Re: [OSM-talk] permament access restrictions and routing
I guess not, this would mean that anybody (for example any tourist in a car) can use this road, as long as he ends the journey there (and parks the car inside the area) - but that is not the case, as only few individuals actually have the permit to go there. vehicle=destination is rougly equivalent to prohibitory traffic sign with "throught traffic forbidden" sign, forbidding entry to people who does not interrupt the journey (load/unload cargo, visit someone, etc ...) inside marked area. Not to "All vehicles forbidden, except those few with a permit". Martin On Mon, Dec 24, 2012 at 4:36 PM, Paul Johnson wrote: > Sounds like perhaps highway=pedestrian, vehicle=destination would be closer > to what we're looking for? > > > On Mon, Dec 24, 2012 at 6:32 AM, Martin Koppenhoefer > wrote: >> >> 2012/12/24 Maarten Deen : >> > On 2012-12-23 23:53, Chris Hill wrote: >> >> >> >> On 23/12/12 15:41, Stan Berka wrote: >> >>> >> >>> Yesterday, I did navigation to my work place here in Warsaw using >> >>> Osmand >> >>> o my G2. The route I was given led through Warsaw Old Town. The >> >>> problem >> >>> with this is, the Old Town is closed to most vehicles, year long, >> >>> except for >> >>> special vehicles (shop supply, city services etc). Thus, the route >> >>> was >> >>> completely useless since I'm not a firetruck driver. How can this >> >>> problem >> >>> be resolved, so OSM routers can generate more useful? >> >> >> >> that sounds like a highway=pedestrian to me. >> > >> > >> > You could add a access=destination to it. Then routing should be done >> > only >> > to addresses on those roads and the roads should not be used to go from >> > A to >> > B. >> >> >> You should't add access=destination to this, because the OP wrote that >> the road is closed for vehicles besides some exceptions like loading >> and emergency vehicles.,Access=destination would be allowing access >> for everyone who has to go to an address in the area. >> >> cheers, >> Martin >> >> ___ >> 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
[OSM-talk] Proposed mechanical edit: Empty Relations
I guess it should be ok if you limit the wipeout only on some carefully selected types of relation (for example type=multipolygon as multipoly without members really does not make sense), perhaps checking for other tags as well (i.e. delete only type=multipolygon without any other tags, or with certain whitelisted key=value parts like natural=forest ...) This will remove useless stuff while keeping stuff that may still be potentially used. I know of one occurrence of empty relation where having empty relation makes sense - in Prague transport, one route was temporarily disabled due to repairs, but was expected to be re-established soon, so the relation "route" remained, but was empty without members (so it won't lose tags). Later it was restored once that line went back to service. Though in this case the empty relation was still member of "network" relation, so it can't be deleted without breaking referential integrity. Martin On Fri, Apr 13, 2012 at 14:31, andrzej zaborowski wrote: > On 13 April 2012 11:12, Paul Norman wrote: >> Through editor errors or other mistakes there are a number of relations in >> OSM which have no members. I propose a mechanical edit to delete these where >> they are not members of some other way. My proposed procedure, documented at >> http://wiki.openstreetmap.org/wiki/Mechanical_Edits/pnorman_imports is to >> identify these with my pgsnapshot database and delete them if they aren't >> referenced by any other relation. >> >> If they are referenced by another relation I will investigate them and deal >> with them manually. >> >> I've investigated a few of them and they appear to be caused by people >> deleting ways from multipolygons but not deleting the relation. I don't want >> to investigate all 5200 by hand so I'm proposing the mechanical edit. >> >> I will filter out ones touched in the last 24 hours to avoid conflicting >> with anyone. > > This thread comes up every some time > (http://lists.openstreetmap.org/pipermail/dev/2009-August/016658.html, > http://lists.openstreetmap.org/pipermail/talk/2010-November/055075.html). > In 2009 I ran a very similar operation with 4.5k relations deleted as > discussed in the first thread I linked. The lesson from it was that > some relations are referenced from outside of the database, mainly > from the OSM wiki. Those should not be deleted automatically either > or it'll upset some users. > > Cheers > > ___ > 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] Orphaned Relations
On Tue, 19 Jul 2011 10:01:24 +0200, Sarah Hoffmann wrote: On Mon, Jul 18, 2011 at 10:28:54PM +0100, MP wrote: On Mon, 18 Jul 2011 16:05:29 +0100, Ed Loach wrote: Relations without members can be used intentionally, Can you give an example, please? Because I've tried and failed to think of any. Perhaps I'm just getting hung up on the name "relation" as something which groups its related members in some way defined by the relation's tags (while not being used as a category). For example in Prague there is relation for transport network which contain empty relation for some tram lines - those are not operational currently due to some constructions, but once these constructions are finished, they can be easily restored to previous state (just dig up some older version of the relation from history, check it and save it) You might just as well delete the relation while the tram line does not exist. You are still able to restore an old version from the history later. There is no difference in the process. Relations should not be used as a garbage dump for information that might be interesting in the future. Yes - if you remember the ID. If you don't, finding the ID can be almost impossible (there is no simple way to search for deleted given its tags) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Orphaned Relations
On Mon, 18 Jul 2011 16:05:29 +0100, Ed Loach wrote: Relations without members can be used intentionally, Can you give an example, please? Because I've tried and failed to think of any. Perhaps I'm just getting hung up on the name "relation" as something which groups its related members in some way defined by the relation's tags (while not being used as a category). For example in Prague there is relation for transport network which contain empty relation for some tram lines - those are not operational currently due to some constructions, but once these constructions are finished, they can be easily restored to previous state (just dig up some older version of the relation from history, check it and save it) There are quite rare uses for empty relation (but being a member of other relation) and mostly these uses are temporary. Since if you download some data by some given bbox, you will get that empty relation (along with some other data). But relations that contain no members and themselves are not members of nay other relation are "invisible" - they are not contained in any bounding box, they do not affect rendering, routing or anything and basically you can download them only by ID (or parse them from planet dump). So these are basically useless Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] We Need to Stop Google's Exploitation of Open Communities
On Tue, 12 Apr 2011 10:58:05 +0200, Erik Johansson wrote: On Tue, Apr 12, 2011 at 1:06 AM, Frederik Ramm wrote: Hi, Bernhard Zwischenbrugger wrote: Where can I access the OSM data? I know it is possible to download the hole planet, setup a database,... but that's not an easy task. Not easy, but possible, and done by literally hundreds of people all over the world. I would place downloading the planet on the same level as compiling a kernel. Not really hard, but something most Ubuntu users are scared of doing. Downloading is the easy part (if you have reasonably fast connection), but making actually some use to it is the worse part - you need lot of disk space depending on what you want to do (tens of GB with OSM3S, hundreds of GBs for importing into postgres, etc ...) and lot of CPU time (probably over a day) to convert/import the planet into something that can be further used. And if you get "lucky", the import process gets killed at 85% by OOM killer or some other fault and you can start again. But it is same for wikipedia - merely reading it on web is simple, but if you want an offline copy, you'll start with 6.7GB compressed XML dump and after spending some time with attempts to import this into local database, soon you'll find out that you need to download some .sql files to speed up the import (so wikimedia won't need to recalculate all the links between articles) and you need to somehow download all the images that are used (you need to write some script for that or perhaps check the net to find if someone already did that). Not exactly easy either. But unlike wikipedia, in OSM there are extracts, which have usually some reasonable size (you can pick a continent of interest, or just the country, or for some large countries like france and germany, extracts for regions are also available). Cutting smaller parts like entire city from the extract with osmosis is then quite easy and fast (several minutes, perhaps an hour at maximum) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] XAPI and using a home server
Triggered by your request, I've made a deployable version of OSM3S. It needs only modest hardware requirements (1 GB RAM and 40 GB hard disk space for the entire world). See This could be interesting, though having two simultanously updated copies may not be the best But I assume you can update semi-manually (from time to time run a script that will download and apply all minute diff from last update until now and once the database is finished updating, then query it... until you think the data are too old and re-run the update again) and then you will need only one database, right? http://wiki.openstreetmap.org/wiki/OSM3S/install It has a different syntax than XAPI and partly different capabilities, but if you work with map data and don't need metadata about users, you likely can use it. Hmmm ... in some cases the username and time may come handy (especially for queries like "find wrongly tagged stuff by user X before time T, where T is time when I send him message that he is doing something wrong") Is there option to turn these fields on at cost of some extra disk space? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Why isn't any XAPI server available ?
More XAPI servers running on good hardware is the only realistic solution. Well, there could perhaps be another solution, like running your own XAPI server - the minutely diffs are usually less than 100Kb, so the required bandwidth to download from planet.openstreetmap.org would be less than 2 Kb/second in average. But the question is - how large would be the planet database on disk (how large would it get once you import the planet dump) I guess the database would be in order of tens of gigabytes, probably over 100 GB ... And how much memory you need on the machine to run some reasonable queries (if 4 GiB works for main XAPI server, would it be usable on machine with only 1 GiB of memory?) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Why I don't use JOSM (was Re: Non-map-based OSM editor)
On Mon, 24 Jan 2011 18:24:48 +1100, Steve Bennett wrote: On Mon, Jan 24, 2011 at 10:25 AM, Anthony wrote: I'm not quite sure why, but I really don't like JOSM. In my case I think it boils down to: 1) Complicated, idiosyncratic user interface. My brain just doesn't have space to learn new interfaces. Everything needs to behave the same. 2) Fragility. I tried fiddling with a plug in once and everything just broke. 3) Poor performance. (On my environment anyway.) Performance is better than in the past (quadbuckets, varous other improvements, ...), but there are still places when JOSM is slow, especially once you have hundred of thousands or even millions of primitives loaded (even if you work only on small part of them at once). But Potlatch is much slower, at least for me, once there are several thousand primitives in view it will become quite unuseable. 4) General preference for online tools (so I don't need to synchronise environments across different computers) I am not sure for newer potlatch, but the few times I was forced to use it (why the hell there is undelete api available only for Potlatch and not as XML?), it had not remembered any settings (so I have for example to turn off the photo underlay again and again and again ...) Martin It looks like a fantastically powerful tool, and I wish all those who use it the very best. But I'm probably one of those few people who actually moved from JOSM *to* Potlatch. Steve ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Changeset 6234414
Does anybody have idea what it is? Seems to be large lot of untagged lines in middle of the ocean and I don't see what it could be (coral reefs? some under-sea features?). http://www.openstreetmap.org/browse/changeset/6234414 What should be done with similar uploads, if the ways are untagged and it is not very clear what they should be? Google translate told the summary is "map made in autocad and being passed on to the OSM" which isn't very explanatory about source and purpose of the data. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] tile downloader
> Maybe we could work around this by automatically changing the link for > the stored tiles? This would also harm "friendly" projects with small > tile-download-rates though. If it is technically possible to identify > this application they could also be filtered out. But what prevents the "evil" applications to set same user agent as good ones? Not a good idea. This would break many things, for example the slippy map preview in JOSM, without managing to do what is intended (stop bulk downloaders - as if one downloader does not work, user will just pick another that will - or wait for update to existing one) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Changeset ignoring API limits?
