Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On Fri, Feb 4, 2011 at 1:11 PM, Ed Avis e...@waniasset.com wrote: Stephen Gower socks-openstreetmap.org@... writes: What do you do when a road has completely gone? I make a way with not:name and no other tags. That doesn't work - it has to have a recognised highway tag. Peter says it will be changed some time soon so that any highway tag (eg highway=abandoned) will get it included in the comparison. Richard ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 9 February 2011 14:44, Richard Mann richard.mann.westoxf...@gmail.comwrote: On Fri, Feb 4, 2011 at 1:11 PM, Ed Avis e...@waniasset.com wrote: Stephen Gower socks-openstreetmap.org@... writes: What do you do when a road has completely gone? I make a way with not:name and no other tags. That doesn't work - it has to have a recognised highway tag. Peter says it will be changed some time soon so that any highway tag (eg highway=abandoned) will get it included in the comparison. Actually we are going to accept an value for the highway tag, so it could be highway=banana or highway=not or whatever. This will reduce ongoing maintenance as people invent new highway tags anyway. Just start tagging as you see fit and it will start working on the next release. Not sure when that will be exactly though. Regards, Peter Richard ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller peter.miller@... writes: What do you do when a road has completely gone? Actually we are going to accept an value for the highway tag, so it could be highway=banana or highway=not or whatever. Cool, so we can add spurious ways tagged with not:name=xxx and highway=no to suppress the check. There are a couple of cases like this in central London which I will retag soon. -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Andy wrote: I've just checked a few well-mapped areas - Tendring, Hull and Edinburgh - and the not:name is running at 2%, 3.1% and 1.9% of all the roads. I'm just catching up as I was out all of yesterday (always seems to be loads of emails when I'm not keeping current) so this may have been covered already, so if so sorry. In Tendring most of the not:name tags are from after I went out surveying all the apostrophe discrepancies and the name I'm using in OSM is the name on the road sign. I also have looked this morning at the 4 new roads I have to go check in the district. Two seem to have added names to service roads, unfortunately at opposite ends of the district (Brightlingsea and Dovercourt). One is a road I marked as highway=road, but missed the FIXME=stub and hadn't noticed it since. It wasn't named in the last set of data, so this has helped me locate an actual road I'd missed. The fourth is a new road (recently built) that I had already surveyed: http://www.openstreetmap.org/edit?lat=51.843074lon=1.073781zoom=18 and the OS have the name wrong (they missed an L from MITCHELL), so I've tagged it not:name. Even bing only shows it as a building site so I was quite proud when I added it all last June in pre-bing days. From the foundations I can see in bing though I seem to have estimated the location of buildings slightly out though. Having said that I have also added roads based on a combination of StreetView, OS Locator and Bing recently in the Bridgnorth district of Shropshire. I have added source:name=OS_OpenData_Locator where I have done so in much the same way as we used to add source=NPE to roads traced from the out of copyright OS maps (and I find realigning these a bit of a pain, too). I chose that area as I have some (admittedly a little outdated) knowledge of the area - I worked in Bridgnorth for a couple of years and we spent many weekends enjoying visiting areas of Shropshire when I used to live in Wolverhampton; I have had friends that lived in Bridgnorth and someone I was at school at was a councillor there. Looking at OSM Analysis I'm guessing I added about 359 such roads which increases coverage from about 29% to about 68%. While doing so I found a number of obvious typos in the OSM data on the few roads which were mapped, and there are roads I didn't touch because my recollection of the name agrees with OSM but I'm not sure enough to add a not:name tag without a survey. I have a feeling in one case the OSM name might be the local name everyone uses rather than the official name. I agree that we need to grow the community, but will having traced roads (whether NPE, Streetview, Bing or whatever) affect this greatly? I know it will upset existing active mappers in an area if it is done without consideration to their ongoing efforts, but in quiet areas of the country might the addition of roads change some people's perception from OSM is rubbish in my area to Gosh, OSM has got my street on it - perhaps I can add my house/the path I take to cut to the shops/whatever? Take for example Skobbler - in Colchester there are a few missing road reports which need a survey (too new to be on bing/os/fake reports), which is great (I've cleared away all rubbish reports to see the useful ones that remain). But what about someone trying to use Skobbler in an area where most of the roads are missing? Will they bother reporting lots of missing roads or get the impression OSM data is of no use? Too many questions, and judging by the many emails yesterday lots of different views. Ed ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Hi, I would back the use of such a bot but in area's that could do with the help and there are only a few of these now. There are a couple of small towns in Aberdeenshire that could do with a little boost. Just be careful. The rest will follow with a little patience I think it is a skill that may be useful to learn and a preparation for other data sets such as hopitals, churches, schools. This could be used as an overlay and not necessarily an import and very useful for other less developed countries There are a large number of motivations. If we have more publicity we can have more people I would also like to point out the news of aerial image tracing http://opengeodata.org/ leading to http://www.bing.com/community/site_blogs/b/maps/archive/2011/02/03/automatically-detect-roads-with-bing-aerial-imagery.aspx Which leads to the video. As far as I can make out, all data sources are incorrect in different ways. Some GPS surveys can be a mess,street name markers misspelled or wrong. Satellite imaging can be incorrectly georeferenced or out of date as are any data sources you can think of. I remember hearing that the district councils are getting together to release their official street name data. When this happens then OSM and OS data will converge so that we are all using the same names and then the street signs can be changed to the correct name. Cheers Bob ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
I agree with you 100% on this. I think if OSM is street-level complete (preferably with postcodes as well) then it will be picked up by a lot more developers for their iPhone and Android apps and the amount of feedback we could get would be a 100 times greater than now. A standardised, OSM hosted, bug reporting api could also be offered to developers so they don't end up building their own versions of MapDust. Kevin On 4 February 2011 01:14, Kai Krueger kakrue...@gmail.com wrote: Also, we are getting some of the most comprehensive on the ground verification and improvement reports from applications like the Sat-Nav Skobbler bugs with MapDust. The more complete the map is, the more people will use things like that and give feedback on errors. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
I would agree with this too; A fuller map will greatly increase usage and visibility. However I don't think that a bot is necessarily the way to go. What would be useful is a simple way for users to specify a bounding box/polygon and download an OSM file with the OS data (or OS updates to OSM data such as adding a name where one does not exist), which they can then manually check and upload as a changeset. This would allow data to be quickly imported semi-automatically in areas which are blank in osm, without the risk of a bot overwriting data one someones' patch. Andy On Fri, 4 Feb 2011 11:00:18 +, Kevin Peat ke...@kevinpeat.com wrote: I agree with you 100% on this. I think if OSM is street-level complete (preferably with postcodes as well) then it will be picked up by a lot more developers for their iPhone and Android apps and the amount of feedback we could get would be a 100 times greater than now. A standardised, OSM hosted, bug reporting api could also be offered to developers so they don't end up building their own versions of MapDust. Kevin On 4 February 2011 01:14, Kai Krueger kakrue...@gmail.com wrote: Also, we are getting some of the most comprehensive on the ground verification and improvement reports from applications like the Sat-Nav Skobbler bugs with MapDust. The more complete the map is, the more people will use things like that and give feedback on errors. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 4 February 2011 10:58, Bob Kerr openstreetmapcraigmil...@yahoo.co.ukwrote: Hi, I would back the use of such a bot but in area's that could do with the help and there are only a few of these now. There are a couple of small towns in Aberdeenshire that could do with a little boost. Just be careful. The rest will follow with a little patience I think it is a skill that may be useful to learn and a preparation for other data sets such as hopitals, churches, schools. This could be used as an overlay and not necessarily an import and very useful for other less developed countries You raise a good point there. I am aware that we still have many place names missing (available as open data in the NatGaz file released by Traveline/DfT). There is also a DB of all schools in the UK which is possibly appropriately licensed - we would need to check that one. Then there is an update of NaPTAN - we did one import and the source data has changed significantly and it has only been imported for part of the country anyway. There will also be many other DBs coming available and we need a way to sync to them while still maintaining a life! There are a large number of motivations. If we have more publicity we can have more people I would also like to point out the news of aerial image tracing http://opengeodata.org/ leading to http://www.bing.com/community/site_blogs/b/maps/archive/2011/02/03/automatically-detect-roads-with-bing-aerial-imagery.aspx Which leads to the video. As far as I can make out, all data sources are incorrect in different ways. Some GPS surveys can be a mess,street name markers misspelled or wrong. Satellite imaging can be incorrectly georeferenced or out of date as are any data sources you can think of. Agreed. I remember hearing that the district councils are getting together to release their official street name data. When this happens then OSM and OS data will converge so that we are all using the same names and then the street signs can be changed to the correct name. Journalists talk about 'the data deluge' and to my mind we need to be prepared for it. Clearly the current government is committed to opening up all sorts of datasets. Street name gazeteers? lamp post locations and IDs etc etc. Regards, Peter Cheers Bob ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller peter.miller@... writes: I am aware that we still have many place names missing (available as open data in the NatGaz file released by Traveline/DfT). Place name nodes are a useful thing to have for address searches and should not raise any objections about messing up existing data, provided they're added only if there isn't an existing place node within 10km. So I would suggest importing this data if it is of reasonably high quality. -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Ed Avis wrote: This is the one thing that perturbs me too about using the OS data. Back in the days when we only had Yahoo, I would not tag the name on a way until I had walked all the way down it checking for footpaths. If I'd only explored part of the way, I would tag the name on that part only. Then the noname map was a guide to unsurveyed areas. But now there isn't such a clear indication. So I did hesitate about adding name=Newton Road to a street which I had not visited. But then I considered the folk living on that street typing its name into Nominatim and getting no results. There is no reason to make them wait for a thorough ground survey of their street before it appears in OSM at all. We must make the best map we can now, and improve it later. I am thoroughly bemused by the widespread attitude of if it isn't in OSM, it doesn't exist. OS OpenData is a fantastic resource and, believe it or not, you can use it outside OSM. In most ways it's easier to use than OSM: the data quality is consistent, everything is neatly layered, you don't get some roads drawn at 1pt/500m and others at 1pt/2m, the .shp format is an industry standard, and, of course, OS/CO chose a sane licence. And people _are_ using it. I spotted a major national charity the other day which has just started using VectorMap District. There are iPhone apps using it. I produce three or four maps for Waterways World every month using Meridian2, and am just about to make a couple of cycling maps for West Oxfordshire District Council using VectorMap District. So. The data consumer has two choices. a) They can have consistent, reliable data with national coverage right now. That's OS OpenData. b) Or they can have glorious rich data which doesn't have national coverage _yet_. That's OSM. If we import OS OpenData into OSM, then OSM becomes national *but* is no longer rich. Because no-one is out there trudging the streets, because no-one cares for the data, it becomes sterile. I note, once again, that the Worcester estates that someone helpfully traced from StreetView last year are still bereft of most of the footpaths and half of the road names. So we lose the chance of a real, community-built map of the UK. Meanwhile we have helped precisely no-one, because the complete data *already* exists and is *already* usable. Can't we actually have a go at doing it ourselves and finishing the UK this year? Say right, let's look at a bot in spring 2012, but we have a year to get this right? I'll do a town this weekend if you will. cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5992349.