Re: [OSM-legal-talk] FW: OSM place name data from Turkey
" "Version 1" would seem to imply not moved, or am I mixing something up?" I understood it so that the version 1s might well be already moved as part of the import process. This is at least what has been done in Haiti when importing place names and peaks and other objects from the US military topo maps. If it would turn out that the names of them wouldn't be public domain (which they are) I would certainly not want to loose the place= and natural=peak tags as it's taken a good amount of work to figure out their locations. Cheers, -Jaakko Sent from my BlackBerry® device from Digicel -- Mobile: +509-37-26 91 54, Skype/GoogleTalk: jhelleranta -Original Message- From: Simon Poole Date: Mon, 29 Apr 2013 12:51:37 To: Reply-To: "Licensing and other legal discussions." Subject: Re: [OSM-legal-talk] FW: OSM place name data from Turkey ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Paul I was basing my thoughts on this statement > There a total of 31k objects, of which 28k are version 1. "Version 1" would seem to imply not moved, or am I mixing something up? Simon / / Am 29.04.2013 11:40, schrieb Paul Norman: > > Clarification on numbers: > > > > Assuming every node has been moved, we'd be talking about 26k place or > mountain peak nodes I can definitely keep, about 3k I can restore from > the existing redactions, and about 3k that I'm not sure about. > > > > Now, it's entirely possible a bunch of nodes haven't been moved from > the HGK positions, in which case they'd be completely removed. I won't > know that until I go ahead with the SQL to identify the nodes. > > > > *From:*Henning Scholland [mailto:o...@aighes.de] > *Sent:* Monday, April 29, 2013 2:31 AM > *To:* Licensing and other legal discussions. > *Subject:* Re: [OSM-legal-talk] FW: OSM place name data from Turkey > > > > Am 29.04.2013 11:27, schrieb Simon Poole: > > > > I would agree that there is some value in having "naked" place > nodes. However considering that at best we are talking about 2-3k > such nodes surviving it is a question if doing an imagery based > "add a place" drive or similar for Turkey after the redaction > wouldn't be more efficient. > > Simon > > Yes, maybe this would be a better solution. > > Henning > > > > ___ > legal-talk mailing list > legal-talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/legal-talk ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Clarification on numbers: Assuming every node has been moved, we'd be talking about 26k place or mountain peak nodes I can definitely keep, about 3k I can restore from the existing redactions, and about 3k that I'm not sure about. Now, it's entirely possible a bunch of nodes haven't been moved from the HGK positions, in which case they'd be completely removed. I won't know that until I go ahead with the SQL to identify the nodes. From: Henning Scholland [mailto:o...@aighes.de] Sent: Monday, April 29, 2013 2:31 AM To: Licensing and other legal discussions. Subject: Re: [OSM-legal-talk] FW: OSM place name data from Turkey Am 29.04.2013 11:27, schrieb Simon Poole: I would agree that there is some value in having "naked" place nodes. However considering that at best we are talking about 2-3k such nodes surviving it is a question if doing an imagery based "add a place" drive or similar for Turkey after the redaction wouldn't be more efficient. Simon Yes, maybe this would be a better solution. Henning ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Am 29.04.2013 11:27, schrieb Simon Poole: Am 29.04.2013 11:14, schrieb Henning Scholland: Am 29.04.2013 10:42, schrieb Simon Poole: However in the current case I doubt that there is actually something useful for OSM left once the names are gone. If the information "There is a village" stays in OSM, it would be useful at all. If you are travelling through Turkey, it's good to know, where the next village is. In most cases you find there a little store or at least you find there people, that can help you. I think if a place-node was moved, then at least the place=*is verified by another source and could stay in OSM with the new coordinates. I would agree that there is some value in having "naked" place nodes. However considering that at best we are talking about 2-3k such nodes surviving it is a question if doing an imagery based "add a place" drive or similar for Turkey after the redaction wouldn't be more efficient. Simon Yes, maybe this would be a better solution. Henning ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Am 29.04.2013 11:14, schrieb Henning Scholland: > Am 29.04.2013 10:42, schrieb Simon Poole: >> However in the current >> case I doubt that there is actually something useful for OSM left once >> the names are gone. > If the information "There is a village" stays in OSM, it would be > useful at all. If you are travelling through Turkey, it's good to > know, where the next village is. In most cases you find there a little > store or at least you find there people, that can help you. > > I think if a place-node was moved, then at least the place=*is > verified by another source and could stay in OSM with the new > coordinates. > > I would agree that there is some value in having "naked" place nodes. However considering that at best we are talking about 2-3k such nodes surviving it is a question if doing an imagery based "add a place" drive or similar for Turkey after the redaction wouldn't be more efficient. Simon ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Am 29.04.2013 10:42, schrieb Simon Poole: However in the current case I doubt that there is actually something useful for OSM left once the names are gone. If the information "There is a village" stays in OSM, it would be useful at all. If you are travelling through Turkey, it's good to know, where the next village is. In most cases you find there a little store or at least you find there people, that can help you. I think if a place-node was moved, then at least the place=*is verified by another source and could stay in OSM with the new coordinates. Henning ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Am 29.04.2013 10:18, schrieb Paul Norman: >> From: Simon Poole [mailto:si...@poole.ch] >> Sent: Sunday, April 28, 2013 11:58 PM >> To: legal-talk@openstreetmap.org >> Subject: Re: [OSM-legal-talk] FW: OSM place name data from Turkey >> >> Hi Paul >> >> Has anybody from the TR community tried to get permission from HGK (with >> a pointer that the data is freely available elsewhere and that removing >> it would add up to deleting and re-adding exactly the same data)? Having >> such permission would seem to be the best solution right now. > People have tried contacting other agencies, but to the best of my > knowledge, no one has had any success with HGK. > > To be clear, it's not the UN who needs to be contacted to get permission, > it's HGK. Yes, I read the thread on the mailing list and it doesn't seem as if there is actually a 1-to-1 replacement for the data from any other place, so that part of the discussion seems to be moot. > >> 2nd question why would somebody re add the HGK data if the same data is >> available from a different agency? Potentially the solution would be to >> redact and add the OK data at the same time. > We don't have the technical means to do anything but a redaction through the > bot, and I don't see us developing it. > > How about this. My understanding of the workflow of the user is that they > took the HGK data (names, object type and location) and then moved it to > agree with imagery, then uploaded, creating v1 of the nodes. > > The names obviously have to go, but if they've verified the object type and > location against imagery, could we keep that? I don't see a reason why not, the LWG has taken the position in the past that the IP is not tied to the OSM object in question, so for example moving an object substantially would clearly not just technically overwrite any IP in the original coordinates. However in the current case I doubt that there is actually something useful for OSM left once the names are gone. Simon ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
> From: Simon Poole [mailto:si...@poole.ch] > Sent: Sunday, April 28, 2013 11:58 PM > To: legal-talk@openstreetmap.org > Subject: Re: [OSM-legal-talk] FW: OSM place name data from Turkey > > Hi Paul > > Has anybody from the TR community tried to get permission from HGK (with > a pointer that the data is freely available elsewhere and that removing > it would add up to deleting and re-adding exactly the same data)? Having > such permission would seem to be the best solution right now. People have tried contacting other agencies, but to the best of my knowledge, no one has had any success with HGK. To be clear, it's not the UN who needs to be contacted to get permission, it's HGK. > 2nd question why would somebody re add the HGK data if the same data is > available from a different agency? Potentially