I looked at changeset http://www.openstreetmap.org/browse/changeset/5853571 and I noticed the page says "Has the following 79290 nodes: ... Has the following 15862 ways:" Isn't there supposed to be a limit of 5 elements per changeset (this one contains about 95000 elements)? I looked at http://www.openstreetmap.org/api/capabilities and I see there line: ... so the limits weren't changed Does anybody know how it is possible for that changeset to bypass these API limits? Or is it some bug in the element-counting code? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Exceeded API bandwidth limit, now what?
This bandwidth limit cutoff happened to me twice - once today and once about two weeks back when fixing duplicate data from imports. The workflow for fixes are generally to download relatively small piece of affected area (I use either script on planet extracts to discover which areas contain these duplicates or look at Matt's duplicate nodes map) and download the data - since you download data from duplicated imports, you download everything usually two or four times. You delete extra copies, do some small fixes to tags or some minor cleanup, upload (usually several thousands of deleted duplicate nodes per request) and then move to next part of the area. Generally, I request one small, but quite stuffed (often close to 5 nodes limit) area every 5-10 minutes or so (download (few secs), fix (about 2 minutes), upload (few secs or few mins of waiting for server), repeat). After few hours I got this ... bandwidth limit exceeded. I don't know what the limits are, but ocassionaly they hit even people who want to edit the data in large scale (me :), not just some leechers that want to get the city by scrapping server with repetitive requests. So maybe it could be slightly raised ... at least for people who upload fixed data back ... Note for Michal Migurski: for downloading relatively fresh areas outside of europe (for areas inside euroope it is usually simpler to cut it out from geofabrik dumps) with limits approximately size of a slightly largish city I use the service at http://78.46.81.38/api/interpreter - the data are about one day old, plus if you want, you can limit what data you need (by tags, etc ...) Martin. > On 14/09/10 03:01, Michal Migurski wrote: > > > > I'm downloading London, in small sections. I just exceeded my API > bandwidth limit. > > > > If you want an entire city please use planet, or a planet extract, rather > than downloading from the api. > > > > Can anyone tell me what the limit is so I know not to exceed it, and how > long I have to wait until I'm allowed back in? > > > > The limit is not some sort of game where you try and download at exactly > the maximum rate allowed - it's a way of cutting off the people who are > downloading vastly more than average. > > It basically cuts off the top fraction of a percent of users - the same > sort of people that got banned by hand before when they were noticed. > > Tom > > -- > Tom Hughes (t...@compton.nu) > http://compton.nu/ > > > ___ > 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 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?
> > 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] [Osmf-talk] ODBL vote (was Re: Enough is enough: disinfecting OSM from poisonous people)
> In the specific case of the OSM database, if you wanted to start doing > this, you would probably need to establish a per-object licensing flag. This > would require significant code changes and I assume you're not volunteering > to do that. We can assume that user won't change his/her mind too often, so for each user we need just maintain a list like: edits from 1. 1. 2004 to 10. 12. 2008 are PD edits from 10.12.2008 to 7. 11. 2009 are only Odbl edits from 7. 11. 2009 to now are PD ... we get few lines in DB for this variable licensing, and not license tag for each of user's object in database (which could be thousands, perhaps even millions of edited primitives for very active users) ... it is then problem of whoever wants to extract PD subset from the data to parse this information and extract really only the PD data ... Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Why quality is more important than routing speed
> I don't share your pessimism. I've mapped maxspeed=* quite a bit. > Compared to name=*, it's no harder to map, and it is of increasing > importance. I think we'll get far more than 8% of road names tagged in > the long-term future, and I think the same of maxspeed=*. Well ... maybe not - since at least in my country, most roads have no speed limits posted on them (so basically, the national "default" limits, like max. 50 in city, etc ... apply) Having 8% of roads tagged with speed limits may be the most you'll ever have for some areas (cities tend to have more speed limit signs on the roads, many country roads have none ...) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Changed highway=*_link meaning?!
IMHO the tag should contain information from what to what the link is (like between redidential and secondary, fron tertiary to primary, etc ...), so renderers can properly decide how much important the link to primary is - it is probably more impornant if it connects to another primary than case when it connects just to some residential road. Even then we would have complicated cases with some spaghetti junctions where roads of many types connect together and now you don't know how to tag. I think it could be solved by some "link_to" tag, so for primary_link connecting from primary to tertiary you will have highway=primary_link + link_to=tertiary (and basically it could be the same as highway=tertiary_link + link_to=primary - both will be connection between primary and tertiary) Renderers could then make better decisions how to draw stuff if they know type of "both ends" of highway link. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] New users have to sign up to the ODbL
> You can override it by adjusting the Accept-Language settings in your > browser. But how we do language *detection* is a separate from how we > should do content delivery once we've done the detection. Theoretically, but if you are in some sort of internet cafe outside your country, you may be unable to do it (I got once stuck with OSM site in French - no way to change it, since the browser preferences were also in French, thus not legible for me, sometimes the preferences are adminbistratively locked, etc ...) Some way to override Accept-language (without having to login) would be nice. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Some measure to prevent duplicate uploads of same data ...
While API 0.6 have implemented object versioning, preventing accidentally overwriting someone else's changes, with introduction of atomic uploads now I see many problems with duplicate data. These come often with imports of data or generally if someone uploads any new data without modifying any existing data (like if someone just traces hundreds of buildings from ortophoto, or alike ) Since in JOSM (and possibly in other tools) the atomic upload is the default method, that user presses some "upload" button and in few seconds all the changes are uploaded to the server, which then starts processing it (this could take some time for larger changes) and once it is finished, it will send new node ID's back to the editor. Unfortunately, sometimes while waiting for server to process the uploaded data, the connection will timeout, so the user sees some error message - thinking the upload failed, he presses "upload" again, starting to push new copy of all the objects to the server. Later, the server want to return ID's from first upload, but nobody is listening on the orher end anymore. Ultimate result is sometimes having 2 to 4 identical copies of some data, sometimes it is thousands of duplicate nodes and ways. Suggestion for one possible countermeasure: after server receives complete succesful atomic upload from user, compute SHA1, MD5, or some other checksum of the uploaded XML. Store it and if user tries uploading exactly the same thing again (because he thinks the upload have failed, which is not true), send him just some error message instead, like: "You have already uploaded this data". Or alternatively, send the user whatever result was there from the last upload (either new set of ID's, or some error message in case that previous upload failed because of some error) I think perhaps last 2 or 3 checksums could be stored in case someone have multiple parallel uploads in multiple editors. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Should roads be connected to all types of crossing ways?
On 15/02/2010, Pieren wrote: > On Sun, Feb 14, 2010 at 9:54 PM, Richard Weait wrote: > > Your router example still works. Even if you and I aren't going to > > turn left at "highway=railway; railway=narrow_gauge" we had best give > > way to traffic there. > > > > > > No, routing is a bad example. We have seen recently on this list someone > saying that he did not want to put a node because it was not allowed to > "turn left" or "turn right" for his car at one intersection. Think more > topology than Router. I prefer Tom's definition. Maybe not for car, but people on foot following that road can turn left or right (which would be impossible if one of the roads would be in tunnel or on bridge), so even in this case the roads should be connected. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Request for user block
On 04/02/2010, John Smith wrote: > On 4 February 2010 21:02, Maarten Deen wrote: > > This of course comes after blocking on email address proves ineffective. > > And we even have blacklists for blocking certain "get for free e-mail" > > domains from ever registering. > > I've dealt with this on a project before in a similar manner, I can't > say with any certainty how well it worked but the problems we were > thought might occur due to abuse so far to date haven't come into > fruition. Going down this path did cause extra labour as we've had a > number requests to allow free domain emails because their ISP didn't > offer email addresses, so this could potentially increase labour, > allowing individual email addresses and/or just replying to emails > about it. There are many small cheap ISP's here who don't give email addresses at all (you have freemail, so there is not great demand on this service), some of larger ISP's have their own freemail and they give paid customers also emails from same domain, so essentially, same situation as freemail (anybody can get freemail). Some ISP's give email account for extra money or their account is so limited, that it is basically useless (1 MB mailbox, etc ...) Also, some "better" ISP's will give you more than one mailbox (sometimes you can get dozens of mailboxes, sometimes "only" aliases, sometimes real mailboxes, often you can remove unused aliases and create new ones, so the amount of mail accounts is practically infinite), so even blocking nonfree emails may not always work. I don't think this is good idea. While this would reduce chance of some idiots for signing up for new account if they are blocked, this would affect also many innocent people who have just their freemail account and some people may still get around this quite easily. I think blocking IP addresses would be more effective (with most ISP's you get only one), though the block should be applied carefully and probably not permanently (shared or dynamic IP's) I think similar blocking scheme that is on wikipedia (when blocking IP's you can choose whether to block also already registered accounts or not, whether to block creation of new accounts and you can choose block duration) should be good enough. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Google launches "Indigenous Mapping" workshop
> But what if that population then consists entirely of Map Maker users? > Is that really beneficial for OSM? I know what you're saying, but it > is reasonable to expect Map Maker users to jump ship to OSM? Is that > even what we should be hoping for? If just one of them discovers OSM somehow, finds out it is better (they can have the map data back, unlike mapmaker where they can 'only' view results on google, plus more types of data are supported - footways, various POIs ), then he may tell others about it and then maybe entire such mapmaker community will gradually come to OSM. :) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Revert tools?
I need to revert part of http://www.openstreetmap.org/browse/changeset/3050941 - someone edited live in Potlatch and screwed things up - he moved many nodes to some silly position, then someone tried to fix it by restoring one of affected ways... so I'd like to revert all node moves. What is current status of the revert tools? In JOSM I can select the affected nodes by searching "user:afonin type:node", but I can't revert them to earlier version in history Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Name, owner, and tenant of building
> >Just tell me how you'd like the tags to be rendered. Maybe, if name > >isn't given on a building, another tag should be rendered instead. > > > Perhaps 'name', 'user' and 'owner' in that order? Tag 'user' was misused few times by some editors in the past to contain name of user who made the edit, so I think we should use some another name of tag to avoid confusion. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Merchandise
On 25/10/2009, Ævar Arnfjörð Bjarmason wrote: > On Sun, Oct 25, 2009 at 3:59 PM, MP wrote: > > > Some java applet, that will download OSM XML. [...] > > The problem is that if you want something that looks like the OSM maps > that's going to rely on a lot of things (mapnik et al) that aren't > Java programs, so it really can't viably run on someone's computer > without installing a lot of things. Mapnik is good for displaying on screen, but for generating PDF and printing it is better to generate vector-based, not image-based map. Converting OSM XML to some set of vector commands (stored as PDF operators) is not trivial considering number of map features and other factors, but it can be done without any external dependencies - OSM XML goes in, PDF with map goes out :) Well, Java applet is one way, there are many another - like running some PHP script, while having some PDF generating extensinon installed on server. > Even if it were just an applet having to run an applet automatically > turns a lot of people away from the service. > > It's not *that* computationally expensive to do this sort of thing anyway:) No, not for generating once, but once people start requesting new atlas every ten seconds in average and server needs over minute to generate it, then you have trouble. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Merchandise
Well, wikipedia have now instaled extension that allows you to order printed books via pediapress - you pick up some articles, then you see how many pages it is and you can download it as PDF or order printed copy. We could perhaps do something similar with OSM - someone chooses scale and area, then atlas is created. OK, this would probably be computationally more intensive per page than "just some article with some words and few pictures" - question is, can we offload this task to user machines? Some java applet, that will download OSM XML, then let user customize it (whether to have a legend or not, street index, whether to include more detailed plan of some interesting areas in the atlas, etc ...) and generate the PDF and send it back for printing could do the job, while keeping server load at acceptable values (since whoever wants the PDF will use his/her CPU power to generate it :). The only caveat is that pediapress now only prints in black and white - color is planned for future. So either push them to make also colored books, or find some other company that can do this :) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Weird changes in map?
