Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
Gervase Markham wrote: >Sent: 14 May 2008 2:01 PM >To: talk@openstreetmap.org >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help provide >"completeness" tools > >Frederik Ramm wrote: >> Once we have a few applications in place that get viewed by *many* >> people, we could just have a button somewhere along the margin of the >> page that says: "I know the area and what I see here looks correct". > >Would it not make more sense to have a button saying "This map is >incorrect in some way", and it opens a text box optionally inviting you >to say what is wrong. These notes can then be stored, and if someone >comes to redo that area of the map, can provide guidance. > I think we need both. We need a method of communicating to OSMers that they can, for the time being, ignore an area and a separate ability for viewers of the map to "notify" errors and omissions to OSMers exactly as you describe, the latter perhaps being limited to areas that have been marked by the former as "complete". Cheers Andy >Gerv > > >___ >talk mailing list >talk@openstreetmap.org >http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk > >No virus found in this incoming message. >Checked by AVG. >Version: 8.0.100 / Virus Database: 269.23.16/1431 - Release Date: >13/05/2008 7:55 PM ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
Frederik Ramm wrote: > Once we have a few applications in place that get viewed by *many* > people, we could just have a button somewhere along the margin of the > page that says: "I know the area and what I see here looks correct". Would it not make more sense to have a button saying "This map is incorrect in some way", and it opens a text box optionally inviting you to say what is wrong. These notes can then be stored, and if someone comes to redo that area of the map, can provide guidance. Gerv ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
>I get this same view. All too often I look at a place and think wow, that >looks complete, but when I drum down into the data a bit it its clear that >there are general gaps and the density of streets is not what you would >expect. That's why I was testing out a completeness metrics method. But I >agree with you, what we are really after hear is a simple way to convey a >level of map usefulness and relevance to the user. A bit of an aside, but I was thinking that a useyourpaths.info type approach might be useful for countryside mapping, to encourage the (it has to be said) rather slow takeup of countryside mappers into OSM. One could almost have a "league table" of counties/parishes/regions where people could be encouraged to map all their parish paths - for OSM/Freemap - and have their parish appear on the top of the league for their county. Comments? Nick ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
David Earl wrote: >Sent: 13 May 2008 6:48 PM >To: Sebastian Spaeth >Cc: OSM >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help provide >"completeness" tools > >On 13/05/2008 15:35, Sebastian Spaeth wrote: >> Frederik Ramm wrote: >>> Once we have a few applications in place that get viewed by *many* >>> people, we could just have a button somewhere along the margin of the >>> page that says: "I know the area and what I see here looks correct". >> >> Given that this will be the default very soon ( :-) ), I'd rather have >> the notes API where people can click and say: "there are streets missing >> here, I know that". No warm fuzzy feeling, but more helpful in >> identifying weak spots. > >My main motivation in wanting this kind of facility is not so much to >help _us_ identify what areas need attention, rather to help our >_consumers_ know whether they can have any confidence in what they are >looking at. That's why I think there needs to be a very straightforward, >not overly onerous, but useful, metric, even if this has more levels >accessible to those in the know. It also means that wiki solutions just >don't cut it (I've been updating completeness pages for the areas I;ve >been doing since I started, but it doesn't help someone looking at the >map). > >Often anyone of reasonable intelligence can tell somewhere is not >complete because only the main streets are there, but I have come across >quite a number of places where a reasonable number of apparently random >residential streets have been done, and whose density would suggest to >someone who doesn't know the area that it is os plausible, when in fact >it may only be 30% complete (for roads and names) or less. I get this same view. All too often I look at a place and think wow, that looks complete, but when I drum down into the data a bit it its clear that there are general gaps and the density of streets is not what you would expect. That's why I was testing out a completeness metrics method. But I agree with you, what we are really after hear is a simple way to convey a level of map usefulness and relevance to the user. Cheers Andy > >David > > >___ >talk mailing list >talk@openstreetmap.org >http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk > >No virus found in this incoming message. >Checked by AVG. >Version: 8.0.100 / Virus Database: 269.23.16/1430 - Release Date: >13/05/2008 7:31 AM ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On 13/05/2008 15:35, Sebastian Spaeth wrote: > Frederik Ramm wrote: >> Once we have a few applications in place that get viewed by *many* >> people, we could just have a button somewhere along the margin of the >> page that says: "I know the area and what I see here looks correct". > > Given that this will be the default very soon ( :-) ), I'd rather have > the notes API where people can click and say: "there are streets missing > here, I know that". No warm fuzzy feeling, but more helpful in > identifying weak spots. My main motivation in wanting this kind of facility is not so much to help _us_ identify