Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
2010/8/18 Steve Bennett : >[...] > > Seriously, guys, creating nodes with zero tags attached, and > attempting to express meaning through them? That's dumb. It's hard > enough managing the various confusing meanings of actual tags, without > having to mindread your way through the absence of them... > > Steve > +1 For showing that the road continues I make a small stump of highway=road, fixme=continuation which meaning should be quite obvious to others I hope. It renders, so you can even see on the map that "aha, there's a road here that just hasn't been mapped yet" http://osm.org/go/0eeuQyGu-- This also gives the benefit of routers correctly adressing roundabouts (so it says "take the second exit" instead of saying "take the first exit" just because the real first exit isn't mapped yet) or junctions (so it says "at the junction turn left" instead of saying "at the end of the road, at the T-crossing, turn left" just because the road straight ahead isn't mapped yet). Konrad ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On Mon, Aug 16, 2010 at 5:05 PM, Aun Johnsen wrote: > In many cases I have seen empty (as in truely empty) nodes left around as a > result of failed imports, or failed uploads of changesets. Or Potlatch bugs. Seriously, guys, creating nodes with zero tags attached, and attempting to express meaning through them? That's dumb. It's hard enough managing the various confusing meanings of actual tags, without having to mindread your way through the absence of them... Steve ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Le mardi 17 août 2010 à 21:10, Pierre-Alain Dorange a écrit : > Empty node has no information, it must not be used to mean something ; > that's all. Yes, that's basically what I said in my first post. -- Renaud Michel ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Renaud MICHEL wrote: > > Is there any part of the earth which is "really empty" in terms of not > > having any possible landuse or natural tag to describe it? > > But putting nodes everywhere with "fixme missing landuse" would be a big > waste of time and resource. But it's not the purpose... Empty node has no information, it must not be used to mean something ; that's all. -- Pierre-Alain Dorange ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Le mardi 17 août 2010 à 19:06, vous avez écrit : > Is there any part of the earth which is "really empty" in terms of not > having any possible landuse or natural tag to describe it? But putting nodes everywhere with "fixme missing landuse" would be a big waste of time and resource. -- Renaud Michel ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Frederik Ramm wrote: > To me, the logical equivalent would be covering every > unmapped place in "fixme"s.) As said before, you can use "note" instead. But an empty region is diffreent from a road not finished (the initial three dots). As a road not finished (a todo job) is not empty for the user who made the 3 dots, so a fixme can also suit perfectly ? I never encounter 3 dots "to continue" a road, but sure a would assume it's a mistake and delete them is they got no tags. Now i know the meanings but i found this very strange... -- Pierre-Alain Dorange ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On Tue, Aug 17, 2010 at 12:58 PM, Renaud MICHEL wrote: > But if you simply see an empty region and have never been there (nor have > aerial images) then you have no idea if there is something missing there, or > if it is really empty. Is there any part of the earth which is "really empty" in terms of not having any possible landuse or natural tag to describe it? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Le mardi 17 août 2010 à 10:28, Peter Wendorff a écrit : > > Wrong because someone could try to build an object from them in the > > next step. > > As I said, that would be no problem, as long as this "someone" does not > download my changes in between - he already has his already uploaded > nodes in his editor. No, when they upload their ways, referring to the already uploaded nodes, they will get a conflict because the nodes they though still existed were deleted. > > [...] > > > > To me, the logical equivalent would be covering every unmapped place > > > > in "fixme"s.) > > Would that be wrong? > For me that sounds completely acceptable: missing data in the map should > be fixed - it's in some cases not worse than wrong data. > A white area in the map of course is sometimes obviously a todo hint, > but I don't see, why the fixme note would be wrong there. It is wrong if you don't know if there is actually something to fix. If you know there is some road somewhere in an empty region, you can add fixme's, you can even put the road name in the fixme if you know it. But if you simply see an empty region and have never been there (nor have aerial images) then you have no idea if there is something missing there, or if it is really empty. -- Renaud Michel ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Am 16.08.2010 15:13, schrieb Maarten Deen: But that automatically means that if you upload in multiple chunks and something goes wrong after nodes have been uploaded, it can occur that the nodes will be created but the ways not. And since most nodes that belong to ways have no tags, you end up with a bunch of untagged nodes. This can be solved by splitting the upload into spacial correlated pieces. A really smart unloader would cluster the data into tiles of 1x1km and upload each one individually (nodes & ways in one diff). Even for huge ways it is possible to upload 150 nodes and a way and then another 150 nodes and a new version of that way covering all 300 nodes. I remember some garmin map creation software that even creates those tiles based upon the number of objects in it, so that in areas with less data there would be also a decreased number uploads. Peter ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