I encountered way http://www.openstreetmap.org/browse/way/30722618 >From history it seems it was probably some forest before someone screw it up to form some sort of circlish object. Is there currently any working tool to see the way as it was before and perhaps revert it? It have five revisions and I don't know which one was the last good one (and how to easily restore it in JOSM). Technically, you have to revert the way and also some of the 700 nodes, which will make it a bit difficult. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] view blocks received?
> Actually, could someone explain how this is supposed to protect against > vandalism? > > It appears that if an account is blocked, a vandal can simply create any > number of alternate accounts and continue his "work" - probably much > faster than anybody can hand out blocks. Won't help much against determined attacker, but may deter some drive-by vandals or experimentators (if the block will ensure also short time block of new account creation from blocked user's IP, it may be relatively effective). Also, it could be used to block bots that start to break things due to some error - block first to stop further damage from the bot if it is still running (even if you write message to author right as you see the damage, it may take some tim before he reads it and stops the bot, perhaps hour, perhaps even more), then talk with bot author later to sort things out (and then unblock the bot once author fixes it). It is not 100% cure for all vandalism, but may be effective perhaps for 80 or 90% of cases. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Landuse areas etc. abutting highways
> For a road, we can either choose to map it as a linear object (this is the > common case), or we can map its geometry more exactly by using an area. In > both cases, however, the object in our database represents the entire road > (i.e. not only the middle line). Because in reality, there is no gap between > the road and the areas next to it, there shouldn't be one in the database > either. > > In other words, we should keep the topology intact, even if we choose to > simplify the geometry. This would be hard to do properly render in the renderers, as they will render the road with non-zero width and to render things correctly, they should "push" the boundaries of touching landuses so they will touch the rendered road borders. It is IMHO easier to learn renderers to support proper width tag and add that tag to the street between. With proper micro-mapping, even the street between could be mapped as an area, but that could be perhaps a bit too much of detail. But a) could be used as acceptable temporary solution until someone with better information (like having aerial photography) remaps it as b) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Landuse areas etc. abutting highways
> If you don't want to do micro mapping the best approach is to create a > multipolygon relations for the farm and one for the golf course. use > the portion of the highway in the polygon as outer way and delete the > duplicate ways. While this may look reasonable, this is IMHO a bad idea. I've seen once an area where someone made this by sharing parts of landuse=residential (the area) with many roads and forests touching it. I tried to fix it in JOSM (the lines were very rough and there was usable aerial imagery for that area to draw more precise boundary), but the mess was too great and I ultimately gave up (tried to move stuff from relations to create separate polygon but it just ended up as a big mess). Unless current tools improve a lot their multipolygon handling, I think this way sharing in multiple multipolygons should be rather avoided. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Landuse areas etc. abutting highways
> In practice, almost all mapping seems to use approach (a) - but would > approach (b) be easier for subsequent editing and addition of detail, and > rather clearer as it avoids superimposed ways and potential editing errors? I think that the correct way is b) - three separate lines. Since if the border of the farm also is somehow displayed (for example if it is fenced, thus having fenced=yes or barrier=fence tags), in the first case (all lines joined to one line) the border would go through center of the road, resulting in artifacts like fence in the middle of road. b) is easier for editing, more reflecting the reality and surprisingly, it also renders well - if the road is thicker than the gap, then it looks exactly like a), if not, then there may be some gap, but often there is at least some sidewalk or something, so the gap is justifiable - one can view the gap as sort of sidewalk :) And if the renderer knows proper width of the road (width=... tag) and can support it, there would be no or very minimal gap. If not, then the size of the gap will at least give you hint of size of the road (or exactly, size of the space where the road is). Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] shop=groceries?
> It looks as though farlokko has reverted some (maybe all) back to > shop=groceries. Some, but not all. I reverted all greengrocers from that changeset back to groceries in Czech republic, in all cases where the shop name does not suggest it is actually a greengrocer (names like "Ovoce - zelenina", etc ...) I have not touched nodes outside Czech republic (approx. half of the nodes was inside Czech. rep.) as I am not familiar with types of shops abroad. These may perhaps still need some fixing. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] shop=groceries?
I notices few days ago user farlokko changed many shop=groceries into shop=greengrocer worldwide. The changeset is http://www.openstreetmap.org/browse/changeset/2562959 I think this change is wrong, at least for most nodes in czech republic - I know about nodes that I've added and only small part (perhaps one out of ten) of them are actually greengrocers, according to my knowledge. Most of them are ordinary grocery stores. Some of them even have no or very little selection of fruit and vegetables. The greengrocer is shop that sells fruits and vegetables (in czech language usually called "Ovoce a zelenina") and no other type of food - according to what is at http://wiki.openstreetmap.org/wiki/Tag:shop%3Dgreengrocer and what would the name suggest. The groceries (I found shop=groceries at rejected features, though it was widely used, JOSM has it in presets, etc ..) should be used for shops selling general food (not only fruit and vegetables), but that do not sell anything other than food (like shop=convenience) and are small (so shop=supermarked won't fit to them) - at least this is what I think. In czech these are called "Potraviny", or "Večerka" if they have closing time very late in night. So the question is how to tag shops selling only food that are small? Should shop=groceries be used (and perhaps somehow added to map features or proposed features, or some other tag should be used? And should that changeset that converted shop=groceries into shop=greengrocer be reverted? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] UserActivity / Vandalism etc.
I got http://svn.openstreetmap.org/applications/utils/gary68/useractivity.pl and I notices some bugs: You are missing #!/usr/bin/perl at first line I fixed it and then I got: Undefined subroutine &main::bzopen called at ./useractivity.pl line 1168. You have to use "use Compress::Bzip2;" also in useractivity.pl - doing it in included osm.pm is not enough :) Also, considering the help printed out when called without enough parameters, the part "Mode = [N|P|D|S|PD]" is not very descriptive. Could be better :) Martin On 18/09/2009, Gary68 wrote: > > PLEASE X-POST TO TALK-GB, i am not registered there. tnx! > > hi, > > i just finished version 2 of UserActivity, a program that monitors > actions in a certain area. there are new reports and drawn maps to be > seen in/via the wiki: > > http://wiki.openstreetmap.org/wiki/UserActivity > > new: > - map mode > - detailed map mode > - ways are monitored > - working time slots are monitored > > so, now the next vandal can come :-)) ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] [OSM-dev] Footnav - new 3D project aimed at countryside users
> I'm glad to see that this project is moving on and especially that > it's using Qt :) However, I get a compile error with revision 2: Me too. In SRTM.h there is: #include But in maths.h I see this instead: #ifdef __APPLE__ #include #else #include #endif Quite inconsistent, considering in both cases probably the same header should be included :). On my system I have only but not Once I fixed that and compiled it, I tried to load N50E014.hgt, but I am getting "Specified bounds not in height file" ... and every time I have to enter all five fields in dialog over and over. And once I manage to get it loaded (after studying source to find out what should I enter in these boxes) it seems not to work (only black screen shown) I guess the program is not yet in a functional state :) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM Mailing List Reply To Header
> > Also most people don't want the reply sent too them twice as if they > > are posting they are on the list, and hence should receive the answer > > twice. > > How do you know? Did you ask them? I _want_ to get direct copies. In case > people don't want to read mails twice they can tell the list "don't send > me a second copy" or tell their mail reader "deduplicate my mail please". Well, Thunderbird, which I use as mail client can't deduplicate the mail. Especially if the mail that went through list have few extra headers and a footer added, so these mails are no longer the same. So I end up with two almost identical mails in my inbox. As for replying, there is usually no button "reply to list" in mail clients I use (not in thunderbird or in gmail), so whether I press reply or reply to all, I have to manually edit To: so there would be only list. > > If someone want to reply off list then they can change but the default > > should be on-list. > > Right. Any sane mail client does that. Based on the list-header, without > an annoying reply-to header which was never meant to do want you want it > to do. Neither thunderbird or gmail does that. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Brainstorming: Simple Revert-Tools
> > be added at the right side of every changeset line. One click might > > report "succeed" or "conflict/failed, do it manually". > > Until this is possible I want to try to (or at least collect all > necessary information to) write an external tool for this. Adding "revert" to main site could attract vandals (ok, let's just revert stuff) or experimentators (what does this button do?). I think better would be to have it as external tool that can either revert or prepare file for reverting (so that it would be loaded, checked and uploaded in JOSM) - reverting should be easy, but not that easy, that it would suffice to click one button to revert. > Maybe such a tool could block such revert-reverts (A revert's Bs > changeset, B reverts As revert) and only allow them to others, to avoid > revert-wars. Maybe this should only happen after two rounds? How do you (automatically) distinguish revert from ordinary edit (or "almost-revert" from ordinary edit)? I think you can't at least not reliably (you can't rely on the commit message, etc ...) So for avoiding revert war we need to use same mechanisms as for avioding any other unwanted edits - block users/IP's from editing, etc ... Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Brainstorming: Simple Revert-Tools
On 02/09/2009, Peter Körner wrote: > > I'm against voting. Voting is a way to take responsibility away from the > > individual. I think that in most cases we should strive to have > > individuals responsible for everything (just like with mapping - you > > don't suggest something which the community then votes upon, you just map). > > But this could lead to reverting as a extended form of vandalism (which > is much more effective!) They you can use the tool again and just revert the revert. Currently, messing things up things is easy, even without any revert tool, reverting them (without asking someone with DB access to do the dirty job) is not so easy. With the tool, reverting will become much easier. And it would help not only against vandalism, but also against bugs introduced by some bot that did not work as the author expected (like bot uploading new copy of all objects with "new" tags, instead of updating the objects, leading to thousands of duplicated stuff in the map). Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Undo request button for changesets
> A brilliant idea. Close inspection of a changeset should also be able to > better find out whether the changeset actually applies to a certain area > or whether it just has a huge bounding box. Currently, we have "ID, Saved at, Comment, Area" fields in the changeset list, so perhaps adding columns with number of ways, nodes and relations changed/added/deleted would make it easier to guess the nature of a changeset. Unusually high number of moved nodes (in relation to other edits in changeset suggest thatr someone may have moved something, perhaps accidentally). Unusually high number of deletion suggest possible vandalism (though it can be case that something big got just bolldozed in reality, but it is worth checking), etc .. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Undo request button for changesets
I think the worst case to revert is mass deletion - only Potlatch have some facilities to find deleted lines given a BBOX, and only lines, not POIs or relations. Such deletion is quite hard to spot. As for reverting changesets - there is list of nodes/ways/relations that was either added, changed or deleted in the changeset XML, so for changed stuff, you can retrieve the previous version from history (using the version from changeset as the "id of old version"), for new stuff, you add action="delete" into the XML and for deleted stuff, you recreate the last version (could be a bit tricky for deleted ways with nodes, as you'll have to mangle the ID's a bit ...) It shouldn't be probably hard to write such a thing ... perhaps some of existing tools can be just a little adapted to produce this. Load this into JOSM, use the conflict resolution in there to sort things out (if someone tries to fix the thing manually) and upload . changeset effectively reverted. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM TrustPoints
> Fixing subway stations in Rome wouldn't be prohibited to newcomers if > they limit themselves for that day to Rome, encouraging them to focus > on an area rather than jumping around the globe changing things (True > value of OSM is _local_ knowledge!). They could be first only allowed > to map within some kilometers around their home, but that might be too > restrictive and demotivating. As for number of daily affected nodes > the number shouldn't be too low, but low enough to prevent moving or > deleting whole cities (whether accidental or malicious) or importing > some data without consulting the community (about legal and technical > issues) and having some help from experienced members. I personally often travel along some long route (100-200 km in my case), then I start adding POIs and other stuff I saw along the route once I get home. And there are people travelling much longer routes (long road routes across US/australia or even flights - one user in one day can easily add some details on road to some european airport and then another details on road from one airport in south america to nearest city centre for example) I don't think we should put any hard limits on the bbox. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM TrustPoints
What about not adding restrictions to the rights, but implement some warning system? We can use some "trust points" that are awarded for good mapping (objects that last on the map) Then warning points in each upload for possibly bad things (deleted nodes, removed tags, accidentally moving many nodes by same offset, etc ) Then publish some feed or list of such edits above certain threshold - so they can be spotted and either marked as false positives (by trusted users) which will cause them to "lose" warning points (the more trusted user mark it, the more point lost), ultimately moving them off the list, or dealt with (reverted, etc ... - which will remove trust points from the user) Newcomer that deletes even single node can be spotted that way - and even older "trusted" people could be spotted if they start some massive vandalism or big error. As for "importing google maps" issue, we can mark some places that are known to have no coverage from allowed sources (Yahoo, french cadastre, ) and every node there can trigger some warning points Point is, that suspicious edits (caused by lack of skill, human error or vandalism...) gets listed for people to review, while not limiting anybody. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Making an offline OpenStreetMap CD/DVD ?