what areas need attention, rather to help our _consumers_ know whether they can have any confidence in what they are looking at. That's why I think there needs to be a very straightforward, not overly onerous, but useful, metric, even if this has more levels accessible to those in the know. It also means that wiki solutions just don't cut it (I've been updating completeness pages for the areas I;ve been doing since I started, but it doesn't help someone looking at the map). Often anyone of reasonable intelligence can tell somewhere is not complete because only the main streets are there, but I have come across quite a number of places where a reasonable number of apparently random residential streets have been done, and whose density would suggest to someone who doesn't know the area that it is os plausible, when in fact it may only be 30% complete (for roads and names) or less. David ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
Frederik Ramm wrote: > Once we have a few applications in place that get viewed by *many* > people, we could just have a button somewhere along the margin of the > page that says: "I know the area and what I see here looks correct". Given that this will be the default very soon ( :-) ), I'd rather have the notes API where people can click and say: "there are streets missing here, I know that". No warm fuzzy feeling, but more helpful in identifying weak spots. ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On Tue, May 13, 2008 at 12:57 AM, Andy Robinson (blackadder-lists) <[EMAIL PROTECTED]> wrote: > Some of us map out an area completely in one go rather than doing it > piecemeal. Even if I come across some existing roads in a new area I ignore > them and do a new survey so that the whole area makes logical sence to me. > That way I can work out where the landuse areas are behind the houses and > the extent of school areas etc etc. So for me a reasonable level of > completeness is easy to decide and annotate. That's how I work too. So when I mark the area that's complete on the new system, I'll mark everywhere that I did as complete. That's a no-brainer. But let's scale this further, since what you, me and David have done, while interesting, is still a small amount of what's there. I would guess that only a proportion of the mappers will use this system, let's say a similar proportion to those who map in landuse=residential. But the key is that not everyone will use it. So how do we scale the efforts of this subset of people who *do* want to use the completeness system to provide measurements of the rest of the data? How do you, me and David (say) work out which areas of Glasgow and St. Louis are complete? Maybe we can look at the road density and guess. But that could be automated. My original point was that we can look at areas that we don't know particularly well and it's much easier to spot the problems than confirm which bits are done. With the proposed system, it'll take me 15 minutes to mark everywhere I thoroughly mapped, and then I want to do something useful. So I can mark 20-something square kilometres of London as complete (to the "95th percentile of completeness"), and *much more* as definitely incomplete (i.e. Dave's renderings of unnamed roads) and much of it 'unassessed' where it could be anything from the 60th to 95th percentile. Anyway, that's just my take on it. Cheers, Andy ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide"completeness" tools
Andy Robinson (blackadder-lists) wrote: > Verification is a whole new ballgame. I think throughout this discussion there is tendency to get hung up on the word "complete", which has been used as a shorthand but is being interpreted differently. In everyday use it has an implication of perfection and that there is nothing more to be said. I think what we should be talking about is an area being "filled in", without implying perfection or immutability. You should expect as high a proportion of mistakes in a filled-in area as in an incomplete one, but fewer omissions. However, if there is a blank space on the map, you can assume that it really is empty in a filled-in area, but you would not know if it was in an incomplete area. Measures of quality and guarantees of correctness require filled-inness, but I think should be regarded as more advanced concepts. I agree with Andy, we should walk before we run - start with an implemention of filled-inness - verification, etc. can come later. Chris ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide"completeness" tools
Andy Robinson (blackadder-lists) wrote: > This demonstrates that really there is no one method that fits all. I map > urban areas out completely in one go, including the cycleways and footways > because just about all are accessible by bike. Occasionally I have a footway > I have to go back and do on foot but these are few and far between and they > nag at me if I leave them for very long. And of cause you have visited all the pubs and restaurants and tested their wares in the name of research so you could log them ;) I think that some secondary means for users ( who do not have the ability to fix a problem themselves ) to report problems? While a 'This looks OK' sounds a good idea - it only looks OK for what the user is looking for. But a 'So and so is missing' would at least provide prompts to help fill in the gaps. -- Lester Caine - G8HFL - Contact - http://home.lsces.co.uk/lsces/wiki/?page=contact L.S.Caine Electronic Services - http://home.lsces.co.uk EnquirySolve - http://enquirysolve.com/ Model Engineers Digital Workshop - http://medw.co.uk// Firebird - http://www.firebirdsql.org/index.php ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On Monday 12 May 2008, Skywave wrote:
> Freek recently created this image which shows how much of the AND data is
> untouched:
> http://www.vanwal.nl/osm/author_density_nl_20080502_full.png(warning 3 MB
> image).
