as cycleway=lane. Speed limits should be mapped as maxspeed=*. The number of lanes should be mapped as lanes=*. I hope you can convince them to map these sorts of things in addition to whatever hazard level tagging system you come up with. Personally I don't mind if they add some sort of subjective hazard level tag as well as these objective tags, but I think the objective tags will be much more useful in the long term. I don't think we have any widely spread tags for traffic levels. It'd be nice for them to contribute, but maybe is beyond the scope of OSM. At the very least some sort of note=* explaining *why* the road is considered hazardous would be nice. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Big sponsors (was: WolframAlpha uses OpenStreetMap data)
On Fri, Jun 18, 2010 at 12:54 AM, SteveC wrote: > OTRS is a horrible system, whereas uservoice is easypeasy. > >From whose perspective? Send an email, wait 3 minutes and 42 seconds, receive a response that your issue has been resolved and thanking you for your report. That's my last experience with Wikipedia's feedback system, which uses OTRS. Doesn't get any more easypeasy than that. Certainly better than go to http://osm.uservoice.com/ , click on "sign up", click on "signup", pick a username, pick a password, type a query, hit search, hit "create new idea", hit "suggest it", and maybe receive a response one day (I guess I should confirm my email?). Who's going through the list of suggestions and sending responses to the people who made them? > As I said on IRC just now - I don't care about uservoice specifically, > there are lots of similar services around but the workflow is very nice. > http://en.wikipedia.org/wiki/Comparison_of_issue_tracking_systems ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Big sponsors (was: WolframAlpha uses OpenStreetMap data)
On Fri, Jun 18, 2010 at 12:40 AM, SteveC wrote: > > On Jun 17, 2010, at 10:36 PM, Anthony wrote: > > Anyway, if your point in this thread is just defending yourself against > what you see as an attack from Frederick, feel free to ignore this. But if > you'd like some ideas on how to create a system to allow OSM users to easily > "get feedback quickly from a real human being", I think you ought to look > beyond Uservoice. > > I do! > > I do. I do. I do! > > I keep saying that. The point though is that uservoice, in about 5 minutes, > would provide a _ton_ of feedback and engagement especially from novice > users who will otherwise just go away. As would OTRS. Or even just an unmoderated, anyone-can-post-without-being-subscribed mailing list staffed by some faithful volunteers who could provide more friendly responses than http://permalink.gmane.org/gmane.comp.gis.openstreetmap/50322 (which is not to pick on Richard - his response was to an OSM regular and not a newbie - the fact of the matter is that a newbie wouldn't have even found the list in the first place). See https://lists.wikimedia.org/mailman/listinfo/unblock-en-l ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Big sponsors (was: WolframAlpha uses OpenStreetMap data)
On Thu, Jun 17, 2010 at 11:46 PM, SteveC wrote: > that's a matter of opinion, any reasonable person I've ever met would > prefer uservoice to trac or otrs or whatever. > For what? From a glance at the two, OTRS and uservoice don't even seem to be in the same category. The two pieces of software seem to provide solutions for much different problems. Uservoice looks like it would be good for the big picture issues that affect lots of people. But after reading http://opengeodata.org/the-importance-of-timing-to-feedback I thought of OTRS, and I don't really see how Uservoice is in any way applicable. But maybe it's because I'm just not familiar enough with Uservoice. If Uservoice is the answer, how hard would it be to set up an email to Uservoice gateway? I'm sure I'm not alone in that I don't want to mess with signing up with Uservoice (or trac, for that matter) just to send feedback and get a response (especially given that Uservoice is not being hosted by OSM). You've commented about the interface of OTRS, but the fact of the matter is that the people submitting the feedback just don't see that interface. The interface is email, though it wouldn't be difficult to make a quick web form which can send the email (along with information like "what zoom level they were at, if they clicked on a particular location for the problem, etc."). Anyway, if your point in this thread is just defending yourself against what you see as an attack from Frederick, feel free to ignore this. But if you'd like some ideas on how to create a system to allow OSM users to easily "get feedback quickly from a real human being", I think you ought to look beyond Uservoice. I mean, even osm-talk works better than Uservoice in those terms, even if you do have to put up with an admonishment about not using trac (see http://permalink.gmane.org/gmane.comp.gis.openstreetmap/50322). But subscribing to osm-talk is another hurdle I don't think you want people to have to jump through to just ask a question or make a suggestion. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Big sponsors (was: WolframAlpha uses OpenStreetMap data)
On Thu, Jun 17, 2010 at 10:31 PM, SteveC wrote: > I think we can do better > Well then, feel free. On Thu, Jun 17, 2010 at 10:32 PM, SteveC wrote: > > On Jun 17, 2010, at 8:27 PM, Anthony wrote: > > I'd expect the company sending all that feedback at us to provide some > manpower, at least > > temporarily, to at least sort through the feedback and issue responses to > the simple ones. > > But the problem is all that manpower and resources would be seen as a > 'takeover'. Perhaps rightly. > The company wouldn't be doing anything that anyone else can't do. Anyone can set up an email address to take feedback on OSM, respond to that feedback, create accounts to make edits, etc. > On Thu, Jun 17, 2010 at 9:58 PM, SteveC wrote: > > lets say you were a very large firm and you wanted to scale OSM to that > level, how would you do it? > > > > Gradually. > > Yeah that's Fredericks point: we're already growing. I say: grow faster! > I've found it to be not particularly useful to tell other people how fast they should do something, at least people who you're not paying to do something for you. But your mileage may vary. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Big sponsors (was: WolframAlpha uses OpenStreetMap data)
On Thu, Jun 17, 2010 at 9:50 PM, SteveC wrote: > > On Jun 17, 2010, at 7:45 PM, Anthony wrote: > > > OTRS? > > huh? Does anyone think it would be a good idea to set up OTRS for OSM? If your question was what OTRS is, http://lmgtfy.com/?q=otrs It's the system that Wikimedia uses for this sort of thing (providing prompt responses to feedback issues utilizing a team of volunteers), and in my experience they generally provide very quick responses. On Thu, Jun 17, 2010 at 9:55 PM, Ian Dees wrote: > Do we really have the mapping capacity (in person-hours and server > hardware) to handle 300M people a day providing feedback? > No, but there's no way 300 million people a day are going to provide feedback. Additionally, I'd expect the company sending all that feedback at us to provide some manpower, at least temporarily, to at least sort through the feedback and issue responses to the simple ones. On Thu, Jun 17, 2010 at 9:58 PM, SteveC wrote: > > lets say you were a very large firm and you wanted to scale OSM to that > level, how would you do it? Gradually. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Big sponsors (was: WolframAlpha uses OpenStreetMap data)
OTRS? On Thu, Jun 17, 2010 at 9:30 PM, SteveC wrote: > Well let me take that back a bit - actually even doing some very simple > cleanup of the interface and having a feedback mechanism *at all* would be a > good first step, as people jumped on my recent OGD post in the comments: > >http://opengeodata.org/the-importance-of-timing-to-feedback > > > > On Jun 17, 2010, at 7:27 PM, SteveC wrote: > > I think you're concentrating on tiles, but that's not really the > bottleneck I would jump on first. > > > > The conversation goes like this: > > > > "steve we have 300 million people a day look at our site and we would > like to send their edits and feedback to OSM" > > > > Really it's the API we're talking about. Tiles are just a CDN problem. > > > > > > On Jun 17, 2010, at 7:18 PM, Frederik Ramm wrote: > >> Steve, > >> > >>> They would like to link to us directly but don't think a) we can > >>> handle the load and b) don't think it would be a good user experience > >>> to dump people on to osm.org, what with the site design. > >> > >> To paraphrase (not specifically Wolfram, but the unnamed other megacorps > you're chatting with): > >> > >> 1. they'd like to link to us directly but our infrastrucutre is too > weak; > >> > >> 2. they would not want to give us a shitload of money to improve our > infrastructure, but could imagine hosting something; > >> > >> 3. there is fear that the community would view this negatively. > >> > >> To which I say, I don't think the community has anything against someone > doing a glorified maps.cloudmade.com; if they have really fast servers and > maybe even a CDN, can do lots of styles and make the tiles and services > available under a free-for-all policy. That would be great, and would - if > given sufficient long-term promise by whoever it is - allow us to reduce our > tile serving to an experimental capacity, freeing up resources for the core > database which obviously we must keep operating ourselves. > >> > >> But there is a logical problem here and that has nothing to do with us > at all. You say that many would like to link to OSM directly if only OSM had > sufficient resources. Now assume that some big guy with many enemies, say > Google, or Microsoft, were to offer super-fat tile serving for OSM as I > outlined above. We would then scale back our own tile ops to a minimum, and > their server would be the main OSM tile server, and whenever you go to > www.osm.org your browser says "connecting to osmtile.google.com" or some > such. > >> > >> I think that the community would be less of a problem - I don't think > many would care if our tiles came from MS or Google or so as long as they > were unrestricted and the data remained free. But all those other big guys, > of whom you say that they would like to link to us - would *they* want to > send their users to get tiles from Google, MS or someone else? Or would the > "we'd like to link to you but your infrastructure cannot take the load and > anyway your front page is ugly" then be replaced with "we'd like to link to > you but you must understand that the 'sponsored by XYZ' on the shiny front > page is a problem"? > >> > >> Of course things would be even worse if the big sponsor wanted to put > the tiles or service under a non-open license (e.g. a license with a > "noncommercial" component"). That, I think, would reduce overall usefulness > rather than improving it. Any funded tile serving would have to be more open > than what we can currently offer, not less. > >> > >> Bye > >> Frederik > >> > >> -- > >> Frederik Ramm ## eMail frede...@remote.org ## N49°00'09" > E008°23'33" > >> > > > > Steve > > > > stevecoast.com > > > > Steve > > stevecoast.com > > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Price for OSM survey
On Tue, Jun 15, 2010 at 4:15 PM, Roland Ramthun wrote: > Am Dienstag, den 15.06.2010, 15:54 -0400 schrieb Richard Weait: > > [...] > > Wow. Every respondent will be getting $50-equivalent? I sure > > misunderstood the original announcement. > > Right, this was somewhat unclear. > > The survey is part of a university thesis about online communities and > there will be 1-2 prizes with a value of around 50€ each or maybe 50€ in > cash. > By answering the survey you take part in a draw of these two prizes. I'd say cash or at least a cash option. International shipping could get expensive depending on the location of the recipient. Might also be tariffs to deal with. Then again, sending an international money order can be just about as expensive. Hopefully the winner has paypal! ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
By the way, I assume we should break this out to an off-list discussion, or on a different list, or something. My apologies to those who don't like a lot of traffic on talk. OSM-verbose, anyone? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
On Mon, Jun 7, 2010 at 10:55 PM, John Smith wrote: > On 8 June 2010 12:46, Anthony wrote: > > How else are you going to describe what your object is? I don't see a > > uuid:lamp_post in your list of examples. I guess that would be, > > uuid:man_made? I don't think many people are going to figure that out. > In > > fact, I can't even really figure it out. > > If I had all the details nutted out this wouldn't still be a proposal :) > > In your next email you suggest they are highway tags, so uuid:highway=* > I just think it'd be a lot easier to say uuid=*, uuid_type=lamp post. Because uuid:highway=* is incredibly non-intuitive. Of course, that doesn't work because then you can't have multiple uuids on a single object without creating relations or unnecessarily duplicating objects. The OSM database doesn't readily support arrays of structured data. On Mon, Jun 7, 2010 at 11:01 PM, John Smith wrote: > If someone needed to add a description can't they just use the note=* > tag? Which then makes it a display issue. No, because there's no easy way to tie the note to the the uuid. What if you want two notes, one of which is a note for uuid:building and one of which is a note for uuid:operator? Again, can't do it easily in OSM, because the OSM database doesn't readily support arrays of structured data. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
