Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
Paweł Marynowski writes: When you make import, people are starting make notes about imprecise data. I would note that, for decades, the best way to get correct information on the Internet has been to post incorrect information. I don't think that has changed, so yeah, let's import data even if its quality is not 100%. -- --my blog is athttp://blog.russnelson.com Crynwr supports open source software 521 Pleasant Valley Rd. | +1 315-600-8815 Potsdam, NY 13676-3213 | Sheepdog ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
No one does imports without audit of random samples. We do it as well. The percentage of errors checked in tested area corresponds to the average content errors created by OpenStreetMap mappers. Of course, this must be checked each time, if we import data. We should never believe in the quality of whole geodatabase, if we check only the part. Best regards, Marek Dnia 3 listopada 2014 4:38 Russ Nelson nel...@crynwr.com napisał(a): Paweł Marynowski writes: When you make import, people are starting make notes about imprecise data. I would note that, for decades, the best way to get correct information on the Internet has been to post incorrect information. I don't think that has changed, so yeah, let's import data even if its quality is not 100%. -- --my blog is at http://blog.russnelson.com Crynwr supports open source software 521 Pleasant Valley Rd. | +1 315-600-8815 Potsdam, NY 13676-3213 | Sheepdog ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
Hi, On Thu, Oct 23, 2014 at 01:32:19PM +0200, Marc Gemis wrote: So don't you expect pressure from companies using OSM for navigation and geocoding to add more addresses ? Or do you expect them to use many different datasources ? I dont get this? pressure? On what basis? You get the OSM data on a take it or leave it basis. If it doesnt suit your need you are free to spend an arbitrary amount of money to buy datasets which better suit your needs. OSM comes without any guarantees. Me personally i love addresses which might be the cause for 40K addresses i have added to OSM in the past years. I am absolutely optimistic that addresscompletion for example for Germany wont take more than 3 years. So by 2017 we'll have all Adresses. Its just a matter of time and i am very relaxed about this. When i started with OSM in 2007 i wouldnt have thought that we'll have that amount of Data in 2014 just 7 years later. Flo -- Florian Lohoff f...@zz.de signature.asc Description: Digital signature ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
OSM is not a melting pot for the world's open data and while there is a place for imports, every import will have to be carefully thought about and evaluated. Streamlining that process is not necessarily in the best interest of OpenStreetMap. What do you mean? Clearly it *is* in the interest of OSM unless you have a very different definition of OSM. In some areas of the map you would not see most of the data (or any data, really) were it not for imports. Nearby my home area Czech community did an awesome import of addresses and buildings. Now compare it with the Polish side: http://www.openstreetmap.org/#map=16/49.8719/18.5786 What is your vision for this specific area if there was no building/address import? Do you expect that in 1 or 5 or 10 years you would have the same level of coverage by local mappers? Regarding streamlining and imports in general - to be honest I don't see much point in the current (quite complex and lenghty) process for analyzing and blessing the good imports. Instead of putting up more walls between data and OSM the project should heavily invest in official backend (as in - *built-in into osm.org*) tools for detecting and manging (e.g. reverting) changes to the database. At this point there are tons of 3rd-party scripts but that will not scale and will lead to more calls for careful imports and that will of course lead to more barriers for people interested in importing data. Paweł ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
For me it is not unthinkable that once a company starts using OSM data for a commercial offering, they might start demanding to add that data. Of course they cannot demand anything from the community, but they can import the data, or some people might feel a need to import data, or... It's hard to explain in a foreign language and via mail. But for me, allowing (address) imports is already a form of obeying that marketing demand. Then OSM is no longer about people surveying and drawing from aerial images. regards m On Fri, Oct 24, 2014 at 12:31 PM, Florian Lohoff f...@zz.de wrote: Hi, On Thu, Oct 23, 2014 at 01:32:19PM +0200, Marc Gemis wrote: So don't you expect pressure from companies using OSM for navigation and geocoding to add more addresses ? Or do you expect them to use many different datasources ? I dont get this? pressure? On what basis? You get the OSM data on a take it or leave it basis. If it doesnt suit your need you are free to spend an arbitrary amount of money to buy datasets which better suit your needs. OSM comes without any guarantees. Me personally i love addresses which might be the cause for 40K addresses i have added to OSM in the past years. I am absolutely optimistic that addresscompletion for example for Germany wont take more than 3 years. So by 2017 we'll have all Adresses. Its just a matter of time and i am very relaxed about this. When i started with OSM in 2007 i wouldnt have thought that we'll have that amount of Data in 2014 just 7 years later. Flo -- Florian Lohoff f...@zz.de -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) iQIVAwUBVEoqZZDdQSDLCfIvAQoqfw//aT5y1P1GRJ69bJgLE1TnzDPKT3p9hJhU xr6lUPh/ZP5KKPdLqwpjyTRvLaVTf1+YfHgUkg032NF7gj17ygH8wXRODTIU5JVa 6zlkCMQC5XlQALTJtJ6vqjxfNzc/mxfJui7xfsoe6RP9OuRzHN2tZTJBx5nMBf4A +NUVNpw2rq4J97oZKeBnrPDZ24gzOEZ5zUSlvSs5oQlA4lc5XHlqF2IZO0OyPsnt rqMexE9t0wwsFFlZapw270EH3hIebohOHuHHfmdKA9EikjmHRHU9vrtxwPMGpxDV L7fWfrupuFjkL6M4ZjIuIFCHXCWLvNSFalN5grf6ospD0JJY+tdSoo26fiuHgX88 CQmxijgbFNOjrJjPmkMJM0GGJRE9LuA9qyQmA6j9laTXwV6Cx6e6HHxcPOsv/Tzz sKC7nlez9e/RqtQjB7B8AG1Dj7Hk/XlYDZa/T9+oIxj8JaxehuvV9gUi9ETNLm9l HLvVeJ5Aujj42Hu42S/6p6T25h/nBS4Fg1VdqfKgudfiW8eoBGy4iUVqfRsd6CE8 xf8wQlTGlYg0FuwpvldgIRUFhc72c4S8XYehGGI6YS9XV09NqTrcVqQBnqm4d9RU f8CMRkG4H8f7YRB7PUqTUAacvE/R7K8gFf4mnATC82QKcKZc4JT7usvhwRPiEdpm 7P1eB1jdTZQ= =cnQK -END PGP SIGNATURE- ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
2014-10-24 13:17 GMT+02:00 Marc Gemis marc.ge...@gmail.com: For me it is not unthinkable that once a company starts using OSM data for a commercial offering, they might start demanding to add that data. indeed, could happen from time to time ;-) Of course they cannot demand anything from the community, but they can import the data, or some people might feel a need to import data, or... But for me, allowing (address) imports is already a form of obeying that marketing demand. if you have address information in a suitable license to import it into OSM, you can do this. All that is required is that you check the data regarding its quality (accuracy, currentness etc.), document your plan, explain how you will merge existing data and discuss this with the local comunity and the experts on the import list. Then OSM is no longer about people surveying and drawing from aerial images. ? We'd still need to survey in order to do the maintenance work and correct errors. That's the main reason why imports are often seen as problematic: they will not create a comunity, rather there are indications that the growth of the comunity will slow down when everything looks completed, and then you'll have the same errors in our db than all the other people that have imported from the official source, and no-one near to fix them or feeling responsible for it. If the amount of errors is significant, more data will actively draw people away (fixing bugs is less appealing to many mappers than setting the first foot steps on white map canvas). cheers, Martin ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
? We'd still need to survey in order to do the maintenance work and correct errors. That's the main reason why imports are often seen as problematic: they will not create a comunity, rather there are indications that the growth of the comunity will slow down when everything looks completed, and then you'll have the same errors in our db than all the other people that have imported from the official source, and no-one near to fix them or feeling responsible for it. If the amount of errors is significant, more data will actively draw people away (fixing bugs is less appealing to many mappers than setting the first foot steps on white map canvas). I totally agree with you, But, I have the impression that people that do the imports don't do them to create a community. they do it to have the data in OSM. I've read things like we don't need to do updates, we can wait until the imported source is updated, they have thousands of paid people to update the data. Pawel Paprota's reaction is along the same lines for me: we could not have accomplished this amount of data without an import. Again no focus on the community or how the extra data is going to attract new mappers. This is a different OSM than I have in mind. regards m ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
On Fri, Oct 24, 2014 at 5:05 AM, Martin Koppenhoefer dieterdre...@gmail.com wrote: We'd still need to survey in order to do the maintenance work and correct errors. That's the main reason why imports are often seen as problematic: they will not create a comunity, rather there are indications that the growth of the comunity will slow down when everything looks completed, and then you'll have the same errors in our db than all the other people that have imported from the official source, and no-one near to fix them or feeling responsible for it. If the amount of errors is significant, more data will actively draw people away (fixing bugs is less appealing to many mappers than setting the first foot steps on white map canvas). I disagree that import do not create community. We build a strong community around an import, and address import. That community continues to grow. We even discovered that having an address node makes it easier to new mappers to add POI information to the node and to nudge it to the entrance. It is correct that imports can create more maintenance work. But then it gives us motivation to add encourage more mappers. If users of our data want addresses, demanding it isn't going to make it happen. If they want specific features, we give them ways to help us achieve those goal. Such as help fund OSM, help us grow the community, fund the development of betters tools, help us get our story out. etc. We shouldn't be afraid, we should embrace these requests. Clifford -- @osm_seattle osm_seattle.snowandsnow.us OpenStreetMap: Maps with a human touch ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
2014-10-24 14:14 GMT+02:00 Marc Gemis marc.ge...@gmail.com: I totally agree with you, But, I have the impression that people that do the imports don't do them to create a community. they do it to have the data in OSM. Don't you think that more data = more people eager to add notes about what's missing? Eg. nearly noone wants to add bunch of notes about missing addresses. When you make import, people are starting make notes about imprecise data. At least that's my observation regarding data in Poland. -- *Paweł Marynowski* user:Yarl Stowarzyszenie OpenStreetMap Polska http://osm.org.pl/ http://fb.com/osmpolska/ ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
Hi, On 10/23/2014 08:57 AM, Marc Gemis wrote: Wouldn't it be great that we would have import tools that can be used by many groups ? As far as addresses are concerned, there's OpenAddresses (the US version, not the older Swiss project) that collects openly licensed address data and to my knowledge they're also looking into how that data can then be combined with OSM for geocoding, without actually importing it in OSM. To my mind that's an excellent solution; you can dump address data into the pool and have it processed without interfering with the manually surveyed data that is in OSM. And users can be given the choice of using only the manually surveyed data, or only the government data, or a mixture of both. OSM is not a melting pot for the world's open data and while there is a place for imports, every import will have to be carefully thought about and evaluated. Streamlining that process is not necessarily in the best interest of OpenStreetMap. Frankly, I don't get all the brouhaha about addresses. Yes, geocoding is commercially interesting, but is it interesting for us as a project? Are the mappers in OSM doing what they do because they always wanted to have a house-level free geocoder? I very much doubt that. Technology wise, I find it almost insulting to reduce OSM to a geocoding database and measure OSM in how many addresses it has. There's so much more to the data we collect than merely placing a latitude and longitude against an address. Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09 E008°23'33 ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
On Thu, Oct 23, 2014 at 9:39 AM, Frederik Ramm frede...@remote.org wrote: Frankly, I don't get all the brouhaha about addresses. Yes, geocoding is commercially interesting, but is it interesting for us as a project? Are So don't you expect pressure from companies using OSM for navigation and geocoding to add more addresses ? Or do you expect them to use many different datasources ? And when the mission statement is The world's best addressable map, don't you expect that people will feel the need to add more addresses, faster, in a shorter period of time ? BTW, I'm just an ordinary mapper without real vision of what OSM has to be. I'm happy to continue mapping the way I do now: surveys, because this means exploring the world around me. I don't need imports. It's just that when OSMF wants more imports (do they ?) they should support that process with the necessary tools, doing so would to improve the quality of the imports IMHO. regards m ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map)
I'd go along with this to some extent; I certainly don't think OSM should be primarily about addressing as Steve's email appears to indicate. It should be about whatever we as mappers want it to be. If addressing's your thing, then fine, do it, but if it isn't, that's fine also. I don't think we should be singling out one mission at all - other than to gather free geodata. Instead, we as mappers should be contributing whatever personally turns us on, so to speak. There are enough people with enough diverse interests that we end up with a map showing a wide range of things. There are still a good number of things besides addressing that are missing. Plenty of footpaths even here in the UK are still missing, for instance - to mention my own personal area of interest. By focusing on one aim you're going to potentially turn people off whose mapping interests lie elsewhere. Nick -Frederik Ramm frede...@remote.org wrote: - To: talk@openstreetmap.org From: Frederik Ramm frede...@remote.org Date: 23/10/2014 08:44AM Subject: Re: [OSM-talk] Addresses are a tiny fraction of what we do (was: The world’s best addressable map) Hi, On 10/23/2014 08:57 AM, Marc Gemis wrote: Wouldn't it be great that we would have import tools that can be used by many groups ? As far as addresses are concerned, there's OpenAddresses (the US version, not the older Swiss project) that collects openly licensed address data and to my knowledge they're also looking into how that data can then be combined with OSM for geocoding, without actually importing it in OSM. To my mind that's an excellent solution; you can dump address data into the pool and have it processed without interfering with the manually surveyed data that is in OSM. And users can be given the choice of using only the manually surveyed data, or only the government data, or a mixture of both. OSM is not a melting pot for the world's open data and while there is a place for imports, every import will have to be carefully thought about and evaluated. Streamlining that process is not necessarily in the best interest of OpenStreetMap. Frankly, I don't get all the brouhaha about addresses. Yes, geocoding is commercially interesting, but is it interesting for us as a project? Are the mappers in OSM doing what they do because they always wanted to have a house-level free geocoder? I very much doubt that. Technology wise, I find it almost insulting to reduce OSM to a geocoding database and measure OSM in how many addresses it has. There's so much more to the data we collect than merely placing a latitude and longitude against an address. Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09 E008°23'33 ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk ___ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