> there are many empty nodes in the osm database. > > In the IRC channel i was told, that there are users who paint > empty nodes in the map to mark things like > "road is not mapped, but continues here" > > Is there already a decision about dealing with empty nodes? Yes, there is http://wiki.openstreetmap.org/wiki/Untagged_unconnected_node thanks to user 'Head' -- Jonas Stein ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On 16.08.2010 22:34, Frederik Ramm wrote: Hi, Peter Wendorff wrote: Deletion of attribute-less objects will never be a problem, as long as nobody tries to get information from that objects (not included from my point of view). Wrong because someone could try to build an object from them in the next step. As I said, that would be no problem, as long as this "someone" does not download my changes in between - he already has his already uploaded nodes in his editor. [...] To me, the logical equivalent would be covering every unmapped place in "fixme"s.) Would that be wrong? For me that sounds completely acceptable: missing data in the map should be fixed - it's in some cases not worse than wrong data. A white area in the map of course is sometimes obviously a todo hint, but I don't see, why the fixme note would be wrong there. As I said, I am not religious about this particular "personal touch" that people may have in mapping. What I dislike is the basic idea of creating rules that everyone must follow (combined with "but what's the PROBLEM in following my rule?"). Well - I didn't want to give a rule for that. In contrast I would like to leave it as it is - allowing you (and others) to use empty geometry to describe todo's and - on the other hand - allowing everybody to delete objects that has no observable information for the database. The latter includes allowing bots deleting empty nodes, just because they are often created by faults, too. We must create rules only as a last resort; only where there is no other way but for everyone to do the same. The real art is to identify the places where one must have rules, and leave anything else alone. Every extra rule makes OSM less good. That's my basic message - the "..." is just an example. If you want, don't take it as a rule - more like a hint: if you use empty nodes, there are people, who think, they are useless and can be deleted; and as they don't have information for everybody except you, that's true for the biggest part of the community. regards Peter ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Renaud MICHEL wrote: > I don't understand your message. > I think no one has suggested to add empty tags. Its seems (according to this thread) that some people used to add three empty nodes (no tag) to signal a way to be continued, instead of useing "fixme or note tag)... It's a very strange strategy. But most of emtpy nodes (no tags) are made by mistake (JOSM error or massive import error). -- Pierre-Alain Dorange ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On 16/08/2010, Frederik Ramm wrote: > Martin, > > MP wrote: > > > You will make some specific check for three continuous dots? Well, > > there are areas with thousands of such orphaned nodes and trying to > > check if there are somewhere three dots in a line in such areas > > wouldn't be easy - either the check will be very slow, prone to > > errors, or you will need some complex and sophisticated algorithm to > > make it at least somewhat reliable. > > > > Yes, I think a re-think is in order regarding the validators or at *least* > those with a direct influence on editing like the JOSM validator. When they > were introduced, people were relatively sure about what they were doing and > the validator was just an "ummm, not sure if this is right...?" voice, to be > taken with a grain of salt. But nowadays, too many take the validators for > gospel, and refrain from making legitimate edits because a validator flags > them up. Today, validators should be more cautious about what they flag. Perhaps alter the default validator settings and make some potentially more destructive check turned off. In case of solitary untagged nodes, there could be some minimum-age setting (default could be day or perhaps few days) and validator will report only older nodes as untagged and unconnected - since if the node is too new, there is chance that it is part of some import and the way will be added later (it can be sometimes even few days gap between adding lot of nodes and ways that use these nodes in case of large imports over slow connections) - so the node should not be deleted, at least not immediately. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Le lundi 16 août 2010 à 22:35, Dave F. a écrit : > It looks like some people left their common sense at home today. > Deliberately adding empty tags, indeed. What a ridiculous notion. Please > stop it. I don't understand your message. I think no one has suggested to add empty tags. We were talking about nodes without any tag, and many people suggested that if those nodes are supposed to be useful for anything, then they should be tagged with (at least one) meaningful tag. -- Renaud Michel ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On 17 August 2010 06:34, Frederik Ramm wrote: > (Personally I think a fixme is too strong - it sounds like there is > something "broken" that needs to be "fixed" whereas I simply want to point > out that there's something there which has not yet been mapped. To me, the > logical equivalent would be covering every unmapped place in "fixme"s.) Would note=* be better? ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On 16/08/2010 15:23, Pierre-Alain Dorange wrote: Willi wrote: Looks like making a mountain out of a molehill. +1 I agree also. It looks like some people left their common sense at home today. Deliberately adding empty tags, indeed. What a ridiculous notion. Please stop it. Dave F. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Hi, Peter Wendorff wrote: Deletion of attribute-less objects will never be a problem, as long as nobody tries to get information from that objects (not included from my point of view). Wrong because someone could try to build an object from them in the next step. Of course the tactics like "I make three dots to mark for myself, that it needs further work there" will fail in that cases. But what's the problem in adding a fixme-attribute to that data? There is no "problem" in adding a fixme attribute. Many people indeed do it because they like it better. Some people put three dots because they like that better. We don't have to force everyone to do it the same way. If I do three dots and I find that another mapper in my area removes them, I'll talk to him and then he'll hopefully understand. If I map in an area where my dots are removed all the time, I'll probably start using a fixme. (Personally I think a fixme is too strong - it sounds like there is something "broken" that needs to be "fixed" whereas I simply want to point out that there's something there which has not yet been mapped. To me, the logical equivalent would be covering every unmapped place in "fixme"s.) As I said, I am not religious about this particular "personal touch" that people may have in mapping. What I dislike is the basic idea of creating rules that everyone must follow (combined with "but what's the PROBLEM in following my rule?"). We must create rules only as a last resort; only where there is no other way but for everyone to do the same. The real art is to identify the places where one must have rules, and leave anything else alone. Every extra rule makes OSM less good. That's my basic message - the "..." is just an example. Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09" E008°23'33" ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
At 2010-08-16 11:44, Sebastian Klein wrote: If that's the case, I observe: Deletion of attribute-less objects will never be a problem, as long as nobody tries to get information from that objects (not included from my point of view). The point of this discussion: There are people who get information from 3 empty nodes at the end of a way. ___ ... Is this documented somewhere? I've removed untagged nodes that are scattered around from the various bugs and anomalies that have been cited. If I've removed someone's graphical ellipses, I'm sorry, but I've had no reasonable way to understand that they meant anything. I think this is a poor fit for the OSM tagging data model. Acceptance of it could be used to defend people drawing other untagged constellations, like question marks and exclamation points, instead of using the fundamental accepted scheme (tagging) to communicate what they are doing. It is another way of saying "fixme=continue" on that way. But it is easy in JOSM, for example, to create a toolbutton that does this with one mouseclick. Although I consider this style a little old fashioned, we shouldn't destroy other people's work, just because we don't like the way they are doing things. Agreed, though we don't have to condone it's continued use. So why not let validator detect these cases when searching for dup nodes? There shouldn't be a problem with that... It's certainly quite a bit harder to decide if there are nearby untagged nodes from the same changeset. Wouldn't it make more sense to write a bot to find these and tag them than to burden the validator forever with this special code? Should we at least vote on it's continued use? Do people that use it really feel that strongly? Sebastian ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk -- Alan Mintz ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Peter Wendorff wrote: I thought about the history feature the OSM has and wonder, if the deletion of empty nodes normaly should be a problem. Please correct me, if I'm wrong, but AFAIK the OSM works as follows: - an object is created, normally - but not necessary with attributes. - an object can be deleted any time - if a deleted object is touched again by changing it's attributes, it will be "recreated" so it's present in the database again. That is correct. If that's the case, I observe: Deletion of attribute-less objects will never be a problem, as long as nobody tries to get information from that objects (not included from my point of view). The point of this discussion: There are people who get information from 3 empty nodes at the end of a way. ___ ... It is another way of saying "fixme=continue" on that way. Although I consider this style a little old fashioned, we shouldn't destroy other people's work, just because we don't like the way they are doing things. So why not let validator detect these cases when searching for dup nodes? There shouldn't be a problem with that... Sebastian ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Hi. I'm new to this list, invited to it because I sayed something to this topic in IRC. I'm not answering to Frederiks Mail in particular, but didn't get the whole topic, so I cannot answer to the original question simply. I thought about the history feature the OSM has and wonder, if the deletion of empty nodes normaly should be a problem. Please correct me, if I'm wrong, but AFAIK the OSM works as follows: - an object is created, normally - but not necessary with attributes. - an object can be deleted any time - if a deleted object is touched again by changing it's attributes, it will be "recreated" so it's present in the database again. At least that's something somebody explained me a few month's ago. If that's the case, I observe: Deletion of attribute-less objects will never be a problem, as long as nobody tries to get information from that objects (not included from my point of view). Of course the tactics like "I make three dots to mark for myself, that it needs further work there" will fail in that cases. But what's the problem in adding a fixme-attribute to that data? I think, at long sight we need a kind of atomicity for changes, but even today I don't see a problem at deleting empty, non-tagged nodes as they contain no useful situation - perhaps except for the one, who created it. I'm not sure wether empty nodes should be deleted automatically or semi-automatically, but there are a lot of bots fixing bugs at tagging automatically - for me empty nodes are most useless of all parts. Please consider: Everything in this Mail is based on the assumption about deleting/adding above. If that assumption is not true, forget the rest. Regards Peter Wendorff On 16.08.2010 19:29, Frederik Ramm wrote: Martin, MP wrote: You will make some specific check for three continuous dots? Well, there are areas with thousands of such orphaned nodes and trying to check if there are somewhere three dots in a line in such areas wouldn't be easy - either the check will be very slow, prone to errors, or you will need some complex and sophisticated algorithm to make it at least somewhat reliable. Yes, I think a re-think is in order regarding the validators or at *least* those with a direct influence on editing like the JOSM validator. When they were introduced, people were