On Mon, Jun 7, 2010 at 10:51 PM, John Smith wrote: > On 8 June 2010 12:50, Anthony wrote: > > On Mon, Jun 7, 2010 at 10:46 PM, Anthony wrote: > >> > >> Really, I think we need a better example than a lamp post, or at least > the > >> node ID of an actual lamp post in OSM. > > > > http://osmdoc.com/en/tag/highway/street_lamp > > > > Still not sure what the use case would be, though :). > > > > Tagging what's on the ground? There are some very decroative, and > possibly historical lamp posts in existence. > > If people are tagging individual trees I don't see why other actual > objects can't be either as long as they are of interest to someone or > a group of people > Fair enough. So the description is optional. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
On Mon, Jun 7, 2010 at 10:46 PM, Anthony wrote: > Really, I think we need a better example than a lamp post, or at least the > node ID of an actual lamp post in OSM. > http://osmdoc.com/en/tag/highway/street_lamp Still not sure what the use case would be, though :). ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
On Mon, Jun 7, 2010 at 10:18 PM, John Smith wrote: > You might be confusing a couple of issues here, when you look at OSM > tags you are viewing a simplified database, that is the raw data, what > you are describing is presentation of that data in a more human > friendly way, this isn't the same thing as free form text. > I'm not the one who brought up free form text, you are. I highly doubt anyone would seriously want to document, beyond what > already exists in the OSM DB, every possible object anyone would want > to link to, eg someone takes a picture of a lamp post because it looks > interesting for a photo and wants to link it to the OSM object that > might describe the location and height of the object. > Do we have such objects in OSM? In any case, I don't think "anyone" will want to document "what already exists in the OSM DB". I would expect someone creating an ID to link to (i.e. the person who uploaded the picture) to put in a brief description of what they're linking to, though (at the very least "a cool lamp post"). Otherwise, we have no basis to maintain the link, and they might as well just link to the node. Really, I think we need a better example than a lamp post, or at least the node ID of an actual lamp post in OSM. If on the other hand the object is worth commenting on or further > describing, this is where something like wikipedia or freebase would > be useful, you link the wikipedia/freebase IDs to the OSM ID and then > you can write a three page essay on the object. > A three page description would be way too long. I might be wrong, but I don't think there is a specific need for free > form text but there is a need to link the text to a map object. > How else are you going to describe what your object is? I don't see a uuid:lamp_post in your list of examples. I guess that would be, uuid:man_made? I don't think many people are going to figure that out. In fact, I can't even really figure it out. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
> >> If you want to interlink databases, eg wikipedia, you would simply > >> extend upon the work I've done for UUID to OSM object lookup table, > >> you'd add one more table and then use the UUID as the key field and > >> link other object IDs from other databases to it. > > > > Do you have a link? I'm not familiar with that proposal. > > I was hoping to have something done before I brought it up, but at > present it's still in my head. I have for the time being included your > previous wording about using a wiki on the proposal, but left it open > because while I don't think a wiki would be the best solution, I'd be > interested to know if you have a good reason for needing freeform > text. > I'm not sure. The description would be fairly freeform. I can't think of anything else that would be as freeform as that. I call it a wiki but I imagine most of the data would be structured. In fact, in many ways I think it'd be better structured than OSM data, which doesn't currently allow for arrays or sets or type checking. Not much different from http://www.openstreetmap.org/browse/way/27940475, except that "edit" wouldn't require flash, and it'd be formatted a little better (*). If the OSM database could better support structures, and especially arrays of structures, it could even use the OSM database for storage. But I'm not sure that's going to happen. For those things that are supported by the OSM database, the wiki would use the API. (*) For instance, instead of "addr:housenumber=411" and "addr:street=Elm Street", it'd say "Address: 411 Elm Street". ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
On Mon, Jun 7, 2010 at 8:16 PM, John Smith wrote: > On 7 June 2010 23:39, Anthony wrote: > > Either way, I could see someone going around removing uuid=* tags from > > places where they couldn't find the QR code in the store window. > > UUIDs aren't just so you can slap a QR code in some shop window Just to clarify, I don't mean to imply that they'd be right, just that I could see it happening. > Also just because something isn't > on the ground shouldn't mean it should be deleted just because someone > is overly zealot about only having things in the database that are on > the ground, otherwise we might as well start deleting half the state > and country borders that have nothing on the ground. > To be clear, I agree. > If you want to interlink databases, eg wikipedia, you would simply > extend upon the work I've done for UUID to OSM object lookup table, > you'd add one more table and then use the UUID as the key field and > link other object IDs from other databases to it. > Do you have a link? I'm not familiar with that proposal. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
On Mon, Jun 7, 2010 at 9:18 AM, John Smith wrote: > On 7 June 2010 23:12, Anthony wrote: > > The only thing I'm really afraid of is that these tags would violate what > > some people seem to believe is a rule - the supposed "map only what's on > the > > ground" rule. Do the website=* and wikipedia=* tags violate this rule? > > Using that logic, source=* from aerial imagery, would need to be > removed, Well, I'm sure even the most ardent "map only what's on the ground" proponents will make exceptions for things which are legally required. As for source tags that *aren't* legally required, I'd actually argue myself that as metatags they should be on the changeset, not the element. Either way, I could see someone going around removing uuid=* tags from places where they couldn't find the QR code in the store window. > although if wikipedia starts linking to OSM objects do we > need to also link to wikipedia objects? > I'd suggest that we should have a single website=* or uuid=* link to an all-inclusive wiki (can't use Wikipedia as that single website because of their notability rules), and that all other linking to any other websites should be done by adding an external link from that wiki page. So, no. On Mon, Jun 7, 2010 at 9:18 AM, John Smith wrote: > While you can embed a UUID in a URL I would suggest it gets it's own > tag since lots of objects already get a website tag. > I'd prefer that. But if the "let's try to get consensus for a new tag" process fails, there's always "shove it into an already accepted tag" option. (At which point it'd probably be website:uuid=*, or even website:uuid:building/operator/etc=*) ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging OSM objects with UUIDs
On Mon, Jun 7, 2010 at 3:58 AM, Peter Körner wrote: > I don't think that your uuid-tags will be more stable either - they are > subject to thousands of editors that will sometimes delete tags they > don't know or that won't take over those tags to the ways when they > create polys for them. > What about the website=* tag? Do people tend to maintain them? Because website=u.osm.org/c9f0f516-7da2-4cf0-a455-b09f90792c7d would work equally well and wouldn't require any new tags. I don't think the uuid tags would be treated perfectly. But they do seem like they'd be better than nothing, even if only some portion of the mappers maintain them. Also, I'm one of those mappers that will sometimes delete tags I don't know. But not if they're given an objective explanation in the wiki. The only thing I'm really afraid of is that these tags would violate what some people seem to believe is a rule - the supposed "map only what's on the ground" rule. Do the website=* and wikipedia=* tags violate this rule? Anthony ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Giving everything a unique ID
On Thu, Jun 3, 2010 at 10:44 AM, Emilie Laffray wrote: > On 3 June 2010 15:38, Serge Wroclawski wrote: > >> On Thu, Jun 3, 2010 at 5:39 AM, Pieren wrote: >> So is the permanent object the node? Is the permanent object the POI? >> What if the POI moves? If I tag the public library as a POI node, then >> do a building trace, that's one POI- but what if the library moves (as >> my local library is planning on doing). Does that permanent object >> move with the library, or does it stay with the building? >> >> > The idea behind John's idea is that the permanent UUID is linked to your > library. So if your library moves, you need to move the UUID tags to the new > building. It is meant to be associated with the "moral" entity like a > library, a shop, etc... "Moral entity" might not be the best term but it is > close, I think. > Yes. The way I see it, the "permanent object"/"moral entity" would be whatever you describe in the text. So if you put in the text "the Texas School Book Depository", the uuid should move when the book depository moves. If you put in the text "the building where Oswald shot Kennedy", the uuid shouldn't move when the book depository moves. I'd think in most cases you'd choose the former rather than the latter, but in some cases you might really want to link to the latter (such as the example given). ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Giving everything a unique ID
On Thu, Jun 3, 2010 at 10:07 AM, Anthony wrote: > Please note that I've abandoned the functionality of having multiple uuids > on a single element (e.g. uuid:building and uuid:shop). > Hmm, on second thought, maybe that's not such a hot idea. There might be two different stores which are combined into one, and obviously we'd want to keep both uuids (otherwise they wouldn't be "permanent"). Still, I'd prefer uuid=1;2;3 to uuid:*=*. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Giving everything a unique ID
On Thu, Jun 3, 2010 at 2:26 AM, jamesmikedup...@googlemail.com < jamesmikedup...@googlemail.com> wrote: > Hi, > here is a humble suggestion, instead of giving *everything* unique id, > we might focus on making some form of permalink that is usable upon > request. Like for wikipedia articles etc, that we can link to and be > relatively sure that the link will still be there. Some form of watch > tool that would inform the user that the permalink he created is > broken. > It would be easier to maintain a list of "don't break me" links than > to rework the whole system. > mike > Okay, here's a plan. I took some of the detail from http://wiki.openstreetmap.org/wiki/Proposed_features/UUID but adapted it a bit: *From the URL for any node/way/relation (or lat/lon pair) one can click on a button "make permanent link". *The node/way/relation gets tagged with uuid=*, where * is generated using an algorithm expected to create a universally unique id (I'll let someone more expert determine how, but I was thinking some sort of hash on the xml of the feature itself plus the time). If a lat/lon pair is created then a new node is created with that uuid. *A wiki page is set up where http://domain/wiki/UUID has text, links, and a slippy map. The slippy map highlights the element which has the UUID. The text is meant to be brief - only enough to uniquely identify the "thing" (perhaps the description text could even be duplicated in a uuid_description tag). Links would be used for the actual interesting data about the "thing". *If more than one element points to the same UUID, this is an error - use a relation if you want to do this. *Mappers are encouraged to check http://domain/wiki/UUID before deleting or repurposing nodes/ways/relations. *A bot goes through regularly checking for additions, deletions, drastic changes, duplicate UUIDs, etc., and adds them to a list for people to manually check/fix. *All external sources are encouraged to point to UUIDs, not to the element id. Please note that I've abandoned the functionality of having multiple uuids on a single element (e.g. uuid:building and uuid:shop). I felt that this overcomplicates things from the standpoint of someone clicking on "make permanent link" - they shouldn't have to know anything about the internal workings of OSM and I want to maintain the flexibility to tag *anything*, not just a predetermined list of things. I suppose this could be allowed for advanced users who want to do things by hand, but it's not in this plan. Anthony ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Software goes on, brain goes off...