> You can also strip out all of the data/tags which you are not interested > in rendering. > > Obviously you would not want to upload any of this 'tainted' data back > into the OSM database. If you do now want to edit, just use the data to display a map (and perhaps do some routing, etc , you need less data (no created_by or other useless tags and things like highway=residential can be translated to some index into specific lookup table) GpsMid (app to display OSM maps on a cellphone and route in them) can take ~72Mb bzipped dump and produce ~36Mb .jar with the app and (almost) all the "usable" data in the OSM. So maybe there is some way to get 6.2 GB planet dump and produce some ~3Gb file that would contain entire OSM in a manner that would allow easy displaying and perhaps routing. Though nobody have written such program yet :) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Explaining to NASA why the ASTER data should be freely licensed
> As it turns out the first clause is (apparently) to facilitate > tracking of how the data is used and so that they can announce > updates, and the second is to ensure proper attribution. I've asked > them permission to quote their complete reply but that's basically it. What about derived data? SRTM is used to generate hillshades and contour lines for example. ASTER data would be good for that too. Do they have some less strict terms about distributing such derived data (like requiring only attribution), or is their policy for it the same? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Making an offline OpenStreetMap CD/DVD ?
> I'm curious if it'd be better to pre-generate all the images before, and > then you're just serving static files, or if it'd be better to have some > sort of web server / database / the whole tile rendering shebang on there. Pre-generating probably won't be the good way to go, I saw some figures for OSM tileserver and tiles for whole world at largest zoom were in order of several hundred terabytes (I think it was like 10^10 tiles for whole world) - though only tiny fraction of them are actually generated and cached. You will end up with map of either quite small area or map that is missing some of the zoom levels Whole world dump (bzip2 compressed) is 6.2 GB, but for practical use you have to convert it to some better format (large bzipped file does not allow random access or queries like "return ways at this bbox"), which will likely make the resulting file larger. Theoretically you may fit on dual layer DVD with whole world, but I doubt it If you limit yourself on one continent, you can probably fiot that on a DVD quite fine. Perhaps you can combine it - use pre-generated map for very high zooms (whole world) and then generate the maps for the continent that is present on the DVD on the fly (with some cache) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] New (better?) source of contours
> They just mention that the data is free of charge, but don't make > further comments as to the license of it. I tried to actually download anything ... and I failed ... seems their server is way too overloaded currently. But in the process, there is some "agreement page" saying this: # I agree to redistribute the ASTER GDEM only to individuals within my organization or project of intended use or in response to disasters in support of the GEO Disaster Theme. (Required) # When presenting or publishing ASTER GDEM data, I agree to include "ASTER GDEM is a product of METI and NASA." Seems not like very open license, though perhaps something better could be negotiated ... Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] full history of a way?
On 14/06/2009, Richard Fairhurst wrote: > > MP wrote: > > So currently, if I want to know "how was the way drawn in time T", > > I had to get history of the way, then find out which nodes were used > > by the way in time T, then get history from each of the nodes and > > pick appropriate node from the history, given the time. Then > > combine this into output XML which I can view > > > > Is there some automated tool that will do that? > > > Ahem, Potlatch. Doesn't have any truck with any of this XML crap though. ;) Well, potlatch have feature of reverting to earlier version of way, but doesn't that reverting have a side effect of actually reverting the way? I want only examine the older version, since I know only the newer is correct. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] full history of a way?
> If you often have this kind of query, consider downloading and suitably > preparing some old planet files for quick access to old data with proper > referential integrity. I don't think this is possible for me, as having several old planet files in state where it could be quickly searched (so probably unpacked) would probably need several hundred GB of disk space. Which is too much for me. Luckily I need this type of query only rarely. > > Perhaps this could be improved in 0.7, by having some > > "[way|relation]/#id/history-full" call, which will return > history of > > referenced way (or relation), including history of all node (ways) > > that was referenced at least in one of the historic or current > > versions. Sort of combination of /history and /full > > > > I'd rather hope someone builds some kind of "time machine" service that has > these capabilities (plus a "give me this area as it was last year" API > call). There is no pressing reason to burden the central API with that once > we figure out how to publish full history dumps. There are no full history dumps currently - having such dump would enable this type of query quite easily. I think the daily diffs could be used for help if the changes are fresh (less than 14 days). Are there any nearby plans for full history dumps? This shouldn't probably be very hard, instead of having only one node/way/relation in the result you will have many with same ID and different timestamp as the history progresses. You'll have just to mark deleted objects somehow (those that have history, but they are not in current data) Basically, once the initial dump will be done, you can just concatenate any changes :) Well, there could be slight problem with pre-0.6 data (no order in relations) and pre-0.5 (extra segment elements in addition to nodes and ways), so this maybe need some conversions. I am not sure what was before 0.4 (could be some problems too...) So, I guess I'll have to write some tool to query the server for the node histories then, as it probably is currently the most effective way considering there won't be many queries... Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] full history of a way?
I tried to download full history of a way The call to http://www.openstreetmap.org/api/0.6/way/34068057/full will fetch the current way with all its nodes (so I can open it in JOSM for example and look at it) The call to http://www.openstreetmap.org/api/0.6/way/34068057/history shows history of the way, but does not provide any nodes. So currently, if I want to know "how was the way drawn in time T", I had to get history of the way, then find out which nodes were used by the way in time T, then get history from each of the nodes and pick appropriate node from the history, given the time. Then combine this into output XML which I can view Is there some automated tool that will do that? Perhaps this could be improved in 0.7, by having some "[way|relation]/#id/history-full" call, which will return history of referenced way (or relation), including history of all node (ways) that was referenced at least in one of the historic or current versions. Sort of combination of /history and /full While you will still need some tool to extract the information you want, you'll need only one request to the server instead of one request per each node in the way you want to examine. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How to tag an archway ?
> > someone on the french list asked how we could tag an "archway" ? > > Something like this: > > http://commons.wikimedia.org/wiki/File:Place_des_Vosges_archway.jpg > > > > I didn't find anything on the wiki or tagwatch. Maybe it's even > > questionable to have a special tag since it is either a normal > > extension of a pedestrian place or a normal sidewalk but under a > > building. > > Thanks for you advices. > > > > Pieren > > > > IMO it's a simple sidewalk. So: > > highway=* + footway=both/left/right Perhaps add "tunnel=yes", as it is sort of a tunnel ... though open on one side Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Wanted feature for API 0.7 ??
I thought of another wanter feature for 0.7 API Retrieving deleted objects, similarly like it is done in potlatch. Currently only potlatch can do this and since potlatch does not work well with larger areas (it is way too slow) and does not support many features that JOSM have (WMS, plugins, loading/saving to disk, gpx tracks, gpx waypoints ), it is not good for more advanced users Potlatch can also only resurrect deleted ways, it could be sometimes useful to be able to resurrect deleted nodes in this way too. Basically, the API call should have bbox as input and return all deleted ways (and perhaps also nodes) as output. I think it should return last existing version of that objects before deletion. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] How do we specify relative importance of features across all types of features?
> Place of Worships: > Cathedral amenity=place_of_worship importance=regional/national > Church amenity=place_of_worship importance=urban > Chapel amenity=place_of_worship importance=suburban You can use some tag object_size, where you specify size of the object: Catherdalamenity=place_of_worship object_size=100m Church amenity=place_of_worship object_size=30m Chapel amenity=place_of_worship object_size=8m The larger object, the more "important" it is ... The "object size" is not subjective (except for errors in guessing the size) and would make good heuristics about what to show at different zoom levels (large buildings are usually more important than small ones) While there are exceptions to this (perhaps some small chapel that is very famous), I think this could be good starting point and it is not subjective but objective. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Which software for SonyEricsson phones?
I am using GPSMid (http://gpsmid.sourceforge.net/) on SE W200i - I don't have bluetooth GPS, so I use it only for displaying the map (which it does quite fine), but it is also capable of using GPS (should work with bluetooth and integrated cellphine GPSes), logging tracks and making waypoints (making and exporting wayponts is something that I am using frequently, so I can confirm this surely works) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] zones for motorway/in town/outof town?