(Blue is untouched AND, green and red have relatively more changes by the
community.)
I also did an image showing the number of different (last) authors per area
for a large part of Europe (untouched AND is not so interesting outside the
Netherlands...):
http://www.vanwal.nl/osm/density/europe_1000_080513_num_authors.png (6 MB)
(Red is one author for all nodes covered by a pixel, green to blue depict an
increasing number of authors, up to around 17.)
Central London clearly has the largest number of contributions from different
people.
Secondly, I thought the average data age might show some interesting patterns
(min. data age turned out to give rather noisy pictures).
http://www.vanwal.nl/osm/density/western_europe_500_080502_avg_age_value.png
("Lava" colour map: black = old --> red --> yellow --> white = latest
contributions, compare to the dark-red AND import for a reference, this was
September 2007. Also note that dark colours have a second meaning: they
depict low node density.)
Now, London gets quite dark at some spots... More remarkably, this picture
shows that data imports dramatically decrease mapping activity (or so it
seems): not only the Netherlands show relatively little activity, also
Osnabrück looks quiet (compare for example to the Ruhr area or the area
between Brussels and Antwerp). Still, in my opinion, these imported areas are
far from complete.
I think pictures like these can give at least some impression of the current
state of affairs, but a human-maintained measure for completeness is still
necessary.
--
Freek
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide"completeness" tools
Jeffrey Martin wrote: >Sent: 12 May 2008 9:38 PM >Cc: talk@openstreetmap.org >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help >provide"completeness" tools > >I'm very far from this in Korea, but I would guess in time some parts of >the UK >will need to be rechecked at some point. How can we make a system >for rechecking an area? Maybe the completeness should be retired >after a period of time. Verification is a whole new ballgame. I tried a bit in my local area, going out on foot with the printed OSM map at highest zoom and annotating the missing bits and pieces for addition. It increased the richness of the data considerably but made me realise that there is a second phase that will still add a huge amount of data. Personally I think we have more pressing things to worry about that verification right now. Something perhaps for a year or two hence. Cheers Andy ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide"completeness" tools
Florian Lohoff wrote: >Sent: 12 May 2008 8:29 PM >To: Inge Wallin >Cc: talk@openstreetmap.org >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help >provide"completeness" tools > >On Mon, May 12, 2008 at 07:06:20PM +0200, Inge Wallin wrote: >> Yes, but it would be good if we planned for at least some of the >subsequent >> levels right from the start. > >I support this idea - I would call them levels as some areas might work >completely different. I would call the different completenesses like > >roadcomplete >cyclewaycomplete >roadnamecomplete >landusecomplete > There is no reason why contributors could not decide what the list should be. It's clear then to everyone what has and has not been mapped and gives the mapper a sense of achievement even if they weren't ever interested in mapping areas for instance. Someone can add that level of completeness later. >etc ... I have no idea about the granularity but in the end it comes >down to a map never containing ALL interesting data but it would be >interesting to note which subset it contains or is complete. > >The completeness is more or less a feeling of the primary mapper of >this specific area so it should be taken with a grain of salt i guess. Yes, Its not a substitute for secondary verification. Cheers Andy > >But this information would be interesting for a map bug tracker which >could assign higher prioritys to bugs in "complete" areas or even refuse >to accept bugs in incomplete areas. > >Flo >-- >Florian Lohoff [EMAIL PROTECTED] +49-171-2280134 > Those who would give up a little freedom to get a little > security shall soon have neither - Benjamin Franklin > >No virus found in this incoming message. >Checked by AVG. >Version: 8.0.100 / Virus Database: 269.23.16/1429 - Release Date: >12/05/2008 6:14 PM ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide"completeness" tools
Inge Wallin wrote: >Sent: 12 May 2008 6:06 PM >To: talk@openstreetmap.org >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help >provide"completeness" tools > >On Monday 12 May 2008 18:06:59 Chris Morley wrote: > >> Starting with a single level of completeness makes sense, but I think it >> should be public roads, named where feasible. > >I have a different view. I think we should have a leveled scheme from the >beginning. I suggest the following: > >Level 1: All the highways (using OSM lingo) usable by cars within an area >are >mapped >Level 2: All highways are mapped and named >Level 3: All highways down to cycleways are mapped (and named if feasible). >-- >Level 4: Level 3 + all amenities within a given set. >Level 5: Level 4 + maybe some other amenities or buildings or whatever we >decide. > >These levels map(!) nicely to how I work myself when I map out an area: >start >with the roads and streets for the cars, and name them. Then later go back >to >the area and add the cycleways. Sometimes I map a whole lot of other areas >before coming back to map the cycleways. Regarding