the solution would be to > redact and add the OK data at the same time. We don't have the technical means to do anything but a redaction through the bot, and I don't see us developing it. How about this. My understanding of the workflow of the user is that they took the HGK data (names, object type and location) and then moved it to agree with imagery, then uploaded, creating v1 of the nodes. The names obviously have to go, but if they've verified the object type and location against imagery, could we keep that? If so, could we then delete the village names but leave the other tags, redacting the versions of the node with the names? I'd also want to verify that the nodes were moved by matching against the original data. Of course, v2 and later nodes might be too complicated to sort out. About 90% of the nodes are v1, and all the nodes redacted so far have been v1. It'd likely be possible to apply the same logic to the already-redacted nodes. Does this sound legally sound? ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Hi Simon, I license of the same data would be from the UN. I think some clarification needs to be made, but Suha Ulgen is the right person to track that down (member of the OSM-tr community and in the appropriate UN working group). One thing I was not clear on was that there was a legal license from the UN source, at least other UN agencies tend to accept the license from the data provider meaning their clearing houses often can be an assortment of licenses and each would have to be looked at individually. -Kate On Mon, Apr 29, 2013 at 1:57 PM, Simon Poole wrote: > Hi Paul > > Has anybody from the TR community tried to get permission from HGK (with > a pointer that the data is freely available elsewhere and that removing > it would add up to deleting and re-adding exactly the same data)? Having > such permission would seem to be the best solution right now. > > 2nd question why would somebody re add the HGK data if the same data is > available from a different agency? Potentially the solution would be to > redact and add the OK data at the same time. > > Simon > > Am 29.04.2013 05:31, schrieb Paul Norman: > > A user in Turkey brought in about 30k place nodes (or mountain peak > nodes) > > from HGK, a Turkish government agency. > > > > HGK prohibits actions besides personal use[1] and this is clearly > > incompatible with the ODbL. As I indicated, this means they need to be > > removed,[2] which in technical terms means redacting them. Identifying > the > > material to be redacted will not be easy. > > > > A couple of people have indicated that there may be other sources > available > > which have village locations. Unfortunately, this does not change the > status > > of the HGK data. It is an odd situation where the data is not available > > under an open license from one agency, but the exact same data is openly > > licensed from another agency. > > > > == Questions == > > > > LWG: My understanding is we need permission or a suitable license from > the > > data source and that finding an alternate source would only allow us to > > bring in the new data, not keep the old data. Is this correct? > > > > LWG+DWG: If someone re-uploads the HGK data because they disagree with > the > > redaction, they will of intentionally and knowingly uploaded copyrighted > > material without permission or a license. Does the DMCA oblige us to do > > something about this? If they do this, should we do something even if we > > aren't required to under the DMCA? > > > > [1]: > http://lists.openstreetmap.org/pipermail/talk-tr/2013-April/000291.html > > as well as a Bing translation of their FAQ > > > > [2]: > http://lists.openstreetmap.org/pipermail/talk-tr/2013-April/000293.html > > > > -Original Message- > > From: Paul Norman [mailto:penor...@mac.com] > > Sent: Sunday, April 28, 2013 8:15 PM > > To: 'Suha Ulgen' > > Cc: 'OSM Mikel Maron'; 'Schuyler Erle'; 'Mikel Maron'; 'Kate Chapman' > > Subject: RE: OSM place name data from Turkey > > > >> From: Suha Ulgen [mailto:m...