> ...but you would need some kind of gis database/functions to evaluate the > polygon data. > the easy way of reading just keys and values like with most of the other > features in the osm database would not be possible. for more advanced use of OSM data you need function "in which polygons does this point/way belong" anyway (you need it to find out in which country/city/suburb/place you are) so I don't consider it such a big problem. But probably OSM could use a special handling for these polygons, as currently OSM API will return polygon only if one of its nodes is inside the requested area. It won't return polygons if entire request area is inside it. (It won't return even way that lies partially inside, but all of its nodes are outside - like line crossing the edge of request area) This is problem even with current data - if you request part of map inside city centre of large city, you won't get the city borders. Same applies for any large landuse/natural areas, like big lakes, forests, etc ... or administrative boundaries Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] zones for motorway/in town/outof town?
> > The problem now is that you cant actually work with polygons as for > > example a motorway could pass over a zone-30. > > Sorry, I maybe didn't make myself clear. "Polygon" rules do not apply > for motorways. Is there any country, where a highway inside a city has > different speed limit from the highway outside of the city? Even if yes, For example Czech republic (80 km/h highways inside city, 130 km/h outside - Law 361/2000, paragraph 18, part 4) In the few rare cases where highway goes over/under some zone with speed limit, the highway can either: 1) have explicitly set speed limit 2) you can have information about this exception - relation saying that some ways (the highway) does not belong to the zone - relation can be used for that. 3) The zonal restriction can contain information about types of roads in the polygon on which it should be applied (so saying that the zonal restriction is valid only for highway=residential and living_street, would exclude also primary/secondary road going through that zone, so it won't need to be tagged explicitly if the limit is not applied on it) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] zones for motorway/in town/outof town?
> > Well, we can have layering (layer=-5 ... 5) like for any other tags, > > so we could have small zone with different limits within one large > > zone. For clashes with "default" rules from place=... and alike, we > > can define some rules of precedence (speed limits on individual ways > > have highest priority, followed by rules from speed limit zones, rules > > from place=..., then default country rules). And if zones with same > > priority overlap? Then it is bad mapping (as are typos like > > highway=residental or other bugs) and someone needs to fix it in the > > database. > > And what about using 'multipolygon' relation? That is another possibility, though for nested zones you'll need two ways - one as inner side of the multipolygon and one as the nested zone. Though if inside the nested zone are default rules, or the bondary is not completely shared, using multipolygon could be better. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] zones for motorway/in town/outof town?
> Could we not have different polygons for Speed Limited Zones. That may > or may not be the same as the city limits. We could even tag these > zones with maxspeed, So that when applying we don't have to go and > look up what that means. The problem is if the zones overlap, which > one applies? Well, we can have layering (layer=-5 ... 5) like for any other tags, so we could have small zone with different limits within one large zone. For clashes with "default" rules from place=... and alike, we can define some rules of precedence (speed limits on individual ways have highest priority, followed by rules from speed limit zones, rules from place=..., then default country rules). And if zones with same priority overlap? Then it is bad mapping (as are typos like highway=residental or other bugs) and someone needs to fix it in the database. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] zones for motorway/in town/outof town?
I wonder, can we have at some place (wiki?) some definition file that
will specify these per-country default limits in some machine-readable
way?
This could look like this:
country(cz) {
maxspeed=90
(highway=motorway|trunk) {
maxspeed=130
foot=no
bicycle=no
}
is_in_polygon(place=city|hamlet|village|town) {
maxspeed=50
}
is_in_polygon(place=city|hamlet|village|town) && (highway=motorway|trunk) {
maxspeed=80
}
}
the example is simplified from reality (there are different speed
limits for trucks/buses/etc...), but if we would have something like
this, it would help when building routing software. Perhaps some XML
could be used instead of these made-up C-like syntax :)
Also, I think in USA there are different limits in different states,
so we could have rules also for smaller regions than countries in few
cases.
> Often one driving through a city will pass one "start of the place" sign
> (with place name) and several times "start of built up zone" and "end of
> built up zone" before the "end of the place" sign. So we would need
> different polygons for built up zones than for place administrative
> boundaries (which are IMHO not less important, unless we want OSM be
> a road map only).
either use landuse=residential for these or some other tag specific to
poland (and perhaps countries with similar driving rules) and then use
rule like
is_in_polygon(place=city|hamlet|village|town) &&
is_in_polygon(landuse=residential || place=build_up_area) {
maxspeed=50
}
Martin
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] zones for motorway/in town/outof town?
> a) motorway: that's very clear, therea are no or very high limits. > b) "city" areas with limited speed and some restrictions > c) everything else, mostly out of town. In Czech republic there are different rules for motorways inside "city" area (maxspeed=80) and outside (maxspeed=130). So case d) identifying a) is easy and if we are able to identify what is inside and outside city - b), we can identify roads and motorways inside/outside city. There was similar discussion on czech mailing list. One suggestion is to use administrative boundaries. There was suggestion to use landuse=residential, but that is not possible - often the "city" area is not corresponding to landuse=residential. Suggestion for administrative area is pretty good, but there are still some caveats - sometimes the boundary follow the road and in few cases, sign for "end of village" and sign for "beginning of village" from other side do not match (so basically, one side of the road is "inside" the village, while the other is not.) > (In practice, the test for a "built up area" seems to be "does it have street > lights?".) Test for street lights won't work, many small villages in Czech. Rep. are signes as build-up area (with all the implied speed limits and such), but they don't have street lights at all. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Re verting Changes....
> It's WAY too easy for a newbie (and I was once) to try to scroll in > Potlatch like they scroll in OpenLayers, munge a way or a node by > moving it inappropriately. Then they panic, are unable to find the > Undo or don't know about the Undo, or are just so scared that they > immediately close the Window (so they don't accidentally save their > changes). What about ading some 10 seconds delay before actual uploads to server, so such panicked actions would have the desired effects? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Satellite for OSM
> "In addition to actively pursuing further experiments for the MOD and BNSC, > the consortium is also seeking new applications to which the technology can > be applied." > > i.e. it's as much a research project as a commercial operation... so maybe > your idea of "let's just ask them" could work. Perhaps they can give us some photography from times when the satellite is idle (moving over areas where nobody wants currently photography of them) either for free or at some reduced cost > Sound great, but in the mean time we can of course buy commercial photography > including the right to derive mapping at a cost of about $17 per sq km So if we manage to photograph over 2300 square km of area of our choice in the week of rented satellite, then the satellite would end up being cheaper (and more up to date) than commercial photography. I guess that could be worth it. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Satellite for OSM
> 2.5m sounds about the same as Y!, so its even enough for rudimentary > building mapping. but thats the black-and-white figure, the colour > resolution is about 5m. :-) You can merge them and get something with almost same quality as 2.5m color resolution. > also, would it be worth it as a PR stunt for qinetiq to just use up > whatever spare capacity they have when maneuvering or between clients > and give us whatever gets photographed...? anyone know anyone at > qinetiq? Maneuvering? The satellite does not have fuel for maneuvers - it have some fuel to make small corrections on the orbit and shift perhaps few meters to avoid collision with debris and other satellites, but not fuel to make any significant changes to the orbit. Landsat have 16 days period in which the orbital pattern repeats: http://landsathandbook.gsfc.nasa.gov/handbook/handbook_htmls/chapter5/chapter5.html I suspect TopSat will have similar orbital characteristics, so you'll have to choose the time when the satellite is flying over area of interest. Also, their website says "The satellite is designed to return its data directly to a mobile ground station immediately after collecting". I suspect the satellite have little or no means of storing the photos, so it will just broadcast it "down" to mobile station. So I suspect you'll need to put up some mobile station somewhere near the photographed area. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Satellite for OSM
> http://en.wikipedia.org/wiki/TopSat > http://www.qinetiq.com/home/defence/defence_solutions/space/topsat.html > > Apparently you can rent it for £25k a week... easily within the ambition of > donate.openstreetmap.org. How large part of earth could be imaged in that timeframe? Topsat have 2.5m resolution, which is quite fine for most areas, though less than aerial imagery ... > MP's point about what you do with the vast quantities of data that you get > is well-observed, of course. But we like a challenge. One thing is having the data on ground - entire world (510,072,000 km²) from Topsat in 2.5m resolution will have ~ 245Tb of uncompressed data (you'll get to about 1/3 of that if you discard imagery with just sea), which is lot, but perhaps still manageable. But you have to either store some non-trivial part of it on the satellite (that is not as easy as on earth where you can buy some server with RAID and plug it into wall) while the satellite does not have direct visibility of the earth contyrol center where it can relay stored images (and then you have some means to transmit large amount of the data while the satellite flies over the earth control center) or have multiple ground stations or bunch of another satellites that relay the continuously transmitted data. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Satellite for OSM
> To lunch a small Satellite is not very expensive. > Are there plans for an OSM satellite? But it is very expensive to build one - if you want satellite to provide some satellite imagery, you'll have to put up some sensors, power source, computing power, some means to transport data back to earth (computer to store the data and send them once satellite passes over ground control center, or multiple satellites, once to make imagery, anothers to help relaying the data) and of course some ground control centre Plus, you won't get probably much higher resolution than landsat. Hi-res aerial imagery is usually done from airplanes from about 1 km height. Imagery satellites like landsat orbit around 700 km. Actually launching the satellite to orbit is only fraction of whole costs. A small fraction. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Zonal restrictions.
> Except it's not a geographic area, but rather a set of streets with that > restriction. If a bridge or tunnel without the restriction goes > over/under a street with the restriction you'll have a problem. In that case, that bridge can have differen speed limits set directly on the way. Just define that tags on individual ways override tags set in the zone and this will solve the issue. > So while it may work for many zonal restrictions to use an area > (although even then there are still issues) you need another way as > well due to the bridge/tunnel issue, and that other mehod can only be > relations. I've seen many zones with some limitations (mostly speed limit or some parking limitations), but such zones rarely contain any bridges or tunnels. If they do, they can be either tagged individually, or we can also make new relation that will contain the zone and roads that are NOT in the zone (relation containing those few exceptions like the bridges, tunnels, etc ... so they won't need individual tagging) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Is there any "desktop app" for OSM data?
I wonder whether there is a desktop app for offline viewing of Openstreetmap data. Something in which I'll put either planet dump or some of its excerpts (like one country or continent) and then I can view (and perhaps also search, route, measure distances, etc ) the map. Editors (josm, merkaartor, ...) are unusable for that (they can't process OSM data in size of entire countries, at least not with reasonable speeds) and it seems that most software on http://wiki.openstreetmap.org/wiki/Software is mostly either for small devices (cellphones, PDA's) or require internet connection (since they fetch tiles from tileserver or something like that) Any tip for program, that will be suitable for offline viewing/searching/routing of OSM data on desktop PC (or larger notebook)? (I run Linux, though I may try using Wine for Windows programs ) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] Delay before the data is visible?