relatively sure about what they were doing and the validator was just an "ummm, not sure if this is right...?" voice, to be taken with a grain of salt. But nowadays, too many take the validators for gospel, and refrain from making legitimate edits because a validator flags them up. Today, validators should be more cautious about what they flag. Bye Frederik ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Martin, MP wrote: You will make some specific check for three continuous dots? Well, there are areas with thousands of such orphaned nodes and trying to check if there are somewhere three dots in a line in such areas wouldn't be easy - either the check will be very slow, prone to errors, or you will need some complex and sophisticated algorithm to make it at least somewhat reliable. Yes, I think a re-think is in order regarding the validators or at *least* those with a direct influence on editing like the JOSM validator. When they were introduced, people were relatively sure about what they were doing and the validator was just an "ummm, not sure if this is right...?" voice, to be taken with a grain of salt. But nowadays, too many take the validators for gospel, and refrain from making legitimate edits because a validator flags them up. Today, validators should be more cautious about what they flag. Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09" E008°23'33" ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Willi wrote: > Looks like making a mountain out of a molehill. +1 > All this could easily be avoided by following some simple polite rules: > - Don't leave deliberately untagged objects. > Use at least a note or fixme to avoid that others waste their time. > - Don't touch anything which has been changed within the last day by another > mapper. > The person might still uploading or working on it. > - Don't touch more complex objects for up to one month after change by > another mapper. > - Consider to contact the mapper at least for complex objects. > > Improvements of these rules are highly welcome. Don't wait one month in this > case ;) Those "rules" seems very fine. -- Pierre-Alain Dorange ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
> > In the IRC channel i was told, that there are users who paint > > empty nodes in the map to mark things like > > "road is not mapped, but continues here" > > > > I do this occasionally, and I'm sure I haven't made this up but got the > practice from someone/somewhere else - when a way is drawn and you know it > goes on but haven't mapped it, you put three "dots", just as you do in > written language: > > --- . . . I've seen "fixme=road continues" tag attached on the last node of the way in these cases. Three dots are quite prone to deletion ... > I'm not religious about it but I think it is pretty elegant because it does > not require language to explain it - or at least that's what I thought until > I heard from several people that they "delete empty nodes on sight" without > further thought. I delete them too, but not without further thought. Usually, if the node is old enough (week or more) then it is not part of some ongoing import (or the import failed) and I usually delete them. Looking at the changets in which the node belong help with determining why the orphaned node is there ... > I mean - an empty node somewhere in the middle of town which has sat there > for ages, ok, but if you saw something like the above, where the three nodes > clearly hint at a way continuation - would you really remove them? I'd think > that a bit careless. using "fixme=road continues" is more stable solution for marking this and unlike three dots, this case will show up as warning in validator - so you can get hint that something needs improvement, in this case probably a survey or tracing from aerial imagery or whatever. Three dots are easy to get deleted accidentaly and easy to miss. > In fact, I'm going to fix the JOSM validator to detect these cases and not > complain - it is too easy for people to thoughtlessly hit "fix errors" and > thus inadvertently remove information. You will make some specific check for three continuous dots? Well, there are areas with thousands of such orphaned nodes and trying to check if there are somewhere three dots in a line in such areas wouldn't be easy - either the check will be very slow, prone to errors, or you will need some complex and sophisticated algorithm to make it at least somewhat reliable. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Looks like making a mountain out of a molehill. All this could easily be avoided by following some simple polite rules: - Don't leave deliberately untagged objects. Use at least a note or fixme to avoid that others waste their time. - Don't touch anything which has been changed within the last day by another mapper. The person might still uploading or working on it. - Don't touch more complex objects for up to one month after change by another mapper. - Consider to contact the mapper at least for complex objects. Improvements of these rules are highly welcome. Don't wait one month in this case ;) Happy mapping Willi ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Jonas Stein wrote: > > Yes, but JOSM has a chunked upload mode where it uses multiple > > transactions for one upload. > > Can someone confirm, that josm first uploads an empty node > and then the content? Can't confirm ; be it would be very strange to upload an empty node, then the tag associate... or an empty node then then way associated. But it coudl happen for ways... > If split uploads contain this emty ones at the moment lets summarise: > > There could be empty nodes during splitted uploads. > So nobody should delete nodes that are younger then a given timespan. The splitted upload has a very short life. I notice that in my area at a moment a create lot to empty nodes (would have been associated with a building) when i try to upload a big changeset (but JOSM failed) so then i use the split upload function from JOSM and never got the same problem. >From those expériences, i could say that big upload with JOSM could lead to dupplicate nodes and empty nodes ; split upload are safer. > ps: uploads should not split within empty nodes, > but thats no topic here for now. It can't be done, there is circompstance that would lead to empty nodes upload. Imagine a large way (more than 300 nodes). If JOSM upload with a split parameter of 200, then it would split during the way+nodes upload and if for any reason the upload stop there empty nodes have been transmit to the server... -- Pierre-Alain Dorange ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