@suhaulgen.com] > >> Sent: Sunday, April 28, 2013 7:56 PM > >> Subject: OSM place name data from Turkey > >> > >> Paul, > >> > >> The Turkish OSM community is very distressed about your recent > >> unilateral action. Apparently you are erasing the place names in the > >> Turkish dataset stating that the source data which you identify as > >> belonging to the Turkish General Command for Mapping (Harita Genel > >> Komutanligi - HGK) is not ODbL-compatible. > > This issue was raised on the talk-tr@ mailing list. The conclusion, as > you > > mentioned, was that the HGK data was not ODbL compatible. Not being ODbL > > compatible, we cannot distribute it. > > > >> The counter argument is that Turkish gazetteer data has been submitted > >> by HGK to the UN Group of Experts on Geographic Names (UNGEGN) and is > >> therefore in the "public domain". > > The list raised the possibility that there might be an alternate source > of > > Turkish place names. Unfortunately, this doesn't change the status of the > > HGK sourced material, which has explicit restrictions against > distribution. > > > > Do you anticipate getting permission from *HGK* to distribute their data? > > > >> Please STOP "redacting" the Turkish place names. I'll talk to the > >> UNGEGN Secretariat tomorrow and revert ASAP. > > It should not be technically possible to use the normal revert tools on a > > redaction. In any case, do not reintroduce the data downloaded from HGK > that > > we do not have permission to redistribute. > > > > > > ___ > > legal-talk mailing list > > legal-talk@openstreetmap.org > > http://lists.openstreetmap.org/listinfo/legal-talk > > > > ___ > legal-talk mailing list > legal-talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/legal-talk > ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Hi Paul Has anybody from the TR community tried to get permission from HGK (with a pointer that the data is freely available elsewhere and that removing it would add up to deleting and re-adding exactly the same data)? Having such permission would seem to be the best solution right now. 2nd question why would somebody re add the HGK data if the same data is available from a different agency? Potentially the solution would be to redact and add the OK data at the same time. Simon Am 29.04.2013 05:31, schrieb Paul Norman: > A user in Turkey brought in about 30k place nodes (or mountain peak nodes) > from HGK, a Turkish government agency. > > HGK prohibits actions besides personal use[1] and this is clearly > incompatible with the ODbL. As I indicated, this means they need to be > removed,[2] which in technical terms means redacting them. Identifying the > material to be redacted will not be easy. > > A couple of people have indicated that there may be other sources available > which have village locations. Unfortunately, this does not change the status > of the HGK data. It is an odd situation where the data is not available > under an open license from one agency, but the exact same data is openly > licensed from another agency. > > == Questions == > > LWG: My understanding is we need permission or a suitable license from the > data source and that finding an alternate source would only allow us to > bring in the new data, not keep the old data. Is this correct? > > LWG+DWG: If someone re-uploads the HGK data because they disagree with the > redaction, they will of intentionally and knowingly uploaded copyrighted > material without permission or a license. Does the DMCA oblige us to do > something about this? If they do this, should we do something even if we > aren't required to under the DMCA? > > [1]: http://lists.openstreetmap.org/pipermail/talk-tr/2013-April/000291.html > as well as a Bing translation of their FAQ > > [2]: http://lists.openstreetmap.org/pipermail/talk-tr/2013-April/000293.html > > -Original Message- > From: Paul Norman [mailto:penor...@mac.com] > Sent: Sunday, April 28, 2013 8:15 PM > To: 'Suha Ulgen' > Cc: 'OSM Mikel Maron'; 'Schuyler Erle'; 'Mikel Maron'; 'Kate Chapman' > Subject: RE: OSM place name data from Turkey > >> From: Suha Ulgen [mailto:m...@suhaulgen.com] >> Sent: Sunday, April 28, 2013 7:56 PM >> Subject: OSM place name data from Turkey >> >> Paul, >> >> The Turkish OSM community is very distressed about your recent >> unilateral action. Apparently you are erasing the place names in the >> Turkish dataset stating that the source data which you identify as >> belonging to the Turkish General Command for Mapping (Harita Genel >> Komutanligi - HGK) is not ODbL-compatible. > This issue was raised on the talk-tr@ mailing list. The conclusion, as you > mentioned, was that the HGK data was not ODbL compatible. Not being ODbL > compatible, we cannot distribute it. > >> The counter argument is that Turkish gazetteer data has been submitted >> by HGK to the UN Group of Experts on Geographic Names (UNGEGN) and is >> therefore in the "public