How large is the current delay before uploaded data became visible? I've uploaded some changes in JOSM, provided a comment ... and when I re-downloaded the area again, my changes were not there. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Offline editing during downtime
I don't think that is so easy. To avoid conflicts, when replacing node/way/relation that is on server with newer version, server sends internal ID/revision number of the way and you have to pass this back (so two people won't overwrite the same way with their version and "last upload wins" - in 0.6, first one upload, second one gets a conflict and will have to resolve it first before being able to upload) You will have to add these revision numbers to all objects for upload to work, changing 0.5. to 0.6 won't do it ... Martin > AFAIK, the .osm files that JOSM is able to save to disk are pretty much the > same for 0.5 and 0.6. Your best shot is to save a 0.5 .osm file, open it up > with JOSM on monday, and try to upload it. The api version auto-detect > feature of the latest JOSM releases *should* handle this issue gracefully. > > If that fails, open the .osm file with a text editor, and change the first > line to read "0.6" instead of "0.5". ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Bogus GPS Uploads
> If you open JOSM and download GPS data for this area: > http://www.openstreetmap.org/?lat=41.067&lon=-76.766&zoom=9 > > You will find a trace for a very long elliptical area, oriented > East-West.. I am very familiar with this area, and I can't imagine how > anyone would have generated such a trace by any legitimate means. Flying in a plane? But then the trace probably does not correspond to anything on earth and is probably worthless. Unless it is some air route. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Tagging dangerous areas
> after the wembley mapping party last year i heard suggestions of a > "locals=angry" tag. maybe we should expand that to include > "locals=violent" or "locals=heavily_armed"? What happened at the Wembley mapping party? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] [tagging] Feature Proposal - Voting - power_output
Simple proposal for adding power_output tag to power generators allowing specifying output power. Could be used to distinguish small (like single pylon acting as wind power plant) and large (like 2000+ MW) power plants and perhaps render/not render them according to zoom level, level of detail, etc.. May be also nice information to show in map, perhaps. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] [Spam] Re: Wikipedia & canvec/geobase import
>> Hmm ... such link would be of very limited use, due to the "not >> aggregate in a mapping or geographic application" clause. Isn't there >> any similar service with less restrictive policies? If there is any, >> we can perhaps start convincing authors in 3D warehouse to upload >> their models also there :) > > I agree that it would be interesting to have a 3D warehouse with more > generous licensing, however I also believe that there is a clear distinction > between adding links to copyright material into a Work and adding the > copyright material itself. We can create links to websites containing > copyright material from within OSM and don't think twice about it because For links to websites, most users want for the link to show up in their browser of choice as preferred action (and then read whatever is there), so that is OK. For links to models, the desired outcome may be different. Some people may want to see/download the link in browser, but many others wou;ld like to see it directly in the map. > the legal right to create links to copyright work was established in the > early days of the web and that is all I am proposing. > > It would however be against their terms to create a 3D city by combining the > OSM data and the 3D warehouse data into a single model or 'experience' and > that is why it would become interesting to have data in a separate 3D > warehouse with share-alike terms. If we link to such a model, in end it probably would end up being displayed (and thus violating terms) in the app directly, as the app will see the link, so it will maybe download and show the model, not knowing anything about google's terms (person who put the link in database won't breach terms, application programmed to display all models won't breach the terms either, but end user who run the app on the database will... in the end Google may be hunting end users of whatever too will utilize the links). While it seems a good idea at a glance, as andrzej zaborowski said, it "may be misleading (people loading the link somewhere might assume the model is distributed under the same terms as the map data and not read the sketchup license)". So I think we either should look for some free repository (wikimedia commons may be good, though they currently accept only images, video and sound files) or create one ourselves. As OSM plan some collaboration with wikipedia, maybe as part of it we can negotiate possibility of uploading 3D models to wikimedia commons and then using them/linking to them. This would solve our "free repository" problem quite nicely. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Wikipedia & canvec/geobase import
>> what about having a pop-up bubble on features, where the default shows >> the osm wiki page discribing the feature. >> (ie, if your hovering over an unnamed school, the osm article for >> 'school' would be shown. If it was labled, the official school website >> would be shown? That would depend on application using the maps, whether the application will somehow link to OSM wiki, copy the description from there or provide their own. I think we should provide description only to objects that are unusual or somewhat different from standard (like unusually small/large snowshack, etc ...) Maybe we could have database where key would be name of the feature (like highway=residential) and language (like "en") and as a value there would be some nice human-readable description including perhaps some link(s) to wikipedia, etc ... Well, sometimes in the database, combination of keys could be used (amenity=parking + parking=underground would have different description than amenity=parking + parking=surface) We now have something like that for mappers (like http://wiki.openstreetmap.org/wiki/Tag:amenity%3Dparking) but we could have similar descriptions also for "end users". We could use perhaps another namespace in wiki for that. > I suggest we include a Sketchup link for structures that are in the > Google Sketchup 3D Warehouse. There is not yet even a tag for sketchup > that I am aware of but it would be a 'bridge' between OSM and a > project where structures are described in great detail. Note that > Google's terms of service make it clear that they don't own the > resulting models but do prohibit end-users from distributing the > content of the 3D Warehouse content within 'mapping or geographic > applications'. To my mind distributing a link to a model is ok, > distributing the model is not. Hmm ... such link would be of very limited use, due to the "not aggregate in a mapping or geographic application" clause. Isn't there any similar service with less restrictive policies? If there is any, we can perhaps start convincing authors in 3D warehouse to upload their models also there :) Martin > "end users may not aggregate the Content obtained from the 3D > Warehouse for redistribution, and may not use or distribute Content > obtained from the 3D Warehouse in a mapping or geographic application > or service without Google’s prior authorization" > http://www.google.com/intl/en/sketchup/3dwh/tos.html > > The level of detail in Google Sketchup is impressive and there are > many 'projects' around the world: > http://sketchup.google.com/3dwarehouse/ ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Bulk import as a data layer (like gpx currently is)
Well, I thought we would have one layer per each mass import (so you can look at what was imported) - one layer for TIGER import, one layer for AND import, etc ... and the data would ge trimported both in that special layer and in ordinary layer, where everybody can edit it. In future, we can have layers even at different servers not operated by OSM (for things that are not exactly within scope of OSM map), like layers with wifi hotspots, etc ... Technically, these layers will merely redirect to another server. Martin > Er, no. Anything that someone might ADD to the map should be, if > imported, in the same place. Otherwise, someone will fire up their > editor, not see the data that already been imported, and start to add > it. No better way to piss off a volunteer than to waste their time. > > Yes, there are problems with this approach. We need to solve them, > not run away from them. We'll just create worse problems if we try to > avoid solving this problem. We ALREADY have to deal with it with the > TIGER import. Neither that data, nor the edits made to it, are going > to be exiled from the database into this proposed "bulk import data > layer". > > -- > Russ Nelson - http://community.cloudmade.com/blog - > http://wiki.openstreetmap.org/wiki/User:RussNelson > r...@cloudmade.com - Twitter: Russ_OSM - > http://openstreetmap.org/user/RussNelson > > > ___ > 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] California bill to limit detail on online mappingtools
> I'm not sure that blurring is what would happen, simply not showing the > data would be what happens. But you're right, by having a blank or > blurred section of an otherwise detailed map just shows that there is > something of interest there. Well, even if that stupid law would get to reality, it won't have any effect on data we supply - as we don't provide any satellite imagery, we won't have to blur anything (plus OSM does not reside in california). We could theoretically even mark all the ducts or map these "sensitive" sites in ridiculous amount of detail, thought I doubt anybody wants to do it. > The more sensible option would be to map the area in high enough detail > to not be obviously "blurred" but not so high as it'll be useful. Take > for example what Ordnance Survey do with sensitive military sites, map > enough of it to not have a blank bit but label it as something > as innocuous as possible. Or do not label it at all - there are many maps with unnamed buildings and you'll never know from the map if it is just another factory area, warehouse or some secret military installation, even if all buildings, roads and tracks between them are mapped very accurately (though you may discover that when you approach that area in person). I remember from one discussion about this law, that someone said that in later revision of that law the blur will have to be replaced by targeting mark, to better reflect reality :) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Map tag in Wikipedia
>> Right now you can get a static dump + image dump and produce a fully >> working copy of a Wikimedia wiki. If all the maps relied on map tiles >> hosted somewhere else that would break a lot of things for the static >> dump content wise. > > I would have to learn more about how these static dumps are made. > The Swedish Wikipedia (300,000 articles) contains no uploaded > images. It fetches all images from Wikimedia Commons that > contains 4 million files. Does a static dump of sv.wikipedia > require a complete copy of Wikimedia Commons? Or can you get a "complete copy of Wikimedia Commons" = several TB of data. When wikimedia commons had 2 million files, total size of all files in it was 2 TB. Now it have 4 million files, so I assume it is about 4 TB or images. Not all of them are used in articles. > separate dump of only the necessary images? Could the same http://meta.wikimedia.org/wiki/Wikix can do that > selection work for map tiles? We really don't want a tile > generator to "upload" every tile through a MediaWiki. :-) there is slight problem - unlike images in wikipedia, map tiles are frequently updated independently of wiki content. Get a planet dump and you can generate all the tiles you need :) > Since these static dumps were produced in June 2008, perhaps this > system doesn't need to be our top priority. > > It is also possible, if the live maps for a city is at z=12, the > static dump could show maps at z=8 thus saving lots of space by > omitting the tiles for deeper zoom levels. After all, a static > dump is already a compromise. I guess no need for tiles - every map can be replaced by single image like this: http://tile.openstreetmap.org/cgi-bin/export?bbox=14.173,49.969,14.63,50.193&scale=27&format=png for the static wiki versions. Few queries and we are done :) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
[OSM-talk] OSM license and optional PD
I have proposed how a checkbox that allows users to release their contribution as public domain could be implemented and how we can solve the "derived data problem" - we won't solve it technically, we will leave it to the people wishing to take out PD subsection of data, but we could make it easier for them to determine which data are PD and which are not by adding few simple features (I think max. few hours of coding of one person) to OSM plus perhaps writing some guidelines how to do it. See http://wiki.openstreetmap.org/wiki/Optional_public_domain_licensing and either add the comments there or express them in this list. Also, allowing people to release their contributions as PD would make it easier to transition to new license, at these peoples contribution are automatically compatible with the new license. This also minimize risk of data deletion, as currently if you improve/fix data made by others, you risk your contributions can be deleted or reverted later due to adoption of new license. If you know the data are in PD, you can fix them without any fear of that :) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] immutable=yes Fwd: DEC Lands
> In that case, the data should not be in OSM but should instead be pulled > in on another level - for example, create transparent tiles to show on > top of OSM tiles, or make a shapefile and pull that in through Mapnik. Well, if the data won't be in OSM (neither in dumps or in things received from API), how would people see them? Slippy map is not all we have. Plus, this would make harder to make data derived from it (if appropriate). >> How do people feel about me importing this data (with all of their >> metadata), adding an immutable=yes tag, with the intent of tracking >> their dataset, and deleting --outright-- any changes made by OSM >> editors. While the data may be very accurate and usually they shouldn't be edited, I don't think they should be really immutale - someone may still want to correct tags or names (even "100% correct" data from authorities often have at least few mispelling in them). The data could be watched for changes, but I don't think forbidding any changes or automatically reverting them without examining the change is the way to go. Maybe the immutable tag could get some support in editor that will show some big warning when you (probably accidentally) try to edit these data (so it would be more of a hint that the data should not be changed), but I won't forbid it in general. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM license change: A license to kill? -> How to make a nightmare come true!