the levels: I can >understand if some people would switch level 2 and 3 and I would be fine >with >that. This demonstrates that really there is no one method that fits all. I map urban areas out completely in one go, including the cycleways and footways because just about all are accessible by bike. Occasionally I have a footway I have to go back and do on foot but these are few and far between and they nag at me if I leave them for very long. Cheers Andy > >When an area reaches level 3, I would say that it is "finished" for the >purpose of finding the way (remember it's Open _Street_Map :-) ). All extra >information is a bonus, and necessary for certain types of maps, but for >the >general user or a router, level 3 should be all that is needed. > >> This covers the essential >> basic requirements for a number of potential applications for example, >> routing, delivery, estate agents, bus services. The points of interest, >> etc. are clearly desirable and but may not be always collected. For >> instance, tracing from arial photos and naming from an out of copyright >> gazeteer; or imports like TIGER. (Also I'm not sure I collected all of >> them when I started 2 years ago.) Start basic and have these in a >> subsequent level. > >Yes, but it would be good if we planned for at least some of the subsequent >levels right from the start. > > -Inge > >> Chris >> >> ___ ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
David Earl [mailto:[EMAIL PROTECTED] wrote: >Sent: 12 May 2008 11:06 PM >To: Andy Allan >Cc: Andy Robinson (blackadder-lists); talk@openstreetmap.org >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help provide >"completeness" tools > >On 12/05/2008 22:51, Andy Allan wrote: >> On Mon, May 12, 2008 at 8:16 PM, David Earl <[EMAIL PROTECTED]> >wrote: >> >>> I think it is terribly hard to know whether you have all the footpaths, >>> and I think we'd hardly ever mark anywhere "complete" if we did that. >> >> I think it's terribly hard to know when a map is correct and complete, >> regardless of what you're considering. > >There will always be some unintentional errors, but I am confident >enough of mapping villages and sections of towns systematically to be >able to assert that I have completed it to that measure of completeness >- that I have visited every street and got every name possible. (But I >don't mind using some other word for it if you like, e.g. a confidence >level or some such). +1, I don't mind what words we use either. > >I think I would have a much harder time being systematic about >footpaths, especially the rural ones, so I wouldn't have the same degree >of confidence in my mapping of a footpath network. Maybe if that's what >I specialised in my confidence would grow, but the concept of junctions >where you can note the other routes from from need attention seems a >much less well defined concept for footpaths. > >But the main point about footpaths is that using that as the only >measure would be very dispiriting because it would be so hard to >complete any reasonable areas to that standard, and completeness at the >street level is very useful for lots of purposes that doesn't require >footpaths so is worth showing to consumers. So maybe we have a list of all the main way types (highway, waterway etc) and we put a completeness level/confidence level on each one. Too many though and contributors will loose interest. Perhaps just: Roads Cycleways/Bridleways Footways Rivers/Major Streams/Canals Railways Cheers Andy > >David > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
Andy Allan [mailto:[EMAIL PROTECTED] wrote: >Sent: 12 May 2008 10:52 PM >To: David Earl >Cc: Andy Robinson (blackadder-lists); talk@openstreetmap.org >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help provide >"completeness" tools > >On Mon, May 12, 2008 at 8:16 PM, David Earl <[EMAIL PROTECTED]> >wrote: > >> I think it is terribly hard to know whether you have all the footpaths, >> and I think we'd hardly ever mark anywhere "complete" if we did that. > >I think it's terribly hard to know when a map is correct and complete, >regardless of what you're considering. > >In fact, as something I've floated with some people before, I think >the idea of "completeness" is the wrong way round. I think we should >be considering where a map is *incomplete*. > >Think about it. If you are presented with a map of your village and >asked whether it's right, it's very unlikely that you know all the >roads and all the names and how everything connects and be sure of >yourself. But it's quite likely that what you'll spot (if anything) is >a mistake or a missing road. I can do this with TIGER stuff for >example - I can't tell you if the map is correct, but I can definitely >find bits that are definitely wrong. Some of us map out an area completely in one go rather than doing it piecemeal. Even if I come across some existing roads in a new area I ignore them and do a new survey so that the whole area makes logical sence to me. That way I can work out where the landuse areas are behind the houses and the extent of school areas etc etc. So for me a reasonable level of completeness is easy to decide and annotate. I accept where many users touch the same area, especially areas with Y! imagery then this approach probably is not workable. Cheers Andy > >I've been considering what I'd do to Wandsworth and Fulham (my local >areas) if someone asked me to mark which areas are correct. I think >there's very little value in me doing so, since most roads I've only >been down once and hardly likely to check the name from memory. But >there's a couple of bits that are definitely wrong, and I can point >them out easily. > >It's just another way of thinking about it, but I think that >neutral/wrong is probably more useful than neutral/complete when >considering maps. And it certainly cuts out trying to define >'complete', since whichever reason something is wrong (name, >connectivity, missingness) it's very easy to state why it's wrong. And >rather than more and more being complete (for this, that and the other >definition of complete) there'll be fewer and fewer bits that are >wrong. > >Nobody ever looks at a map and remarked how many bits were correct. >Nor does any software product keep a list of lines of code that are >working. Or it's an 'exception driven' way of considering things. > >Cheers, >Andy > >No virus found in this incoming message. >Checked by AVG. >Version: 8.0.100 / Virus Database: 269.23.16/1429 - Release Date: >12/05/2008 6:14 PM ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Florian Lohoff schrieb: | On Mon, May 12, 2008 at 07:06:20PM +0200, Inge Wallin wrote: |> Yes, but it would be good if we planned for at least some of the subsequent |> levels right from the start. | | I support this idea - I would call them levels as some areas might work | completely different. I would call the different completenesses like | | roadcomplete | cyclewaycomplete | roadnamecomplete | landusecomplete | | etc ... I have no idea about the granularity but in the end it comes | down to a map never containing ALL interesting data but it would be | interesting to note which subset it contains or is complete. | | The completeness is more or less a feeling of the primary mapper of | this specific area so it should be taken with a grain of salt i guess. I would like to point out the system Munich, Bremen, and other city-communities use on our wiki to mark their completeness. it's not leveled, just different areas (an not necessarily complete as an indicating scheme). - -- Dirk-Lüder "Deelkar" Kreie Bremen - 53.0952°N 8.8652°E -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFIKK3gFUbODdpRVDwRAgWOAKCpL+HsiNT53a3U8wHnymX9BWRtXQCgxc8Y tqFo2yOQSvpWbz+9tNr7YPw= =CEPn -END PGP SIGNATURE- ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On 12/05/2008 22:51, Andy Allan wrote: > On Mon, May 12, 2008 at 8:16 PM, David Earl <[EMAIL PROTECTED]> wrote: > >> I think it is terribly hard to know whether you have all the footpaths, >> and I think we'd hardly ever mark anywhere "complete" if we did that. > > I think it's terribly hard to know when a map is correct and complete, > regardless of what you're considering. There will always be some unintentional errors, but I am confident enough of mapping villages and sections of towns systematically to be able to assert that I have completed it to that measure of completeness - that I have visited every street and got every name possible. (But I don't mind using some other word for it if you like, e.g. a confidence level or some such). I think I would have a much harder time being systematic about footpaths, especially the rural ones, so I wouldn't have the same degree of confidence in my mapping of a footpath network. Maybe if that's what I specialised in my confidence would grow, but the concept of junctions where you can note the other routes from from need attention seems a much less well defined concept for footpaths. But the main point about footpaths is that using that as the only measure would be very dispiriting because it would be so hard to complete any reasonable areas to that standard, and completeness at the street level is very useful for lots of purposes that doesn't require footpaths so is worth showing to consumers. David ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On Mon, May 12, 2008 at 8:16 PM, David Earl <[EMAIL PROTECTED]> wrote: > I think it is terribly hard to know whether you have all the footpaths, > and I think we'd hardly ever mark anywhere "complete" if we did that. I think it's terribly hard to know when a map is correct and complete, regardless of what you're considering. In fact, as something I've floated with some people before, I think the idea of "completeness" is the wrong way round. I think we should be considering where a map is *incomplete*. Think about it. If you are presented with a map of your village and asked whether it's right, it's very unlikely that you know all the roads and all the names and how everything connects and be sure of yourself. But it's quite likely that what you'll spot (if anything) is a mistake or a missing road. I can do this with TIGER stuff for example - I can't tell you if the map is correct, but I can definitely find bits that are definitely wrong. I've been considering what I'd do to Wandsworth and Fulham (my local areas) if someone asked me to mark which areas are correct. I think there's very little value in me doing so, since most roads I've only been down once and hardly likely to check the name from memory. But there's a couple of bits that are definitely wrong, and I can point them out easily. It's just another way of thinking about it, but I think that neutral/wrong is probably more useful than neutral/complete when considering maps. And it certainly cuts out trying to define 'complete', since whichever reason something is wrong (name, connectivity, missingness) it's very easy to state why it's wrong. And rather than more and more being complete (for this, that and the other definition of complete) there'll be fewer and fewer bits that are wrong. Nobody ever looks at a map and remarked how many bits were correct. Nor does any software product keep a list of lines of code that are working. Or it's an 'exception driven' way of considering things. Cheers, Andy ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
[OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
Freek recently created this image which shows how much of the AND data is untouched: http://www.vanwal.nl/osm/author_density_nl_20080502_full.png(warning 3 MB image). I think the z18 idea is good idea. On Mon, May 12, 2008 at 11:36 PM, David Earl <[EMAIL PROTECTED]> wrote: > On 12/05/2008 21:02, Cartinus wrote: > > On Monday 12 May 2008 21:16:38 David Earl wrote: > >> So I think Inge is right - we need different measures for our own use. > >> But on the public map, all streets with names seems a pretty good > >> achievable and useful thing to show. > > > > I don't think it is a good idea to call that just "complete". I think it > > should be made very clear that it is just the road network that is > complete > > and not the map. > > Fine. > > > With the import of the AND data the road network in the Netherlands is > mostly > > complete. Now we often hear: "Why do you still need to map more? It is > > complete with the AND data, isn't it?" > > > > Having more levels or categories of completeness makes it clear that we > are > > not finished yet after we put the roads in the database. > > I think we are violently agreeing here! > > David > > > > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk > ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On 12/05/2008 21:02, Cartinus wrote: > On Monday 12 May 2008 21:16:38 David Earl wrote: >> So I think Inge is right - we need different measures for our own use. >> But on the public map, all streets with names seems a pretty good >> achievable and useful thing to show. > > I don't think it is a good idea to call that just "complete". I think it > should be made very clear that it is just the road network that is complete > and not the map. Fine. > With the import of the AND data the road network in the Netherlands is mostly > complete. Now we often hear: "Why do you still need to map more? It is > complete with the AND data, isn't it?" > > Having more levels or categories of completeness makes it clear that we are > not finished yet after we put the roads in the database. I think we are violently agreeing here! David ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
Hi, > I'm very far from this in Korea, but I would guess in time some > parts of the UK will need to be rechecked at some point. How can we > make a system for rechecking an area? Maybe the completeness should > be retired after a period of time. Once we have a few applications in place that get viewed by *many* people, we could just have a button somewhere along the margin of the page that says: "I know the area and what I see here looks correct". That would not give us any safety but if, when looking at a part of the map, you knew that within the last 6 months 178 people had clicked on "this looks correct" then that would perhaps give you at least a warm fuzzy feeling ;-) Bye Frederik -- Frederik Ramm ## eMail [EMAIL PROTECTED] ## N49°00'09" E008°23'33" ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
I'm very far from this in Korea, but I would guess in time some parts of the UK will need to be rechecked at some point. How can we make a system for rechecking an area? Maybe the completeness should be retired after a period of time. -- http://bowlad.com ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On Mon, May 12, 2008 at 07:06:20PM +0200, Inge Wallin wrote: > Yes, but it would be good if we planned for at least some of the subsequent > levels right from the start. I support this idea - I would call them levels as some areas might work completely different. I would call the different completenesses like roadcomplete cyclewaycomplete roadnamecomplete landusecomplete etc ... I have no idea about the granularity but in the end it comes down to a map never containing ALL interesting data but it would be interesting to note which subset it contains or is complete. The completeness is more or less a feeling of the primary mapper of this specific area so it should be taken with a grain of salt i guess. But this information would be interesting for a map bug tracker which could assign higher prioritys to bugs in "complete" areas or even refuse to accept bugs in incomplete areas. Flo -- Florian Lohoff [EMAIL PROTECTED] +49-171-2280134 Those who would give up a little freedom to get a little security shall soon have neither - Benjamin Franklin signature.asc Description: Digital signature ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On 12/05/2008 20:02, Andy Robinson (blackadder-lists) wrote: > David Earl wrote: >> Sent: 12 May 2008 7:10 PM >> To: [EMAIL PROTECTED] >> Cc: talk@openstreetmap.org >> Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help provide >> "completeness" tools >> >> On 12/05/2008 18:06, Inge Wallin wrote: >>> On Monday 12 May 2008 18:06:59 Chris Morley wrote: >>> >>>> Starting with a single level of completeness makes sense, but I think it >>>> should be public roads, named where feasible. >>> I have a different view. I think we should have a leveled scheme from >> the >>> beginning. I suggest the following: >>> >>> Level 1: All the highways (using OSM lingo) usable by cars within an area >> are >>> mapped >>> Level 2: All