(1) It's not exactly more reliable with dedicated bulk upload scripts > either. If the API takes too long to check the uploaded osmChange for > validity, the TCP session appears to timeout. The script/JOSM never receives > the OK from the API, including the new object IDs. The next time you hit > upload to resume, it will reupload that failed chunk in its entirety, > leading to (in my example) 5k duplicate objects on the server. > I have often tried to upload smaller or larger chunks from slow/unstable connections, and have experienced various problems, I have even come to the point where I have had ~100 items left when the connection have timed out, with the result that the entire upload had to be done again. The only way to fix it afterwards is to do a code validation of the area, but with these unstable lines that I suffer from time to time, that doesn't solve much either. Having JOSM or similar intelligently combine nodes and ways in these chunks, as well as accepted that "all but the last" have been transferred correctly, than that would have helped imensely on the end result. This makes sense in in certain situations. E.g. you upload 2 objects > and you get "precondition failed" on the last 100 of them. That would mean: > (1) - upload 19899+i objects > (2) - upload is aborted by server -> get the server error > (3) - fix the problem (e.g. download way and fix it) > (4) - i++ > (5) - goto (1) > > If you have a slow connection, this is not acceptable. > > This is exactly what I am talking about ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On Mon, Aug 16, 2010 at 9:51 AM, Dave Stubbs wrote: In the IRC channel i was told, that there are users who paint empty nodes in the map to mark things like "road is not mapped, but continues here" >>> >>> I do this occasionally, and I'm sure I haven't made this up but got >>> the practice from someone/somewhere else - when a way is drawn and you >>> know it goes on but haven't mapped it, you put three "dots", just as >>> you do in written language: >>> >>> --- . . . > > I've no idea where I first came across it, but I have also used it as > far back as 2006. We didn't have any aerial imagery to trace, so if > nobody had walked down that way with a GPS you had no idea where it > was going, but you might have a name or something from the end, so you > put in a small stub road and dot dot dot. There also wasn't much of a > map viewer, or styled editing, so three dots was a lot more obvious > when editing than a tag. Here's some documentation on it from 2007, along with some other styles in common use back then (e.g. the arrowhead made from untagged segments) http://wiki.openstreetmap.org/wiki/Talk:Proposed_features/More_to_be_Mapped ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
> Yes, but JOSM has a chunked upload mode where it uses multiple > transactions for one upload. Can someone confirm, that josm first uploads an empty node and then the content? If split uploads contain this emty ones at the moment lets summarise: There could be empty nodes during splitted uploads. So nobody should delete nodes that are younger then a given timespan. ps: uploads should not split within empty nodes, but thats no topic here for now. -- Jonas Stein ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
> --- . . . > In fact, I'm going to fix the JOSM validator to detect these cases and > not complain - it is too easy for people to thoughtlessly hit "fix > errors" and thus inadvertently remove information. -1 i fear that is the wrong approach. Please dont ignore the results of this Thread. Dont ignore the community it would be a new bug in the validator. -- Jonas Stein ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On 16/08/2010 07:48, Maarten Deen wrote: On Mon, 16 Aug 2010 08:19:42 +0200, Frederik Ramm wrote: Hi, Jonas Stein wrote: In the IRC channel i was told, that there are users who paint empty nodes in the map to mark things like "road is not mapped, but continues here" I do this occasionally, and I'm sure I haven't made this up but got the practice from someone/somewhere else - when a way is drawn and you know it goes on but haven't mapped it, you put three "dots", just as you do in written language: --- . . . Hi That seems like a pointless waste of time that to be & gives no clarity as to what your intentions are. The way I do it: Draw a small section of the way for as for as you could see it. Tag it correctly for what it is. add fixme tag with 'stub of way, please resurvey' http://wiki.openstreetmap.org/wiki/Key:fixme Cheers Dave F. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Peter Körner wrote: Am 16.08.2010 02:50, schrieb Lennard: On 16-8-2010 1:41, M∡rtin Koppenhoefer wrote: We've had several cases today, where people came swooping in and deleted nodes from an ongoing upload, where the ways hadn't been uploaded yet. So why are you uploading Nodes without ways? The POST /api/0.6/changeset/#id/upload call is atomic in a transaction. Why not split your upload into multiple OSC Parts and post thems via this call. That way no imcomplete data would be visible to other users at any time. Yes, but JOSM has a chunked upload mode where it uses multiple transactions for one upload. This makes sense in in certain situations. E.g. you upload 2 objects and you get "precondition failed" on the last 100 of them. That would mean: (1) - upload 19899+i objects (2) - upload is aborted by server -> get the server error (3) - fix the problem (e.g. download way and fix it) (4) - i++ (5) - goto (1) If you have a slow connection, this is not acceptable. Sebastian ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