> as you said: "comments should explain things that *aren't* in the > code", not repeat the code (incorrectly) in english. your example of a > bad comment doesn't answer my question: if you are reading code and > you do not understand why it is written the way it is, don't you read > the comments to find out? Comments allow also to see what is the code about - without needing to fully understand what is in. Comments like "this function does fast fourier transform" usually are enough to understand what the function does (if you know what FFT is) without need to look at the code and all the bloody mathematical stuff inside (if you don't know about FFT you won't have much idea about what the code does after reading it anyway). Same can be done for GPL for instance - "if you distribute GPL'd program, you must give people complete source code and give them the same rights to program you have". Maybe too simple (full details of how can source be distributed, what exactly is source, etc ... are in the license), but good enough for most people to have idea what they can and cannot do with GPL'd stuff. We need the same for ODBL. >>> did you come out of steve's evil basement portal of dooom? :-P >> >> I don't understand why people think steve has an evil portal of doom in his >> basement. It's in his attic. > > if thats in his attic, what were all those ghastly and inhuman screams > coming from his basement? Eh ... interdimensinal portal to the attic? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM license change: A license to kill? -> How to make a nightmare come true!
> I want to correct something here, there is this view of 100,000 users > needing consent. The number is in fact far smaller for people who ever > made an edit (about 30% of the users). It's vastly smaller still for > anyone who has edited anything significant. It's an easier problem Considering the 'derived data', the damage may be more than just few insignificant nodes. If someone uncontactable creates a node that later gets used as crossing point of three major roads, their tiny contribution could lead to deletion of all three roads cause of derived data (no matter how many of these three roads were drawn by the man who created the node). Disagreement from (or unsuccessful attempts to contact) one person that made "cosmetic" changes (like changing from highway=minor to highway=residential) to many streets in london in early days of this project could lead easily to deleting (or reverting to ancient version) half of downtown London for example. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM license change: A license to kill? -> How to make a nightmare come true!
> 3. Define a way for feedback from the community. Maybe some unoffical > votes would have given an impression on how well a particular idea would > have worked. Maybe put up a poll like: Do you think OSM should change license for all data from cc-by-sa to odbl? ( ) Yes, I agree ( ) Yes, but the license needs to be improved a bit first ( ) I am not sure ( ) I don't care, my contributions are PD. ( ) I need more information ( ) No, it is bad idea ... along with link to license text + explkanation why the change would be necessary, what about possible data deletion, etc Mail everybody with link to the poll - this will both inform people that something is happeing and also you will get feedback about how many people are willing to accept the new license. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM license change: A license to kill? -> How to make a nightmare come true!
> Yes. At least when you expect 10 people to go along and the issue > has the potential to break OSM apart, it would not be a bad idea to send > monthly information about the state of things. Hmm ... perhaps sometimes it would be good to mass-email all members when it is about changes with possibly devastating (mass deletion) effect. Not everybody reads various blogs or parts of wiki around OSM, but almost everybody reads their email. > >> But I guess it's > >> their own fault if 5 people fail to scan blogs at a different site > >> for half-year old entries. Not worth a notification or a prominent hint > >> in the wiki. > > > > Click "BLOG" on http://openstreetmap.org/ > > "Hardware Upgrade Appeal: Thank you" So the information about the license was at the blog, accesible by clicking on that hideous link at the right time to notice that... Reminds me of The Hitchhiker's Guide to Galaxy - the post informing that your house is scheduled for demolition was posted on a posterboard in local townhall. OK, the posterboard was at far end of cellar. Ok, it was hidden under door with sign "do not open, tiger inside". And the door was blocked by old library full of books. But it was there... I personally had no idea about the license change before it got posted on this list few days ago and I am contributing to OSM for more than year and half... Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan - what data would need deleting
OK, so lets assume that some data would have to be deleted (hopefully not lot of them, otherwise it would probably kill the project and spawn some forks with "complete" cc-by-sa data). Where there is the exact line between deleted and kept data is on another debate, but I wonder the way how the data would be deleted: - Would people know in advance what data are going to be deleted/reverted, so they can perhaps delete and redraw them themself? - If yes, how that would be done and how long before marking the data and actual deletion? - After deletion, would there be some marker like "there is something deleted from here"/"this way was reverted" for people to see, so they can improve the data again? - Would there be some "log" summarizing what data was/are going to be deleted from where after/before the deletion? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
>>Also, technically, when "mixing licenses", we won't have mashup of >>cc-by-sa and odbl, we will have mashup of cc-by-sa without consent to >>relicense later under odbl and cc-by-sa with consent to relicense >>later under odbl. > > I guess that would work. The resulting collection would be distributable > under > CC-BY-SA only. If all of the old work without the extra consent is deleted, > and > all work derived from it (see earlier discussion), then you could distribute > the > result under the ODbL. I think this should be the way we'll be going instead of deletion. While it will postpone the moment from which all the data will be ODbLed, perhaps maybe year or alike in future, no or very little deletion would be necessary. > ... PD mappers just have better to do mapping with JOSM > and save their incontestable edits as an own local copy for the future needs > to > be uploaded into OSM PD-repository or something. Main OSM database is not a > place to store PD data to be extracted out afterwards. I suppose it is not > even > OK to add POIs with Potlatch and read them back to JOSM for making a local > copy. Why not? Unless someone modifies the data in meantime, what is wrong about reading back your own data as PD? > One question: All edits of PD mappers could of course be tranferred under the > new license even without asking us. But is is possible to connect our PD > declarations in the wiki with our OSM data in a reliable way so that the > tranfer > could be automatic? User names used for mapping and in the wiki are not the > same thing, or are they? ... No, they are not. In my case they are different (the one I picked for wiki was already taken by someone else for mapping, so I had to use another name for mapping). While I have on my wiki page (User:Bilbo) the template that tells everybody my edits are PD, the template actually does not tell which user's edits. So in my case, the matching username in mapping is completely different person. Best would be to add the ability directly to OSM (perhaps to the user settings somewhere) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
> If we follow the rule of reverting incompatible changes only 2 is > reverted to 1 (A’s scribble gets added back in). 3 is considered an > independent change. We end up with both a scribble and a neat road in > the same area. This situation likely won’t be easy to detect until > after the changes, when validators will gleefully litter the map with > warnings about overlapping ways. That is perhaps worse than "we'll just delete/revert 5% of data". Because this revertion/deletion would cause another perhaps 20% of data (and it won't be always easy to spot them in the remaining 95%) to be left in more or less inconsistent state that would need some work to get consistent again. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
On Tue, Mar 3, 2009 at 23:50, Gervase Markham wrote: > On 03/03/09 18:39, Matthias Julius wrote: >> It is not that simple. What if those 5% is half of South Africa? You >> certainly can not interpolate overall OSM growth to re-surveying South >> Africa. > > ...which is why this is an unanswerable question. Let's go through the > exercise, see what the percentage actually is and where the data is, and > then decide what to do. Thayt is the worst thing - now you don't know who will agree to new license and who don't (unless you have some magic crystal ball). So you don't know which data are going to be removed and how much of them would it be until the last moment. You can only guess or estimate that, though once we start gathering consent for new license, we'll be having lower bound for the estimate - we will know for some data that these will stay there for sure (data for people wioth consent for new license) and the rest maybe yes, maybe no. I thought of one improvement - in addition to allowing people to consent to new license, allow them also to (completely voluntary) agree to Public domain their contributions. Some of the people on wikipedia (though not nearly a majority) does that for their contributions and many photos on wikimedia commons are under PD, so I assume some contributors may like it here too - then give them the possibility. Some tools to extract only PD subset of data could be added later if necessary (export list of users agreeing to PD would make this possible). And as wikipedia offers their complete dump with entire history, we maybe can offer planet dump with entire history in it too, so it could be easier to pick up only the PD contributions (or basically to dig through history for any reason without querying the main server) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
On Tue, Mar 3, 2009 at 23:47, Roland Olbricht wrote: > And what to users who do not log in with a browser? Send them email. If they don't respond in some time (few weeks?) by visiting their account, deny them access to uploading new data. That will make them look in their acount and try to figure out what is wrong. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
> It would be great to require that only free software could use OSM > maps. I saw other peoples agreement on this when we discussed > someone's 3D viewer for OSM data, and the #1 comment on this mailing > list was "we shouldn't glorify the use of non-free software". > > Proprietary routing software on OSM data was seen as something outside > of this community, not necessarily unacceptable of itself, but > certainly something that needed to be discarded and replaced with a > free alternative. We have now tool to convert OSM data to garmin format (Mkgmap). The tool is opensource. Garmin can do routing (at least I assume it can, I don't posses any garmin devices or software myself) and is closed source. Would the new license make mkgmap unusable/illegal with odbl'd data? What about distributing maps converted to garmin format (assuming they are still under ODbL). If that would be allowed, then it won't prevent using closed-source SW for routing over OSM data, as anyubody can simply get converted maps and upload them to their garmin device. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
>> - if you decide to delete contributions and those contributions are >> only part of the history of objects, do you rollback to a previous >> version of these objects ? > > Rollback to the last version before any changes incompatible with the > new licence are made. This could be perhaps "optimized": if user A creates some highway=road, user B changes it to residential and user C changes it to secondary. A and C agrees to new license, B won't. But contribution of B was completely removed by C's edit, so it won't be necessary to revert to highway=road in this case. Basically, if the edits of "incompatible users" got later reverted or altered so their contribution is not there anymore, there is no need to rollback, just delete their revision from history. This could help in cases where user B just make lot of mistakes that got later reverted/corrected. Technically, for ways we would have problems with restoring old revision, since the nodes referenced by the old revision could have been moved/deleted in the meantime, so that would possibly create some invalid data. > There is the idea floating around that modifications to existing data > are insubstantial, and successive contributions could potentially be > kept without issue, but I think it is safest to remove them. Perhaps for really minor changes, like alterations to created_by or conversion from "true" to "yes" or alike we could make an exception. Or in cases where the object was "completely modified" from the last "license-incompatible" version. > In the interests of keeping it clean, any reverts made due to > incompatible changes would not be kept in the history. Would there be at least some information like "this object was reverted because of new license" (which would signal that the object perhaps need to be re-improved somehow) and for deleted objects information that "something was deleted from here"? > A backup can be kept of the old database of CC-by-sa compatible data. > It might come in handy if some non‐responders pipe up and say “yes”, or > the “no” voters change their minds. Won't be of much use after longer time, since the missing data are probably first to get readded and merging contribution of people who changed their mind with the parts that was restored by remapping the affected area in meantime would be difficult and won't be posible to automate. Also, what if someone who disagrees to new license deletes some data (either because that data is wrong or is replaced by something else that he draws). Will the deleted data get restored? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] It's all too fast...