On Wed, Jun 2, 2010 at 7:27 AM, Nic Roets wrote: > When I map, I just want to create a useful map. And when I write > software it should be backward compatible with old data and forward > compatible with new data and still give reasonable results. I don't > want to waste time on finding the legal status of everything. > Then don't use tags which indicate the legal status. Easy peasy, right? The rest of your post focused on what a "routing engine" should do. But this discussion isn't about the routing engine. Different routing engines are going to have different rules. This is about the data, and the data should be unambiguous. That means not using the same tag for "illegal" and "bad for the environment" and "unsafe". ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Fw: Software goes on, brain goes off...
On Wed, Jun 2, 2010 at 10:16 AM, John Smith wrote: > On 3 June 2010 00:15, Anthony wrote: > > They could always require people to log in. Or require people to log in > if > > they want to put up with the annoying terms and conditions only once. > > This woman is claiming she wasn't warned the route might be bad, so to > ensure that people don't forget they agreed to view the warning only > once, everyone will be forced to view it every time regardless. > Really? Do you have a link for that? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Fw: Software goes on, brain goes off...
On Wed, Jun 2, 2010 at 9:49 AM, John Smith wrote: > Worst than that, it could lead to even more stupid terms and > conditions annoying you over and over and over again after every route > request. They could always require people to log in. Or require people to log in if they want to put up with the annoying terms and conditions only once. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Software goes on, brain goes off...
On Tue, Jun 1, 2010 at 9:38 PM, John Smith wrote: > On 2 June 2010 10:23, Anthony wrote: > > You seem to have missed the rest of my post. I was arguing that a road > with > > no pavement but with a shoulder is *not* unsafe. OTOH, if the road has > no > > shoulder, and traffic traveling at 55 mph, and only 1 car a day, I'm not > > walking down it. > > My mother often goes for walks on roads that have 3 or 4 times that > amount of traffic at that speed, and she isn't the only one. It's > perfectly safe to do so because there is room to get off the road and > you can usually hear them coming, especially when it's a B-Double* > instead of a car, I've never heard of any pedestrians being clipped or > killed. > Room to get off the road. That's what I was referring to as a "shoulder". The amount of traffic nor the speed they travel at nor type of traffic > doesn't inherently make walking on a road unsafe. > Agreed. 100%. > I have no idea of the legality of walking along roads outside towns, > but hitchhikers do it often and I don't think they get arrested. > Here's the (relevant) law in Florida: "Where sidewalks are not provided, any pedestrian walking along and upon a highway shall, when practicable, walk only on the shoulder on the left side of the roadway in relation to the pedestrian's direction of travel, facing traffic which may approach from the opposite direction. " So, technically, here in Florida, walking on the roadway when there is a shoulder available (and practicable) would be illegal. Interestingly, "shoulder" does not seem to be defined in the law, but I've always assumed it meant the part of the right of way (paved or unpaved) which was able to be walked upon and which was not part of the roadway. However I think it would be a great idea to indicate the difference > between legally disallowed and just not a good idea due to personal > safety, I don't think re-using the foot tag is a good idea, because it > might be legal, but not safe to do late at night because you'll get > mugged etc etc etc. > Best case scenario, if we really want to be able to produce adequate walking directions for people unfamiliar with the route, would be to map the entire right of way as one or more areas with surface=* designations. Anything short of that is probably going to be insufficient, because there is so much variation as to what people would consider safe enough (between different people, and even between different times and scenarios - if I'm taking my kids with me I might want a 30 ft. wide shoulder but if I'm walking alone a much smaller one would be acceptable). ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Software goes on, brain goes off...
On Tue, Jun 1, 2010 at 11:44 AM, Nic Roets wrote: > On Tue, Jun 1, 2010 at 5:22 PM, Anthony wrote: > >> Ideally, yes. But routing software can't possibly process the logic > >> correctly in cases like these. Some roads may not have a pavement, but > >> they are safe for pedestrians due to the lack of traffic. In other > >> cases extreme footways should not be used because of crime. > > > > What does lack of traffic matter? Unless you mean absolutely no traffic, > I > > don't think that makes much difference. If the road is unsafe to walk > on, > > I'm not going to walk down it whether there's 1 car a day or 10,000. If > > there's a low enough speed limit maybe. > > By that logic you should never leave your house. What if a storm > suddenly appears and you get hit by lightning ? > You seem to have missed the rest of my post. I was arguing that a road with no pavement but with a shoulder is *not* unsafe. OTOH, if the road has no shoulder, and traffic traveling at 55 mph, and only 1 car a day, I'm not walking down it. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Software goes on, brain goes off...
On Tue, Jun 1, 2010 at 10:23 AM, Nic Roets wrote: > On Tue, Jun 1, 2010 at 3:32 PM, Matt Williams wrote: > > The road should simply be marked as having no pavement/sidewalk. > > Something like pavement=yes/no is a start at least. It's best to avoid > > subject assessments like how dangerous a road is. > > Ideally, yes. But routing software can't possibly process the logic > correctly in cases like these. Some roads may not have a pavement, but > they are safe for pedestrians due to the lack of traffic. In other > cases extreme footways should not be used because of crime. > What does lack of traffic matter? Unless you mean absolutely no traffic, I don't think that makes much difference. If the road is unsafe to walk on, I'm not going to walk down it whether there's 1 car a day or 10,000. If there's a low enough speed limit maybe. If you don't trust your own opinion, ask a few locals if they would > advise a tourist to walk there. If they say no, then tag them with > foot=no and add a note describing why you did it. > So because a few locals wouldn't advise a tourist to walk there you're going to tag the road equivalently to one that is illegal to walk on? I think we've gotta do better than that. I'd prefer the ambiguous foot=dangerous to foot=no. Especially if you're saying that high traffic + no pavement = dangerous (I've walked on plenty of roads with high traffic and no pavement - as long as they have a shoulder I wouldn't tag them as "foot=no, never, way too dangerous", I'd tag them as "foot=try to find a better route, but if you must, use caution". How about "foot=destination"? :) I'm kidding, but it'd be better (and more accurate) than foot=no. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Software goes on, brain goes off...
On Tue, Jun 1, 2010 at 9:32 AM, Matt Williams wrote: > On 1 June 2010 13:33, Jason Cunningham wrote: > > 1... What's the correct way of tagging a street as 'dangerous/suicidal' > for > > pedestrians in OSM? (Couldnt find an answer in the wiki) > > Recently come across a road in my area (London, UK) that had no pavement > and > > which clearly should be avoided by pedestrains, but there were no > > restrictions in place for pedestrians (apart from common sense). The UK > also > > does not have restrictions on pedestrains being on roads that some other > > countries have. > > > > So, in my opinion, foot=no would be wrong because it incorrectly > indicates > > pedestrians are not allowed. I guess foot=dangerous would be useful for > > routing software, but is there agreed way of tagging these problem roads. > > The road should simply be marked as having no pavement/sidewalk. > Something like pavement=yes/no is a start at least. It's best to avoid > subject assessments like how dangerous a road is. > Hmm, is "shoulder" a fairly universal term? Because shoulder=no would be much more daunting to me than pavement=no. Also, if the street really is dangerous/suicidal, is there any chance you can report it to the govt. so they can either fix it or ban pedestrians? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Software goes on, brain goes off...