highways are mapped and named >>> Level 3: All highways down to cycleways are mapped (and named if >> feasible). >> >> That's a very car-centric view of the world. Why "down" to cycleways? >> Who are you to say something usable by a car is more important than >> something usable by a bike? > > I was actually going to argue that even the footways should be on. That's > why I was going for the tile approach because I felt that building up from > little pieces was more logical that an all encompassing area. If you have a > few footpaths in your area not completed then its not really complete, > whereas small tile can be "signed off" and holes wouldn't matter, they would > just get filled in later as you or someone else gets to them. I wasn't being entirely serious. I think it is terribly hard to know whether you have all the footpaths, and I think we'd hardly ever mark anywhere "complete" if we did that. So I think Inge is right - we need different measures for our own use. But on the public map, all streets with names seems a pretty good achievable and useful thing to show. David ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
David Earl wrote: >Sent: 12 May 2008 7:10 PM >To: [EMAIL PROTECTED] >Cc: talk@openstreetmap.org >Subject: Re: [OSM-talk] [OSM-dev] Developers requested to help provide >"completeness" tools > >On 12/05/2008 18:06, Inge Wallin wrote: >> On Monday 12 May 2008 18:06:59 Chris Morley wrote: >> >>> Starting with a single level of completeness makes sense, but I think it >>> should be public roads, named where feasible. >> >> I have a different view. I think we should have a leveled scheme from >the >> beginning. I suggest the following: >> >> Level 1: All the highways (using OSM lingo) usable by cars within an area >are >> mapped >> Level 2: All highways are mapped and named >> Level 3: All highways down to cycleways are mapped (and named if >feasible). > >That's a very car-centric view of the world. Why "down" to cycleways? >Who are you to say something usable by a car is more important than >something usable by a bike? I was actually going to argue that even the footways should be on. That's why I was going for the tile approach because I felt that building up from little pieces was more logical that an all encompassing area. If you have a few footpaths in your area not completed then its not really complete, whereas small tile can be "signed off" and holes wouldn't matter, they would just get filled in later as you or someone else gets to them. Cheers Andy > >David > >___ >talk mailing list >talk@openstreetmap.org >http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk > >No virus found in this incoming message. >Checked by AVG. >Version: 8.0.100 / Virus Database: 269.23.16/1427 - Release Date: >11/05/2008 1:08 PM ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On 12/05/2008 18:06, Inge Wallin wrote: > On Monday 12 May 2008 18:06:59 Chris Morley wrote: > >> Starting with a single level of completeness makes sense, but I think it >> should be public roads, named where feasible. > > I have a different view. I think we should have a leveled scheme from the > beginning. I suggest the following: > > Level 1: All the highways (using OSM lingo) usable by cars within an area are > mapped > Level 2: All highways are mapped and named > Level 3: All highways down to cycleways are mapped (and named if feasible). That's a very car-centric view of the world. Why "down" to cycleways? Who are you to say something usable by a car is more important than something usable by a bike? David ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On Monday 12 May 2008 18:06:59 Chris Morley wrote: > Starting with a single level of completeness makes sense, but I think it > should be public roads, named where feasible. I have a different view. I think we should have a leveled scheme from the beginning. I suggest the following: Level 1: All the highways (using OSM lingo) usable by cars within an area are mapped Level 2: All highways are mapped and named Level 3: All highways down to cycleways are mapped (and named if feasible). -- Level 4: Level 3 + all amenities within a given set. Level 5: Level 4 + maybe some other amenities or buildings or whatever we decide. These levels map(!) nicely to how I work myself when I map out an area: start with the roads and streets for the cars, and name them. Then later go back to the area and add the cycleways. Sometimes I map a whole lot of other areas before coming back to map the cycleways. Regarding the levels: I can understand if some people would switch level 2 and 3 and I would be fine with that. When an area reaches level 3, I would say that it is "finished" for the purpose of finding the way (remember it's Open _Street_Map :-) ). All extra information is a bonus, and necessary for certain types of maps, but for the general user or a router, level 3 should be all that is needed. > This covers the essential > basic requirements for a number of potential applications for example, > routing, delivery, estate agents, bus services. The points of interest, > etc. are clearly desirable and but may not be always collected. For > instance, tracing from arial photos and naming from an out of copyright > gazeteer; or imports like TIGER. (Also I'm not sure I collected all of > them when I started 2 years ago.) Start basic and have these in a > subsequent level. Yes, but it would be good if we planned for at least some of the subsequent levels right from the start. -Inge > Chris > > ___ > talk mailing list > talk@openstreetmap.org > http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On Monday 12 May 2008 18:06:59 Chris Morley wrote: > a) Being able to expand the boundary after each session is likely to be > a great motivator. I suspect that this progressive "taming of the > wilderness" or "making order out of the void" is what drives many mappers. > > b) Applicability to small or irregular areas (route to work?) might > encourage more users. Andy was talking about zoom 18 squares, which map less than 100x100m of the real world each at the latitude of London and Amsterdam. This would be sufficiently detailed to do both those things. -- m.v.g., Cartinus ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