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On Thu, Feb 03, 2011 at 11:18:19AM +, Peter Miller wrote: I use the following method. If the OS name is different from the streetsign and general usage I put it in not:name If it is apparently a valid alternative I put it on alt_name I am not clear why anything else is required. What do you do when a road has completely gone? Mascall Avenue in Oxford has completely gone. There's a new housing estate, with a road network that doesn't match what was there before, so there's no way to mark with old_name or not:name or anything. s ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 4 Feb 2011, at 12:55, Stephen Gower wrote: What do you do when a road has completely gone? Mascall Avenue in Oxford has completely gone. There's a new housing estate, with a road network that doesn't match what was there before, so there's no way to mark with old_name or not:name or anything. I've a similar situation in my area. There are some disused roads that are visible on the ground if you know where to look, but others have been completely erased from the ground but still exist in the Locator data. James ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 4 February 2011 12:55, Stephen Gower socks-openstreetmap@earth.liwrote: On Thu, Feb 03, 2011 at 11:18:19AM +, Peter Miller wrote: I use the following method. If the OS name is different from the streetsign and general usage I put it in not:name If it is apparently a valid alternative I put it on alt_name I am not clear why anything else is required. What do you do when a road has completely gone? Mascall Avenue in Oxford has completely gone. There's a new housing estate, with a road network that doesn't match what was there before, so there's no way to mark with old_name or not:name or anything. I suggest that the main upload process should spot an attempt to add back a feature which has been removed with the message ' Mascall Avenue was removed by -this- edit on xx-xx- with the comment 'blar blar' Are you sure you want to add it back'. A bot would also ideally do this prior to it getting to the upload stage. Wikipedia has a similar warning for people attempting to re-create a previously deleted article. Peter s ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Richard Fairhurst richard@... writes: If we import OS OpenData into OSM, then OSM becomes national *but* is no longer rich. I can't agree with that at all. A blank area is not any more 'rich' than one with basic details complete. We do not improve the map by leaving blank bits. There are parts of OSM today which have only very basic street layout, worse than the OS data. Do you suggest deleting them to make the map 'richer'? Because no-one is out there trudging the streets, because no-one cares for the data, it becomes sterile. I note, once again, that the Worcester estates that someone helpfully traced from StreetView last year are still bereft of most of the footpaths and half of the road names. And estates that nobody traced from StreetView, and were left blank... are mostly still blank. Perhaps we do need some better motivating tool to encourage people to visit areas which have been only traced from OS. When things are traced from Yahoo the old 'nonames' tiles straightforwardly directs mappers to go and visit them on foot. An equivalent for OS might be useful. I have been thinking, too, about a 'noname buildings' layer for large buildings with no name or address. Can't we actually have a go at doing it ourselves and finishing the UK this year? Say right, let's look at a bot in spring 2012, but we have a year to get this right? I think that would be a fair compromise - I think we have already waited long enough, but I'm willing to go with your suggestion. It does not imply at all that in 2012 we should just give up on mapping parties and community outreach and all the other good things you have talked about. Only that we should consider using all the other resources we have too, to make the best map possible. -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Brief expansion of previous point: Ed Avis wrote: So I did hesitate about adding name=Newton Road to a street which I had not visited. But then I considered the folk living on that street typing its name into Nominatim and getting no results. Let's see what Wikipedia does (and when you have to take usability lessons from the bracket-infested rulefest that is Wikipedia, you _know_ you've got problems): http://en.wikipedia.org/wiki/Frog_Pellets Frog Pellets From Wikipedia, the free encyclopedia Wikipedia does not have an article with this exact name. [Start the Frog Pellets article], using the [Article Wizard] if you wish, or [add a request for it]. Brilliant. Why aren't we doing that? OpenStreetMap hasn't mapped Newton Road, Chigwell yet. Why not add it? We can know where Newton Road is by keeping OS Locator in a separate table. We can use this to take the user directly to the right place in Potlatch, with Bing imagery automatically selected, nice tutorial video showing them how to do it. (I've asked MapQuest to work on tutorial videos because, guess what, none of you lot have had the imagination to think if I were a new user, what would help me most?.) We get a new OSM user and _real_, unique local data with the footpaths, the pubs, the postboxes, all of that. Or we could just write a bot. cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5992534.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Richard, I don't think we need a bot for this as the current tools seem quite adequate to me. If the missing streets are added this year then that would be great. Building a community is ideal but I think outside the successful parts of the country we are not going to get a lot of people wanting to do ground surveys of whole towns. Mapping is a geeky activity and in large parts of the SW, Wales and the North there just aren't that many of those people about. As I said before I think the only realistic approach in those areas is to get the maps onto as many devices as we can (for which they need to be street level complete) and then cultivate feedback from satnav/smartphone users to add some richness to the data. Hopefully some of those people would join the OSM community along the way. Kevin On 4 February 2011 12:15, Richard Fairhurst rich...@systemed.net wrote: Can't we actually have a go at doing it ourselves and finishing the UK this year? Say right, let's look at a bot in spring 2012, but we have a year to get this right? I'll do a town this weekend if you will. cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5992349.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller peter.miller@... writes: What do you do when a road has completely gone? Mascall Avenue in Oxford has completely gone. There's a new housing estate, with a road network that doesn't match what was there before, so there's no way to mark with old_name or not:name or anything. I suggest that the main upload process should spot an attempt to add back a feature which has been removed with the message ' Mascall Avenue was removed by -this- edit on xx-xx- with the comment 'blar blar' Are you sure you want to add it back'. A bot would also ideally do this prior to it getting to the upload stage. Wikipedia has a similar warning for people attempting to re-create a previously deleted article. Hmm, just possibly, but I think relying on user intelligence and good judgement is better. (The mapper needs to have those anyway.) Nobody would blindly trace a street from OS through the middle of some existing features - at least I hope not! And any putative bot should definitely behave better than that too. I would suggest that whoever removed Mascall Avenue from the map should have mapped what replaced it - a brownfield site or whatever - to avoid future confusion. It's most unlikely for features to be removed from OSM and replaced with an entirely blank canvas. The Wikipedia parallel is not exact: an article is deleted from Wikipedia usually not because the thing it refers to doesn't exist in the real world, but because it is not 'notable' or 'verifiable' or has been merged into another article. OSM has a much simpler criterion of whether the thing exists. -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, ?000 UK roads quickly using a Bot?
On Fri, Feb 04, 2011 at 01:25:31PM +, Ed Avis wrote: I would suggest that whoever removed Mascall Avenue from the map should have mapped what replaced it - a brownfield site or whatever - to avoid future confusion. For what it's worth, we did - there's now a landuse=residential; access=private polygon around the area. Now it's open, we could probably infer the building polygon as well - what was Mascall Avenue goes straight through the middle of this. s ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 2 February 2011 21:10, Peter Miller peter.mil...@itoworld.com wrote: It could do the following: 1) Add names to existing roads in osm where there is a single un-named ways in osm with a bounding box which matches that of a single entry in os locator. 2) In addition... it might be able to also add roads to osm from os vector district, snapping them into existing roads as required where the existing roads align neatly with os streetview. It would only do this if there were no ways close by on either side. Complex situations will be left to humans. Humans could also sometimes prepare an area for analysis by the bot, splitting ways as appropriate, adjusting alignment of existing roads and dealing in advance with situations we know the bot will have difficulties with. Edits would be made as individual changesets, referenced to the mapper operating of the bot. Each edit would be 'signed off' by the mapper who would be able to see the proposed changes visual prior to accepting them. I'd be happy to review a few of these edits for Gwynedd, I've no objection if it works. On another ITO-analysis note, Peter you are driving me slightly potty because new road errors keep popping up. I presume this is because of new OS data? Just when I thought I had all of Southwark bar the north east ear sorted, another six errors pop up. Grr. Tom -- http://tom.acrewoods.net http://twitter.com/tom_chance ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 08:47, Tom Chance t...@acrewoods.net wrote: On 2 February 2011 21:10, Peter Miller peter.mil...@itoworld.com wrote: It could do the following: 1) Add names to existing roads in osm where there is a single un-named ways in osm with a bounding box which matches that of a single entry in os locator. 2) In addition... it might be able to also add roads to osm from os vector district, snapping them into existing roads as required where the existing roads align neatly with os streetview. It would only do this if there were no ways close by on either side. Complex situations will be left to humans. Humans could also sometimes prepare an area for analysis by the bot, splitting ways as appropriate, adjusting alignment of existing roads and dealing in advance with situations we know the bot will have difficulties with. Edits would be made as individual changesets, referenced to the mapper operating of the bot. Each edit would be 'signed off' by the mapper who would be able to see the proposed changes visual prior to accepting them. To be clear, ITO are not proposing to write this bot but we would be happy to encourage and support it to happen if there is a general mood that it would be useful and achievable. I'd be happy to review a few of these edits for Gwynedd, I've no objection if it works. On another ITO-analysis note, Peter you are driving me slightly potty because new road errors keep popping up. I presume this is because of new OS data? Just when I thought I had all of Southwark bar the north east ear sorted, another six errors pop up. Grr. Don't worry - this will only happen once every three months when the OS publish an updated OS Locator file. We updated to the latest OS Locator version (dated November 10) last night, hence a few places have fallen off their '100% perch'! We should be due another one reasonably soon I guess and may get it up sooner. In time I hope that we will find that some of our 'not:name' reports will have been fixed by the OS. I hear that the OS is getting much more receptive to this whole Open Data thing. One layer I would like ITO to produce would be the reverse OS Locator view, which would be for the OS's use (and our amusement!). It would show all the named roads that are in OSM but which are not in the OS. They would then need to research why that is the case and update their own products (without compromising our license). Regards, Peter Tom -- http://tom.acrewoods.net http://twitter.com/tom_chance ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter I have been using the excellent analysis tool to try to complete roads data in my Borough (Enfield). I am now left with some that need a further very careful walk/bike survey. This is to deal with parts of North Circular and Hertford Road with weird combinations of road/ref names according to OS. I also have several spellers from OS data that are wrong and thus need reporting and will stop achievement of 100% till they are agreed/changed/reloaded (eg Smtyhe Close). I have looked on wiki and Ito blog for info on reporting these not:names but can't find info anywhere. Can you please point me to some info? Keep up the grand work. Cheers STEVE From: Peter Miller [peter.mil...@itoworld.com] Sent: Thursday, February 03, 2011 9:36 AM To: Tom Chance Cc: Talk-GB@openstreetmap.org Subject: Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot? On 3 February 2011 08:47, Tom Chance t...@acrewoods.netmailto:t...@acrewoods.net wrote: On 2 February 2011 21:10, Peter Miller peter.mil...@itoworld.commailto:peter.mil...@itoworld.com wrote: It could do the following: 1) Add names to existing roads in osm where there is a single un-named ways in osm with a bounding box which matches that of a single entry in os locator. 