> We can make sure the existing-people-problem doesn't get worse meantime > by making people creating new accounts agree to dual licensing under > CC-BY-SA and ODbL 1.0. Perhaps give option to agree to ODbL also to existing accounts (though do not make it mandatory for now). This could also solve some problems if people leave the project in the meantime (perhaps because they have already mapped their area of interest or whatever ...) Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
>>I think we should find some way to avoid deleting at all. For some >>transitional time (in which the data will be still under cc-by-sa but >>we will be collecting consent of users for ODbL) mark data coming >>from/derived from people uncontactable/disagreeing with license with >>some special tag. Let people delete these parts and redraw them from >>scratch (from allowed sources/existing GPS tracks, anything except >>the original data). > > You would have to be very careful about doing that. I don't think it > would work to view the map, see a street tagged 'bad licence', delete > it and then add it back. Even if you were honest enough to close your > eyes, turn around three times and then re-trace it from the aerial > photography, it still looks very suspect. And when deleting the > street you would have to delete all its nodes, including those that > are intersections with other streets, since it obviously doesn't do > anything to delete the way but leave all the nodes there to be > straightaway reconnected. Sometimes (if current data are drawn very inaccurately and do not contain any valuable tags like name, etc..) I do this - delete current data, then draw it again from scratch from aerial photography with greater accuracy. It is faster than trying to move existing vertices around, splitting and merging the ways in the process. Yes, you have to be very catious when redrawing, but I think it may be possible. > Try this thought experiment: suppose a user imported data from Google Well, this is disallowed completely in first place. But here we have good data, just under different (but similar) license. > Since the reason for relicensing is to be ultra-cautious and take care > of certain theoretical legal bogeymen, it makes sense to be ultra-cautious > in removing possibly tainted data. There is no point doing a relicensing > that leaves the project in a more questionable legal situation than before. Well, but how can you then explain to users that half of the data is lost just due to small incompatibilities between cc and odbl? Also, technically, when "mixing licenses", we won't have mashup of cc-by-sa and odbl, we will have mashup of cc-by-sa without consent to relicense later under odbl and cc-by-sa with consent to relicense later under odbl. I think such "mashup" could work for short time (before we persuade all to get consent or delete and replace their data if we have no consent), once we have all "cc-by-sa with consent for odbl", we can just switch to odbl. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
> A little bit more respect to the people that actually did the mapping > work would probably be a very good idea. "We're only loosing 5% of the > data" is a very, very strange attitude for me. Not because of the data > but because of the people behind that data. Losing 5% of data will do much more damage than it looks - as we can probably assume, that the 5% would be rather randomly distributed, random 5% of objects would "disappear". Now you need to go through the remaining 95% and check/remap it, especially for areas that are already mapped "almost completely", to find out what was lost and redraw it. People will be "stuck" for weeks/months checking the data and repairing the damage - and some of them may get frustrated and leave the project. I think we should find some way to avoid deleting at all. For some transitional time (in which the data will be still under cc-by-sa but we will be collecting consent of users for ODbL) mark data coming from/derived from people uncontactable/disagreeing with license with some special tag. Let people delete these parts and redraw them from scratch (from allowed sources/existing GPS tracks, anything except the original data). Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
On Tue, Mar 3, 2009 at 03:39, Iván Sánchez Ortega wrote: > El Martes, 3 de Marzo de 2009, MP escribió: >> Note that if cc-by-sa is somehow abusable, anybody that want to abuse >> the license using some loophole will simply grab last dump srill >> published under cc-by-sa instead of the new license - and then abuse >> the non-enforcability of cc-by-sa to databases. > > ... Which is, IMHO, the reason for the migration to ODbL to be as fast as > possible. > > (If this happens, though, we should start looking for loopholes in other > people's data). What about finding a loophole that will allow convert from cc-by-sa to ODbL without asking anybody? :) I think wikipedia is doing something similar with their transition from GFDL to cc-by-sa >> So if we assume we will contact everybody who has logged/uploaded data >> at least once in last month and we will fail to contact the others - >> how many data will be removed? > > We won't know until we ask. I tried running some statistics on extract of Czech Republic from (~ 78000 km^2, 684 Mb uncompressed) If i take data from all users, that have uploaded/modified at least one node, way or relation in last month, I end up with 72.66% of all the data. If I use last two months instead, I end up with 85.56% of data. That is only current state, not considering any history or possible derivative work, etc When I tried the same for germany, I get 59.82% for one-month-recent contributors and 79.17% for two-month-recent. Even worse. If we assume people without contribution in last two months as unreachable (lack of interest in OSM for them), we lose at least 20% data. If the loss would be such high, I think we'll have another fork very soon. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
>> This seems rather apocalyptic. What do you mean by 'lose everything' and >> how would changing to a different licence avoid that? > > It is my opinion that CC-by-sa poses a high risk of not being enforceable to > databases. That would mean losing the share-alike rights to the data. So you mean the data will become sort of "public domain"? Well, then there is question: what is worse? 1. Have all the data, but risk someone "abusing it"? 2. Or force the license change, therefore enforcing the share-alike rights correctly, but tossing some data away? Note that if cc-by-sa is somehow abusable, anybody that want to abuse the license using some loophole will simply grab last dump srill published under cc-by-sa instead of the new license - and then abuse the non-enforcability of cc-by-sa to databases. As for the possible data loss - since new license is basically still in spirit of cc-by-sa we can assume that most users will agree to license change, if we can contact them. So if we assume we will contact everybody who has logged/uploaded data at least once in last month and we will fail to contact the others - how many data will be removed? Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
On 02/03/2009, Ed Avis wrote: > MP gmail.com> writes: > > >As for the people who can't be reached/refused to accept new license - > >what about tagging such data with some tag like "license=cc_by_sa" to > >warn people that this part is licensed otherwise and keep the data in > >database? > > > I don't think that would work. If some parts of the data are CC-BY-SA, and > some > parts are under a new licence, then the resulting database )or maps derived > from Well, if you need the data for personal use - you can use them even with mixed license. If you need to distribute them, etc ... you could filter the cc-by-sa data out. This would allow the remaining cc-by-sa data to be iteratively deleted and then redrawn under "correct" license. I think this could be viable, if there would be only small part of such data. (so the period in which the data won't be properly distributable will be quite small, perhaps few days till all is redrawn) Martin > it) would be a derived work of both. That means that it can be distributed > only > under CC-BY-SA, and also that it can be distributed only under the new > licence. > The result would be that you cannot legally distribute it at all. > > Presumably OSM chose CC-BY-SA to stop other organizations taking the OSM data > and distributing it under different conditions. Even if only some of the > data > in your work is OSM data licensed CC-BY-SA, you must distribute the whole > work > under that licence, or not at all. What's sauce for the goose is sauce for > the > gander. > > > -- > Ed Avis ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] highway=tertiary[_link?] (was: Re: highway=secondary_link)
http://www.informationfreeway.org/api/0.5/*[highway=tertiary_link] -> 1.3mb file Well, these seems to be quite a lot tertiary_links out there ... For example http://www.openstreetmap.org/?lat=49.95321&lon=11.57331&zoom=16 (not rendered though, you have to use some editor to see it) So I'd say we should have tertiary links too. Martin On 02/03/2009, Andrew Chadwick (email lists) wrote: > Eugene Alvin Villar wrote: > > > > Would adding also highway=tertiary_link be too much? :-) > > I'm not sure I can think of any examples in the areas I'm familiar with. > Perhaps that's just due to local road design though: link-like > structures seem to be reserved for faster, more multi-lane road designs. > > Not having link roads: a concrete criterion for highway=tertiary? :-) ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM 3D, featured image
Nice images, but I found no software for generating the data (I'd like to see another country in that way), so usability is somewhat limited. But it could be nice image of what can be done, so I think we should get one nice image from them to featured images. Maybe it'll inspire someone to produce similar tool, but an opensource one. Martin > Other than the terrain images most of the city views are pretty flat. > Maybe ask them to produce a render of somwhere with plenty of OSM data > and interesting terrain. Oh, and perhaps ask them to license that image > CC-BY-SA or something. > > Cheers > > > Andy ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] License plan
> This will be a particular problem if this happened with someone who made > lots of changes and then went off in a huff for some reason. What about data donated by varous organizations? In these cases, the user that uploaded them usually just "merely" converted the data from another format (possibly adding some manual cleanup, and thus copyrightable, etc ...), but the major contribution is from the organization itself. This won't be problem in case of PD sources (for example from various government agencies, as their output is "automatically" PD in many countries), but may pose a problem with private companies donating data (they may or may not be happy with the license change) As for the new license - it's 10 pages of legalese, so I'd expect once we start asking users about it, there would be some human-readable excerpt nearby. As for the people who can't be reached/refused to accept new license - what about tagging such data with some tag like "license=cc_by_sa" to warn people that this part is licensed otherwise and keep the data in database? This should be acceptable for most contributors (no data lost) but may be a bit troubling for people who want to use the data (if cc-by-sa is not good for them they would have to filter these out). Also, those cc-by-sa parts could be replaced in time by "properly licensed" data (someone would redraw/reimport the affected area or alike ...). Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM wiki local copy
There is special page to request export of data as XML dump http://wiki.openstreetmap.org/wiki/Special:Export And a list of all pages: http://wiki.openstreetmap.org/wiki/Special:AllPages I have tool that makes complete XML dump using these two pages quite effectively (tested on wikipedia). It need about one request per 1000 or so pages plus the server merely loads text from DB and sends it to user, not needing to parse it or whatever, so I think it won't cause heavy loads, unless used en-masse. Still, best would be if whoever maintains the wiki would put up the XML dump by themselves - If they won't, maybe I can either make the dump and put it somewhere and/or publish the tool to make it. Martin > Currently there is not a database dump of the wiki content. As I recall, > trawling the whole of the wiki with Wget just overloads the server causing > problems for normal users and hence why it is disallowed. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] OSM on The Reg
> The concept of an editor is already too complex for most people. > This guy searches for command buttons corresponding to it's > first desire: "add a road", or "add a cycleway", or "change road", Then let them add road (highway=road) and then let the "geeks" figure out if it is secondary, residential, unclassified or whatever... Maybe add some "complete novice" mode in potlatch, with things like "add road", "add river" ... I guess no more than 15 functions ... and perhaps make it a bit foolproof, so users won't accidentally move existing roads by dragging ways/nodes elsewhere when clicking on the map furiously. Maybe the last button in row would be to enter "expert mode", aka old potlatch Perhaps change the "created_by" to "Potlatch novice mode" or alike for data created in the novice mode, so we can spot bugs created by inexperienced users easier. Martin ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