** Adding josm-dev to the list. Please post technical follow-ups there. ** On 16-8-2010 12:15, Peter Körner wrote: The POST /api/0.6/changeset/#id/upload call is atomic in a transaction. Why not split your upload into multiple OSC Parts and post thems via this call. That way no imcomplete data would be visible to other users at any time. Uploading 30k+ objects in a single chunk with JOSM(1) is just too unreliable to make that workable. So either we have to split the data in smaller chunks by hand, or use JOSM's native chunked upload mode. If you have 40k nodes and 5k ways, and upload in 5k chunks, you will upload 8 chunks with nodes, and 1 chunk with ways. Each chunk is atomic, and that's where atomicity ends, as far as the API is involved. JOSM makes no attempt to sort the data in a smart way, to keep all nodes and associated ways and relations close together, in the same chunk when possible. I asked about such a feature before(2), but nothing has come of it as of yet. If such a sorting feature is added to JOSM, the chunk size should be a soft size, able to vary slightly if that means related objects end up in the same chunk. May I point out smarter-sort.py(3)(4) as an example? Sorted uploads would mostly prevent these 'fields of empty nodes' that appear to other mappers during a chunked upload, limiting the opportunity they have to wreak havoc on an ongoing upload by 'helpfully' deleting the nodes. JOSM's chunked upload mode is an answer to API timeout issues, but it does have its own issues to keep in mind. (1) It's not exactly more reliable with dedicated bulk upload scripts either. If the API takes too long to check the uploaded osmChange for validity, the TCP session appears to timeout. The script/JOSM never receives the OK from the API, including the new object IDs. The next time you hit upload to resume, it will reupload that failed chunk in its entirety, leading to (in my example) 5k duplicate objects on the server. (2) http://josm.openstreetmap.de/ticket/4299 (3) http://wiki.openstreetmap.org/wiki/Upload.py (4) http://svn.openstreetmap.org/applications/utils/import/bulkupload/smarter-sort.py -- Lennard ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
> I'd never heard of it before, but I like it too. Maybe the editor > could automatically attach a fixme=yes tag if the user does > triple-click and moves (even better if the editor added fixme=continue > when someone does triple-click / move a short way / triple-click / > move a short way / triple-click, but that's probably too much like > hard work). in Josm you could create your own icon in the toolbar for that: 1. make node 2. click on icon - that will set fixme=street contiues here http://wiki.openstreetmap.org/wiki/DE:Anpassen_der_Vorlagen_von_JOSM (somewhere should be the english version, but i did not find it) -- Jonas Stein ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
>> I do this occasionally, and I'm sure I haven't made this up but got >> the practice from someone/somewhere else - when a way is drawn and you >> know it goes on but haven't mapped it, you put three "dots", just as >> you do in written language: >> >> --- . . . > It's the first that I heard of this strategy and I'm not sure if I > would recognize it. I certainly haven't in the past. > > It does raise a question: why not just map a way over it and tag it > with some FIXME? If I map a new area and make photo's and see that there > is a road somewhere that I didn't go, I map the road as far as I can see > it and. +1 >> I mean - an empty node somewhere in the middle of town which has sat >> there for ages, ok, but if you saw something like the above, where the >> three nodes clearly hint at a way continuation - would you really >> remove them? I'd think that a bit careless. > > I would at least tag the nodes with a FIXME. Personally I do make a > point of looking at the history of a stray node, but that is far from > failsafe. +1 its documented here and well-established. http://wiki.openstreetmap.org/wiki/Key:fixme Only if a node contains information it carries information. An empty node is as usefull like 42. OSM is a community-project, so it means no one should insert information that can not be used by others. If s.b. searches for missing streets in the region how should one search for hints by other mappers? Search for all empty nodes? May be others use 4 empty nodes for a building and 5 for an missing area and then there are trillions of empty nodes from broken software. Many applications make usefull things with fixme-nodes * there are markers on gps-devices * websites highlighting fixme * josm is shipped with some extra support for fixme-nodes * and lots of more Another well-established way is to use openstreetbugs: http://openstreetbugs.schokokeks.org/ Everyone can write here "street contiues 2km to west" -- Jonas Stein ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Am 16.08.2010 02:50, schrieb Lennard: On 16-8-2010 1:41, M∡rtin Koppenhoefer wrote: We've had several cases today, where people came swooping in and deleted nodes from an ongoing upload, where the ways hadn't been uploaded yet. So why are you uploading Nodes without ways? The POST /api/0.6/changeset/#id/upload call is atomic in a transaction. Why not split your upload into multiple OSC Parts and post thems via this call. That way no imcomplete data would be visible to other users at any time. Peter ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On Mon, Aug 16, 2010 at 9:51 AM, Dave Stubbs wrote: > In the world of validators, slippy maps, JOSM post mappaint, aerial > imagery and masses of POI/addresses it makes less sense, and is less > visually obvious. But I still like it :-) I'd never heard of it before, but I like it too. Maybe the editor could automatically attach a fixme=yes tag if the user does triple-click and moves (even better if the editor added fixme=continue when someone does triple-click / move a short way / triple-click / move a short way / triple-click, but that's probably too much like hard work). Richard ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On Mon, Aug 16, 2010 at 7:48 AM, Maarten Deen wrote: > On Mon, 16 Aug 2010 08:19:42 +0200, Frederik Ramm > wrote: >> Hi, >> >> Jonas Stein wrote: >>> In the IRC channel i was told, that there are users who paint >>> empty nodes in the map to mark things like >>> "road is not mapped, but continues here" >> >> I do this occasionally, and I'm sure I haven't made this up but got >> the practice from someone/somewhere else - when a way is drawn and you >> know it goes on but haven't mapped it, you put three "dots", just as >> you do in written language: >> >> --- . . . >> >> I'm not religious about it but I think it is pretty elegant because >> it does not require