2) In addition... it might be able to also add roads to osm from os vector district, snapping them into existing roads as required where the existing roads align neatly with os streetview. It would only do this if there were no ways close by on either side. Complex situations will be left to humans. Humans could also sometimes prepare an area for analysis by the bot, splitting ways as appropriate, adjusting alignment of existing roads and dealing in advance with situations we know the bot will have difficulties with. Edits would be made as individual changesets, referenced to the mapper operating of the bot. Each edit would be 'signed off' by the mapper who would be able to see the proposed changes visual prior to accepting them. To be clear, ITO are not proposing to write this bot but we would be happy to encourage and support it to happen if there is a general mood that it would be useful and achievable. I'd be happy to review a few of these edits for Gwynedd, I've no objection if it works. On another ITO-analysis note, Peter you are driving me slightly potty because new road errors keep popping up. I presume this is because of new OS data? Just when I thought I had all of Southwark bar the north east ear sorted, another six errors pop up. Grr. Don't worry - this will only happen once every three months when the OS publish an updated OS Locator file. We updated to the latest OS Locator version (dated November 10) last night, hence a few places have fallen off their '100% perch'! We should be due another one reasonably soon I guess and may get it up sooner. In time I hope that we will find that some of our 'not:name' reports will have been fixed by the OS. I hear that the OS is getting much more receptive to this whole Open Data thing. One layer I would like ITO to produce would be the reverse OS Locator view, which would be for the OS's use (and our amusement!). It would show all the named roads that are in OSM but which are not in the OS. They would then need to research why that is the case and update their own products (without compromising our license). Regards, Peter Tom -- http://tom.acrewoods.net http://twitter.com/tom_chance ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller wrote: Don't worry - this will only happen once every three months when the OS publish an updated OS Locator file. We updated to the latest OS Locator version (dated November 10) last night, hence a few places have fallen off their '100% perch'! We should be due another one reasonably soon I guess and may get it up sooner. In time I hope that we will find that some of our 'not:name' reports will have been fixed by the OS. I hear that the OS is getting much more receptive to this whole Open Data thing. One layer I would like ITO to produce would be the reverse OS Locator view, which would be for the OS's use (and our amusement!). It would show all the named roads that are in OSM but which are not in the OS. They would then need to research why that is the case and update their own products (without compromising our license). Peter In my own area there are not too many discrepancies, but some OS names ARE wrong compared to the ground details. Have we got a way of flagging the name:os so that we can identify these and flag them as 'different' rather than 'missing' -- Lester Caine - G8HFL - Contact - http://lsces.co.uk/wiki/?page=contact L.S.Caine Electronic Services - http://lsces.co.uk EnquirySolve - http://enquirysolve.com/ Model Engineers Digital Workshop - http://medw.co.uk// Firebird - http://www.firebirdsql.org/index.php ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller wrote: Any thoughts? Very, very sceptical. We are slowly coming out of a dismal winter and getting back into the season when we can do real surveying. That is, and always will be, OSM's strength. If a bot can fix OSM by mashing it up with OS data, it can just as easily post the result to a third-party server without interfering with OSM. There is ample evidence that imports delay, or stop, the development of a mapping community. It is not enough to say a local mapper approved it - well-intentioned but rogue local mappers can seriously foul up an area, as plenty of people here can testify. If you merely want complete, (largely) accurate OS data but with the advantages of the OSM toolset, it is trivial to process VectorMap or Meridian2 into .osm files and run your favourite OSM utilities over them. Speaking locally: If anyone were to propose running this on the West Oxfordshire/Cotswolds area I would revert it as a matter of course, _unless_ very good reasons had been agreed in advance on the oxon-cotswolds@ mailing list. cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5988237.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
+1 PHILLIP BARNETT SERVER MANAGER 200 GRAY'S INN ROAD LONDON WC1X 8XZ UNITED KINGDOM T +44 (0)20 7430 4474 F E phillip.barn...@itn.co.uk WWW.ITN.CO.UK Please consider the environment. Do you really need to print this email? -Original Message- From: Richard Fairhurst [mailto:rich...@systemed.net] Sent: 03 February 2011 10:15 To: talk-gb@openstreetmap.org Subject: Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot? Peter Miller wrote: Any thoughts? Very, very sceptical. We are slowly coming out of a dismal winter and getting back into the season when we can do real surveying. That is, and always will be, OSM's strength. If a bot can fix OSM by mashing it up with OS data, it can just as easily post the result to a third-party server without interfering with OSM. There is ample evidence that imports delay, or stop, the development of a mapping community. It is not enough to say a local mapper approved it - well-intentioned but rogue local mappers can seriously foul up an area, as plenty of people here can testify. If you merely want complete, (largely) accurate OS data but with the advantages of the OSM toolset, it is trivial to process VectorMap or Meridian2 into .osm files and run your favourite OSM utilities over them. Speaking locally: If anyone were to propose running this on the West Oxfordshire/Cotswolds area I would revert it as a matter of course, _unless_ very good reasons had been agreed in advance on the oxon-cotswolds@ mailing list. cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5988237.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb Please Note: Any views or opinions are solely those of the author and do not necessarily represent those of Independent Television News Limited unless specifically stated. This email and any files attached are confidential and intended solely for the use of the individual or entity to which they are addressed. If you have received this email in error, please notify postmas...@itn.co.uk Please note that to ensure regulatory compliance and for the protection of our clients and business, we may monitor and read messages sent to and from our systems. Thank You. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Richard Fairhurst wrote: Speaking locally: If anyone were to propose running this on the West Oxfordshire/Cotswolds area I would revert it as a matter of course,_unless_ very good reasons had been agreed in advance on the oxon-cotswolds@ mailing list. I think I'm probably with you on my area of the Cotswolds. A lot of the 'missing' roads are simply where a name is misspelt, or a section has not kept the name when someone split it. I'm left with the -'s- problem and some roads which I know are right on OSM ( hence the question about flagging that ;) ) -- Lester Caine - G8HFL - Contact - http://lsces.co.uk/wiki/?page=contact L.S.Caine Electronic Services - http://lsces.co.uk EnquirySolve - http://enquirysolve.com/ Model Engineers Digital Workshop - http://medw.co.uk// Firebird - http://www.firebirdsql.org/index.php ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On Wed, Feb 2, 2011 at 9:10 PM, Peter Miller peter.mil...@itoworld.com wrote: ITO have been offering a service to compare osm road names with os locator road names for a while now[1] which has encouraged a lot of activity - and has even led to Andy to obsession.[2] I have also suffered from a bout of urgent mapping myself while completing all of Suffolk to 95% in the past few weeks! Can I suggest that for our sanity we should consider developing a bot to do some of this work for us? This would also allow us to get the rest of the 250,000 remaining roads in place in less than the 13 months Andy estimates will be required? I'd rather you thought again, and I feel embarrassed that my blog post has had such a radically different outcome that I was intending. The ITO analysis is a useful healthcheck of the OSM community in different areas, and a useful way to motivate mappers in partly-mapped areas and encourage them to drive the quality of OSM higher in well-mapped areas. For example, I normally go for a quick blast on my bike in the morning, but this morning I took some detours and did some mapping to investigate the discrepancies - collecting bike parking, finding closed pubs and checking on building developments along the way. If we take the healthcheck analogy, the equivalent solution to finding high blood pressure would be a bot to go around letting blood - when the better way of doing it is to encourage the patient to get some exercise. That's similar to my intentions in this project - while we could all sit at home blindly copying from OS data (or, worse still, sit at home writing software to blindly copy from OS data, shudder), we could instead be figuring out how to get more mappers, and motivate more mappers, to go do some mapping. Completing the road network is a by product of having a healthy mapping community. So what I hoped would happen were that people would read my post, spring up on this list and say things like: I know some sustrans rangers / scouts / guides in Powys and so I've ordered them some promotional leaflets to get them interested [1] I like your thinking in London, we're organising some pub meetups in Inverclyde next month I've found a list of libraries in my area and I'm running some workshops in Bolton I normally go for a quick blast on my bike in the morning, but this morning I took some detours[2] I find the talk of bots disheartening, and it misses the key challenge entirely - it's not hard to check a street sign, or transcribe lettering between OS Locator and Potlatch2. It's hard to recruit, motivate and retain mappers, and that's what we should be putting our efforts in, for that's the proper measure of progress. Cheers, Andy [1] Available for free from http://shop.opencyclemap.org/products/openstreetmap-promotional-leaflets [2] See what I did there? ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 02/02/2011 21:10, Peter Miller wrote: ITO have been offering a service to compare osm road names with os locator road names for a while now[1] which has encouraged a lot of activity - and has even led to Andy to obsession.[2] I have also suffered from a bout of urgent mapping myself while completing all of Suffolk to 95% in the past few weeks! Can I suggest that for our sanity we should consider developing a bot to do some of this work for us? This would also allow us to get the rest of the 250,000 remaining roads in place in less than the 13 months Andy estimates will be required? This bot would do a number of repetitive tasks for us within the bounding box in which it was authorised to operate by a contributor. It could do the following: 1) Add names to existing roads in osm where there is a single un-named ways in osm with a bounding box which matches that of a single entry in os locator. 2) In addition... it might be able to also add roads to osm from os vector district, snapping them into existing roads as required where the existing roads align neatly with os streetview. It would only do this if there were no ways close by on either side. Complex situations will be left to humans. Humans could also sometimes prepare an area for analysis by the bot, splitting ways as appropriate, adjusting alignment of existing roads and dealing in advance with situations we know the bot will have difficulties with. Edits would be made as individual changesets, referenced to the mapper operating of the bot. Each edit would be 'signed off' by the mapper who would be able to see the proposed changes visual prior to accepting them. Any thoughts? Peter [1] http://www.itoworld.com/product/data/osm_analysis/summary [2] http://www.gravitystorm.co.uk/shine/archives/2011/02/02/the-london-streets-challenge/ ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb I personally don't think it's a great idea. There are many aspects to this, so I'll just take the ones which occur to me right now. * Automating road completion is not a huge impossiblity. In a desultory way I have been playing with name assignment to VectorMap District data, and I'm sure that approached in a more systematic and determined fashion it would feasible to produce a programmatic way to assign Locator names to the VDM data set. I have a pretty good idea of the major issues, and the outline algorithms to do this. The main step I have not tried is sticking combinations of linestrings together to maximise fit to a locator box. Connectivity is the other main issue. However, I think such data would be pretty much useful on their own, or could be mashed-up with OSM data for a given application (e.g., a garmin map). I don't see huge immediate utility in putting such data into OSM, as opposed to making it available in such a way that it could be integrated with OSM data. * Relying on OS data reduces the range of sources and validation of OSM. I'm currently experimenting http://sk53-osm.blogspot.com/2011/01/simulating-urban-atlas-using-osm.html with the Nottingham area to see how close I can get to the Urban Atlas mapping done for the EEA. My gut feel is that OSM data which is sourced through combinations of ground-survey and aerial imagery can provide similar levels of accuracy in terms of areas, and higher levels of reliability in terms of landuse classification. In other words, once OSM data start to get very detailed they provide a separately surveyed source of data which is distinct from OSGB. If we populate huge swathes of the country from OSGB data we lose this value (but see next). * Huge swathes of the country *are to all intents and purposes* only populated with OSGB data. Most towns in Northern England appear to have been largely traced: obvious examples: Darlington, Middlesborough, Bolton, Oldham, Rochdale, Grimsby, many parts of South Yorkshire. In many cases tracers using OS StreetView have not bothered to add road names (from time to time, I have been doing sweeps through Oldham fixing some of these). The few journeys I have made in these places indicate a poor level of quality. Adding data does not add mappers. * Many areas traced from Yahoo have received little or no ground survey. Most of Merseyside, Greater Manchester and West London fall into these categories. In some cases a concerted effort has been made to at least add street names from OS StreetView. See comment above. * We are still nowhere near 1 dedicated mapper per City/District which seems to be the minimum to give a mapped area 'life'. Of course we really want a group of people maintaining an area, but I'd settle for A lifeless but complete map
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