On Mon, May 31, 2010 at 7:01 PM, Anthony wrote: > On Mon, May 31, 2010 at 5:40 PM, John Smith wrote: > >> On 1 June 2010 07:29, Frederik Ramm wrote: >> > Hi, >> > >> > Nakor wrote: >> >> Did Google add their notice after the fact? >> > >> > I am trying to make it a habit to read articles before I reply to them >> > and have already found it saves me some embarassment. >> >> In this case it doesn't matter if there is a notice or not, I >> personally wouldn't go and play on a busy road just because some >> mapping software suggests it's a good idea :) >> > > She wasn't playing, she was walking to her destination. I can't tell from > the pictures whether it was her fault for following the route, her fault for > walking on the wrong side of the road, her fault for not staying close > enough to the side of the road, the government's fault for not banning > pedestrians, the government's fault for setting too high of a speed limit, > or the car driver's fault. From the aerials it does look as though there > was enough room on at least one side of the road to walk (it wasn't a paved > sidewalk, but whatever). > > Ridiculous that she'd try to blame Google, though. > Looking more closely, there is a sidewalk, which turns into a "cycleway" ( http://www.openstreetmap.org/browse/way/40967519), about 50 feet from the roadway, on the southbound side which is the same side she was walking on. And if she had been using the sidewalk while heading north on Main St, it would have led her directly to that sidewalk. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Software goes on, brain goes off...
On Mon, May 31, 2010 at 5:40 PM, John Smith wrote: > On 1 June 2010 07:29, Frederik Ramm wrote: > > Hi, > > > > Nakor wrote: > >> Did Google add their notice after the fact? > > > > I am trying to make it a habit to read articles before I reply to them > > and have already found it saves me some embarassment. > > In this case it doesn't matter if there is a notice or not, I > personally wouldn't go and play on a busy road just because some > mapping software suggests it's a good idea :) > She wasn't playing, she was walking to her destination. I can't tell from the pictures whether it was her fault for following the route, her fault for walking on the wrong side of the road, her fault for not staying close enough to the side of the road, the government's fault for not banning pedestrians, the government's fault for setting too high of a speed limit, or the car driver's fault. From the aerials it does look as though there was enough room on at least one side of the road to walk (it wasn't a paved sidewalk, but whatever). Ridiculous that she'd try to blame Google, though. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
2010/5/31 Ian Dees > I don't think anyone has suggested that we leave out things I'd they > aren't signposted. Nathan, who started this thread, has done exactly that, and he's gone around removing route relations where the routes were not signed on the ground. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Mon, May 31, 2010 at 3:13 PM, Ian Dees wrote: > If they are not marked, how do the locals know what and where they are? > They look at a map! ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Mon, May 31, 2010 at 3:13 PM, Ian Dees wrote: > On Mon, May 31, 2010 at 2:06 PM, Gustav Foseid wrote: > >> On Mon, May 31, 2010 at 9:01 PM, Ian Dees wrote: >> >>> >>> No borders? No national parks? No nature reserves? No voltage on power lines? No named farms (unless the owner puts up a sign)? No names for peaks? >>> >>> Except for borders, all of those things are verifiable on the ground. I >>> think that is Frederik's point. >>> >> >> How do you, on the ground, verify the name of a farm? > > > You ask the owner. > What's the point of the phrase "on the ground"? If asking the owner counts as being "verifiable on the ground", anything verifiable is verifiable on the ground (just "ask X", where X is the name of someone who can verify it. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Mon, May 31, 2010 at 1:57 PM, Frederik Ramm wrote:
> Hi,
>
>
> Anthony wrote:
>
>> By these definitions, something that is able to be confirmed as true or
>> false in an official online source is actually *more* verifiable than
>> something written on a street sign in a place where Google Street View has
>> not yet visited. It certainly is verifiable, and it is not necessarily "on
>> the ground".
>>
>
> Something that is available from an official online source but not
> verifiable on the ground should not - in my personal opinion - be included
> in OSM.
>
> For the simple reason that we cannot improve the data - how should we if
> there is not reference on the ground? So the data will just sit there and be
> left to rot, or left to wait for another update by those who keep it. But
> OSM is not a "mirror" for official data. I don't want data that OSMers
> cannot work with; such data would only be in OSM for ease of retrieval, and
> I don't view OSM as some data dumpster for the world's geodata.
>
Well, I think there's a difference between being "verifiable on the ground"
and having "reference on the ground". The name of a lake or a river has
"reference on the ground", even if the name of the lake or river isn't
printed on a sign. A county border which is defined with reference to
roads, rivers, fences, etc. has a "reference on the ground", even if the
name of the county isn't printed on any of those features.
I can agree that OSM should not include data which has absolutely no
reference to physical (relatively) non-movable features on the ground. But
once the feature is there, when is it okay to tag it with features which
aren't strictly "on the ground"?
Maybe the answer is "never". Personally I think the correct answer is more
like "usually not", at least "usually not" in highly developed areas which
tend to put signs on everything. But I think it's confusing if you refer to
that as encompassed by the "verifiability" rule. It's really a separate
rule altogether.
On Mon, May 31, 2010 at 2:33 PM, John F. Eldredge wrote:
> Well, some people in the traffic-jam discussion seem to be taking the
> viewpoint that if something is not verifiable by people in other
> geographical locations, without actually visiting the location under
> discussion, then it should not be classified as being verifiable at all.
>
Huh? So in any location where we don't have good aerials and street views
we can't map at all? Or do photographs count (in which case, just take a
photograph of the traffic jam)?
Actually, wrt the traffic jams I'd just as well they not be in OSM, for much
the same argument that Frederik just presented ("the data will just sit
there and be left to rot, or left to wait for another update by those who
keep it"). But as far as I'm concerned, so long as people use tags and
features which don't collide with the things I'm interested in, I really
don't care what unmaintainable crap they add to the database.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Mon, May 31, 2010 at 12:05 PM, Andrew wrote: > If anything is unclear on the ground the mapper needs to provide a source. > That > way other mappers can judge whether the source is legitimate. > That's a great point. I hate fixing an area of map which is already in place and coming across information which I can't verify. Do I copy the information blindly, assuming that the previous contributor must have verified it, or do I delete it? If everything is either "on the ground" or has a link to a source, this would greatly simplify that dilemma. Not completely, though. It's sometimes quite difficult to verify road names, especially when the road can't be reached by foot. I guess the safest thing to do then is delete the names and let someone else go back and add them, especially if the roads have been significantly changed (I'm thinking of the Tampa Airport Interchange Project which I see Nathan has gone in and cleaned up quite nicely). And then there's the road designations of primary/secondary/etc. Not only is that not "on the ground", it's not even verifiable. In those cases I just guess. Or mark it as "highway=road" (see http://www.openstreetmap.org/browse/way/47280788). ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Mon, May 31, 2010 at 2:45 AM, Frederik Ramm wrote: > Hi, > > > Anthony wrote: > >> I guess the suggestion to "map what's on the ground" is good advice as >> long as it's not exclusionary. But my beef is with people who tell us to >> "map what's on the ground" to the exclusion of everything that isn't on the >> ground. >> > > Problem is that whatever is not on the ground is not verifiable; I'd have > to take the mapper's word for it. And this opens the door to people > inventing stuff. > That's precisely the reasoning that I'm arguing against. With all due respect, it just doesn't make any sense. You have to either drastically redefine the meaning of "verifiability" (able to be confirmed as true or false by other mappers) or the meaning of "on the ground" (less clear, but roughly encapsulated in relatively non-movable property in a public place). By these definitions, something that is able to be confirmed as true or false in an official online source is actually *more* verifiable than something written on a street sign in a place where Google Street View has not yet visited. It certainly is verifiable, and it is not necessarily "on the ground". So, I don't know if you've got different definition(s), or you just don't follow that logic. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Sun, May 30, 2010 at 6:48 PM, Nathan Edgars II wrote: > On Sun, May 30, 2010 at 9:28 AM, Anthony wrote: > > In any case, more important than the etymology of the phrase "map what's > on > > the ground" is what it means and whether or not it's good advice. In > terms > > of its use in excluding verifiable information I think it is quite > > problematic. When a route isn't written "on the ground" that's exactly > when > > it's most useful to have it identified in a map. > > Not really; maps are primarily used for navigation, whether > computer-routed or human-read. If the map shows that Long Street is > the A1889, someone using the map will be looking for the A1889. But if > Long Street is not marked "on the ground" as the A1889, that > designation is about as relevant as the fact that it was once the > route of the A1. In other words, if we know for sure that Long Street > is officially the A1889, it might make sense as a separate > ref_unmarked=A1889 tag, like old_ref=A1, but using the same tagging > for signed and unsigned routes helps nobody. > Agreed. Using the exact same tagging would be inappropriate. You might as well be taking that from the private message I sent you earlier today, where I said as much (I said we should map unsigned routes and add a tag of signed=yes/no). But you're looking at only half of the picture. Yes, if you know where you want to go on the map and you want to find it in the real world, you want to have access to the signs that are on the ground. Of course, for that purpose we're better off mapping most of the signs as nodes, not as ways. But what if you're looking for A1889? If it's not on the map, and it's not on the ground, you're not going to find it. I guess the suggestion to "map what's on the ground" is good advice as long as it's not exclusionary. But my beef is with people who tell us to "map what's on the ground" to the exclusion of everything that isn't on the ground. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Sun, May 30, 2010 at 9:19 AM, John Smith wrote: > On 30 May 2010 23:17, Anthony wrote: > > From what I can tell, it was actually the solution to such an edit war. > How > > "map what the people on the ground say" turned into "map what's on the > > ground", I can't figure out. > > Seems like it would logically go the other way round, from map what > was on the ground to map what people on the ground say... > I'm not sure what it means to "logically go" one way or the other, but the earliest reference I can find to any sort of "on the ground rule" is November/December 2007, and it's that one quoted at http://wiki.openstreetmap.org/wiki/Disputes . If you can find an earlier one I'd be quite interested. In any case, more important than the etymology of the phrase "map what's on the ground" is what it means and whether or not it's good advice. In terms of its use in excluding verifiable information I think it is quite problematic. When a route isn't written "on the ground" that's exactly when it's most useful to have it identified in a map. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Sun, May 30, 2010 at 3:40 AM, John Smith wrote: > On 30 May 2010 15:39, Anthony wrote: > > "If the dispute can not be resolved through discussion, then the simple > > default rule is that whatever name, designation, etc are used by the > people > > on the ground at that location are used in the non-localized tags." > > Isn't that kinda asking for an edit war where there is disputed > territory with different names by different languages possibly in > multiple languages? > >From what I can tell, it was actually the solution to such an edit war. How "map what the people on the ground say" turned into "map what's on the ground", I can't figure out. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Sun, May 30, 2010 at 1:35 AM, Anthony wrote: > On Sat, May 29, 2010 at 6:46 PM, Nathan Edgars II wrote: > >> Right now, the only mention of the "on the ground" rule on the wiki is >> here: http://wiki.openstreetmap.org/wiki/Disputes#On_the_Ground_Rule >> Should a separate page be created about how it applies more generally? >> > > Well, that all depends. How does it apply more generally? > > Should a route relation only be used when there are signs along the route? > How often do the signs need to be placed? > "If the dispute can not be resolved through discussion, then the simple default rule is that whatever name, designation, etc are used by the people on the ground at that location are used in the non-localized tags." That's interesting, because that says pretty much the exact opposite of what the "map what's on the ground" people argue for. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] On the ground rule on the wiki
On Sat, May 29, 2010 at 6:46 PM, Nathan Edgars II wrote: > Right now, the only mention of the "on the ground" rule on the wiki is > here: http://wiki.openstreetmap.org/wiki/Disputes#On_the_Ground_Rule > Should a separate page be created about how it applies more generally? > Well, that all depends. How does it apply more generally? Should a route relation only be used when there are signs along the route? How often do the signs need to be placed? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Google Hiring 300 Temps to Fix Map Errors
OSM: using 10,000 people to do what Google does with 300. ;) On Fri, May 21, 2010 at 11:50 AM, Ian Dees wrote: > > http://www.techflash.com/seattle/2010/05/google_hiring_300_temp_workers_in_kirkland_to_pinpoint_bugs_in_google_maps.html > > Anyone care to come up with a press release that says something like > "OpenStreetMap volunteers, numbering in the 10s of thousands, fixing a free > map for free." > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk > > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] closedstreetmap.org
It'd be perfect for the closed source fork of OSM which comes into place after the content is moved from CC-BY-SA to DbCL. I'll take it. On Wed, May 19, 2010 at 8:04 PM, SteveC wrote: > If anyone wants the above domain name, let me know as it expires in a > month. > > Yours &c. > > Steve > > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] [OSM-legal-talk] Proposed human readable contributor terms
On Wed, May 19, 2010 at 2:14 PM, andrzej zaborowski wrote: > Can you submit copyrighted data if you agreed to the new > Contributor Terms, even if it's under a free license? > Obviously only if it's a free license which is compatible with DbCL. That probably includes CC0 and not much else. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] River boundaries , not Post code areas
On Tue, Apr 6, 2010 at 5:04 AM, John Smith wrote: > On 6 April 2010 18:00, Igor Brejc wrote: > > I don't know about NSW and Vic case, but in the above cases the official > > Well we're talking about this specific case and the border is based on > the southern bank of the main flow of the river. > Do you have a legal document or court case or something stating that? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Google Wave si great for organizing OSM mapping parties
Maybe some people will be more interested in creating an account *after* or *during* the meetup. I like the entusiasm, Valent. If you can manage to introduce people to OSM who otherwise wouldn't be interested, this is great. Maybe you can act as a liason between the interested parties on Wave and those on the wiki. This way the many individuals who wish to avoid Google can participate too. Anthony On Apr 5, 2010 6:22 PM, "John Smith" wrote: On 6 April 2010 05:37, Frederik Ramm wrote: > which they don't fully realise. ... I found it weird that someone was suggesting people that wanted to be involved with a mapping party don't want to get an OSM account... ___ talk mailing list talk@openstreetmap.org http://lis... ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Post code areas
On Fri, Apr 2, 2010 at 8:57 AM, Lester Caine wrote: > Do you add the 'postcode' as a tag to > each house, or just create a relation for a post code? > Neither. I let people look up postcodes using lookup tables, not maps. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Post code areas
On Thu, Apr 1, 2010 at 6:34 PM, Frederik Ramm wrote: > There is a post code database that is officially maintained but it is not > free. > That's...interesting. Wouldn't any accurate description of the post codes necessarily be a derivative of that official database? If you're not sure of the answer maybe this is a question for the legal list, but what methods, if any, are allowed to create a free database from an officially defined (non-physical) non-free one? Basically, you just launder the data through a bunch of different others until it can no longer be traced to the original source? In the end, the source is always necessarily that officially maintained database. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Post code areas
On Thu, Apr 1, 2010 at 2:16 PM, Frederik Ramm wrote:
> Post code areas are tremendously useful in Germany because they are
> commonly used as a cheap machine readable form of location descriptor
> ("enter your post code to find the nearest band branch" etc).
>
That's a good use for post code centroids, not post code areas. And post
code centroids are best stored in a simple lookup table, not OSM.
The actual areas are basically only useful for reverse geocoding (click a
spot on the map and get the postal code). But whether or not that's even
possible is highly dependent on whether or not the post office provides such
information. For some post offices, such information is not meaningful.
What is the postal code for the middle of a highway? Maybe there is one
defined, which represents what the postal code would be if there were a post
box there. But maybe there isn't. It depends on the post office.
I have no idea where Germany fits in that classification, which is why I
asked the questions above.
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Post code areas
Hi Frederik, What are the sources for the post code areas? How often are they updated? How are they defined (by reference to houses, by reference to geographical features, by lat/lon, something else)? Will this data be integrated into other OSM data, or is it basically just a separate layer? Anthony On Thu, Apr 1, 2010 at 5:49 AM, Frederik Ramm wrote: > Hi, > >we're thinking about importing post code areas in Germany. Are post > code areas being mapped in other countries already, and if so, using > what tagging schema? > > I was thinking of creating multipolygon boundary relations with > boundary=post_code_area or so. > > Bye > Frederik > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Editor without relation-support makes sense?
On Thu, Mar 25, 2010 at 3:48 PM, Seventy 7 wrote: > > > Tirkon wrote: > > > > I found some discussions within OSM, that it would make sense to > offer > > > > an OSM editor especially for beginners. To make it easy enough, they > > > > should not confuse the beginner with complicated stuff like relations > > > > and thus not show and support them. But does that make sense? > > No, quite the reverse. The use of relations should only increase over time, > but what is needed from the next generation of editors is to hide all of > this background database detail. > > For instance, I don't want to create a relation of type multipolygon, I > want to create an area; I don't want to create a relation of type route, I > want to add a way to a route and so on. > Excellently put. > Imagine a proper WYSIWYG editor - you would simply draw an island in a lake > and say "let it be grass" and the editor will do all the stuff with > multipolygons in the background (creating them if necessary) to make it so. > The user shouldn't need to know. Yes. I'd like very much, for instance, to be able to create a multipolygon lake using a paint (and eraser) tool. The editor should then be able to create the way(s) and put them into a multipolygon. This would make it much easier to trace lakes. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Mon, Mar 22, 2010 at 3:32 AM, Lester Caine wrote: > We NEED as a mater of > urgency an agreed method of MANAGING groups of ways that at a low zoom > level > define a single linear object, but at higher zoom levels show that the > 'boundaries', carriage ways and structure are physically distinct. When a > structure like this gets moved, then all of the fine detail moves as well - > but > with the option to 'unsnap' a boundary or other sub element if necessary. > Any ideas for how this would work? There has to be a very good reason for REMOVING any data, and the assertion > that > we can in general 'remove multiple ways' is only acceptable if the project > is > also going to adopt the rule 'we will never map detail'? Perhaps it is time > for > a split in the project, and those of us who need to maintain details such > as > road structures and property boundaries set up our own fork with tools that > allow things that the simplify everything camp don't thing we should have? > I share your frustration but hope it doesn't come to that. What tools do you think are needed? What expertise can we tap into to produce these tools? Maybe if we get the tools finished first, we'll find there's no need to fork. What would be the best editor to learn in order to provide assistance? Or would you suggest starting on a new editor from scratch? Either way, if there's something I can help with, let me know. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Mon, Mar 22, 2010 at 12:42 AM, John Smith wrote: > On 22 March 2010 14:32, Anthony wrote: > > By reading the legal definition, of course. Same way I'd determine what > the > > border is in the first place. > > How many borders in the US are there exactly? > 3. Fortunately we have 4 editors. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Mon, Mar 22, 2010 at 12:28 AM, John Smith wrote: > On 22 March 2010 14:15, Anthony wrote: > > YOU said that I "meant re-use the road as part of a relation". But in > fact > > I did not. My position on that is that sometimes that is a good idea. > And > > sometimes it isn't. It's really case-dependent. If a boundary is > legally > > And how would you know which ones are legally defined this way exactly? > By reading the legal definition, of course. Same way I'd determine what the border is in the first place. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Sun, Mar 21, 2010 at 11:59 PM, John Smith wrote: > On 22 March 2010 13:53, Anthony wrote: > > True. What's your point? > > That you haven't actually done much work on boundaries to figure this > all out for yourself and the pitfalls of some of the suggestions you > are making... > What suggestions am I making? I made a suggestion to delete the TIGER boundaries, because they suck. I also made a suggestion to use boundary relations for administrative boundaries, because boundaries are generally shared (with other boundaries). YOU said that I "meant re-use the road as part of a relation". But in fact I did not. My position on that is that sometimes that is a good idea. And sometimes it isn't. It's really case-dependent. If a boundary is legally defined as the centerline of a road, then yes, I think you should use a single way, because if you fix one, you should fix the other. On the other hand, if it is just a coincidence that the two lines are in the same position, you shouldn't use a single way. In fact, in most such situations the lines wouldn't even coincide, they'd be off, if even by a centimeter or two. While considering your points I did think of another situation, which is that a boundary is defined as the "current" centerline of a road (i.e. as of the time of adoption of the definition). In that case I suppose it is best to use duplicate ways - one for the road and one for the boundary. But even in that case you should *still* use a boundary relation. In any case, I don't think you've seen the mess created by the TIGER import of administrative boundaries. This is especially true with regard to CDPs, which coincide with absolutely nothing outside of what the census decided to use for census purposes (but have names which are annoyingly similar to real administrative areas which have completely different legal definitions). ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Sun, Mar 21, 2010 at 11:40 PM, John Smith wrote: > On 22 March 2010 13:31, Anthony wrote: > > 1) How so? In the worst case scenario you have an equal-sized mess. Can > > you give an example? > > Because you are trying to hit a moving target... > What does that mean? > > 2) In most cases of road-realignment you generally *want* to move the > > boundary at the same time you move the road. If a road centerline and a > > boundary line exactly coincide, it's almost surely because the boundary > line > > is *legally defined* as the road centerline. (If some of the lines > > coincided by pure coincidence, then you can and should use duplicate > lines, > > but even that doesn't stop you from using a boundary relation.) > > While I don't know about the US specifically, it has happened in > Australia where boundaries that coincided with the centre of the road > weren't moved when the roadway was realigned. > I'm sure it has happened in the US as well. On top of that you have boundaries of local governments change when > state governments redefine the local government boundaries, you have > boundaries of postcodes that change, are split or merged, road > alignment doesn't mean the boundary moves with it. > Postcodes are a whole different story. At least in the US (which is what we're talking about in this thread, as the thread is about TIGER), postcodes are not defined by boundaries in the first place. They are defined as a set of delivery points. In any case, let's stick to the topic of TIGER administrative boundaries. Yes, these boundaries change. And the roads change too. And sometimes, when the road changes the boundary does not. On the other hand, sometimes, when the road changes the boundary does. Just because some boundaries coincide with the centreline, doesn't > mean they are legally bound to the centreline... > True. What's your point? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Sun, Mar 21, 2010 at 10:24 PM, Mike N. wrote: > In your point b), do you mean that if we did use boundary relations that > there would not be an issue with boundaries and roads being co-mingled and > mis-edited? > The use of boundary relations doesn't prevent people from mis-editing. I'm not sure what you mean by co-mingling, but I'm sure the use of boundary relations doesn't prevent that either. Are you familiar with boundary relations and how they work? Anything you can do with a single closed way you can do with a boundary relation. So by that fact any mistake or mess you can make with a single closed way can be made with a boundary relation. One of the biggest reasons to use boundary relations has nothing to do with roads. It's the fact that a border is generally shared by multiple boundaries. A single way can be used for a state border, a two different county borders, a city border, and a township border, instead of having 5 duplicate ways (and that's not at all an unique type of situation - it's something that happens all the time at state borders). On Sun, Mar 21, 2010 at 10:28 PM, John Smith wrote: > On 22 March 2010 12:24, Mike N. wrote: > > In your point b), do you mean that if we did use boundary relations that > > there would not be an issue with boundaries and roads being co-mingled > and > > mis-edited? > > The problem with this is when boundaries or roads move independent of > each other, such as for road-realignment, the whole thing becomes a > bigger mess. > 1) How so? In the worst case scenario you have an equal-sized mess. Can you give an example? 2) In most cases of road-realignment you generally *want* to move the boundary at the same time you move the road. If a road centerline and a boundary line exactly coincide, it's almost surely because the boundary line is *legally defined* as the road centerline. (If some of the lines coincided by pure coincidence, then you can and should use duplicate lines, but even that doesn't stop you from using a boundary relation.) ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Sun, Mar 21, 2010 at 8:41 PM, Anthony wrote: > How will boundary relations help what? > If you were asking how they help in general, see http://wiki.openstreetmap.org/wiki/Relation:boundary In particular, they "Make it easier to stitch all the parts of a border to each other" and "Avoid multiple duplicated ways above each other" ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Sun, Mar 21, 2010 at 5:35 PM, Mike N. wrote: > >How about 4: delete the TIGER imported administrative boundaries? > > > >In my experience a) they're not very good, and b) we should be using > boundary relations anyway. > How will boundary relations help? They must still refer to a closed way > in order to define the administrative boundary. > How will boundary relations help what? Boundary relations refer to multiple (non-closed) ways which form a closed border. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Administrative boundaries along roads
On Sun, Mar 21, 2010 at 11:39 AM, Jeff Spirko wrote: > Hi, all, > > Many of the administrative boundaries in my area follow roads (or vice > versa). (E.g. http://osm.org/go/Zcll6ubE?layers=B000TTF ) It seems > like the TIGER import has a separate list of nodes for the two ways > (one administrative and one road), but the nodes are at identical > locations. This makes Maplint and Potlatch complain about duplicated > nodes, making it hard to distinguish real map lint from this type. > > It seems like the TIGER data for the administrative boundaries is > pretty good, so I don't want to touch those nodes. > > What's the best thing to do with this situation: > 1. Leave it as-is. > 2. Combine the nodes so they're not duplicated. (Do bots sometimes > do this? How would a person do it in Potlatch?) > 3. Offset the nodes of the road so it follows the same line but no > longer forms map lint. > How about 4: delete the TIGER imported administrative boundaries? In my experience a) they're not very good, and b) we should be using boundary relations anyway. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] What do you wish you'd known?
On Sat, Mar 20, 2010 at 11:05 PM, Martin Koppenhoefer < dieterdre...@gmail.com> wrote: > 2010/3/21 Anthony : > > How to add USGS high res imagery as a layer for tracing. > > you could file a josm-ticket to have them in the basic configuration > (already as preset like yahoo). (add the wms-url to the ticket) > > cheers, > Martin > I'm actually not sure how to get it to work in josm. In Merkaartor I use http://imsortho.cr.usgs.gov:80/wmsconnector/com.esri.wms.Esrimap/USGS_EDC_Ortho_Florida? which lets me pick the county. In any case, this is what I wish I'd known. Now that I know it, it's not an issue any more. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] What do you wish you'd known?
On Fri, Mar 19, 2010 at 6:45 PM, SteveC wrote: > What are the thing or things you know know that you wish you'd known when > you started with OpenStreetMap? > How to add USGS high res imagery as a layer for tracing. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Google StreetMap embedding Flickr photos
On Mon, Mar 1, 2010 at 1:46 PM, John Smith wrote: > On 2 March 2010 04:36, Anthony wrote: > > Wikipedia is a whole different beast. It'll likely be replaced by Google > > when and if Google come out with a breakthrough in natural language > > processing. It looks like that breakthrough has already come in terms of > > creating a 3D model of the world. All that's left is taking lots of raw, > > dumb pictures. No sense in wasting manpower over something a bunch of > > unmanned dirigibles could do. > > Yes and we've been told for years that in 5 years time computers will > program themselves. While computers may be able to crunch numbers, > they can't just mash a bunch of information together and make > something useful and interesting. > Great point. I think the power really comes when we combine the two methods, massive collaboration and technology. I'd love to see some tools to facilitate automated high res orthography tracing with direct input from human mappers. Micromapping is uncharted territory, and is a place OSM could potentially shine. Unfortunately, one advantage Google has over OSM (as exemplified by this very thread, actually), is that they avoid copyright paranoia. OTOH, there is a lot of high res ortho which is in the public domain, albeit mostly in the United States. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Google StreetMap embedding Flickr photos
On Mon, Mar 1, 2010 at 12:51 PM, John Smith wrote: > On 2 March 2010 03:28, Anthony wrote: > > Yeah. Anyone know how they're doing this? It's stuff like this that > makes > > me think that a free, non-profit project is always going to be many steps > > behind the big boys when it comes to this domain. > > Google like all companies has limited growth potential with employees > alone, which is why they're trying to branch out into crowd sourcing, > they've already have one failure on this front with their attempt to > replace wikipedia. > Wikipedia is a whole different beast. It'll likely be replaced by Google when and if Google come out with a breakthrough in natural language processing. It looks like that breakthrough has already come in terms of creating a 3D model of the world. All that's left is taking lots of raw, dumb pictures. No sense in wasting manpower over something a bunch of unmanned dirigibles could do. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Google StreetMap embedding Flickr photos
On Mon, Mar 1, 2010 at 12:08 PM, wrote: > Google tech is really nice > Yeah. Anyone know how they're doing this? It's stuff like this that makes me think that a free, non-profit project is always going to be many steps behind the big boys when it comes to this domain. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Google StreetMap embedding Flickr photos
On Mon, Mar 1, 2010 at 11:32 AM, wrote: > This photo is (and, IIRC, has always been) CC-SA. Since this is 'Share > Alike', does it now mean that Google Street View is now a derivative work, > and therefore CC-SA? > No. It's only CC-SA if they decide to release it under CC-SA. > Mungewell. > (who doesn't really mind them using the photos, but likes license > compliance). > Send them a DMCA takedown notice. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Thoughts on OSM design, and looking forward and back