language to explain it - or at least that's what I >> thought until I heard from several people that they "delete empty >> nodes on sight" without further thought. > > It's the first that I heard of this strategy and I'm not sure if I > would recognize it. I certainly haven't in the past. > > It does raise a question: why not just map a way over it and tag it > with some FIXME? If I map a new area and make photo's and see that there > is a road somewhere that I didn't go, I map the road as far as I can see > it and. > I've no idea where I first came across it, but I have also used it as far back as 2006. We didn't have any aerial imagery to trace, so if nobody had walked down that way with a GPS you had no idea where it was going, but you might have a name or something from the end, so you put in a small stub road and dot dot dot. There also wasn't much of a map viewer, or styled editing, so three dots was a lot more obvious when editing than a tag. In the world of validators, slippy maps, JOSM post mappaint, aerial imagery and masses of POI/addresses it makes less sense, and is less visually obvious. But I still like it :-) Dave ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
In many cases I have seen empty (as in truely empty) nodes left around as a result of failed imports, or failed uploads of changesets. Most of these nodes should be removed, and the validator plugin in JOSM allows this in an easy way. On the other hands, nodes with only user information tags, such as note, comment, and FIXME should be kept, or the information within the tags should be dealt with. These nodes can work as reminders for people tagging in a certain area, such as "need to complete this road", or "trace this lake", etc. A On Mon, Aug 16, 2010 at 7:48 AM, Maarten Deen wrote: > On Mon, 16 Aug 2010 08:19:42 +0200, Frederik Ramm > wrote: > > Hi, > > > > Jonas Stein wrote: > >> In the IRC channel i was told, that there are users who paint > >> empty nodes in the map to mark things like > >> "road is not mapped, but continues here" > > > > I do this occasionally, and I'm sure I haven't made this up but got > > the practice from someone/somewhere else - when a way is drawn and you > > know it goes on but haven't mapped it, you put three "dots", just as > > you do in written language: > > > > --- . . . > > > > I'm not religious about it but I think it is pretty elegant because > > it does not require language to explain it - or at least that's what I > > thought until I heard from several people that they "delete empty > > nodes on sight" without further thought. > > It's the first that I heard of this strategy and I'm not sure if I > would recognize it. I certainly haven't in the past. > > It does raise a question: why not just map a way over it and tag it > with some FIXME? If I map a new area and make photo's and see that there > is a road somewhere that I didn't go, I map the road as far as I can see > it and. > > > I mean - an empty node somewhere in the middle of town which has sat > > there for ages, ok, but if you saw something like the above, where the > > three nodes clearly hint at a way continuation - would you really > > remove them? I'd think that a bit careless. > > I would at least tag the nodes with a FIXME. Personally I do make a > point of looking at the history of a stray node, but that is far from > failsafe. > > Regards, > Maarten > > ___ > 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] RFC: what are empty nodes and how should we use them?
On Mon, Aug 16, 2010 at 8:10 AM, Renaud MICHEL wrote: > Maybe most of those empty nodes are remnants from some time ago, when some > editors would delete a way, but not the nodes it contained (I think there > used to be such a bug, even before I started contributing to OSM). Empty nodes can also be left behind if you alter a way to update it and in the process delete some nodes from the way, but then in JOSM instead of doing upload data you choose upload selection with the way that you changed selected. Then those node deletion actions don't get sent to the server and your left with these empty nodes. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Hi, Jonas Stein wrote: In the IRC channel i was told, that there are users who paint empty nodes in the map to mark things like "road is not mapped, but continues here" I do this occasionally, and I'm sure I haven't made this up but got the practice from someone/somewhere else - when a way is drawn and you know it goes on but haven't mapped it, you put three "dots", just as you do in written language: --- . . . I'm not religious about it but I think it is pretty elegant because it does not require language to explain it - or at least that's what I thought until I heard from several people that they "delete empty nodes on sight" without further thought. I mean - an empty node somewhere in the middle of town which has sat there for ages, ok, but if you saw something like the above, where the three nodes clearly hint at a way continuation - would you really remove them? I'd think that a bit careless. In fact, I'm going to fix the JOSM validator to detect these cases and not complain - it is too easy for people to thoughtlessly hit "fix errors" and thus inadvertently remove information. Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09" E008°23'33" ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On Mon, Aug 16, 2010 at 7:42 AM, Jonas Stein wrote: > there are many empty nodes in the osm database. > > In the IRC channel i was told, that there are users who paint > empty nodes in the map to mark things like > "road is not mapped, but continues here" Personally, I delete them. If people want to map the above, the tag "fixme=continues" is pretty common. Steve ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On 16-8-2010 1:41, M∡rtin Koppenhoefer wrote: +1, at least if they are older than some days/weeks. If someone is doing a very big upload it might be (dependant on the way he structures the upload) that he is first uploading nodes and only later the ways. That's why I wouldn't delete recently created empty nodes. We've had several cases today, where people came swooping in and deleted nodes from an ongoing upload, where the ways hadn't been uploaded yet. Okay, it happens, wish JOSM would sort the upload on object locality[1]. In this case it was just a mapper that didn't know what that 'field of nodes' was all about. Then there was another, where someone came in about 40 minutes after a botched upload, while the uploader was still trying to repair this and attempting to reuse those nodes where possible, and started his script to 'helpfully' delete those 'duplicated nodes'. I just wish those scripts would only process nodes that have been idle for a few days, and not very fresh ones. [1] http://josm.openstreetmap.de/ticket/4299 -- Lennard ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On Mon, 16 August, 2010 9:36:50 AM, Nathan Edgars II wrote: > There's still such a bug: http://trac.openstreetmap.org/ticket/2700 This often happens to me in Potlatch. Here's my usual use case. An intersection of two streets needs to be converted to a roundabout. 1. Split each street at the intersection. 2. For each of the 4 ways, select the way and the intersection, hit backspace and reposition the end of the way where it should intersect the roundabout, forming a diamond. 3. Create a new way for the roundabout, incorporating each of the 4 ends of the ways. 4. Tag the roundabout appropriately (e.g. junction=roundabout, highway=residential, source=nearmap, maxspeed=50) 5. Save and quit Potlatch. I then come back some time later and find that there is an empty node in the place where the intersection originally was. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Renaud MICHEL-2 wrote: > > Hello > Le dimanche 15 août 2010 à 23:42, Jonas Stein a écrit : >> Validators claim empty nodes are defective, but are they right? >> Is painting with empty nodes data that we want to have in the database? >> Are there any empty nodes that make sense, or is a empty always node >> nonsense? > > I'd say the validators are right, because a node that is neither part of a > way, nor part of a relation, and has no tags is simply useless: we have no > idea why it sits there. > > If you put a note as a reminder of something, then you should at least put > a > note tag on it. > Note that JOSM's handling of nodes with only note tags is suboptimal: it puts them under the same heading (but a different subheading) as ones with only created_by (the latter should be treated as truly untagged). Renaud MICHEL-2 wrote: > > Maybe most of those empty nodes are remnants from some time ago, when some > editors would delete a way, but not the nodes it contained (I think there > used to be such a bug, even before I started contributing to OSM). > There's still such a bug: http://trac.openstreetmap.org/ticket/2700 -- View this message in context: http://gis.638310.n2.nabble.com/RFC-what-are-empty-nodes-and-how-should-we-use-them-tp5426109p5426331.html Sent from the General Discussion mailing list archive at Nabble.com. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
2010/8/16 Renaud MICHEL : >> Are there any empty nodes that make sense, or is a empty always node >> nonsense? > > I'd say the validators are right, because a node that is neither part of a > way, nor part of a relation, and has no tags is simply useless: we have no > idea why it sits there. +1, at least if they are older than some days/weeks. If someone is doing a very big upload it might be (dependant on the way he structures the upload) that he is first uploading nodes and only later the ways. That's why I wouldn't delete recently created empty nodes. cheers, Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
On 16 August 2010 07:42, Jonas Stein wrote: > Are there any empty nodes that make sense, or is a empty always node nonsense? The only thing I can think of , is that when I upload a way, the nodes go first, then the way joins them all up. Is it possible for somebody else to get the data while I'm still in the process up loading? If so, they may get empty nodes from that, while if they load again later the whole way may be there. I don't know if this is possible - we'd need to ask a dev how the upload works. I know that I've had a couple of uploads crash, and when I looked at the data later, a lot of nodes were there but the ways weren't. Stephen ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Yeah, I would tend to agree that empty nodes should be regarded as errors. That is how I spotted a problematic changeset the other day that really needed reverting. I know OSM emphasizes distributed tagging/usage but an empty node has absolutely no value to anyone except (maybe) the person who created it. And even then it is a pretty bad way of leaving a reminder for yourself. For the sake of the community, there should at least be a note or FIXME tag or *something* on it otherwise I view it as doing nothing but taking up space in the database. Toby ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] RFC: what are empty nodes and how should we use them?
Hello Le dimanche 15 août 2010 à 23:42, Jonas Stein a écrit : > Validators claim empty nodes are defective, but are they right? > Is painting with empty nodes data that we want to have in the database? > Are there any empty nodes that make sense, or is a empty always node > nonsense? I'd say the validators are right, because a node that is neither part of a way, nor part of a relation, and has no tags is simply useless: we have no idea why it sits there. If you put a note as a reminder of something, then you should at least put a note tag on it. Maybe most of those empty nodes are remnants from some time ago, when some editors would delete a way, but not the nodes it contained (I think there used to be such a bug, even before I started contributing to OSM). -- Renaud Michel ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] RFC: what are empty nodes and how should we use them?
there are many empty nodes in the osm database. In the IRC channel i was told, that there are users who paint empty nodes in the map to mark things like "road is not mapped, but continues here" Is there already a decision about dealing with empty nodes? I think we should discuss that on this list in english language, as it concerns all mappers worldwide. Validators claim empty nodes are defective, but are they right? Is painting with empty nodes data that we want to have in the database? Are there any empty nodes that make sense, or is a empty always node nonsense? Kind regards, -- Jonas Stein ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