I am listening to the feedback. Lets look that this from my perspective. I have spent a lot of time filling out Suffolk with data from the OS.This started in Ipswich with using the ITO compare tiles to spot errors and omissions in the town itself. This started over Xmas I decided to experimented using Potlatch 2, OS Open Data and Bing aerial to map a completely unmapped town (Lakenheath) to OSM as far as I could without visting the place. It worked very well and I then added Brandon and finished off Mildenhall. Did I remove anyone's work? No. Did I overwrite anyone's street names? certainly not. I then did the same for a bunch of smaller places across Suffolk that weren't mapped out completely. I then worked on the name discrepancies where OS Locator and OSM disagreed to decide who was right. If OSM was right then I put the wrong spelling in not:name; if OSM was wrong then I would correct the name in OSM. A couple of us at ITO then did the same in parts of Cambs using the same method. I have been tempted to do Thapston, Northants which I know reasonably well. I could follow the same process as I have done above and it would take about 10 hours. I would prefer to run a bot of the town, add all the geometry from os vector district, add new names from OS Locator and then do a clean-up pass on discrepencies which would take half the time. I look at Bradford and Leeds with 4,000 roads missing each and think - this is not a good use of time. Some of the work has to be done by people, but it can be speeded up considerably using tools to do the starting work. Will this discourage mapping? I don't believe so - there will still be loads of landuse and paths to add, it will just free people up to do that work rather than repetitive jobs that a computer can do. By getting the UK mapping up to a good basic standard then more people will start using it and will thereby be encouraged to add more detail where they need it. Am I proposing running it all over the country? Certainly not, it would be used by a mapper as part of a longer process to fix up an area of interest to them. Fyi, ITO will soon have to give the OS £13,500 for another's years use of their ITN dataset and then additional usage and printing fees during the year. As such I really want to get to the point where we can say 'no thanks'! Regards, Peter On 3 February 2011 10:38, Andy Allan gravityst...@gmail.com wrote: On Wed, Feb 2, 2011 at 9:10 PM, Peter Miller peter.mil...@itoworld.com wrote: ITO have been offering a service to compare osm road names with os locator road names for a while now[1] which has encouraged a lot of activity - and has even led to Andy to obsession.[2] I have also suffered from a bout of urgent mapping myself while completing all of Suffolk to 95% in the past few weeks! Can I suggest that for our sanity we should consider developing a bot to do some of this work for us? This would also allow us to get the rest of the 250,000 remaining roads in place in less than the 13 months Andy estimates will be required? I'd rather you thought again, and I feel embarrassed that my blog post has had such a radically different outcome that I was intending. The ITO analysis is a useful healthcheck of the OSM community in different areas, and a useful way to motivate mappers in partly-mapped areas and encourage them to drive the quality of OSM higher in well-mapped areas. For example, I normally go for a quick blast on my bike in the morning, but this morning I took some detours and did some mapping to investigate the discrepancies - collecting bike parking, finding closed pubs and checking on building developments along the way. If we take the healthcheck analogy, the equivalent solution to finding high blood pressure would be a bot to go around letting blood - when the better way of doing it is to encourage the patient to get some exercise. That's similar to my intentions in this project - while we could all sit at home blindly copying from OS data (or, worse still, sit at home writing software to blindly copy from OS data, shudder), we could instead be figuring out how to get more mappers, and motivate more mappers, to go do some mapping. Completing the road network is a by product of having a healthy mapping community. So what I hoped would happen were that people would read my post, spring up on this list and say things like: I know some sustrans rangers / scouts / guides in Powys and so I've ordered them some promotional leaflets to get them interested [1] I like your thinking in London, we're organising some pub meetups in Inverclyde next month I've found a list of libraries in my area and I'm running some workshops in Bolton I normally go for a quick blast on my bike in the morning, but this morning I took some detours[2] I find the talk of bots disheartening, and it misses the key challenge entirely - it's not hard to check a street sign, or transcribe lettering
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Thanks for your thoughts on this. I won't respond to individual points not because I don't think they are valid, but because I think it will be useful to hear from others and let the conversation develop. I am of course aware that there is ia lot of concern about the proposal in the comments so far. One point of clarification though - I did indeed consider that I and other could be 'locked out' of OSM due to my use of OS Open Data, however that is no longer the case given that the OS have adopted the Open Government License. Regards, Peter On 3 February 2011 11:03, Jerry Clough : SK53 on OSM sk53_...@yahoo.co.ukwrote: On 02/02/2011 21:10, Peter Miller wrote: ITO have been offering a service to compare osm road names with os locator road names for a while now[1] which has encouraged a lot of activity - and has even led to Andy to obsession.[2] I have also suffered from a bout of urgent mapping myself while completing all of Suffolk to 95% in the past few weeks! Can I suggest that for our sanity we should consider developing a bot to do some of this work for us? This would also allow us to get the rest of the 250,000 remaining roads in place in less than the 13 months Andy estimates will be required? This bot would do a number of repetitive tasks for us within the bounding box in which it was authorised to operate by a contributor. It could do the following: 1) Add names to existing roads in osm where there is a single un-named ways in osm with a bounding box which matches that of a single entry in os locator. 2) In addition... it might be able to also add roads to osm from os vector district, snapping them into existing roads as required where the existing roads align neatly with os streetview. It would only do this if there were no ways close by on either side. Complex situations will be left to humans. Humans could also sometimes prepare an area for analysis by the bot, splitting ways as appropriate, adjusting alignment of existing roads and dealing in advance with situations we know the bot will have difficulties with. Edits would be made as individual changesets, referenced to the mapper operating of the bot. Each edit would be 'signed off' by the mapper who would be able to see the proposed changes visual prior to accepting them. Any thoughts? Peter [1] http://www.itoworld.com/product/data/osm_analysis/summary [2] http://www.gravitystorm.co.uk/shine/archives/2011/02/02/the-london-streets-challenge/ ___ Talk-GB mailing list Talk-GB@openstreetmap.orghttp://lists.openstreetmap.org/listinfo/talk-gb I personally don't think it's a great idea. There are many aspects to this, so I'll just take the ones which occur to me right now. - Automating road completion is not a huge impossiblity. In a desultory way I have been playing with name assignment to VectorMap District data, and I'm sure that approached in a more systematic and determined fashion it would feasible to produce a programmatic way to assign Locator names to the VDM data set. I have a pretty good idea of the major issues, and the outline algorithms to do this. The main step I have not tried is sticking combinations of linestrings together to maximise fit to a locator box. Connectivity is the other main issue. However, I think such data would be pretty much useful on their own, or could be mashed-up with OSM data for a given application (e.g., a garmin map). I don't see huge immediate utility in putting such data into OSM, as opposed to making it available in such a way that it could be integrated with OSM data. - Relying on OS data reduces the range of sources and validation of OSM. I'm currently experimentinghttp://sk53-osm.blogspot.com/2011/01/simulating-urban-atlas-using-osm.htmlwith the Nottingham area to see how close I can get to the Urban Atlas mapping done for the EEA. My gut feel is that OSM data which is sourced through combinations of ground-survey and aerial imagery can provide similar levels of accuracy in terms of areas, and higher levels of reliability in terms of landuse classification. In other words, once OSM data start to get very detailed they provide a separately surveyed source of data which is distinct from OSGB. If we populate huge swathes of the country from OSGB data we lose this value (but see next). - Huge swathes of the country *are to all intents and purposes* only populated with OSGB data. Most towns in Northern England appear to have been largely traced: obvious examples: Darlington, Middlesborough, Bolton, Oldham, Rochdale, Grimsby, many parts of South Yorkshire. In many cases tracers using OS StreetView have not bothered to add road names (from time to time, I have been doing sweeps through Oldham fixing some of these). The few journeys I have made in these places indicate a poor level
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 03/02/11 11:04, Peter Miller wrote: Fyi, ITO will soon have to give the OS £13,500 for another's years use of their ITN dataset and then additional usage and printing fees during the year. As such I really want to get to the point where we can say 'no thanks'! As I suspected, the real motivation has now emerged. Let's not compromise our integrity to save ITO some money. -- Cheers, Chris user: chillly ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 10:04, Lester Caine les...@lsces.co.uk wrote: Peter Miller wrote: Don't worry - this will only happen once every three months when the OS publish an updated OS Locator file. We updated to the latest OS Locator version (dated November 10) last night, hence a few places have fallen off their '100% perch'! We should be due another one reasonably soon I guess and may get it up sooner. In time I hope that we will find that some of our 'not:name' reports will have been fixed by the OS. I hear that the OS is getting much more receptive to this whole Open Data thing. One layer I would like ITO to produce would be the reverse OS Locator view, which would be for the OS's use (and our amusement!). It would show all the named roads that are in OSM but which are not in the OS. They would then need to research why that is the case and update their own products (without compromising our license). Peter In my own area there are not too many discrepancies, but some OS names ARE wrong compared to the ground details. Have we got a way of flagging the name:os so that we can identify these and flag them as 'different' rather than 'missing' I use the following method. If the OS name is different from the streetsign and general usage I put it in not:name If it is apparently a valid alternative I put it on alt_name I am not clear why anything else is required. One other technique I have found useful which is worth mentioning is with A roads (such as the A14) where part of that are technically still Huntingdon Road (and I remember it as a kid when it was one lane in each direction with a chicken crossing over the A1). It isn't appropriate these days to have the A14 labeled as 'Huntingdon Road' however Huntingdon Road is not 'wrong' either. In these cases I put the name Hungingdon Road into alt_name and leave name empty. In this way it is cleared from the OS Locator comparison test but isn't rendered on the main map. Regards, Peter -- Lester Caine - G8HFL - Contact - http://lsces.co.uk/wiki/?page=contact L.S.Caine Electronic Services - http://lsces.co.uk EnquirySolve - http://enquirysolve.com/ Model Engineers Digital Workshop - http://medw.co.uk// Firebird - http://www.firebirdsql.org/index.php ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On Thu, Feb 3, 2011 at 11:04 AM, Peter Miller peter.mil...@itoworld.com wrote: This started over Xmas I decided to experimented using Potlatch 2, OS Open Data and Bing aerial to map a completely unmapped town (Lakenheath) to OSM as far as I could without visting the place. It worked very well and I then added Brandon and finished off Mildenhall. Did I remove anyone's work? No. Did I overwrite anyone's street names? certainly not. Did you check that OS OpenData was correct? Certainly not. I've just checked a few well-mapped areas - Tendring, Hull and Edinburgh - and the not:name is running at 2%, 3.1% and 1.9% of all the roads. So for every 100 roads you've added, you could be adding 1 to 3 bogus names - and since we've only got the OS Locator to help spot mistakes, you're making them very hard to find. I would prefer to run a bot of the town, add all the geometry from os vector district, add new names from OS Locator and then do a clean-up pass on discrepencies which would take half the time. Again - missing the point. You're trying to speed up blindly copying OS data into OSM, when you shouldn't be blindly copying it in the first place. Come back when the problem is I've spent a day out mapping and X could make me work faster. Cheers, Andy ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On Thu, Feb 3, 2011 at 11:03 AM, Jerry Clough : SK53 on OSM sk53_...