On Thu, Feb 25, 2010 at 1:04 PM, Dermot McNally wrote: > On 25 February 2010 17:28, Anthony wrote: > > > Are you sure about that? How many people does it take to map the world? > > 1,000? 10,000? 100,000? 1,000,000? More than that? > > > > The more the merrier. But I'm not sure about the whole "we have to" > part... > > > > Alright - let's settle on "we should". In a world where we can't > define when a map is "complete", "have to" is going to be similarly > subjective. > > But the answer to your second question is "more than that" - whatever > the suggested figure is. > Ha, well, my point failed then, because I was thinking more like 10,000. How many does Google have? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Thoughts on OSM design, and looking forward and back
On Wed, Feb 24, 2010 at 1:00 PM, Dermot McNally wrote: > On 24 February 2010 17:19, Tom Hughes wrote: > > > I completely disagree. We're running a project to map the world, > > We agree on that - but I claim that to do so effectively we have to > harness the power of all those people who don't yet "get" what we're > trying to achieve... > Are you sure about that? How many people does it take to map the world? 1,000? 10,000? 100,000? 1,000,000? More than that? The more the merrier. But I'm not sure about the whole "we have to" part... ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Tue, Feb 23, 2010 at 2:33 AM, John Smith wrote: > On 23 February 2010 17:30, Anthony wrote: > > Perhaps they did, but they would be wrong. > > Because of hindsight? > No, because ways aren't powerful enough to build complex data structures. > Relations are recursive - they can contain other relations. Ways can only > > contain nodes. > > You missed the point, I'm just giving examples to show that people who > think everything we need, we already have at our disposal, the > suggestions on this I've made in the past require changes to database > tables etc to work, rather than trying to shoe horn existing tools to > do something they aren't very well suited for. > The power of OSM is that we can create new data types without bugging the DBAs and the people with svn access. When we decided we wanted to map barriers we didn't have to add a barrier table - we just took the existing way structure and gave it a barrier tag. However, as I said above, ways cannot contain other ways, they can only contain nodes. So when it came to building a data structure which contained multiple ways (quintessential example being the multipolygon), ways alone were not sufficient - we needed relations. The idea of building multipolygons (aka "complex multipolygons" in OSM-speak) using relations was really a wonderful idea. Kudos to whoever came up with it. I don't think they were the type of structure that was intended by relations, since relations has that silly term "relation" instead of a more powerful name like "object". But, in any case, the power was discovered - and is being used to great effect right now - without ever having to go through the hassle of creating a new table in the database and new code in the svn. Interestingly, now that we have relations, we don't actually need ways any more - a relation can do everything that a way can do and then some. However, 1) we might as well keep them around for backward compatibility purposes; and 2) they make the job of database optimization a little bit simpler. In any case, my point is that requiring the DBAs and developers to get involved every time you come up with a neat idea for a new data type just doesn't make any sense. You already have the tools to build what you want - show that your design is sound first and then if it catches on you can always convert your relation type to its own table later. If I thought you had a sound design in the first place I would have already shown you how the same thing can be accomplished (just as elegantly) using relations, but so far you haven't convinced me of that. If it turns out there's something in your design which really can't be handled elegantly without adding a new table - fine, but then I'd suggest a more general solution so that we're not once again tied to the developer cycle every time a new idea comes along. In any case, I don't see it happening. Just take the name of the table and put it as the relation type. Then take the fields of the table and make them keys in the relation. Or, if any of those fields are foreign keys into the nodes, ways, or relations table, designate them as members. When you get into the billions of rows and the database starts slowing down, *then* you can talk about splitting those rows out into their own table. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Tue, Feb 23, 2010 at 1:49 AM, John Smith wrote: > On 23 February 2010 16:43, Anthony wrote: > > We've got all the tools we need - nodes and relations. With them we can > > build anything else we want. > > I'm sure people said the same thing about ways and nodes, Perhaps they did, but they would be wrong. > why did we need relations? > Relations are recursive - they can contain other relations. Ways can only contain nodes. > I think an acceptable method of micro mapping lanes will come as soon as > > someone makes a renderer that renders one of the myriad of possible > > solutions. > > Micro mapping isn't just for rendering, in fact it has far bigger > applications in the routing side of things, eg "In 500m merge into the > right lane" etc, a lot of this is just meta information and doesn't > need to be mapped visually to the nth degree. > True, but I don't think people will accept a micro-mapping solution until they can see it. It's too abstract for most people to picture in their minds. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Tue, Feb 23, 2010 at 1:30 AM, John Smith wrote: > On 23 February 2010 16:22, Anthony wrote: > > I only found one (the one about directional information, in the case of a > > one-way road) to be correct. The other 5 were complaints about how the > > current renderers work. > > > > Anyway, I do think there is one major problem with mapping highways as > areas > > right now. It's too time consuming. Other than that, I think it's the > way > > of the future - I'm just not sure how long it's going to be for that > future > > to arrive. > > Both these points are in common, ideally it would be nice to be able > to micro mapping lanes, not just areas a road way covers yet we still > lack tools to do this. > We've got all the tools we need - nodes and relations. With them we can build anything else we want. I think an acceptable method of micro mapping lanes will come as soon as someone makes a renderer that renders one of the myriad of possible solutions. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Mon, Feb 22, 2010 at 7:33 PM, Roy Wallace wrote: > On Tue, Feb 23, 2010 at 9:29 AM, Anthony wrote: > > > >> Exactly. Mapping a way as an area is fine as long as you also > >> represent *the path of travel*. > > > > What path of travel? There are many paths of travel, and generally none > of > > them are properly represented by a line going through the middle of a > > roadway. > > By "path of travel", I mean what is currently represented as a > highway=* way in the OSM database. Tobias already gave 6 reasons why > this is important. > I only found one (the one about directional information, in the case of a one-way road) to be correct. The other 5 were complaints about how the current renderers work. Anyway, I do think there is one major problem with mapping highways as areas right now. It's too time consuming. Other than that, I think it's the way of the future - I'm just not sure how long it's going to be for that future to arrive. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Mon, Feb 22, 2010 at 5:30 PM, David Paleino wrote: > On Monday 22 February 2010 23:26:52, John Smith wrote: > > On 23 February 2010 08:05, David Paleino wrote: > > > I remember someone complaining with me that routers not supporting > > > highway=* + area=yes in the same relation with a "normal" highway=*, > > > might get confused -- and that something like "landuse=road" would be > > > better. > > > > Wouldn't landuse=road bleed colour between the way and the area? > > If you noted the link I included in my mail, I haven't used it. > > I think landuse=road is semantically more correct than highway=* + area=yes > (but this could be debatable too), but the drawback is that renderers have > the > burden of colouring landuse=road the same way of its way. If both are in a > relation, it could probably be done, but I believe it'd take some effort. > Why does the landuse have to be the same color as the way? I'm pretty sure I'd prefer it to be a different color by default. As for semantical correctness, I think that depends on the road. For roads without any lines, where people are allowed to drive as they please subject to a standard rule like "keep right except to pass", I'd say the area is more semantically correct. In most standard cases, though, where a road is lined, simply mapping it as an area is inadequate. In any case, I'd say landuse=highway would be better than landuse=road, and that should represent the entire right of way. If you want *=road, amenity=road or man_made=road would be more appropriate. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Mon, Feb 22, 2010 at 4:36 PM, Roy Wallace wrote: > On Tue, Feb 23, 2010 at 6:11 AM, Tobias Knerr wrote: > > > > ...Way representation is more useful for > ... > > - anything that has directional information, such as oneway roads > > Exactly. Mapping a way as an area is fine as long as you also > represent *the path of travel*. > What path of travel? There are many paths of travel, and generally none of them are properly represented by a line going through the middle of a roadway. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Mon, Feb 22, 2010 at 12:55 AM, Roy Wallace wrote: > On Mon, Feb 22, 2010 at 5:03 AM, Niklas Cholmkvist > wrote: > > I live in a place where I feel the need to map some streets as areas. If > > I start a little of such mapping, will routing software get confused? > > How were you planning to achieve this? > > There is still no consensus that I'm aware of for how to do this > What about http://wiki.openstreetmap.org/wiki/Key:area ? As I said above, so long as the road doesn't have any street lines within it (as most of the residential roads where I live are), you just draw a border around the area and tag it with highway=residential (or whatever) and area=yes. For example, http://www.openstreetmap.org/?lat=28.077444&lon=-82.548096&zoom=18&layers=B000FTF ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Sun, Feb 21, 2010 at 5:36 PM, Anthony wrote: > On Sun, Feb 21, 2010 at 2:03 PM, Niklas Cholmkvist > wrote: > >> I live in a place where I feel the need to map some streets as areas. If >> I start a little of such mapping, will routing software get confused? >> > > If it's not a one-way road, routing software should be pretty much fine, > even if it ignores the area tag and treats the road as a loop. > By the way, according to the wiki, "area=yes, in the context of roads, indicates that the area has no street lines within it". That may or may not pose a problem depending on the streets you are mapping. In my neighborhood none of the minor streets have street lines so an area tag could be used without even ignoring the wiki. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Mapping streets as areas - can I do it now?
On Sun, Feb 21, 2010 at 2:03 PM, Niklas Cholmkvist wrote: > I live in a place where I feel the need to map some streets as areas. If > I start a little of such mapping, will routing software get confused? > If it's not a one-way road, routing software should be pretty much fine, even if it ignores the area tag and treats the road as a loop. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Two different ways with the same nodes?
On Fri, Feb 12, 2010 at 3:34 AM, Shaun McDonald wrote: > Why not offset the nodes a little to make them easier to work with and be > able to see that there is two ways there? > Because that would be tagging for the renderer editor. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Two different ways with the same nodes?
On Thu, Feb 11, 2010 at 4:57 PM, wrote: > > Hi, > > > > Stefan Pflumm wrote: > >> this ways are all highways. > > > > It surely is unusual for two highways sharing the same nodes, and I > > cannot think of an example where this would make sense. But that doesn't > > mean there is none; can you give an example? > > > > Bye > > Frederik > > > Double-decker bridge > The ways should not share nodes, because the ways don't intersect. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] [Tagging] Proposed feature: Gated Communities
On Wed, Feb 3, 2010 at 3:47 PM, Martin Koppenhoefer wrote: > 2010/2/3 Chango640 : > > If you are interested in this proposal, please visit > > http://wiki.openstreetmap.org/wiki/Proposed_features/Gated_community to > see > > full details and discuss. > > Why not use landuse=residential I agree with this, although where feasible I'd rather see landuse=residential only on the residential sections of the gated community. > together with another tag, say community=gated (where community could also > become other stuff like > religious, seniors, female, Adding community=gated seems redundant. Just map the wall or fence itself. ... and or add access=private? > I guess you could put access=private on the wall/fence, but isn't that the default? I'd definitely put access=private on all the roads, parking spaces, parks, etc. Anyway I would suggest to map the extent of the gated area by adding > the fence barrier=fence and the entraces / gates. > Yeah, definitely. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Using editors to indicate license preference.
On Mon, Jan 18, 2010 at 4:24 PM, Roy Wallace wrote: > On Tue, Jan 19, 2010 at 6:24 AM, Liz wrote: > > > >> I think Google and others will quickly rape the PD server > >> > > This assumes that they can find a means to import and check the data. > > This is Google - it's what they do best. > Yeah, Google would probably quickly start importing data from the PD OSM into its products. That alone would be enough to cause me to contribute. I'd love to be able to correct errors that I come across in Google Maps Navigation (the Android app). To compare such a thing to rape is really quite appalling. Not to mention hypocritical, when you consider CloudMade and its obnoxious TOS. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Using editors to indicate license preference.
On Mon, Jan 18, 2010 at 10:57 AM, SteveC wrote: > I'm personally thinking at this stage that a better solution would be for > the license move to ODbL to include a fork as part of the agreement, so > everyone contributes their data as both ODbL and/or PD, not just move to > ODbL. We could host a PD or CC0 server alongside the ODbL one and just see > which dataset grows more quickly. I think Google and others will quickly > rape the PD server while the ODbL server will have a process which is either > manual or automated to pull in the PD data and will always be the better > map. Would that make the PD crowd happy? As long as it's hosted on a server which doesn't contains a Terms of Service asserting that people agree to the ODbL, I guess it's a start. If you decide to do it, CC0 would be a much better choice than PD. But it'd probably cause nothing but headaches for OSM. Which site will new users go to by default? How are you going to phrase the question which asks people which one they want? Which data shows up by default on the main page? I really can't see this becoming more than a half-hearted effort, which is perhaps worse than no effort at all. > Personally I don't see it as much different to just working for Google and > TomTom for free. > As opposed to just working for GeoFabrik and CloudMade for free? C'mon. Releasing the data into the public domain helps out a lot more than *just* Google and TomTom. In fact, it helps out all the people and organizations that releasing data under the ODbL does, *plus* lots of other people and organizations too. Plus you get whatever benefit you get which convinced you to contribute in the first place. So no, you're not *just* working for Google and TomTom for free. But you are helping Google and TomTom for free, so if you think Google and/or TomTom are evil, then I can see why you might not want to release your contributions into the public domain. Personally (without naming names) I can think of much better examples than Google and TomTom, of companies that I'd rather not help out. So I'm not sure if I'd find such a PD fork acceptable or not. If you're going to run a fork, why not run a fork under CC-BY-SA? That would be something a lot more useful, as it's a fork you're already implicitly creating anyway. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Using editors to indicate license preference.
On Sun, Jan 17, 2010 at 12:01 PM, Frederik Ramm wrote: > Anthony wrote: > >> But I think you're missing the fact that sites which try to restrict >> people from copying their databases pretty much universally do not provide >> database dumps. >> > > I don't think that matters at all. Whether or not a database dump is > provided universally, or by accident, or not at all, does not make a > difference regarding rights to the data(base). It is just an implementation > detail. > It doesn't make a difference regarding rights to the data(base), but it does make a big difference to the de facto ability to enforce restrictions on the data(base). This is especially true when copyright rights are explicitly waived. Database rights are not very strong, and they are not universal, which is *why* "sites which try to restrict people from copying their databases pretty much universally do not provide database dumps". And database rights are not even clearly applicable to the OSMF: http://en.wikipedia.org/wiki/Database_Directive";>While copyright protects the creativity of an author, database rights specifically protect the "qualitatively and/or quantitatively [a] substantial investment in either the obtaining, verification or presentation of the contents" OSMF did not obtain or verify the data in the OSM database. I suppose you could say they present it, but the presentation part isn't the part the they want to restrict anyway. I still don't see how setting up a bunch of servers and telling people to make maps gives you a database right, but I have to admit I'm not an expert in database rights, since they don't apply to me here in Florida. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Using editors to indicate license preference.
On Sun, Jan 17, 2010 at 10:51 AM, John Smith wrote: > 2010/1/18 Frederik Ramm : > > whereas if the data is not copyrighted, but given to me under a contract > > that stipulates that I may not put it up on a web site and say "download > and > > use freely" then > > Assuming the data isn't copyrightable, the vector + lat/lon > information may not be, but there is a lot of meta information that > may be. However I'm guessing Anthony will be the first to breach > contract, so it's only a matter of time before we have a willing > candidate :) > I never plan to agree to the contract in the first place, so I won't be breaching it. On Sun, Jan 17, 2010 at 10:50 AM, Ævar Arnfjörð Bjarmason wrote: > > The theory is that the data isn't copyrightable and therefore the > CC-BY-SA didn't apply in the first place since it is a copyright > license. > Whose theory is that? It isn't mine. My theory is that the proposed contributor terms, which explicitly grant essentially everyone permission to do essentially everything, is tantamount to a dedication into the public domain (better, in fact, because it works in some jurisdictions where public domain dedications aren't permitted). On Sun, Jan 17, 2010 at 10:54 AM, Frederik Ramm wrote: > Sites that depend on contracts will usually have something in their terms > and conditions that says "if you're not legally able to enter a contract > then go away". (We have discussed this and most OSMers would not really like > to have terms and conditions that basically shut out minors from the > project. I don't know what will happen.) > Yeah, I can't see that happening. But I think you're missing the fact that sites which try to restrict people from copying their databases pretty much universally do not provide database dumps. A terms and conditions which says "if you're not legally able to enter a contract then go away" only works if it's combined with a technical means to catch people trying to systematically download everything (who can then be charged with computer trespass which is much more severe than a mere breach of contract, and applies to everyone including minors). On Sun, Jan 17, 2010 at 10:54 AM, Ævar Arnfjörð Bjarmason wrote: > You seem to be forgetting that Google has terms of service for map > maker: http://www.google.com/mapmaker/mapfiles/s/terms_mapmaker.html > > "By submitting User Submissions to the Service, you give Google a > perpetual, irrevocable, worldwide, royalty-free, and non-exclusive > license to reproduce, adapt, modify, translate, publish, publicly > perform, publicly display, distribute, and create derivative works of > the User Submission. " > > OpenStreetMap however does not > Not yet, but: "You hereby grant to OSMF and any party that receives Your Contents a worldwide, royalty-free, non-exclusive, perpetual, irrevocable license to do any act that is restricted by copyright over anything within the Contents, whether in the original medium or any other." http://www.osmfoundation.org/wiki/License/Contributor_Terms My comments are based on the axiom that OSMF adopts those contributor terms. If the switch to ODbL (which includes contributor terms, and I was told that this 0.9 draft is likely to be equivalent to the 1.0 release as far as this clause is concerned) does not go through, then everything will be CC-BY-SA, and the whole PD flagging thing makes more sense. If someone wants to bring up this proposal again *after* the failure of the switch to ODbL, fine. But let's handle one misguided and not-well-thought-out proposal at once. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Using editors to indicate license preference.
On Sun, Jan 17, 2010 at 10:49 AM, Frederik Ramm wrote: > whereas if the data is not copyrighted, but given to me under a contract > that stipulates that I may not put it up on a web site and say "download and > use freely" then > > * I am in breach of contract > * anyone who downloads data from my site is not in breach of anything > And what if you're under the age of majority, and can't be bound by a contract in the first place? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Using editors to indicate license preference.
On Sun, Jan 17, 2010 at 10:47 AM, Anthony wrote: > On Sun, Jan 17, 2010 at 10:44 AM, Frederik Ramm wrote: > >> Anthony wrote: >> >>> How exactly does one get protection as a database owner? It's unclear to >>> me how OSMF would get protection as a database owner since they're not the >>> ones actually doing anything. >>> >> >> You're right, just like Google doesn't own any of the data the GMM users >> upload because Google doesn't do anything. >> > > Yes, I know I'm right. And so does Google. It's why Google doesn't host > database dumps or provide an unrestricted API for GMM. > By the way, what am I right about. I asked a question, which I don't know the answer to. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Using editors to indicate license preference.
On Sun, Jan 17, 2010 at 10:42 AM, DavidD wrote: > 2010/1/17 John Smith : > > 2010/1/18 Anthony : > >> I didn't say it's invalid so much as it's redundant. > >> > >> All contributions are effectively PD anyway. > > > > That still isn't the point, people want to produce PD data that is > > readily accessible to all, not PD data shrink wraped with another > > license. > > OSM has masses of CC-BY-SA data and contributors. How will the PD > people deal with that? Well, how will the ODbL people deal with that? If OSM switches to ODbL, all the CC-BY-SA-only data has to go into the trash can anyway. On Sun, Jan 17, 2010 at 10:44 AM, Frederik Ramm wrote: > Anthony wrote: > >> How exactly does one get protection as a database owner? It's unclear to >> me how OSMF would get protection as a database owner since they're not the >> ones actually doing anything. >> > > You're right, just like Google doesn't own any of the data the GMM users > upload because Google doesn't do anything. > Yes, I know I'm right. And so does Google. It's why Google doesn't host database dumps or provide an unrestricted API for GMM. That said, Google does a lot more than OSMF. At least Google chooses what types of data to include/exclude. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