David Earl wrote: > Here is what I was going to do: > > 1. use a set of ways like coastline (i.e. with an "on the right" rule, > because they'd be too long as one way), to define areas of completeness. > By definition, coastline would form one boundary. > > 2. Render these plus coastline to a new set of tiles (possibly only at > one zoom level, say 13 or 14) so that complete areas are transparent and > incomplete areas are semi transparent red, say. Use the ocean tiles > database to avoid putting red over all areas of sea. > > 3. Present these tiles as a layer. For other zooms either combine or > divide, or sample, or simply rescale in the browser - the edges need > only be quite coarse. > > I felt this leveraged most from the tools we already have so would be > the most straightforward to implement. > > I was going to have only one measure of "complete", that is the surveyor > asserts that all publicly-accessible roads are present, with names for > all except for those impossible to determine and when that is indicated, > and all key points of interest from a limited set: schools, pubs, places > of worship; and any railways and significant watercourses. > > I think it would be confusing for a consumer to have lots of variations > in what "complete" means - just an indicator to say "you can't really > trust this area yet" would be simple. > > However, since the boundaries can be tagged, there would not be any > problem about expanding this for internal use to cover degrees of > completeness. I think this is the right approach. I prefer the completeness boundary being a tagged way over the predefined squares approach suggested by Andy because: a) Being able to expand the boundary after each session is likely to be a great motivator. I suspect that this progressive "taming of the wilderness" or "making order out of the void" is what drives many mappers. b) Applicability to small or irregular areas (route to work?) might encourage more users. c) No additional tools or procedures are need by the mapper. Starting with a single level of completeness makes sense, but I think it should be public roads, named where feasible. This covers the essential basic requirements for a number of potential applications for example, routing, delivery, estate agents, bus services. The points of interest, etc. are clearly desirable and but may not be always collected. For instance, tracing from arial photos and naming from an out of copyright gazeteer; or imports like TIGER. (Also I'm not sure I collected all of them when I started 2 years ago.) Start basic and have these in a subsequent level. Chris ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
Re: [OSM-talk] [OSM-dev] Developers requested to help provide "completeness" tools
On 12/05/2008 11:59, Andy Robinson (blackadder-lists) wrote: > Posted to talk and dev > > This post really follows on from my consideration of Completeness metrics. A > number of experienced OSMers who have mapped out there areas have either > contacted me directly or posted through the lists about areas of OSM that > are essentially complete, at least as complete as you would expect from a > conventional on-line map or better. ... I've raised this a number of times in the past and have been thinking about it recently - to the point I was thinking about implementing something. Here is what I was going to do: 1. use a set of ways like coastline (i.e. with an "on the right" rule, because they'd be too long as one way), to define areas of completeness. By definition, coastline would form one boundary. 2. Render these plus coastline to a new set of tiles (possibly only at one zoom level, say 13 or 14) so that complete areas are transparent and incomplete areas are semi transparent red, say. Use the ocean tiles database to avoid putting red over all areas of sea. 3. Present these tiles as a layer. For other zooms either combine or divide, or sample, or simply rescale in the browser - the edges need only be quite coarse. I felt this leveraged most from the tools we already have so would be the most straightforward to implement. I was going to have only one measure of "complete", that is the surveyor asserts that all publicly-accessible roads are present, with names for all except for those impossible to determine and when that is indicated, and all key points of interest from a limited set: schools, pubs, places of worship; and any railways and significant watercourses. I think it would be confusing for a consumer to have lots of variations in what "complete" means - just an indicator to say "you can't really trust this area yet" would be simple. However, since the boundaries can be tagged, there would not be any problem about expanding this for internal use to cover degrees of completeness. David ___ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