@yahoo.co.uk wrote: OS OpenData is out-of-date. The April 2010 StreetView tiles are at least 2 years old, and where I've checked VDM is similarly dated. I have not failed to find a significant change between OS OpenData (and Bing imagery) in detailed surveys I've done this year. Chris Hill has a similar experience. +1. i've had to remove several things which were traced and tagged from OS which are no longer present. for example; [1], which was demolished in 2007 (and deconsecrated some time previously). it's not the use of OS data which is the problem, it's using it in areas of which there's no recent local knowledge and, pretty much by definition, any bot would have no local knowledge ;-) cheers, matt [1] http://www.openstreetmap.org/browse/changeset/6152842 ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller peter.miller@... writes: If the OS name is different from the streetsign and general usage I put it in not:name If it is apparently a valid alternative I put it on alt_name I hadn't realized that your OS Locator data check looks at alt_name too, thanks. Sometimes it is necessary to add both alt_name and not_name. For example OS Locator contains the name 'Edgware Road The Hyde'. In OSM we don't tag with both names concatenated like that, so I added name=The Hyde, alt_name=Edgware Road (chosen since 'The Hyde' is more common in addressing), and then not:name to quieten the check. It isn't appropriate these days to have the A14 labeled as 'Huntingdon Road' however Huntingdon Road is not 'wrong' either. In these cases I put the name Hungingdon Road into alt_name and leave name empty. In this way it is cleared from the OS Locator comparison test but isn't rendered on the main map. That's one way to do it. I think I would prefer to put the name in the name tag but then perhaps have some kind of hint to say that it should not be rendered. (Or a general rule that 'A' roads be shown with their number not a name.) -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Andy Allan gravitystorm@... writes: This started over Xmas I decided to experimented using Potlatch 2, OS Open Data and Bing aerial to map a completely unmapped town (Lakenheath) to OSM as far as I could without visting the place. I've just checked a few well-mapped areas - Tendring, Hull and Edinburgh - and the not:name is running at 2%, 3.1% and 1.9% of all the roads. So for every 100 roads you've added, you could be adding 1 to 3 bogus names Just a quick data point - from my experience in London visiting streets to check the name mismatches between OSM and OS Locator, the ratio of mistakes is about ten to one in favour of OS Locator. In other words when the two disagree, it is almost always OSM which has been wrong on the ground. (Today, after cleanup work, the OSM names are more correct.) This is without counting obvious misspellings such as 'Stret' instead of 'Street', which don't need a resurvey to correct them. (They are plentiful in OSM but somewhat rare in OS Locator.) Come back when the problem is I've spent a day out mapping and X could make me work faster. I think this is exactly the situation. In my experience, using the OS data is both faster and more accurate than manual surveying, and I have done a great deal of both. -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller wrote: I use the following method. If the OS name is different from the streetsign and general usage I put it in not:name If it is apparently a valid alternative I put it on alt_name I am not clear why anything else is required. Probably half of the 'errors' in this area are due to full stops and apostrophes differences. I though OS has generally dropped them, but a few names that are not currently matching because of an apostrophe in one or other and their use or not is probably a religious war area ;) I'm not sure that an alt_name entry is necessary in this case and I wonder how many of the remaining matches would be made if these were simply stripped before comparing? -- Lester Caine - G8HFL - Contact - http://lsces.co.uk/wiki/?page=contact L.S.Caine Electronic Services - http://lsces.co.uk EnquirySolve - http://enquirysolve.com/ Model Engineers Digital Workshop - http://medw.co.uk// Firebird - http://www.firebirdsql.org/index.php ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 11:44, Lester Caine les...@lsces.co.uk wrote: Peter Miller wrote: I use the following method. If the OS name is different from the streetsign and general usage I put it in not:name If it is apparently a valid alternative I put it on alt_name I am not clear why anything else is required. Probably half of the 'errors' in this area are due to full stops and apostrophes differences. I though OS has generally dropped them, but a few names that are not currently matching because of an apostrophe in one or other and their use or not is probably a religious war area ;) I'm not sure that an alt_name entry is necessary in this case and I wonder how many of the remaining matches would be made if these were simply stripped before comparing? Fyi, apostrophe differences are classed as 'minor errors' on the mapping and the summary. On the mapping they are shown in grey rather than a colour. In the list they are in a fainter colour with a not attached. Other punctuation differences are treated as 'major errors' which may not be ideal but will do for now. Peter -- Lester Caine - G8HFL - Contact - http://lsces.co.uk/wiki/?page=contact L.S.Caine Electronic Services - http://lsces.co.uk EnquirySolve - http://enquirysolve.com/ Model Engineers Digital Workshop - http://medw.co.uk// Firebird - http://www.firebirdsql.org/index.php ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
I think automatically importing the OS data for areas where OSM currently has little to no coverage - or coverage merely traced from Yahoo imagery - is a great idea. Many people don't want bots to trample on their 'patch' which they have carefully surveyed, which is fine. Any bot would have to be thoroughly tested. As you said I wouldn't like the data in my area being mucked up by a buggy bot and trying, incorrectly, to be clever in the way that certain well-known commercial software products can be! Nick ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Ed Avis wrote: Lastly, I don't believe that adding data from external sources discourages contributors. Quite the opposite. It is a blank canvas that puts people off. The way to bring in contributors is to show a map with a few missing details that are so tempting to fix 'just one thing'... Is there an example of a road import that has led to an increase in contributors? I thought that in most cases it had the exact opposite effect. IMO the blank canvas is what pulls people in: an area that's 90% already there finds it harder to attract new mappers as it already looks done (filling in all the footpaths, post boxes, pubs, etc, is something you tend to do once you're already hooked). Comparisons to a reference source like OS are still valuable, but more as a way to prioritise mapping: a vague heatmap of patchy areas is more motivating to me than than an exhaustive list of streets. If you want a 1:1 copy of the OS data, why not just use the OS data? -dair ___ d...@refnum.com http://www.refnum.com/ ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 11:51, Ed Avis e...@waniasset.com wrote: I think automatically importing the OS data for areas where OSM currently has little to no coverage - or coverage merely traced from Yahoo imagery - is a great idea. Many people don't want bots to trample on their 'patch' which they have carefully surveyed, which is fine. But there is still a huge part of the country which is very under-mapped. Let's get it complete as soon as possible using the best data we have. (It is my opinion that the OS data is often of higher quality than a typical OSM ground survey, although of course multiple surveys by different people combined with the use of Bing imagery will give the best possible map.) I also think that what is good for ITO is good for the map in general - if there is a need to have a complete map for some government contract or whatever, then that is a need that exists in the real world. Lastly, I don't believe that adding data from external sources discourages contributors. Quite the opposite. It is a blank canvas that puts people off. The way to bring in contributors is to show a map with a few missing details that are so tempting to fix 'just one thing'... Thank you Ed, I think there are two underlying philosophies running here: 1) OSM should either be 'right' or the map should be empty (which by definition is not wrong because there is nothing that is wrong on the map). The only way to make sure it is 'right' before adding data is to visit the site. Or 2) An empty bit of the map is by definition 'wrong' and the in the UK the OS data is almost certainly more right than that and as such an edit that adds OS data as an import to blank areas will make it more 'right'. The map can then subsequently be made even more 'right' by people who know or visit the area. Matt raises a useful side issues that people mapping at a distance should be confident that they are not actually making the map worse by use of out-of-date information. I have suggested a way of adding notes to the background layers or a warning during data upload. Nick has just commented that such a bot would need to be well tested. I agree and it would need to be trialed in small areas to gain confident. There are well established protocols for doing 'type approval' and MOT's on Bots and allowing for feedback both here and in Wikipedia. This one would need to follow the guidelines. There is suggestion raised by a number of people, but refuted by others that imports reduce the number of contributors. I believe that more people will correct an error than go and map a huge blank area of the map. Regards, Peter -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Dair Grant dair@... writes: Lastly, I don't believe that adding data from external sources discourages contributors. Quite the opposite. It is a blank canvas that puts people off. The way to bring in contributors is to show a map with a few missing details that are so tempting to fix 'just one thing'... Is there an example of a road import that has led to an increase in contributors? I thought that in most cases it had the exact opposite effect. I was thinking in particular of the Yahoo tracing. In the beginning OSM was largely blank and the only way to add new ways was to go out and make GPS tracks. I suggest, but cannot prove, that seeing an entirely blank canvas doesn't entice you to start adding to the map, which must necessarily involve adding small bits at a time. Once the Yahoo aerial photos became available and many OSM contributors traced large areas from them (yes, even though they had not visited those areas on the ground!) then anyone could contribute small bits and pieces such as the name of their local street or the pub on the corner. This is how I got started. Now importing ways from OS data is not quite the same as tracing from an aerial photo but I suggest a similar principle applies. IMO the blank canvas is what pulls people in: an area that's 90% already there finds it harder to attract new mappers as it already looks done (filling in all the footpaths, post boxes, pubs, etc, is something you tend to do once you're already hooked). On the other hand what you say here sounds plausible too. However, we do have some real-world evidence. There are towns which have been blank for a long time on OSM. If a blank canvas were a good way to encourage contributors, they would have been filled in by now. The fact that they are still missing suggests that the strategy of deliberately leaving an area blank and hoping for somebody to go out and survey it from scratch is not always effective. The way the OSM project has grown is by the principle of 'do what you can'. Map something roughly using the best of your knowledge, which is still better than leaving it entirely unmapped. Then somebody can come along and improve it later. If you want a 1:1 copy of the OS data, why not just use the OS data? What I want (and I think what others want) is a map in machine-readable form which corresponds to the real world. The source it originally came from does not matter. -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
I'm not sure that mapping parties by non-locals are that different from armchair tracing in terms of community engagement. Rather than a bot, I'd aim for a tool that suggested the OS name if you draw a road or click on an existing road without a name. That would save an awful lot of typing while still requiring review-as-you-go I think we are more likely to build community by using the data, particularly by drawing maps with it (cue ad for Maperitive). It has to be said that you can't draw a map if there's no data. Richard ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On Thu, Feb 3, 2011 at 11:55 AM, Peter Miller peter.mil...@itoworld.com wrote: On 3 February 2011 11:32, Matt Amos zerebub...@gmail.com wrote: On Thu, Feb 3, 2011 at 11:03 AM, Jerry Clough : SK53 on OSM sk53_...@yahoo.co.uk wrote: OS OpenData is out-of-date. The April 2010 StreetView tiles are at least 2 years old, and where I've checked VDM is similarly dated. I have not failed to find a significant change between OS OpenData (and Bing imagery) in detailed surveys I've done this year. Chris Hill has a similar experience. +1. i've had to remove several things which were traced and tagged from OS which are no longer present. for example; [1], which was demolished in 2007 (and deconsecrated some time previously). it's not the use of OS data which is the problem, it's using it in areas of which there's no recent local knowledge and, pretty much by definition, any bot would have no local knowledge ;-) Would it be useful to be able to add annotations to base layers to indicate where they are wrong. For example to add a polygon to a Bing or Yahoo aerial or OS Streetview layer that partly obsures the image and says 'this area has changed' or something similar. This would need to be flagged for re-checking when the source material is updated. it seems to me that this would simply add another place for stuff to be wrong, whether out of date or simply mis-entered or misunderstood. all data is out-of-date the moment it's been surveyed, so while other data sources are useful, i prefer to base my own efforts on knowledge of the area or real surveying. Another approach would be for the submit process to put up an alert 'you are adding a feature which has previously been deleted - the deletion included the following comment in the changeset'. with fuzzy matching of the features to prevent against misspelling, differences in tagging, etc... this sounds very cool, but my SMOP-sense is tingling. ;-) as andy pointed out, i think we're addressing the wrong problem and trying to fix it technically. maybe the best way forward is to address the social problem: what can we do to grow the community? is it just me, or did we used to have more mapping parties? (maybe it's just the winter) do we need to try and reach out to cycling / youth / technology SIGs and hope that some of them find this as addictive as we do? OSM is a wiki, which means it's only mildly annoying when people trace OS / Yahoo / Bing data into my local area (or import a bunch of massively positionally inaccurate bus stops). i can watch the area using the tools available and correct it. but my local area is quite small - how can we get more people to monitor and garden their own local area? finally, if it's a rich, accurate and detailed data set you want then importing OS data isn't going to help. someone will need to put in the extra stuff that's not on OS / Yahoo / Bing. so we're going to need people on the ground surveying or living there anyway... cheers, matt ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter Miller wrote: There is suggestion raised by a number of people, but refuted by others that imports reduce the number of contributors. It has been denied, not refuted. I think the closest there is to real data on the effect is: http://www.asklater.com/matt/wordpress/2009/09/imports-and-the-community-ii Personally I feel the impact of road network imports is more substantial than of any other kind. Importing bus stops or post boxes feels like a step forward, importing a road network feels like a step back (why that should be the case, I don't know: perhaps due to the greater visibility of roads on the default map). -dair ___ d...@refnum.com http://www.refnum.com/ ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 03/02/11 11:51, Ed Avis wrote: Lastly, I don't believe that adding data from external sources discourages contributors. Quite the opposite. It is a blank canvas that puts people off. The way to bring in contributors is to show a map with a few missing details that are so tempting to fix 'just one thing'... I don't think you can generalise this - the blank canvas attracts some people (including me) and repels others. However, I used to be quite active in wikipedia, and the experience there is clearly that there are far, far more people who want to be editors than who want to contribute brand-new content. I suspect if OSM did more to attract gardeners as someone else put it, there could be a big, largely untapped source of contributors out there. Jonathan. -- Jonathan Harley: Managing Director : SpiffyMap Ltd Email: m...@spiffymap.com Phone: 0845 313 8457 www.spiffymap.com Post: The Venture Centre, Sir William Lyons Road, Coventry CV4 7EZ ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Ed Avis wrote: I suggest, but cannot prove, that seeing an entirely blank canvas doesn't entice you to start adding to the map, which must necessarily involve adding small bits at a time. Actually, seeing a blank canvas in my local area was exactly my motivation. But everyone is different, as proved by us having had the exact opposite reasons for starting. :-) However, we do have some real-world evidence. There are towns which have been blank for a long time on OSM. If a blank canvas were a good way to encourage contributors, they would have been filled in by now. The fact that they are still missing suggests that the strategy of deliberately leaving an area blank and hoping for somebody to go out and survey it from scratch is not always effective. Part of that problem is that we have no way to distinguish this area is unmapped from this area is under-mapped - they both look blank. If you could rewind time to the start of OSM, an import strategy I think would have been useful would have been to start with a set of towns and create straight lines between them for roads. That makes it obvious there's something there, and that it can be improved, without having the tiger effect of going from an empty map to one which looks complete but is full of bad data. If you want a 1:1 copy of the OS data, why not just use the OS data? What I want (and I think what others want) is a map in machine-readable form which corresponds to the real world. The source it originally came from does not matter. At the data level, no. But for the longer term impact on motivation, I think it does. Things change over time of course, so perhaps people who were interested in mapping by hand will drift away and people who're more interested in automated maintenance will come in. But I can't say the latter sounds like a very interesting thing to spend my free time on. -dair ___ d...@refnum.com http://www.refnum.com/ ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Dair Grant dair@... writes: There is suggestion raised by a number of people, but refuted by others that imports reduce the number of contributors. It has been denied, not refuted. I think the closest there is to real data on the effect is: http://www.asklater.com/matt/wordpress/2009/09/imports-and-the-community-ii We do also have real data on the effect of not doing imports - the towns which are almost completely unmapped. While importing data from OS may not be ideal, doing nothing and waiting for somebody to go and map it doesn't seem like a successful strategy either, if the past five years are a guide. For prosperous city areas there is no difficulty finding a local mapper who will take on a new hobby to get away from the computer screen for a few hours. But OSM has a real coverage gap in socially disadvantaged areas (Fake SteveC has a pithier name for them). But we want a complete map and not just a map of where the typical OSM contributor lives. If using some of the work already done by the Ordnance Survey helps us get there, that has to be a good thing. -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 11:51, Ed Avis e...@waniasset.com wrote: I think automatically importing the OS data for areas where OSM currently has little to no coverage - or coverage merely traced from Yahoo imagery - is a great idea. I agree. I would be totally opposed to this bot sniffing around Southwark, which we have got very close to 100% through a lot of on-the-ground surveying. I would echo Ed's observation that the OS road names have been much more accurate than the OSM data, mind you. For areas like Southwark with at least a few dedicated mappers willing to alter their commutes and check roads, the manual approach is much, much better. But what about the Lleyn Peninsula in Gwynedd, north west Wales? I've worked on Criccieth and the surrounding area for years, some others have done bits in a few other towns, but most of the county and the peninsula are still very bare after 5-6 years of OSM. I don't see anyone nipping out on their bicycle of a weekend, or altering their commute to work, to finish the basic road network in Gwynedd. I suspect the area has a very low density of IT/geo professionals. Most of the work seems to be done by tourists, like me, who visit specific areas often. If there are tools like the no names map and maybe an un-checked OS Locator copied names map, I don't see the problem with giving those remote rural areas a big boost. If anything, it might make it easier to recruit the sort of local mappers that can happily add a handful of local POIs. Best wishes, Tom -- http://tom.acrewoods.net http://twitter.com/tom_chance ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On Thu, Feb 3, 2011 at 1:17 PM, Ed Avis e...@waniasset.com wrote: Dair Grant dair@... writes: There is suggestion raised by a number of people, but refuted by others that imports reduce the number of contributors. It has been denied, not refuted. I think the closest there is to real data on the effect is: http://www.asklater.com/matt/wordpress/2009/09/imports-and-the-community-ii We do also have real data on the effect of not doing imports - the towns which are almost completely unmapped. While importing data from OS may not be ideal, doing nothing and waiting for somebody to go and map it doesn't seem like a successful strategy either, if the past five years are a guide. For prosperous city areas there is no difficulty finding a local mapper who will take on a new hobby to get away from the computer screen for a few hours. But OSM has a real coverage gap in socially disadvantaged areas (Fake SteveC has a pithier name for them). But we want a complete map and not just a map of where the typical OSM contributor lives. If using some of the work already done by the Ordnance Survey helps us get there, that has to be a good thing. my experience of the OS data traced into my local area is that it's been almost entirely inaccurate. if this is the case where a typical OSM contributor lives then i'd assume that people tracing over OS have introduced several hundred inaccurate features in London alone. perhaps if the people enamoured of tracing OS would organise a mapping party, or reach out to local community groups (people still live in socially disadvantaged areas, right?) then we could create a complete, living map, rather than a road-network-complete, dead one. cheers, matt ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Tom Chance wrote: But what about the Lleyn Peninsula in Gwynedd, north west Wales? I've worked on Criccieth and the surrounding area for years, some others have done bits in a few other towns, but most of the county and the peninsula are still very bare after 5-6 years of OSM. I don't see anyone nipping out on their bicycle of a weekend, or altering their commute to work, to finish the basic road network in Gwynedd. The basic road network _is_ complete on the Llyn Peninsula. A few residential roads are missing, largely in places of the size of Nefyn, Abererch, and Chwilog. I'm very happy to nip out on my bicycle of a weekend next time I'm there, probably this summer (the in-laws have a house in Morfa Nefyn). In doing so I will spot (and survey) footpaths, cut-throughs, pubs, RCN routes, and so on. You don't get that from importing OS data. I'm a bit ambivalent about putting Edern on the map, though. Not sure I want to make it any harder to get a table at the excellent Ship Inn than it already is. ;) cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5988993.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Ed Avis wrote: We do also have real data on the effect of not doing imports - the towns which are almost completely unmapped. So let's go and map them. It's worked very well so far. Am I missing something? cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5989005.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
I'm upset that the blue(95%+) area in the North East turns out to be Darlington, which I believe is tracers that started OS once it was available. I've used the OS Locator tool around Durham, mainly in JOSM, to get towards that 95%, and I turned a bit addicted. It makes me worried that I might go beyond reasonable edits(country roads with no signs, broken signs, my own mistakes/omissions, etc) and just stay at home in my arm chair (it's been cold outside). One village I know isn't done, so it's reminded me to go out and do that. I much rather prefer to create roads than to adjust their locations, so I think it's better I start with my GPS trace rather than OS Locator. Peter, I know the ITO locator layer only shows missing names and that you're not just checking name=* (if the OS has a different name for some reason). What is the list of other name tags that get used? Gregory. On 3 February 2011 14:08, Matt Amos zerebub...@gmail.com wrote: On Thu, Feb 3, 2011 at 1:17 PM, Ed Avis e...@waniasset.com wrote: Dair Grant dair@... writes: There is suggestion raised by a number of people, but refuted by others that imports reduce the number of contributors. It has been denied, not refuted. I think the closest there is to real data on the effect is: http://www.asklater.com/matt/wordpress/2009/09/imports-and-the-community-ii We do also have real data on the effect of not doing imports - the towns which are almost completely unmapped. While importing data from OS may not be ideal, doing nothing and waiting for somebody to go and map it doesn't seem like a successful strategy either, if the past five years are a guide. For prosperous city areas there is no difficulty finding a local mapper who will take on a new hobby to get away from the computer screen for a few hours. But OSM has a real coverage gap in socially disadvantaged areas (Fake SteveC has a pithier name for them). But we want a complete map and not just a map of where the typical OSM contributor lives. If using some of the work already done by the Ordnance Survey helps us get there, that has to be a good thing. my experience of the OS data traced into my local area is that it's been almost entirely inaccurate. if this is the case where a typical OSM contributor lives then i'd assume that people tracing over OS have introduced several hundred inaccurate features in London alone. perhaps if the people enamoured of tracing OS would organise a mapping party, or reach out to local community groups (people still live in socially disadvantaged areas, right?) then we could create a complete, living map, rather than a road-network-complete, dead one. cheers, matt ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb -- Gregory o...@livingwithdragons.com http://www.livingwithdragons.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 14:32, Gregory nomoregra...@googlemail.com wrote:
I'm upset that the blue(95%+) area in the North East turns out to be
Darlington, which I believe is tracers that started OS once it was
available.
I've used the OS Locator tool around Durham, mainly in JOSM, to get towards
that 95%, and I turned a bit addicted. It makes me worried that I might go
beyond reasonable edits(country roads with no signs, broken signs, my own
mistakes/omissions, etc) and just stay at home in my arm chair (it's been
cold outside). One village I know isn't done, so it's reminded me to go out
and do that. I much rather prefer to create roads than to adjust their
locations, so I think it's better I start with my GPS trace rather than OS
Locator.
Peter, I know the ITO locator layer only shows missing names and that
you're not just checking name=* (if the OS has a different name for some
reason). What is the list of other name tags that get used?
name, not:name, alt:name name:en, name{lang} and possiby a few more.
Peter
Gregory.
On 3 February 2011 14:08, Matt Amos zerebub...@gmail.com wrote:
On Thu, Feb 3, 2011 at 1:17 PM, Ed Avis e...@waniasset.com wrote:
Dair Grant dair@... writes:
There is suggestion raised by a number of people, but refuted by
others that
imports reduce the number of contributors.
It has been denied, not refuted. I think the closest there is to real
data
on the effect is:
http://www.asklater.com/matt/wordpress/2009/09/imports-and-the-community-ii
We do also have real data on the effect of not doing imports - the towns
which
are almost completely unmapped. While importing data from OS may not be
ideal,
doing nothing and waiting for somebody to go and map it doesn't seem
like a
successful strategy either, if the past five years are a guide.
For prosperous city areas there is no difficulty finding a local mapper
who will
take on a new hobby to get away from the computer screen for a few
hours. But
OSM has a real coverage gap in socially disadvantaged areas (Fake SteveC
has a
pithier name for them). But we want a complete map and not just a map
of where
the typical OSM contributor lives. If using some of the work already
done by
the Ordnance Survey helps us get there, that has to be a good thing.
my experience of the OS data traced into my local area is that it's
been almost entirely inaccurate. if this is the case where a typical
OSM contributor lives then i'd assume that people tracing over OS have
introduced several hundred inaccurate features in London alone.
perhaps if the people enamoured of tracing OS would organise a mapping
party, or reach out to local community groups (people still live in
socially disadvantaged areas, right?) then we could create a complete,
living map, rather than a road-network-complete, dead one.
cheers,
matt
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb
--
Gregory
o...@livingwithdragons.com
http://www.livingwithdragons.com
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Richard Fairhurst richard@... writes: We do also have real data on the effect of not doing imports - the towns which are almost completely unmapped. So let's go and map them. It's worked very well so far. Am I missing something? 'Worked very well so far' is the part I disagree with. The OSM model has worked well for densely populated, prosperous areas. It's not at all clear that it is working well for remoter ones. If it were, we would not be having this conversation. Also, the progress that has been made in OSM so far has not been purely by people ground surveying with GPS units. In many countries the Yahoo imagery (and now Bing) has turbo-charged the project by allowing large areas to be approximately mapped using your excellent Potlatch tool, and then visited on the ground to add street names and more detail. I suggest that getting a basic map in place using OS data (where nothing exists now) would also help mapping. I don't at all disagree with going out and mapping them! But I don't think that is a reason not to use all the resources that are available to us in addition to our own ground surveys. If someone else has surveyed the area and made their map available to us then why not use it? -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 14:41, Ed Avis e...@waniasset.com wrote: Richard Fairhurst richard@... writes: We do also have real data on the effect of not doing imports - the towns which are almost completely unmapped. So let's go and map them. It's worked very well so far. Am I missing something? 'Worked very well so far' is the part I disagree with. The OSM model has worked well for densely populated, prosperous areas. It's not at all clear that it is working well for remoter ones. If it were, we would not be having this conversation. Also, the progress that has been made in OSM so far has not been purely by people ground surveying with GPS units. In many countries the Yahoo imagery (and now Bing) has turbo-charged the project by allowing large areas to be approximately mapped using your excellent Potlatch tool, and then visited on the ground to add street names and more detail. I suggest that getting a basic map in place using OS data (where nothing exists now) would also help mapping. I don't at all disagree with going out and mapping them! But I don't think that is a reason not to use all the resources that are available to us in addition to our own ground surveys. If someone else has surveyed the area and made their map available to us then why not use it? Statistics says that if there are 1,000 biggish places in a country and a small percentage of slightly obsessive mappers available, a random percentage of whom will meet OSM and get addicted then a percentage of towns around where they live and go will get mapped fast. However... waiting for obsessive mappers to turn up in every unmapped location in the country is an entirely different matter and has a very low probability! Regards, Peter -- Ed Avis e...@waniasset.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Ed Avis wrote: 'Worked very well so far' is the part I disagree with. The OSM model has worked well for densely populated, prosperous areas. It's not at all clear that it is working well for remoter ones. If it were, we would not be having this conversation. I spend half my week in Charlbury (which ain't densely populated) and the other half in Burton-on-Trent (which ain't prosperous). They're both complete, and certainly not entirely through my own efforts. We are doing amazingly well in many rural areas. We're also doing surprisingly well in - I'm trying to avoid Fake SteveC's phrase here - many deprived areas. In other places, sure, we still have a way to go. But we can, and will, get there. Shropshire mapping parties (S Shrops 370th, N Shrops 407th) with a press release in the Shropshire Star? Hell yeah. In the vast rural county of Powys, two people hammering at Newtown for one day will send it 100 places up the stats. It's much, much easier than it looks. Also, the progress that has been made in OSM so far has not been purely by people ground surveying with GPS units. Absolutely. Bing and OS OpenData are great ways of assisting survey, and in many ways remove the need for a GPS unit. But they don't remove the need for a survey. I personally like to print out the OS StreetView maps for an unmapped area before visiting it, and to annotate it - it saves a bit of scribbling and helps me plan my survey. If I'm cycling along a residential distributor road, there might be a cul-de-sac branching off it where I can both see the street sign (to check against the OS) and see all the way to the end (to check there aren't any footpaths I'm missing). If so, that's just saved me a minute cycling up and down there - and, crucially, staved off the boredom. That's great. But I still need to eyeball it to check there's no footpath there. In many countries the Yahoo imagery (and now Bing) has turbo-charged the project by allowing large areas to be approximately mapped using your excellent Potlatch tool :) There is a _lot_ which the tools can do to help and we're still only scratching the surface. Aware that I don't do as much mapping as I like, I try and console myself by thinking that an hour spent on Potlatch will hopefully lead to 100 more hours of mapping productivity for others... well, that's the theory! cheers Richard -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5989147.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 03/02/2011 13:36, Tom Chance wrote: On 3 February 2011 11:51, Ed Avis e...@waniasset.com mailto:e...@waniasset.com wrote: I think automatically importing the OS data for areas where OSM currently has little to no coverage - or coverage merely traced from Yahoo imagery - is a great idea. I agree. I would be totally opposed to this bot sniffing around Southwark, which we have got very close to 100% through a lot of on-the-ground surveying. I would echo Ed's observation that the OS road names have been much more accurate than the OSM data, mind you. For areas like Southwark with at least a few dedicated mappers willing to alter their commutes and check roads, the manual approach is much, much better. But what about the Lleyn Peninsula in Gwynedd, north west Wales? I've worked on Criccieth and the surrounding area for years, some others have done bits in a few other towns, but most of the county and the peninsula are still very bare after 5-6 years of OSM. I don't see anyone nipping out on their bicycle of a weekend, or altering their commute to work, to finish the basic road network in Gwynedd. I suspect the area has a very low density of IT/geo professionals. Most of the work seems to be done by tourists, like me, who visit specific areas often. If there are tools like the no names map and maybe an un-checked OS Locator copied names map, I don't see the problem with giving those remote rural areas a big boost. If anything, it might make it easier to recruit the sort of local mappers that can happily add a handful of local POIs. Best wishes, Tom -- http://tom.acrewoods.net http://twitter.com/tom_chance ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb Well this touches close to home. My mother's family come from North Wales: many of the poorly mapped places in the area either have family associations, relatives still living there or other memories from family holidays or my early childhood. You can look at places I've mapped in the area: some are from short more or less annual visits to the area, others reflect deeper meaning: the cemetery where my uncle and great-grandmother are buried, the village where my cousins' used to run the post office, the house where my mother holidayed before WWII, the open land which caught fire and my father helped beat out the flames. Llyn does have a basic road network (pace Richard). On top of that the OS data is a poor substitute for exploring the rich topography of Llyn, or for engaging people who already know it. So why don't we try and find some local organisations which might have an interest: the local councils may well log GPS traces for their vehicles. There is a centre at Plas Tan-y-Bwlch http://www.eryri-npa.gov.uk/study-centre run by the Snowdonia National Park which runs conferences courses. There is a university at Bangor with a CS department. There is an incipient Welsh Placename Society http://www.bbc.co.uk/news/uk-wales-11800957. There is the Royal Commission http://www.rcahmw.gov.uk/ on Ancient Monuments for Wales which is in beta with the People's Collection Wales http://www.peoplescollectionwales.co.uk/. In most parts of England there are active Welsh Societies (e.g,[1],[2]), perhaps their members would love a talk about OSM and the chance to reminisce about the villages and towns where they grew up. There are local history societies like that for the Nantlle http://www.nantlle.com/home.htmvalley. There is the Urdd http://www.urdd.org/index.php?lng=en. In other words there are lots of people organisations with which we at OSM could engage, or we could just import the OS data. Jerry [1]. http://www.devamedia.co.uk/cymdeithas/nottingham/ [2]. http://www.oxfordshirewelshsociety.com/ ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 15:10, Jerry Clough : SK53 on OSM sk53_...@yahoo.co.ukwrote: On 03/02/2011 13:36, Tom Chance wrote: On 3 February 2011 11:51, Ed Avis e...@waniasset.com wrote: I think automatically importing the OS data for areas where OSM currently has little to no coverage - or coverage merely traced from Yahoo imagery - is a great idea. I agree. I would be totally opposed to this bot sniffing around Southwark, which we have got very close to 100% through a lot of on-the-ground surveying. I would echo Ed's observation that the OS road names have been much more accurate than the OSM data, mind you. For areas like Southwark with at least a few dedicated mappers willing to alter their commutes and check roads, the manual approach is much, much better. But what about the Lleyn Peninsula in Gwynedd, north west Wales? I've worked on Criccieth and the surrounding area for years, some others have done bits in a few other towns, but most of the county and the peninsula are still very bare after 5-6 years of OSM. I don't see anyone nipping out on their bicycle of a weekend, or altering their commute to work, to finish the basic road network in Gwynedd. I suspect the area has a very low density of IT/geo professionals. Most of the work seems to be done by tourists, like me, who visit specific areas often. If there are tools like the no names map and maybe an un-checked OS Locator copied names map, I don't see the problem with giving those remote rural areas a big boost. If anything, it might make it easier to recruit the sort of local mappers that can happily add a handful of local POIs. Best wishes, Tom -- http://tom.acrewoods.net http://twitter.com/tom_chance ___ Talk-GB mailing list Talk-GB@openstreetmap.orghttp://lists.openstreetmap.org/listinfo/talk-gb Well this touches close to home. My mother's family come from North Wales: many of the poorly mapped places in the area either have family associations, relatives still living there or other memories from family holidays or my early childhood. You can look at places I've mapped in the area: some are from short more or less annual visits to the area, others reflect deeper meaning: the cemetery where my uncle and great-grandmother are buried, the village where my cousins' used to run the post office, the house where my mother holidayed before WWII, the open land which caught fire and my father helped beat out the flames. Llyn does have a basic road network (pace Richard). On top of that the OS data is a poor substitute for exploring the rich topography of Llyn, or for engaging people who already know it. So why don't we try and find some local organisations which might have an interest: the local councils may well log GPS traces for their vehicles. There is a centre at Plas Tan-y-Bwlchhttp://www.eryri-npa.gov.uk/study-centrerun by the Snowdonia National Park which runs conferences courses. There is a university at Bangor with a CS department. There is an incipient Welsh Placename Society http://www.bbc.co.uk/news/uk-wales-11800957. There is the Royal Commission http://www.rcahmw.gov.uk/ on Ancient Monuments for Wales which is in beta with the People's Collection Waleshttp://www.peoplescollectionwales.co.uk/. In most parts of England there are active Welsh Societies (e.g,[1],[2]), perhaps their members would love a talk about OSM and the chance to reminisce about the villages and towns where they grew up. There are local history societies like that for the Nantlle http://www.nantlle.com/home.htmvalley. There is the Urddhttp://www.urdd.org/index.php?lng=en. In other words there are lots of people organisations with which we at OSM could engage, or we could just import the OS data. Or do both? Surely the guys at the Royal Commission on Ancient Monuments are not about to head out on bikes to survey the area. What they will do, possibly, is get the boundary correct and put names on the buildings but not until the road network is in place or it will look a bit stupid? Another point... at the 2009 State of the Map there was a big 'ideas' session where I suggested that as one panned the map in OSM it should highlight at the side any relevant OSM wiki pages. So... as one panned over a town or a county it would highlight the page. As an alternative it could make an editor aware of the pages when they press 'edit'. These wiki pages could then be used to guide editors on local issues, challenges and preferences. It could include warning about out-of-date imagery etc. Current I suspect that few edits are even aware that there is a wiki page for Forest Heath and for Suffolk. Or indeed for Wales and Gwynedd. All wiki pages that are associated with an area have a map on the and that map has coordinates in it to allow then to be presented appropriately. None of this is telling me that a bot should not exist. It is saying that it won't be
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 03/02/2011 14:41, Peter Miller wrote:
On 3 February 2011 14:32, Gregory nomoregra...@googlemail.com
mailto:nomoregra...@googlemail.com wrote:
Peter, I know the ITO locator layer only shows missing names and
that you're not just checking name=* (if the OS has a different
name for some reason). What is the list of other name tags that
get used?
name, not:name, alt:name name:en, name{lang} and possiby a few more.
Hopefully also alt_name?
--
Steve
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 03/02/2011 15:24, Peter Miller wrote: Or do both? Surely the guys at the Royal Commission on Ancient Monuments are not about to head out on bikes to survey the area. What they will do, possibly, is get the boundary correct and put names on the buildings but not until the road network is in place or it will look a bit stupid? None of this is telling me that a bot should not exist. It is saying that it won't be welcome everywhere and won't be used by everyone. Regards, Peter Regards, You mis-represent my point. I am not for a moment suggesting that the RCAHMW will do any mapping: they already have a substantial job in stringent financial times. SImilarly, with other organisations. However, we could devote resources to a bot, or we could work at contacting people who may have lots of knowledge and enthusiasm about places which we have, as yet, not mapped in detail. Some of the contacts might yield new mappers, others new resources, others might be a way to get a mapping party off the ground, or just to validate the maps we have. What they have in common is that they involve speaking Welsh or English (or in Brechfa perhaps a bit of Basque) rather than python or perl. Sure as hell, we won't get on the ground engagement if we appear to be OpenSaisMap. Send instant messages to your online friends http://uk.messenger.yahoo.com ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
On 3 February 2011 15:53, Lester Caine les...@lsces.co.uk wrote:
Peter Miller wrote:
name, not:name, alt:name name:en, name{lang} and possiby a few more.
alt:name or alt_name ?
Just to nail things down a little before I clear any differences ...
Where a name is different on a street sign --- add not:name=
( I've a couple of 'road' that have the sign saying 'street' )
Where a name is just laid out differently --- add alt_name=
( Missing/extra dots/apostrophe and the like )
My mistake. It is alt_name not alt:name
Peter
--
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb
___
Talk-GB mailing list
Talk-GB@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Peter, Having read through most of the responses to date it is clear that there are arguments for and against your suggestion. Time to put my view. I'm not in favour, perhaps though not a surprise from the way I have mapped historically. I think imports generally receive mixed interest partly because of the information they hold. For instance I was happy to see bus stops added because they included a lot of additional data which is not always easy to collect on the ground, I was also happy to have the transport data silent in OSM in our area (the Brum import did not include the bus_stop tag) so that the data was there as a reference but not as a definitive attribute. I view data from any other source the same way. OS products are just that, they are products of the OS, there is nothing definitive (in our terms) about any of them. We map what is there in the real world. I'm sure the OS would argue they do exactly the same, but we each produce our own product. The OS might refer to OSM and send out a surveyor to check something if they find a discrepancy, they might do the same with a Local Authority if a query comes in from a member of the public. OSM should work the same way, we use these data sources (and I include imagery and other data sets) as a reference to assist our work, they should not replace our work. Having said that, if a user wishes to add all the data for an area from an OS product (manually or automatically) and then go out and check it on the ground that's fine, that's just a method of mapping, but to blindly import because we can is not in my view helpful at all. At a recent midlands social I mentioned that Penkridge was not mapped, a local mapper took that as a challenge and mapped it. As Andy Allan says, its motivating mappers (and the general public) to go out and collect and update data that will help maintain OSM at the forefront, not automated imports. Cheers Andy From: Peter Miller [mailto:peter.mil...@itoworld.com] Sent: 02 February 2011 21:11 To: Talk-GB@openstreetmap.org Subject: [Talk-GB] Adding a further 250,000 UK roads quickly using a Bot? ITO have been offering a service to compare osm road names with os locator road names for a while now[1] which has encouraged a lot of activity - and has even led to Andy to obsession.[2] I have also suffered from a bout of urgent mapping myself while completing all of Suffolk to 95% in the past few weeks! Can I suggest that for our sanity we should consider developing a bot to do some of this work for us? This would also allow us to get the rest of the 250,000 remaining roads in place in less than the 13 months Andy estimates will be required? This bot would do a number of repetitive tasks for us within the bounding box in which it was authorised to operate by a contributor. It could do the following: 1) Add names to existing roads in osm where there is a single un-named ways in osm with a bounding box which matches that of a single entry in os locator. 2) In addition... it might be able to also add roads to osm from os vector district, snapping them into existing roads as required where the existing roads align neatly with os streetview. It would only do this if there were no ways close by on either side. Complex situations will be left to humans. Humans could also sometimes prepare an area for analysis by the bot, splitting ways as appropriate, adjusting alignment of existing roads and dealing in advance with situations we know the bot will have difficulties with. Edits would be made as individual changesets, referenced to the mapper operating of the bot. Each edit would be 'signed off' by the mapper who would be able to see the proposed changes visual prior to accepting them. Any thoughts? Peter [1] http://www.itoworld.com/product/data/osm_analysis/summary [2] http://www.gravitystorm.co.uk/shine/archives/2011/02/02/the-london-streets-c hallenge/ _ No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1204 / Virus Database: 1435/3420 - Release Date: 02/03/11 ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Richard Fairhurst richard@... writes: The OSM model has worked well for densely populated, prosperous areas. It's not at all clear that it is working well for remoter ones. I spend half my week in Charlbury (which ain't densely populated) and the other half in Burton-on-Trent (which ain't prosperous). They're both complete, and certainly not entirely through my own efforts. Oh, I don't mean to suggest that no well-mapped rural areas exist. Only that coverage is patchy, and that there are many where nothing has been done despite several years of sitting with a blank map waiting for somebody to turn up. How important this is depends on the use you envisage for the map. If OSM is to be used by individuals navigating round their local area, then a map with 50% coverage is half as good as one with 100% coverage. Half as many people can use it, but for those lucky enough to live in a covered area, the map serves their needs. If the example use of OSM is address lookup, for example by businesses, then a map with only 50% coverage scattered across the country is useless. For that particular use it is far better to cover every part of the country, even if at only a basic level of roads and street names, than to have a patchwork of excellent areas and missing areas. Personally, I tend towards the 'world domination' school of thought and would like OSM to replace proprietary addressing products such as the PAF as soon as possible, at least for some applications. So I tend to think that what matters is to get 100% coverage. I enjoy micro-mapping and putting in footpaths as much as anyone else, but I don't want the best to become the enemy of the good. Bing and OS OpenData are great ways of assisting survey, and in many ways remove the need for a GPS unit. But they don't remove the need for a survey. I agree with this, but given the choice between a map that has the basic details complete but needs survey, versus a map that has nothing at all and needs survey, I would take the best we can do right now with what we have. I personally like to print out the OS StreetView maps for an unmapped area before visiting it, and to annotate it - it saves a bit of scribbling and helps me plan my survey. If I'm cycling along a residential distributor road, there might be a cul-de-sac branching off it where I can both see the street sign (to check against the OS) and see all the way to the end (to check there aren't any footpaths I'm missing). If so, that's just saved me a minute cycling up and down there - and, crucially, staved off the boredom. That's great. But I still need to eyeball it to check there's no footpath there. This is the one thing that perturbs me too about using the OS data. Back in the days when we only had Yahoo, I would not tag the name on a way until I had walked all the way down it checking for footpaths. If I'd only explored part of the way, I would tag the name on that part only. Then the noname map was a guide to unsurveyed areas. But now there isn't such a clear indication. So I did hesitate about adding name=Newton Road to a street which I had not visited. But then I considered the folk living on that street typing its name into Nominatim and getting no results. There is no reason to make them wait for a thorough ground survey of their street before it appears in OSM at all. We must make the best map we can now, and improve it later. Also it has to be said that the quality of the OS data, at least for street layout and street names, is pretty good and arguably superior to the typical OSM ground survey (at least as things stood in London a year ago). -- Ed Avis e...@waniasset.com In many countries the Yahoo imagery (and now Bing) has turbo-charged the project by allowing large areas to be approximately mapped using your excellent Potlatch tool :) There is a _lot_ which the tools can do to help and we're still only scratching the surface. Aware that I don't do as much mapping as I like, I try and console myself by thinking that an hour spent on Potlatch will hopefully lead to 100 more hours of mapping productivity for others... well, that's the theory! cheers Richard ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Matt Amos wrote: as andy pointed out, i think we're addressing the wrong problem and trying to fix it technically. maybe the best way forward is to address the social problem: what can we do to grow the community? I wouldn't say it is addressing the wrong problem but an orthogonal problem. Imho we need to actively try to grow the UK community anyway, whether we partially import OS data as a first pass, or not. Data maintanance needs a much bigger community than the one-time initial survey (My guess, looking at the German community would be that the UK community should grow by at least about 3 - 6 times) Also I don't see the decision necessarily hindering or helping an active recruitment drive. It might effect the strategy though. (If we have loads of vast empty spaces, we might for example want to aim to recruit geo-cachers who treat OSM as another GPS game that happens to produce a map. On the other hand if the data has a good base coverage by importing OS data, we might want to aim to recruit for example sat-nav users who are prepared to add the odd missing name or POI in order to get a good, up-to-date cheap sat nav.) Matt Amos wrote: is it just me, or did we used to have more mapping parties? (maybe it's just the winter) do we need to try and reach out to cycling / youth / technology SIGs and hope that some of them find this as addictive as we do? Yes, I think we do need to actively reach out to various groups and magazines to get the word of OSM out onto the streets. One of the biggest jumps in the rate of new mappers ( http://wiki.openstreetmap.org/wiki/File:Osmdbstats4.png ) was in May 2008, which corresponded to whole slew of OSM stories in main stream media in Germany ( http://wiki.openstreetmap.org/wiki/DE:OpenStreetMap_in_the_media#Mai_2008 ) and another jump in March/April 2009 ( http://wiki.openstreetmap.org/wiki/DE:OpenStreetMap_in_the_media#M.C3.A4rz_2009 ), when both techpress, national daily press and TV stations picked up reporting on OSM, which seems to have been followed by a number of local and regional media picking it up too. I have no idea how that was achieved or what triggered it, but trying to repeat something like that in the UK by reaching out to the press might help grow the community quite a bit. Perhaps sending out press releases or contacting editors if they are interested in covering OSM might help? Furthermore, in the German community people seem to be organising OSM booths with large banners on all of the relevant (international) Tradefairs such as for example CeBIT or Linux Tag or the boating and cycling traidfairs, to actively try and spread the word and engage with various special interest groups who might benefit from or have interest in it. Again this has the potential to reach quite a lot of people with comparatively little effort. It might not need to be quite as grand at first, but perhaps offering to write an article in the local cycling magazine, or your university newspaper or your local newspaper, who often are looking for good articles to fill their magazines, might be a great start. Some of this might need some support from a OSMF-gb chapter, in order to organise and pay for banners and promotional material e.g. to use at tradefairs and other relevant events, but it should be possible to do these kind of events in the UK too. At the same time, an active recruitment drive might also allow us to get more in touch with newbies again. Ask them what are the things that would stop them from contributing. What would they expect, what would they need? With live demos at booths, you could also immediately do some usability studies at the same time, to see where the problems really are. For example the events at UCL ( http://www.gravitystorm.co.uk/shine/archives/2010/10/04/quick-and-dirty-usability-testing-of-osm/ ) seems to have straight off triggered a bunch of technical improvements to make it easier for newbies, which was great to see. We just need more of those. In the end, it doesn't matter if we have roads imported or a blank canvas if people simply have never heard of OSM. Kai -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5991118.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
Re: [Talk-GB] Adding a further 250, 000 UK roads quickly using a Bot?
Ed Avis wrote: I think automatically importing the OS data for areas where OSM currently has little to no coverage - or coverage merely traced from Yahoo imagery - is a great idea. Many people don't want bots to trample on their 'patch' which they have carefully surveyed, which is fine. But there is still a huge part of the country which is very under-mapped. Let's get it complete as soon as possible using the best data we have. (It is my opinion that the OS data is often of higher quality than a typical OSM ground survey, although of course multiple surveys by different people combined with the use of Bing imagery will give the best possible map.) +1 I fully agree on this. Many of the problems people complain about with respect to bots is that they degrade very well mapped areas, as e.g. they added an old road back in that was deliberately left out in the ground survey as it no longer exists. Adding new data in areas that are not mapped at all is much less problematic, as we can be sure it hasn't made the situation worse, just not made it immediately as good as it can be after many iterations of improvements. OS Streetview data is very high quality data with good positional and labeling accuracy. Sure it has errors, is out of data and doesn't have all POIs and other metadata. But it is better than most first pass ground surveys done by OSM contributors. According to Muki Haklay's Paper How Many Volunteers Does it Take to Map an Area Well? The Validity of Linus’ Law to Volunteered Geographic Information, on average OSM only reaches its quality only once it reaches 5 - 15 contributors per square kilometer. Does that mean one shouldn't go to mapping parties in different cities or map during one off visits or holidays? After all, one might leave dead data behind without building a community? No, OSM has always been do as good a job as you can now and some one will improve things later. If we can't build the right tools and services to attract people into areas that already have a base coverage, to update and improve things, OSM will be dead data anyway eventually. We need to face these issues sooner rather than later and having an (imported) base map to build upon will allow us to get to this stage quicker. It allows us to focus on these issues of how to attract people into maintanance and gardening data that others have entered before they joined. Also looking at Germany, they don't actually seem to have too many problems attracting new people into data dense areas. Ed Avis wrote: I also think that what is good for ITO is good for the map in general - if there is a need to have a complete map for some government contract or whatever, then that is a need that exists in the real world. Also, we are getting some of the most comprehensive on the ground verification and improvement reports from applications like the Sat-Nav Skobbler bugs with MapDust. The more complete the map is, the more people will use things like that and give feedback on errors. Currently you still get a lot of errors like Leeds is missing 4000 roads or No roads in the next 10 km which are useless, but in areas that were traced from OS data, one can get some good on the ground feedback about missspellings, wrong connectivity, missing POIs and other valuable data. Way more than we have gotten in those areas in the last 6 years. (Yes, MapDust still has a big problem with Signal-to-Noise ratio, but thanks to the high volume there is actually quite a lot of good and helpful data, unless the town just simply doesn't exist.) Kai -- View this message in context: http://gis.638310.n2.nabble.com/Adding-a-further-250-000-UK-roads-quickly-using-a-Bot-tp5986539p5991203.html Sent from the Great Britain mailing list archive at Nabble.com. ___ Talk-GB mailing list Talk-GB@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-gb