domain". > The list raised the possibility that there might be an alternate source of > Turkish place names. Unfortunately, this doesn't change the status of the > HGK sourced material, which has explicit restrictions against distribution. > > Do you anticipate getting permission from *HGK* to distribute their data? > >> Please STOP "redacting" the Turkish place names. I'll talk to the >> UNGEGN Secretariat tomorrow and revert ASAP. > It should not be technically possible to use the normal revert tools on a > redaction. In any case, do not reintroduce the data downloaded from HGK that > we do not have permission to redistribute. > > > ___ > legal-talk mailing list > legal-talk@openstreetmap.org > http://lists.openstreetmap.org/listinfo/legal-talk ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
Re: [OSM-legal-talk] FW: OSM place name data from Turkey
Whoops. Another auto-complete to: mistake. Not that there's anything secret if you've been following the talk-tr@ list, > -Original Message- > From: Paul Norman [mailto:penor...@mac.com] > Sent: Sunday, April 28, 2013 8:32 PM > To: d...@osmfoundation.org; legal-talk@openstreetmap.org > Subject: [OSM-legal-talk] FW: OSM place name data from Turkey ... ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
[OSM-legal-talk] FW: OSM place name data from Turkey
A user in Turkey brought in about 30k place nodes (or mountain peak nodes) from HGK, a Turkish government agency. HGK prohibits actions besides personal use[1] and this is clearly incompatible with the ODbL. As I indicated, this means they need to be removed,[2] which in technical terms means redacting them. Identifying the material to be redacted will not be easy. A couple of people have indicated that there may be other sources available which have village locations. Unfortunately, this does not change the status of the HGK data. It is an odd situation where the data is not available under an open license from one agency, but the exact same data is openly licensed from another agency. == Questions == LWG: My understanding is we need permission or a suitable license from the data source and that finding an alternate source would only allow us to bring in the new data, not keep the old data. Is this correct? LWG+DWG: If someone re-uploads the HGK data because they disagree with the redaction, they will of intentionally and knowingly uploaded copyrighted material without permission or a license. Does the DMCA oblige us to do something about this? If they do this, should we do something even if we aren't required to under the DMCA? [1]: http://lists.openstreetmap.org/pipermail/talk-tr/2013-April/000291.html as well as a Bing translation of their FAQ [2]: http://lists.openstreetmap.org/pipermail/talk-tr/2013-April/000293.html -Original Message- From: Paul Norman [mailto:penor...@mac.com] Sent: Sunday, April 28, 2013 8:15 PM To: 'Suha Ulgen' Cc: 'OSM Mikel Maron'; 'Schuyler Erle'; 'Mikel Maron'; 'Kate Chapman' Subject: RE: OSM place name data from Turkey > From: Suha Ulgen [mailto:m...@suhaulgen.com] > Sent: Sunday, April 28, 2013 7:56 PM > Subject: OSM place name data from Turkey > > Paul, > > The Turkish OSM community is very distressed about your recent > unilateral action. Apparently you are erasing the place names in the > Turkish dataset stating that the source data which you identify as > belonging to the Turkish General Command for Mapping (Harita Genel > Komutanligi - HGK) is not ODbL-compatible. This issue was raised on the talk-tr@ mailing list. The conclusion, as you mentioned, was that the HGK data was not ODbL compatible. Not being ODbL compatible, we cannot distribute it. > The counter argument is that Turkish gazetteer data has been submitted > by HGK to the UN Group of Experts on Geographic Names (UNGEGN) and is > therefore in the "public domain". The list raised the possibility that there might be an alternate source of Turkish place names. Unfortunately, this doesn't change the status of the HGK sourced material, which has explicit restrictions against distribution. Do you anticipate getting permission from *HGK* to distribute their data? > Please STOP "redacting" the Turkish place names. I'll talk to the > UNGEGN Secretariat tomorrow and revert ASAP. It should not be technically possible to use the normal revert tools on a redaction. In any case, do not reintroduce the data downloaded from HGK that we do not have permission to redistribute. ___ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk
