From matthias.straetling at buerotiger.de  Thu Dec 12 06:59:28 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Thu, 12 Dec 2019 07:59:28 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
Message-ID: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>

Dear IANALs,

I want to use polygons (district boundaries) from OSM dataset to select points for a proprietary dataset.
The OSM dataset might be altered trivially (f.e. boundaries might be merged where needed).
The proprietary data isn't allowed to be used freely and is incompatible with ODBL.

The result of the intersections is a geodatabase, which doesn't contain any OSM data.

Is there some kind of share-alike? Does the resulting dataset need attribution like "selected with OSM data"?

Regards,
Matthias


From frederik at remote.org  Thu Dec 12 07:18:41 2019
From: frederik at remote.org (Frederik Ramm)
Date: Thu, 12 Dec 2019 08:18:41 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
Message-ID: <2c1e52e5-969e-d042-0fe0-6b11e0e0a358@remote.org>

Hi,

On 12.12.19 07:59, matthias.straetling at buerotiger.de wrote:
> I want to use polygons (district boundaries) from OSM dataset to select points for a proprietary dataset.
> The OSM dataset might be altered trivially (f.e. boundaries might be merged where needed).
> The proprietary data isn't allowed to be used freely and is incompatible with ODBL.
> 
> The result of the intersections is a geodatabase, which doesn't contain any OSM data.

In my NAL opinion, the result will be derived from OSM data and
therefore inherits the ODbL license. This does, however, not mean that
you have to publish it; but *if* you publish (or "publilcy use") it,
then it has to be available under ODbL. If you just use it internally
then it is still ODbL but that doesn't matter to you.

As an exception to the above, if the number of boundaries you use is
less than 100 - an crucially this could be after the trivial alterations
you mention - then the extract you are making is considered not to be
substantial (see
https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Substantial_-_Guideline)
and therefore does not have to be under ODbL.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"


From dieterdreist at gmail.com  Thu Dec 12 08:08:28 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Thu, 12 Dec 2019 09:08:28 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <2c1e52e5-969e-d042-0fe0-6b11e0e0a358@remote.org>
References: <2c1e52e5-969e-d042-0fe0-6b11e0e0a358@remote.org>
Message-ID: <6F6F67A2-DBC0-4997-9614-BA721DC5E5F8@gmail.com>



sent from a phone

> On 12. Dec 2019, at 08:19, Frederik Ramm <frederik at remote.org> wrote:
> 
> As an exception to the above, if the number of boundaries you use is
> less than 100 - an crucially this could be after the trivial alterations
> you mention - then the extract you are making is considered not to be
> substantial (see
> https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Substantial_-_Guideline)
> and therefore does not have to be under ODbL.


when you write „number of boundaries“, you intend „boundary points“?

Cheers Martin 

From dieterdreist at gmail.com  Thu Dec 12 08:21:31 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Thu, 12 Dec 2019 09:21:31 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
Message-ID: <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>

Am Do., 12. Dez. 2019 um 08:01 Uhr schrieb <
matthias.straetling at buerotiger.de>:

> I want to use polygons (district boundaries) from OSM dataset to select
> points for a proprietary dataset.



>From a practical point of view, boundaries in OSM rarely originate from
surveys, you might be lucky to be able to identify the original source
(most likely open data) which may have a more liberal license than ODbL
(check the history and changeset source tags / object source tags).

Cheers
Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/507afb87/attachment.html>

From matthias.straetling at buerotiger.de  Thu Dec 12 18:49:57 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Thu, 12 Dec 2019 19:49:57 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <2c1e52e5-969e-d042-0fe0-6b11e0e0a358@remote.org>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <2c1e52e5-969e-d042-0fe0-6b11e0e0a358@remote.org>
Message-ID: <trinity-0a4b73bf-fd60-4c50-81df-78c93ee853e9-1576176597380@3c-app-gmx-bap20>

Hi,

> In my NAL opinion, the result will be derived from OSM data and
> therefore inherits the ODbL license. This does, however, not mean that
> you have to publish it; but *if* you publish (or "publilcy use") it,
> then it has to be available under ODbL. If you just use it internally
> then it is still ODbL but that doesn't matter to you.

Oh wait. I don't want to publish the modified OpenStreetMap data.
I want to publish the proprietary dataset, which I've selected using the OpenStreetMap data.
For example, I want to select the non-free points using postcode polygons, derived from OpenStreetMap.

Regards,
Matthias


From matthias.straetling at buerotiger.de  Thu Dec 12 18:50:44 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Thu, 12 Dec 2019 19:50:44 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <6F6F67A2-DBC0-4997-9614-BA721DC5E5F8@gmail.com>
References: <2c1e52e5-969e-d042-0fe0-6b11e0e0a358@remote.org>
 <6F6F67A2-DBC0-4997-9614-BA721DC5E5F8@gmail.com>
Message-ID: <trinity-8bd7ccab-ce9d-43a5-bfcc-cc24401ad931-1576176643973@3c-app-gmx-bap20>

Hi,

> when you write „number of boundaries“, you intend „boundary points“?

No, for example postcodes. I want to merge some of them to new polygons.

Regards,
Matthias


From matthias.straetling at buerotiger.de  Thu Dec 12 18:53:08 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Thu, 12 Dec 2019 19:53:08 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
Message-ID: <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>

> From a practical point of view, boundaries in OSM rarely originate from surveys,
> you might be lucky to be able to identify the original source (most likely open data)
> which may have a more liberal license than ODbL (check the history and changeset
> source tags / object source tags).

But I neither want to merge OSM data or add it to my data. I just want to use it to
select points of my dataset.


From nunocapelocaldeira at gmail.com  Thu Dec 12 19:05:13 2019
From: nunocapelocaldeira at gmail.com (Nuno Caldeira)
Date: Thu, 12 Dec 2019 19:05:13 +0000
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
Message-ID: <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>

https://opendatacommons.org/licenses/odbl/1.0/index.html
the license is quite clear and 4.2 applies to the case you mention. any use
of OSM data (over 100 nodes) combined with proprietary data results in more
open data under ODbL.
we are here to create more open data, not to feed proprietary data than is
lock under their TOS.

On Thu, 12 Dec 2019, 18:53 , <matthias.straetling at buerotiger.de> wrote:

> > From a practical point of view, boundaries in OSM rarely originate from
> surveys,
> > you might be lucky to be able to identify the original source (most
> likely open data)
> > which may have a more liberal license than ODbL (check the history and
> changeset
> > source tags / object source tags).
>
> But I neither want to merge OSM data or add it to my data. I just want to
> use it to
> select points of my dataset.
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/1bf482a1/attachment.html>

From dieterdreist at gmail.com  Thu Dec 12 19:22:09 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Thu, 12 Dec 2019 20:22:09 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
Message-ID: <CABPTjTBG3LuCThm1HcQCGpZ3gJmHWYUXZApo51WaPbCk-=vL+w@mail.gmail.com>

Am Do., 12. Dez. 2019 um 19:53 Uhr schrieb <
matthias.straetling at buerotiger.de>:

> But I neither want to merge OSM data or add it to my data. I just want to
> use it to
> select points of my dataset.



then it may eventually fall under the geocoding guideline:
https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline

Cheers
Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/1a5c5753/attachment.html>

From simon at poole.ch  Thu Dec 12 19:31:25 2019
From: simon at poole.ch (Simon Poole)
Date: Thu, 12 Dec 2019 20:31:25 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
Message-ID: <9183e2bf-16b2-969b-94ee-20bb5897fcd0@poole.ch>

Yes, if a Derivative Database was created in the first place, and that
is not clear at all, see:

/“Derivative Database” – Means a database based upon the Database, and//
//includes any translation, adaptation, arrangement, modification, or any//
//other alteration of the Database or of a Substantial part of the//
//Contents. This includes, but is not limited to, Extracting or//
//Re-utilising the whole or a Substantial part of the Contents in a new//
//Database./

Essentially the question is if the envelopes of the extracted points are
a substantial extract of our data (because they represent the residual
information/data from OSM), while it is obviously possible to construct
cases were this would be the case, in the typical relatively sparse POI
scenario I don't quite see that.

I don't particularly like engaging in these discussion because they tend
to end up being arguments about how many angels can stand on a pin and
are not answerable outside of the detailed specifics of the actual use
case (which are typically not available).

Simon

Am 12.12.2019 um 20:05 schrieb Nuno Caldeira:
> https://opendatacommons.org/licenses/odbl/1.0/index.html
> the license is quite clear and 4.2 applies to the case you mention.
> any use of OSM data (over 100 nodes) combined with proprietary data
> results in more open data under ODbL. 
> we are here to create more open data, not to feed proprietary data
> than is lock under their TOS. 
>
> On Thu, 12 Dec 2019, 18:53 , <matthias.straetling at buerotiger.de
> <mailto:matthias.straetling at buerotiger.de>> wrote:
>
>     > From a practical point of view, boundaries in OSM rarely
>     originate from surveys,
>     > you might be lucky to be able to identify the original source
>     (most likely open data)
>     > which may have a more liberal license than ODbL (check the
>     history and changeset
>     > source tags / object source tags).
>
>     But I neither want to merge OSM data or add it to my data. I just
>     want to use it to
>     select points of my dataset.
>
>     _______________________________________________
>     legal-talk mailing list
>     legal-talk at openstreetmap.org <mailto:legal-talk at openstreetmap.org>
>     https://lists.openstreetmap.org/listinfo/legal-talk
>
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/a57e4e1c/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/a57e4e1c/attachment-0001.sig>

From matthias.straetling at buerotiger.de  Thu Dec 12 19:45:34 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Thu, 12 Dec 2019 20:45:34 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
Message-ID: <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>

> we are here to create more open data, not to feed proprietary data than is lock under their TOS.  

I want to apologize for my misunderstanding: my final product does not contain any OpenStreetMap data.


From matthias.straetling at buerotiger.de  Thu Dec 12 19:54:29 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Thu, 12 Dec 2019 20:54:29 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <9183e2bf-16b2-969b-94ee-20bb5897fcd0@poole.ch>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <9183e2bf-16b2-969b-94ee-20bb5897fcd0@poole.ch>
Message-ID: <trinity-63a42bfd-0c93-4c64-9b10-5c292022707c-1576180469856@3c-app-gmx-bap59>

An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/485bf5ca/attachment.html>

From nunocapelocaldeira at gmail.com  Thu Dec 12 19:53:25 2019
From: nunocapelocaldeira at gmail.com (Nuno Caldeira)
Date: Thu, 12 Dec 2019 19:53:25 +0000
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
Message-ID: <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>

does contain derivate however,which means license applies

On Thu, 12 Dec 2019, 19:46 , <matthias.straetling at buerotiger.de> wrote:

> > we are here to create more open data, not to feed proprietary data than
> is lock under their TOS.
>
> I want to apologize for my misunderstanding: my final product does not
> contain any OpenStreetMap data.
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/96ad6c11/attachment.html>

From kathleen.lu at mapbox.com  Thu Dec 12 22:40:18 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Thu, 12 Dec 2019 14:40:18 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
Message-ID: <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>

No, ODbL does not apply to any database that does not include OSM data.
There are two reasons.

First, this example is analogous to the FAQ here:
https://wiki.osmfoundation.org/wiki/Licence/Licence_and_Legal_FAQ#Can_I_use_OSM_data_and_OpenStreetMap-derived_maps_to_verify_my_own_data_without_triggering_share-alike.3F
Can I use OSM data and OpenStreetMap-derived maps to verify my own data
without triggering share-alike?

Yes, provided that you are only comparing and do not copy any OpenStreetMap
data. If you make any changes to your data after making the comparison, you
should be able to reasonably demonstrate that any such change was made
either from your own physical observation or comes from a non-OpenStreetMap
source accessed directly by you. I.e you can compare but not take!

   - Example 1: You notice that a street is called one name on your map and
   another in OpenStreetMap. You should visit the street and check the name,
   then you are free to put that name in your data as it is your own
   observation.


   - Example 2: You notice that a boundary is different in your data and
   OpenStreetMap. You should check back to original authoritative sources and
   make any correction required.


When someone does example #1 above, they compare OSM data and nonOSM data
and make a list of streets to check in the real world. Neither the nonOSM
data nor the list of streets needs to be licensed under ODbL. You may
*compare* freely.
If I understand your usecase correctly, Matthais, you are essentially
checking your list against OSM boundaries. If something is both on your
list and within the OSM boundary, then you say 'yes, this goes on the
secondary list.' Then you want to publish your secondary list. There is no
OSM data in the secondary list so it is not a Derivative Database.

Second, see the Geocoding Guidelines, which Martin also pointed out -
https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline#The_Guideline
Your example is akin to using OSM polygons for certain areas to geocode.
You already have the lat/long for your points (houses and flats), so what
you are getting from OSM is equivalent to the name of the area you are
filtering against (e.g., all these points are in neighborhood X).
The Geocoding Guidelines specifically state "if only names are provided in
Geocoding Results from OSM -- in particular, latitude/longitude information
from OSM is not included in the Geocoding Results -- *a collection of such
results is not a substantial extract*."
Thus, no ODbL obligations attach.

-Kathleen






On Thu, Dec 12, 2019 at 11:55 AM Nuno Caldeira <nunocapelocaldeira at gmail.com>
wrote:

> does contain derivate however,which means license applies
>
> On Thu, 12 Dec 2019, 19:46 , <matthias.straetling at buerotiger.de> wrote:
>
>> > we are here to create more open data, not to feed proprietary data than
>> is lock under their TOS.
>>
>> I want to apologize for my misunderstanding: my final product does not
>> contain any OpenStreetMap data.
>>
>> _______________________________________________
>> legal-talk mailing list
>> legal-talk at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/legal-talk
>>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191212/35dbf1b6/attachment-0001.html>

From frederik at remote.org  Fri Dec 13 08:48:00 2019
From: frederik at remote.org (Frederik Ramm)
Date: Fri, 13 Dec 2019 09:48:00 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
Message-ID: <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>

Kathleen,

On 12.12.19 23:40, Kathleen Lu via legal-talk wrote:
> No, ODbL does not apply to any database that does not include OSM data.

Are you sure about this? Let me give an example:

> If I understand your usecase correctly, Matthais, you are essentially
> checking your list against OSM boundaries. If something is both on your
> list and within the OSM boundary, then you say 'yes, this goes on the
> secondary list.' Then you want to publish your secondary list. There is
> no OSM data in the secondary list so it is not a Derivative Database.

Let us assume I have a list of all streets in Germany with their
geometry, from a non-OSM source.

I want to divide these into two groups: streets that have at least one
pub, and streets that have no pub.

Using OSM information about the location of pubs, I count the number of
pubs along each street, allowing me to make the desired separation.

I end up with a database of "streets that have at least one pub". This
database does not include OSM data.

In my eyes, though, it is still *derived* from OSM data. It is the
result of an algorithmic process that has made use of OSM data; if you
will, the OSM data residue is in the name/description of my new
database: "roads with pubs". It is derived from OSM; it could not have
been made without OSM.

Do you disagree?

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"


From matkoniecz at tutanota.com  Fri Dec 13 09:39:53 2019
From: matkoniecz at tutanota.com (Mateusz Konieczny)
Date: Fri, 13 Dec 2019 10:39:53 +0100 (CET)
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
Message-ID: <LvycERz--3-1@tutanota.com>

12 Dec 2019, 23:40 by legal-talk at openstreetmap.org:
> No, ODbL does not apply to any database that does not include OSM data.

It is true but misleading to mention here as this database contains transformed OSM data.

> Can I use OSM data and OpenStreetMap-derived maps to verify my own data 
without triggering share-alike?

Mentioning this here is extremely misleading. 

> If something is both on your list and within the OSM boundary, then you say 'yes, 
this goes on the secondary list.' Then you want to publish your secondary list. 

Note "you should be able to reasonably demonstrate that any such change was made
 either from your own physical observation or comes from a non-OpenStreetMap source
 accessed directly by you. I.e you can compare but not take!"

Note "you should visit the street and check the name, then you are free to put that 
name in your data as it is your own observation."

You are allowed to use this "secondary list" only and solely to pinpoint data that is 
likely to be wrong and requires verification. List itself is OSM derived and ODBL apply.
 
> There is no OSM data in the secondary list so it is not a Derivative Database.

This is blatantly untrue. There is OSM data there, only transformed. 

12 Dec 2019, 23:40 by legal-talk at openstreetmap.org:

> No, ODbL does not apply to any database that does not include OSM data. There are two reasons.
>
> First, this example is analogous to the FAQ here: > https://wiki.osmfoundation.org/wiki/Licence/Licence_and_Legal_FAQ#Can_I_use_OSM_data_and_OpenStreetMap-derived_maps_to_verify_my_own_data_without_triggering_share-alike.3F
> Can I use OSM data and OpenStreetMap-derived maps to verify my own data without triggering share-alike?
>
> Yes, provided that you are only comparing and do not copy any OpenStreetMap data. If you make any changes to your data after making the comparison, you should be able to reasonably demonstrate that any such change was made either from your own physical observation or comes from a non-OpenStreetMap source accessed directly by you. I.e you can compare but not take!
>
> Example 1: You notice that a street is called one name on your map and another in OpenStreetMap. You should visit the street and check the name, then you are free to put that name in your data as it is your own observation.
> Example 2: You notice that a boundary is different in your data and OpenStreetMap. You should check back to original authoritative sources and make any correction required.
>
> When someone does example #1 above, they compare OSM data and nonOSM data and make a list of streets to check in the real world. Neither the nonOSM data nor the list of streets needs to be licensed under ODbL. You may *compare* freely. 
> If I understand your usecase correctly, Matthais, you are essentially checking your list against OSM boundaries. If something is both on your list and within the OSM boundary, then you say 'yes, this goes on the secondary list.' Then you want to publish your secondary list. There is no OSM data in the secondary list so it is not a Derivative Database.
>
> Second, see the Geocoding Guidelines, which Martin also pointed out - > https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline#The_Guideline>  
> Your example is akin to using OSM polygons for certain areas to geocode. You already have the lat/long for your points (houses and flats), so what you are getting from OSM is equivalent to the name of the area you are filtering against (e.g., all these points are in neighborhood X). 
> The Geocoding Guidelines specifically state "if only names are provided in Geocoding Results from OSM -- in particular, latitude/longitude information from OSM is not included in the Geocoding Results -- > a collection of such results is not a substantial extract> ." 
> Thus, no ODbL obligations attach.
>
> -Kathleen
>
>  
>
>
>
>
> On Thu, Dec 12, 2019 at 11:55 AM Nuno Caldeira <> nunocapelocaldeira at gmail.com> > wrote:
>
>> does contain derivate however,which means license applies 
>>
>> On Thu, 12 Dec 2019, 19:46 , <>> matthias.straetling at buerotiger.de>> > wrote:
>>
>>> > we are here to create more open data, not to feed proprietary data than is lock under their TOS.  
>>>  
>>>  I want to apologize for my misunderstanding: my final product does not contain any OpenStreetMap data.
>>>  
>>>  _______________________________________________
>>>  legal-talk mailing list
>>>  >>> legal-talk at openstreetmap.org
>>>  >>> https://lists.openstreetmap.org/listinfo/legal-talk
>>>
>> _______________________________________________
>>  legal-talk mailing list
>>  >> legal-talk at openstreetmap.org
>>  >> https://lists.openstreetmap.org/listinfo/legal-talk
>>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/053da570/attachment.html>

From matkoniecz at tutanota.com  Fri Dec 13 09:44:27 2019
From: matkoniecz at tutanota.com (Mateusz Konieczny)
Date: Fri, 13 Dec 2019 10:44:27 +0100 (CET)
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
Message-ID: <LvydHe6--3-1@tutanota.com>




13 Dec 2019, 09:48 by frederik at remote.org:

> I end up with a database of "streets that have at least one pub". This
> database does not include OSM data.
>
> In my eyes, though, it is still *derived* from OSM data. It is the
> result of an algorithmic process that has made use of OSM data; if you
> will, the OSM data residue is in the name/description of my new
> database: "roads with pubs". It is derived from OSM; it could not have
> been made without OSM.
>
Or more extreme. 

(1) I create database of points on the grid of 0.1 cm.
(2) I create two list, first contains what according to OSM is on land,
second contains what according to OSM data is water
(3) I claim that both lists are not ODBL licenced.

Are you going to claim that also this one is legally sound, and not Mapbox-tier 
"attribution hidden behind mystery meat "(i)" navigation 
fulfills ODBL attribution requirement" swindle?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/219e2fe4/attachment.html>

From Lars-Daniel.Weber at gmx.de  Fri Dec 13 18:25:18 2019
From: Lars-Daniel.Weber at gmx.de (Lars-Daniel Weber)
Date: Fri, 13 Dec 2019 19:25:18 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <LvydHe6--3-1@tutanota.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <LvydHe6--3-1@tutanota.com>
Message-ID: <trinity-42a2805d-f223-42fb-b365-f63f832300ee-1576261518091@3c-app-gmx-bap23>

An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/8ebedba8/attachment.html>

From matthias.straetling at buerotiger.de  Fri Dec 13 18:28:38 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Fri, 13 Dec 2019 19:28:38 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-42a2805d-f223-42fb-b365-f63f832300ee-1576261518091@3c-app-gmx-bap23>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <LvydHe6--3-1@tutanota.com>
 <trinity-42a2805d-f223-42fb-b365-f63f832300ee-1576261518091@3c-app-gmx-bap23>
Message-ID: <trinity-63ec25e8-35d9-4d1e-8dc0-0551e01aa838-1576261717948@3c-app-gmx-bap23>

An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/d59042df/attachment.html>

From kathleen.lu at mapbox.com  Fri Dec 13 18:28:50 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Fri, 13 Dec 2019 10:28:50 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
Message-ID: <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>

Hi Frederik,

Here's why I disagree. The meaning of "derived" in a colloquial sense and
the definition of "Derivative Database" are not the same.
While colloquially, it may be fair to interpret "derived" as "made from" or
"could not have been made without", that is not the legal definition of
"Derivative Database".

>From ODbL:
“Derivative Database” – Means a database based upon the Database, and
includes any translation, adaptation, arrangement, modification, or any
other alteration of the Database or of a Substantial part of the
Contents. This includes, but is not limited to, Extracting or
Re-utilising the whole or a Substantial part of the Contents in a new
Database.

So a Derivative Database must include a "translation, adaptation,
arrangement, modification, or any other alteration of the Database or of a
Substantial part of the
Contents." In other words, it has to include in the new database at least a
substantial part of what was in the previous database.
The inference of "with pubs" would not be, in my mind, a "translation,
adaptation, arrangement, modification, or any other alteration of the
Database or of a Substantial part of the
Contents" because it is too minor of an inference.
I view Mattias's usecase as *using* OSM, not *making a Derivative Database
from OSM*.
I would also say that, looking back at the EU Database Directive, I do not
see a case for breach of the restricted rights, particularly the right of
"translation, adaptation, arrangement and any other alteration."

With respect to Mateusz's more extreme example, this is also very
specifically covered in the Geocoding Guidelines: "A collection of
Geocoding Results will be considered a systematic attempt to aggregate data
if it is used as a general purpose geodatabase, regardless of how the
original aggregation was accomplished."

In other worse, if, as in Mateusz's hypothetical, you attempt to abuse the
system to reverse engineer a database that is equivalent to OSM, you make a
Derivative Database. But Mattias has been very clear that is not what he's
doing. He just wants to display the subparts of a list of points he already
has on a different layer than the other subparts.

-Kathleen


On Fri, Dec 13, 2019 at 12:49 AM Frederik Ramm <frederik at remote.org> wrote:

> Kathleen,
>
> On 12.12.19 23:40, Kathleen Lu via legal-talk wrote:
> > No, ODbL does not apply to any database that does not include OSM data.
>
> Are you sure about this? Let me give an example:
>
> > If I understand your usecase correctly, Matthais, you are essentially
> > checking your list against OSM boundaries. If something is both on your
> > list and within the OSM boundary, then you say 'yes, this goes on the
> > secondary list.' Then you want to publish your secondary list. There is
> > no OSM data in the secondary list so it is not a Derivative Database.
>
> Let us assume I have a list of all streets in Germany with their
> geometry, from a non-OSM source.
>
> I want to divide these into two groups: streets that have at least one
> pub, and streets that have no pub.
>
> Using OSM information about the location of pubs, I count the number of
> pubs along each street, allowing me to make the desired separation.
>
> I end up with a database of "streets that have at least one pub". This
> database does not include OSM data.
>
> In my eyes, though, it is still *derived* from OSM data. It is the
> result of an algorithmic process that has made use of OSM data; if you
> will, the OSM data residue is in the name/description of my new
> database: "roads with pubs". It is derived from OSM; it could not have
> been made without OSM.
>
> Do you disagree?
>
> Bye
> Frederik
>
> --
> Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/ab910383/attachment-0001.html>

From matthias.straetling at buerotiger.de  Fri Dec 13 18:32:10 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Fri, 13 Dec 2019 19:32:10 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <LvycERz--3-1@tutanota.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <LvycERz--3-1@tutanota.com>
Message-ID: <trinity-f584154a-fe99-469f-a54b-f3bb6bdb1a32-1576261930556@3c-app-gmx-bap23>

Hi Mateusz,

>> No, ODbL does not apply to any database that does not include OSM data.
> It is true but misleading to mention here as this database contains transformed OSM data.
 
So if I don't merge the postcodes, it's fine?

>> There is no OSM data in the secondary list so it is not a Derivative Database.
> This is blatantly untrue. There is OSM data there, only transformed.

So as soon I'm selecting any data using OSM polygons, it gets transformed OSM data?
They're not even touching on the same layer, since it's a different feature type.

Regards,
Matthias



From frederik at remote.org  Fri Dec 13 18:56:00 2019
From: frederik at remote.org (Frederik Ramm)
Date: Fri, 13 Dec 2019 19:56:00 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
Message-ID: <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>

Hi,

On 13.12.19 19:28, Kathleen Lu via legal-talk wrote:
> “Derivative Database” – Means a database based upon the Database, and
> includes any translation, adaptation, arrangement, modification, or any
> other alteration of the Database or of a Substantial part of the
> Contents.

Interesting. I knew the ODbL text but I have always glossed over this
definition, assuming that "well you know what derived means".

I'll have to ponder this for a while, it changes some assumptions I had
made. It would mean that, for example, a database that contains a count
of all pubs in each municipality, or a database that contains the
average travel time from a building in a city to the nearest hospital,
or a heatmap of ice cream parlours, would not fall under the ODbL
because these, while derived from OSM, do not actually contain a copy of
anything in OSM (and neither could they possibly be used to reassemble
OSM).

I had until now assumed that such works would definitely fall under the
ODbL but you are right, they don't really fit the "Derivative Database"
definition.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"


From dieterdreist at gmail.com  Fri Dec 13 19:04:55 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Fri, 13 Dec 2019 20:04:55 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-f584154a-fe99-469f-a54b-f3bb6bdb1a32-1576261930556@3c-app-gmx-bap23>
References: <trinity-f584154a-fe99-469f-a54b-f3bb6bdb1a32-1576261930556@3c-app-gmx-bap23>
Message-ID: <A7E7BE98-D832-424B-A0D3-45A365CEB812@gmail.com>



sent from a phone

> On 13. Dec 2019, at 19:32, matthias.straetling at buerotiger.de wrote:
> 
> So as soon I'm selecting any data using OSM polygons, it gets transformed OSM data?
> They're not even touching on the same layer, since it's a different feature type.


if you modify your data based on OpenStreetMap data you are creating a derivative database.

Cheers Martin 

From nunocapelocaldeira at gmail.com  Fri Dec 13 19:11:12 2019
From: nunocapelocaldeira at gmail.com (Nuno Caldeira)
Date: Fri, 13 Dec 2019 19:11:12 +0000
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
Message-ID: <CAK5cRsq1sNY3Gyi-63hGZzie5kV9LZ-oWkq6AapuMCD=_kGtDg@mail.gmail.com>

these new Liberal interpretation of ODbL are funny. to bad it's not
documented what we wanted when we changed license. seems to be full of lies

https://wiki.osmfoundation.org/wiki/Licence/Historic/We_Are_Changing_The_License

*"This means that “good guys” are stopped from using our data but the “bad
guys” may be able to use it anyway." *

*" We believe that a reasonable consensus has been built that our current
progress should be to maintain a Share-Alike license (see more below) but
have it written explicitly for data."*

*"Both licenses are “By Attribution” and “Share Alike”." *

*"But what happens if the Foundation is taken over by people with
commercial interests?*

   - *You still own the rights to any data you contribute, not the
   Foundation. In the new Contributor Terms, you license the Foundation to
   publish the data for others to use and ONLY under a free and open license.*


   - *The Foundation is not allowed to take your contribution and release
   it under a commercial license.*


   - *If the Foundation fails to publish under only a free and open
   license, it has broken its contract with you. A copy of the existing data
   can be made and released by a different body.*


   - *If a change is made to another free and open license, it is active
   contributors who decide yes or no, not the Foundation."*



On Fri, 13 Dec 2019, 18:56 Frederik Ramm, <frederik at remote.org> wrote:

> Hi,
>
> On 13.12.19 19:28, Kathleen Lu via legal-talk wrote:
> > “Derivative Database” – Means a database based upon the Database, and
> > includes any translation, adaptation, arrangement, modification, or any
> > other alteration of the Database or of a Substantial part of the
> > Contents.
>
> Interesting. I knew the ODbL text but I have always glossed over this
> definition, assuming that "well you know what derived means".
>
> I'll have to ponder this for a while, it changes some assumptions I had
> made. It would mean that, for example, a database that contains a count
> of all pubs in each municipality, or a database that contains the
> average travel time from a building in a city to the nearest hospital,
> or a heatmap of ice cream parlours, would not fall under the ODbL
> because these, while derived from OSM, do not actually contain a copy of
> anything in OSM (and neither could they possibly be used to reassemble
> OSM).
>
> I had until now assumed that such works would definitely fall under the
> ODbL but you are right, they don't really fit the "Derivative Database"
> definition.
>
> Bye
> Frederik
>
> --
> Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/aa48dc70/attachment.html>

From kathleen.lu at mapbox.com  Fri Dec 13 19:34:54 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Fri, 13 Dec 2019 11:34:54 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAK5cRsq1sNY3Gyi-63hGZzie5kV9LZ-oWkq6AapuMCD=_kGtDg@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
 <CAK5cRsq1sNY3Gyi-63hGZzie5kV9LZ-oWkq6AapuMCD=_kGtDg@mail.gmail.com>
Message-ID: <CAFekdnAF63R6crnjRKG5gXXkhXR9hi+_ED1fRYr4H9evvqqKkA@mail.gmail.com>

Nuno - I think you are operating under the mistaken assumption that a
CC-BY-SA license would mean that uses such as Mattias's would require
sharealike.

Here's CC-BY-SA's definition of a Derivative Work:
*"Derivative Work"* means a work based upon the Work or upon the Work and
other pre-existing works, such as a translation, musical arrangement,
dramatization, fictionalization, motion picture version, sound recording,
art reproduction, abridgment, condensation, or any other form in which the
Work may be recast, transformed, or adapted, except that a work that
constitutes a Collective Work will not be considered a Derivative Work for
the purpose of this License. For the avoidance of doubt, where the Work is
a musical composition or sound recording, the synchronization of the Work
in timed-relation with a moving image ("synching") will be considered a
Derivative Work for the purpose of this License.

Here's CC-BY-SA's definition of a Collective Work:
*"Collective Work"* means a work, such as a periodical issue, anthology or
encyclopedia, in which the Work in its entirety in unmodified form, along
with a number of other contributions, constituting separate and independent
works in themselves, are assembled into a collective whole. A work that
constitutes a Collective Work will not be considered a Derivative Work (as
defined below) for the purposes of this License.

As you can see from these examples (which focus on creative derivatives,
since facts are not even copyrightable in the US and there is no US
database protection law), a "derivative work" needs quite a bit of the
original to qualify. The meaning of a "derivative work" was always much
narrower than what a colloquial understanding of what "derived" might be,
and the change in license did not change that.

-Kathleen



On Fri, Dec 13, 2019 at 11:11 AM Nuno Caldeira <nunocapelocaldeira at gmail.com>
wrote:

> these new Liberal interpretation of ODbL are funny. to bad it's not
> documented what we wanted when we changed license. seems to be full of lies
>
>
> https://wiki.osmfoundation.org/wiki/Licence/Historic/We_Are_Changing_The_License
>
> *"This means that “good guys” are stopped from using our data but the “bad
> guys” may be able to use it anyway." *
>
> *" We believe that a reasonable consensus has been built that our current
> progress should be to maintain a Share-Alike license (see more below) but
> have it written explicitly for data."*
>
> *"Both licenses are “By Attribution” and “Share Alike”." *
>
> *"But what happens if the Foundation is taken over by people with
> commercial interests?*
>
>    - *You still own the rights to any data you contribute, not the
>    Foundation. In the new Contributor Terms, you license the Foundation to
>    publish the data for others to use and ONLY under a free and open license.*
>
>
>    - *The Foundation is not allowed to take your contribution and release
>    it under a commercial license.*
>
>
>    - *If the Foundation fails to publish under only a free and open
>    license, it has broken its contract with you. A copy of the existing data
>    can be made and released by a different body.*
>
>
>    - *If a change is made to another free and open license, it is active
>    contributors who decide yes or no, not the Foundation."*
>
>
>
> On Fri, 13 Dec 2019, 18:56 Frederik Ramm, <frederik at remote.org> wrote:
>
>> Hi,
>>
>> On 13.12.19 19:28, Kathleen Lu via legal-talk wrote:
>> > “Derivative Database” – Means a database based upon the Database, and
>> > includes any translation, adaptation, arrangement, modification, or any
>> > other alteration of the Database or of a Substantial part of the
>> > Contents.
>>
>> Interesting. I knew the ODbL text but I have always glossed over this
>> definition, assuming that "well you know what derived means".
>>
>> I'll have to ponder this for a while, it changes some assumptions I had
>> made. It would mean that, for example, a database that contains a count
>> of all pubs in each municipality, or a database that contains the
>> average travel time from a building in a city to the nearest hospital,
>> or a heatmap of ice cream parlours, would not fall under the ODbL
>> because these, while derived from OSM, do not actually contain a copy of
>> anything in OSM (and neither could they possibly be used to reassemble
>> OSM).
>>
>> I had until now assumed that such works would definitely fall under the
>> ODbL but you are right, they don't really fit the "Derivative Database"
>> definition.
>>
>> Bye
>> Frederik
>>
>> --
>> Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"
>>
>> _______________________________________________
>> legal-talk mailing list
>> legal-talk at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/legal-talk
>>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/ac232b12/attachment-0001.html>

From dieterdreist at gmail.com  Fri Dec 13 19:35:19 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Fri, 13 Dec 2019 20:35:19 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
References: <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
Message-ID: <4672FC1E-0A22-4852-B5BC-90284873C2B1@gmail.com>



sent from a phone

> On 13. Dec 2019, at 19:56, Frederik Ramm <frederik at remote.org> wrote:
> 
> I'll have to ponder this for a while, it changes some assumptions I had
> made. It would mean that, for example, a database that contains a count
> of all pubs in each municipality,



-> adaptation 


> or a database that contains the
> average travel time from a building in a city to the nearest hospital,


-> adaptation 


> or a heatmap of ice cream parlours,


-> adaptation 



> (and neither could they possibly be used to reassemble
> OSM).


is there a requirement in the ODbL that your adaptation or alteration has to be reversible?

Cheers Martin 



From nunocapelocaldeira at gmail.com  Fri Dec 13 19:45:13 2019
From: nunocapelocaldeira at gmail.com (Nuno Caldeira)
Date: Fri, 13 Dec 2019 19:45:13 +0000
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnAF63R6crnjRKG5gXXkhXR9hi+_ED1fRYr4H9evvqqKkA@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
 <CAK5cRsq1sNY3Gyi-63hGZzie5kV9LZ-oWkq6AapuMCD=_kGtDg@mail.gmail.com>
 <CAFekdnAF63R6crnjRKG5gXXkhXR9hi+_ED1fRYr4H9evvqqKkA@mail.gmail.com>
Message-ID: <3169ce79-12b1-a5ef-9bc1-237597e4d11f@gmail.com>

well 
https://wiki.osmfoundation.org/wiki/Licence/Licence_and_Legal_FAQ#The_OpenStreetMap_Geodata_Licence

> Secondly, you *"Share Alike"*. If you do not make any changes to 
> OpenStreetMap data, then you are unlikely to have a "Share Alike" 
> obligation. But, if you _publicly distribute something that you have 
> made_ from our data, such as a _map or another database_, AND you have 
> _added to or enhanced our data_, then we want you to make those 
> additions publicly available. We obviously prefer it if you added the 
> data straight back to our database, but you do not have to, _as long 
> as the public can easily get a copy of what you have done._ If you do 
> not publicly distribute anything, then you do not have to share anything. 

Às 19:34 de 13/12/2019, Kathleen Lu via legal-talk escreveu:
> Nuno - I think you are operating under the mistaken assumption that a 
> CC-BY-SA license would mean that uses such as Mattias's would require 
> sharealike.
>
> Here's CC-BY-SA's definition of a Derivative Work:
> *"Derivative Work"*means a work based upon the Work or upon the Work 
> and other pre-existing works, such as a translation, musical 
> arrangement, dramatization, fictionalization, motion picture version, 
> sound recording, art reproduction, abridgment, condensation, or any 
> other form in which the Work may be recast, transformed, or adapted, 
> except that a work that constitutes a Collective Work will not be 
> considered a Derivative Work for the purpose of this License. For the 
> avoidance of doubt, where the Work is a musical composition or sound 
> recording, the synchronization of the Work in timed-relation with a 
> moving image ("synching") will be considered a Derivative Work for the 
> purpose of this License.
>
> Here's CC-BY-SA's definition of a Collective Work:
> *"Collective Work"*means a work, such as a periodical issue, anthology 
> or encyclopedia, in which the Work in its entirety in unmodified form, 
> along with a number of other contributions, constituting separate and 
> independent works in themselves, are assembled into a collective 
> whole. A work that constitutes a Collective Work will not be 
> considered a Derivative Work (as defined below) for the purposes of 
> this License.
>
> As you can see from these examples (which focus on creative 
> derivatives, since facts are not even copyrightable in the US and 
> there is no US database protection law), a "derivative work" needs 
> quite a bit of the original to qualify. The meaning of a "derivative 
> work" was always much narrower than what a colloquial understanding of 
> what "derived" might be, and the change in license did not change that.
>
> -Kathleen
>
>
>
> On Fri, Dec 13, 2019 at 11:11 AM Nuno Caldeira 
> <nunocapelocaldeira at gmail.com <mailto:nunocapelocaldeira at gmail.com>> 
> wrote:
>
>     these new Liberal interpretation of ODbL are funny. to bad it's
>     not documented what we wanted when we changed license. seems to be
>     full of lies
>
>     https://wiki.osmfoundation.org/wiki/Licence/Historic/We_Are_Changing_The_License
>
>     /
>     /
>     /"This means that “good guys” are stopped from using our data but
>     the “bad guys” may be able to use it anyway." /
>     /
>     /
>     /" We believe that a reasonable consensus has been built that our
>     current progress should be to maintain a Share-Alike license (see
>     more below) but have it written explicitly for data."/
>     /
>     /
>     /"Both licenses are “By Attribution” and “Share Alike”." /
>     /
>     /
>     /"But what happens if the Foundation is taken over by people with
>     commercial interests?/
>
>       * /You still own the rights to any data you contribute, not the
>         Foundation. In the new Contributor Terms, you license the
>         Foundation to publish the data for others to use and ONLY
>         under a free and open license./
>
>       * /The Foundation is not allowed to take your contribution and
>         release it under a commercial license./
>
>       * /If the Foundation fails to publish under only a free and open
>         license, it has broken its contract with you. A copy of the
>         existing data can be made and released by a different body./
>
>       * /If a change is made to another free and open license, it is
>         active contributors who decide yes or no, not the Foundation."/
>
>
>
>     On Fri, 13 Dec 2019, 18:56 Frederik Ramm, <frederik at remote.org
>     <mailto:frederik at remote.org>> wrote:
>
>         Hi,
>
>         On 13.12.19 19:28, Kathleen Lu via legal-talk wrote:
>         > “Derivative Database” – Means a database based upon the
>         Database, and
>         > includes any translation, adaptation, arrangement,
>         modification, or any
>         > other alteration of the Database or of a Substantial part of the
>         > Contents.
>
>         Interesting. I knew the ODbL text but I have always glossed
>         over this
>         definition, assuming that "well you know what derived means".
>
>         I'll have to ponder this for a while, it changes some
>         assumptions I had
>         made. It would mean that, for example, a database that
>         contains a count
>         of all pubs in each municipality, or a database that contains the
>         average travel time from a building in a city to the nearest
>         hospital,
>         or a heatmap of ice cream parlours, would not fall under the ODbL
>         because these, while derived from OSM, do not actually contain
>         a copy of
>         anything in OSM (and neither could they possibly be used to
>         reassemble
>         OSM).
>
>         I had until now assumed that such works would definitely fall
>         under the
>         ODbL but you are right, they don't really fit the "Derivative
>         Database"
>         definition.
>
>         Bye
>         Frederik
>
>         -- 
>         Frederik Ramm  ##  eMail frederik at remote.org
>         <mailto:frederik at remote.org> ##  N49°00'09" E008°23'33"
>
>         _______________________________________________
>         legal-talk mailing list
>         legal-talk at openstreetmap.org <mailto:legal-talk at openstreetmap.org>
>         https://lists.openstreetmap.org/listinfo/legal-talk
>
>     _______________________________________________
>     legal-talk mailing list
>     legal-talk at openstreetmap.org <mailto:legal-talk at openstreetmap.org>
>     https://lists.openstreetmap.org/listinfo/legal-talk
>
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/a311b96e/attachment.html>

From chris_hormann at gmx.de  Fri Dec 13 19:53:32 2019
From: chris_hormann at gmx.de (Christoph Hormann)
Date: Fri, 13 Dec 2019 20:53:32 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
Message-ID: <201912132053.32630.chris_hormann@gmx.de>

On Friday 13 December 2019, Frederik Ramm wrote:
>
> I had until now assumed that such works would definitely fall under
> the ODbL but you are right, they don't really fit the "Derivative
> Database" definition.

My reading of the ODbL has always been that something is either

1) the original Database (or substantial parts of it)
2) a Derivative Database
3) a Collective Database
4) a Produced Work

or if something is neither of these it would be either

5) something that is not protected by law at all so free to use
independent of the license terms (like insubstantial extracts of data).
6) something the ODbL does not grant any rights for and therefore cannot
be legally used by the user based on the ODbL.

So my question would always be if someone considers certain things not
to be a Derivative Database which of the five other above cases applies
instead.

I would kind of assume that for case (5) there are probably already some
court rulings available for to what extent EU database protection
applies to set operations of different databases since this is nothing
specific to spatial databases but also is relevant for many other types
of data.

As i already wrote in

https://lists.openstreetmap.org/pipermail/talk/2019-November/083535.html

existing OSMF community guidelines suggest spatial operations like
ST_Difference() and ST_Intersection() yield Derivative Databases that
are subject to share-alike.

--
Christoph Hormann
http://www.imagico.de/


From geodesy99 at gmail.com  Fri Dec 13 23:56:58 2019
From: geodesy99 at gmail.com (Michael Patrick)
Date: Fri, 13 Dec 2019 15:56:58 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <mailman.51345.1576265714.689.legal-talk@openstreetmap.org>
References: <mailman.51345.1576265714.689.legal-talk@openstreetmap.org>
Message-ID: <CAC54+E6Ud2Uz1OcYdS5TbFJQZz4-_g9bh4-iixCx4aPoYE1_VQ@mail.gmail.com>

There are other writings about ODBL, but this one captures the issues
fairly well:
https://wiki.openstreetmap.org/wiki/ODbL_comments_from_Creative_Commons , a
more in depth treatment can be found in ' Safe to be Open: Study on the
protection of research data and recommendation for access and usage ':
https://www.fosteropenscience.eu/sites/default/files/pdf/835.pdf

Whether it is government, commercial interest or academic researchers, IMHO
it is almost inevitable that 'mixing' will occur as information flows
downstream and back upstream among federated sources. Until somebody
develops and deploys some sort of blockchain provenance that can be
attached to each DB element to keep track of things through pipelines of
analysis/extract transformations, despite the best intentions of data
consumers, it is impossible to be ODBL compliant except in the very
simplest cases, like displaying the end map (with attribution) or using
exclusively ODBL data.

To some extent, extremely permissive licenses ( or 'license free' like the
US Federal ) allow data to easily flow into ODBL, but ironically that
benefit can not be reciprocated. While the https://opendatacommons.org/
seems to be stagnant since 2010 ( last  'News' post, and the Advisory
Council page links mostly go to 404 ) , in contrast the Creative Commons
organization has been continually evolving and adapting.

For me, the point selection via ODBL polygons creates ODBL data.

Michael Patrick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/49578ff6/attachment.html>

From kathleen.lu at mapbox.com  Sat Dec 14 01:03:08 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Fri, 13 Dec 2019 17:03:08 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <201912132053.32630.chris_hormann@gmx.de>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
 <201912132053.32630.chris_hormann@gmx.de>
Message-ID: <CAFekdnDg056fC=5a=FGhEneGBVkK5hDQXy=Lke7WSdDvXRF_Tg@mail.gmail.com>

Hi Christoph,
I think that there is a premise to your list that I do not quite agree
with. ODbL says:

3.1 Subject to the terms and conditions of this License, the Licensor
grants to You a worldwide, royalty-free, non-exclusive, terminable (but
only under Section 9) license to Use the Database for the duration of
any applicable copyright and Database Rights. These rights explicitly
include commercial use, and do not exclude any field of endeavour. To
the extent possible in the relevant jurisdiction, these rights may be
exercised in all media and formats whether now known or created in the
future.

So the rights granted are *all* database and copyright rights, subject to
certain conditions that apply to Produced Works and Derivative Databases.
There are rights that are completely not covered by ODbL are trademark and
patent rights, which would be #6.
But that also means there is a category #7 you have not listed, where the
database and/or copyright rights *are* conveyed by ODbL and no limitations
on the exercise of those rights is placed on the user.
So as far as usecases like Matthias's which we are discussing, my opinion
is that it is #5, but if it is not, it could be #7. But it could not be #6.



On Fri, Dec 13, 2019 at 11:54 AM Christoph Hormann <chris_hormann at gmx.de>
wrote:

> On Friday 13 December 2019, Frederik Ramm wrote:
> >
> > I had until now assumed that such works would definitely fall under
> > the ODbL but you are right, they don't really fit the "Derivative
> > Database" definition.
>
> My reading of the ODbL has always been that something is either
>
> 1) the original Database (or substantial parts of it)
> 2) a Derivative Database
> 3) a Collective Database
> 4) a Produced Work
>
> or if something is neither of these it would be either
>
> 5) something that is not protected by law at all so free to use
> independent of the license terms (like insubstantial extracts of data).
> 6) something the ODbL does not grant any rights for and therefore cannot
> be legally used by the user based on the ODbL.
>
> So my question would always be if someone considers certain things not
> to be a Derivative Database which of the five other above cases applies
> instead.
>
> I would kind of assume that for case (5) there are probably already some
> court rulings available for to what extent EU database protection
> applies to set operations of different databases since this is nothing
> specific to spatial databases but also is relevant for many other types
> of data.
>
> As i already wrote in
>
> https://lists.openstreetmap.org/pipermail/talk/2019-November/083535.html
>
> existing OSMF community guidelines suggest spatial operations like
> ST_Difference() and ST_Intersection() yield Derivative Databases that
> are subject to share-alike.
>
> --
> Christoph Hormann
> http://www.imagico.de/
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/ca05717e/attachment.html>

From kathleen.lu at mapbox.com  Sat Dec 14 01:07:52 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Fri, 13 Dec 2019 17:07:52 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <3169ce79-12b1-a5ef-9bc1-237597e4d11f@gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
 <CAK5cRsq1sNY3Gyi-63hGZzie5kV9LZ-oWkq6AapuMCD=_kGtDg@mail.gmail.com>
 <CAFekdnAF63R6crnjRKG5gXXkhXR9hi+_ED1fRYr4H9evvqqKkA@mail.gmail.com>
 <3169ce79-12b1-a5ef-9bc1-237597e4d11f@gmail.com>
Message-ID: <CAFekdnDwR1U7zaQ7G80dVxMbBb5uK7pKNyf1-FEwS1-HJaxHqQ@mail.gmail.com>

Nuno -
I do not see how Matthias's usecase qualifies as "AND you have *added to or
enhanced our data*" since the houses and flat and their prices are *not*
added to OSM houses or flats, but if this FAQ answer is misleading, we
should rewrite this FAQ answer to more accurate reflect ODbL.

On Fri, Dec 13, 2019 at 11:45 AM Nuno Caldeira <nunocapelocaldeira at gmail.com>
wrote:

> well
> https://wiki.osmfoundation.org/wiki/Licence/Licence_and_Legal_FAQ#The_OpenStreetMap_Geodata_Licence
>
> Secondly, you *"Share Alike"*. If you do not make any changes to
> OpenStreetMap data, then you are unlikely to have a "Share Alike"
> obligation. But, if you *publicly distribute something that you have made*
> from our data, such as a *map or another database*, AND you have *added
> to or enhanced our data*, then we want you to make those additions
> publicly available. We obviously prefer it if you added the data straight
> back to our database, but you do not have to, *as long as the public can
> easily get a copy of what you have done.* If you do not publicly
> distribute anything, then you do not have to share anything.
>
>
> Às 19:34 de 13/12/2019, Kathleen Lu via legal-talk escreveu:
>
> Nuno - I think you are operating under the mistaken assumption that a
> CC-BY-SA license would mean that uses such as Mattias's would require
> sharealike.
>
> Here's CC-BY-SA's definition of a Derivative Work:
> *"Derivative Work"* means a work based upon the Work or upon the Work and
> other pre-existing works, such as a translation, musical arrangement,
> dramatization, fictionalization, motion picture version, sound recording,
> art reproduction, abridgment, condensation, or any other form in which the
> Work may be recast, transformed, or adapted, except that a work that
> constitutes a Collective Work will not be considered a Derivative Work for
> the purpose of this License. For the avoidance of doubt, where the Work is
> a musical composition or sound recording, the synchronization of the Work
> in timed-relation with a moving image ("synching") will be considered a
> Derivative Work for the purpose of this License.
>
> Here's CC-BY-SA's definition of a Collective Work:
> *"Collective Work"* means a work, such as a periodical issue, anthology
> or encyclopedia, in which the Work in its entirety in unmodified form,
> along with a number of other contributions, constituting separate and
> independent works in themselves, are assembled into a collective whole. A
> work that constitutes a Collective Work will not be considered a Derivative
> Work (as defined below) for the purposes of this License.
>
> As you can see from these examples (which focus on creative derivatives,
> since facts are not even copyrightable in the US and there is no US
> database protection law), a "derivative work" needs quite a bit of the
> original to qualify. The meaning of a "derivative work" was always much
> narrower than what a colloquial understanding of what "derived" might be,
> and the change in license did not change that.
>
> -Kathleen
>
>
>
> On Fri, Dec 13, 2019 at 11:11 AM Nuno Caldeira <
> nunocapelocaldeira at gmail.com> wrote:
>
>> these new Liberal interpretation of ODbL are funny. to bad it's not
>> documented what we wanted when we changed license. seems to be full of
>> lies
>>
>>
>> https://wiki.osmfoundation.org/wiki/Licence/Historic/We_Are_Changing_The_License
>>
>> *"This means that “good guys” are stopped from using our data but the
>> “bad guys” may be able to use it anyway." *
>>
>> *" We believe that a reasonable consensus has been built that our current
>> progress should be to maintain a Share-Alike license (see more below) but
>> have it written explicitly for data."*
>>
>> *"Both licenses are “By Attribution” and “Share Alike”." *
>>
>> *"But what happens if the Foundation is taken over by people with
>> commercial interests?*
>>
>>    - *You still own the rights to any data you contribute, not the
>>    Foundation. In the new Contributor Terms, you license the Foundation to
>>    publish the data for others to use and ONLY under a free and open license.*
>>
>>
>>    - *The Foundation is not allowed to take your contribution and
>>    release it under a commercial license.*
>>
>>
>>    - *If the Foundation fails to publish under only a free and open
>>    license, it has broken its contract with you. A copy of the existing data
>>    can be made and released by a different body.*
>>
>>
>>    - *If a change is made to another free and open license, it is active
>>    contributors who decide yes or no, not the Foundation."*
>>
>>
>>
>> On Fri, 13 Dec 2019, 18:56 Frederik Ramm, <frederik at remote.org> wrote:
>>
>>> Hi,
>>>
>>> On 13.12.19 19:28, Kathleen Lu via legal-talk wrote:
>>> > “Derivative Database” – Means a database based upon the Database, and
>>> > includes any translation, adaptation, arrangement, modification, or any
>>> > other alteration of the Database or of a Substantial part of the
>>> > Contents.
>>>
>>> Interesting. I knew the ODbL text but I have always glossed over this
>>> definition, assuming that "well you know what derived means".
>>>
>>> I'll have to ponder this for a while, it changes some assumptions I had
>>> made. It would mean that, for example, a database that contains a count
>>> of all pubs in each municipality, or a database that contains the
>>> average travel time from a building in a city to the nearest hospital,
>>> or a heatmap of ice cream parlours, would not fall under the ODbL
>>> because these, while derived from OSM, do not actually contain a copy of
>>> anything in OSM (and neither could they possibly be used to reassemble
>>> OSM).
>>>
>>> I had until now assumed that such works would definitely fall under the
>>> ODbL but you are right, they don't really fit the "Derivative Database"
>>> definition.
>>>
>>> Bye
>>> Frederik
>>>
>>> --
>>> Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"
>>>
>>> _______________________________________________
>>> legal-talk mailing list
>>> legal-talk at openstreetmap.org
>>> https://lists.openstreetmap.org/listinfo/legal-talk
>>>
>> _______________________________________________
>> legal-talk mailing list
>> legal-talk at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/legal-talk
>>
>
> _______________________________________________
> legal-talk mailing listlegal-talk at openstreetmap.orghttps://lists.openstreetmap.org/listinfo/legal-talk
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191213/7b2864fc/attachment-0001.html>

From matkoniecz at tutanota.com  Sat Dec 14 05:41:36 2019
From: matkoniecz at tutanota.com (Mateusz Konieczny)
Date: Sat, 14 Dec 2019 06:41:36 +0100 (CET)
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
Message-ID: <Lw1vHrl--3-1@tutanota.com>




13 Dec 2019, 19:28 by legal-talk at openstreetmap.org:

> Hi Frederik,
>
> Here's why I disagree. The meaning of "derived" in a colloquial sense and the definition of "Derivative Database" are not the same.
> While colloquially, it may be fair to interpret "derived" as "made from" or "could not have been made without", that is not the legal definition of "Derivative Database".
>
> From ODbL:
> “Derivative Database” – Means a database based upon the Database, and
> includes any translation, adaptation, arrangement, modification, or any
> other alteration of the Database or of a Substantial part of the
> Contents. This includes, but is not limited to, Extracting or
> Re-utilising the whole or a Substantial part of the Contents in a new
> Database.
>
> So a Derivative Database must include a "translation, adaptation, arrangement, modification, or any other alteration of the Database or of a Substantial part of the
> Contents." In other words, it has to include in the new database at least a substantial part of what was in the previous database.
>
Can you point me to legal definition
of "substantial part"?

I would expect that extracting boundary
data for locations is some other database
would count as "substantial part".

But maybe this loophole actually works
if "substantial" is defined to be unreasonably high.
> The inference of "with pubs" would not be, in my mind, a "translation, adaptation, arrangement, modification, or any other alteration of the Database or of a Substantial part of the
> Contents" because it is too minor of an inference.
>
It is certainly translation/adaptation,
but such use may pass as not
"substantial".

I hope that "not substantial" would
work for something like 

(1) not repeated 
extraction of location data for 100 points,
but I would be happy to be pointed to an
unbiased legal analysis of this term.

(2) not repeated extraction of boundaries
from OSM with no more than 100 nodes
in total

but it is hard to be to guess what can
be considered as substantial.
Probably depends on a lawyer and who
pays them to write a legal decision.
> With respect to Mateusz's more extreme example, this is also very specifically covered in the Geocoding Guidelines: "> A collection of Geocoding Results will be considered a systematic attempt to aggregate data if it is used as a general purpose geodatabase, regardless of how the original aggregation was accomplished."
>
> In other worse, if, as in Mateusz's hypothetical, you attempt to abuse the system to reverse engineer a database that is equivalent to OSM, you make a Derivative Database. But Mattias has been very clear that is not what he's doing. He just wants to display the subparts of a list of points he already has on a different layer than the other subparts. 
>
I thought that it was claimed that it
is not translation/adaptation - but
given that argument relies on 
"substantial" part then my example is
not applicable.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191214/699a29b5/attachment.html>

From robert.whittaker+osm at gmail.com  Sat Dec 14 11:42:41 2019
From: robert.whittaker+osm at gmail.com (Robert Whittaker (OSM lists))
Date: Sat, 14 Dec 2019 11:42:41 +0000
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
Message-ID: <CACavFKs-HPjjVf8mSXR_rnA6bWkqisdV1gHUc4E+L-p=7nvHBw@mail.gmail.com>

On Thu, 12 Dec 2019 22:41 Kathleen Lu via legal-talk, <
legal-talk at openstreetmap.org> wrote:

> No, ODbL does not apply to any database that does not include OSM data.
> There are two reasons.
>

I would argue that the dataset here does include some OSM data, as it includes
(albeit limited) information about the regions enclosed by certain features
in OSM. Regardless of that though, I would use the following reasoning,
involving a chain of derivative databases, to argue that final databases
could be subject to ODbL even if they don't directly contain any OSM data,
if OSM data has been used in their creation.

At some point in the process of deriving the new dataset here, an OSM extract
and some proprietary data were combined into a database. To start with,
these would be independent parts, and so the combined database would be a
Collective Database, and so not subject to ODbL.
However, I would argue that the moment you run a query that combines both
parts of the database in a non-trivial way, those parts can no longer be
said to be "independent", and hence they cannot be part of a
Collective Database. The combined OSM extract, proprietary data, and the
output from the query are now a derivative database, and hence subject to
ODbL, thanks to the presence of the OSM data. The data published is then a
substantial extract from this derivative database,
and is thus is a derivative of the derivative. This makes it also subject
to ODbL.

Robert.

>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191214/a059bd51/attachment.html>

From frederik at remote.org  Sat Dec 14 14:48:41 2019
From: frederik at remote.org (Frederik Ramm)
Date: Sat, 14 Dec 2019 15:48:41 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <Lw1vHrl--3-1@tutanota.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CABPTjTBYbW-=z0bc3q-AuGZ7he6zSTZEtJw9x5e677+PV=QNOQ@mail.gmail.com>
 <trinity-cff85c2b-1967-46f0-b819-99285df26c44-1576176788314@3c-app-gmx-bap20>
 <CAK5cRsoFeOjj-toLXuN4L5KLcLVn9Ta_EB_pprE1emsis-ev2w@mail.gmail.com>
 <trinity-ea6b773e-fd2b-43f4-aa00-94e87a172703-1576179934062@3c-app-gmx-bap59>
 <CAK5cRsoq0u1cZfuv+a_t13KZ02E7PP+T6R++XT_S6smo=onZpg@mail.gmail.com>
 <CAFekdnAbDaOPLME_aTMVv0JzF6y_QcCiS+NO0N9bK36xgvN0rg@mail.gmail.com>
 <0fb2cdc5-a3b7-1a58-023e-c2a1158cce4a@remote.org>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <Lw1vHrl--3-1@tutanota.com>
Message-ID: <7006c7db-fdab-57bd-9b1f-35cc271b57ff@remote.org>

Hi,

On 14.12.19 06:41, Mateusz Konieczny wrote:
> Can you point me to legal definition
> of "substantial part"?

There is none, hence:
https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Substantial_-_Guideline

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"


From matthias.straetling at buerotiger.de  Sat Dec 14 18:53:46 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Sat, 14 Dec 2019 19:53:46 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <201912132053.32630.chris_hormann@gmx.de>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <CAFekdnCJF=pZ281sQH3A7vVvMwyYN_Bcg1pDjfdTvsQrLS_dFQ@mail.gmail.com>
 <e4bd650f-2158-e5b5-70b5-6f9a0c824fa8@remote.org>
 <201912132053.32630.chris_hormann@gmx.de>
Message-ID: <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>

>
> existing OSMF community guidelines suggest spatial operations like
> ST_Difference() and ST_Intersection() yield Derivative Databases that
> are subject to share-alike.

Let's take the Collective Database Guideline, you've mentioned:
https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Collective_Database_Guideline_Guideline

"Technically a reference between non-OSM and OSM data can be by a database key or any other method of identifying a specific OSM or non-OSM element that may be used with a database join."

So actually, I just need to create a collective database, put the non-free data in one table and OSM data in another.
For table joining, I'm using ST_Intersects() and I'm fine?

Confusing.


From chris_hormann at gmx.de  Sat Dec 14 19:31:09 2019
From: chris_hormann at gmx.de (Christoph Hormann)
Date: Sat, 14 Dec 2019 20:31:09 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
Message-ID: <201912142031.09237.chris_hormann@gmx.de>

On Saturday 14 December 2019, matthias.straetling at buerotiger.de wrote:
> > existing OSMF community guidelines suggest spatial operations like
> > ST_Difference() and ST_Intersection() yield Derivative Databases
> > that are subject to share-alike.
>
> Let's take the Collective Database Guideline, you've mentioned:
> https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Coll
>ective_Database_Guideline_Guideline
>
> "Technically a reference between non-OSM and OSM data can be by a
> database key or any other method of identifying a specific OSM or
> non-OSM element that may be used with a database join."
>
> So actually, I just need to create a collective database, put the
> non-free data in one table and OSM data in another. For table
> joining, I'm using ST_Intersects() and I'm fine?

No, the quoted guideline says that share-alike does not apply if OSM
data and non-OSM data *do not* reference each other and in specific
other cases.  None of these cases covers references through spatial
relationships.

The idea that your process of intersecting non-OSM data with OSM based
admin polygons results in a collective database is not realistic.  To
me this kind of operation would be a textbook example of something
generating a derivative database - you combine OSM data with non-OSM
data to generate something of additional value compared to either of
these data sets alone.  This is exactly the kind of scenario
share-alike is meant for and why it was chosen as license for OSM.  But
there are of course fairly strong economic interests for this not being
subject to share-alike so people think of ways to interpret the ODbL
accordingly.

--
Christoph Hormann
http://www.imagico.de/


From matthias.straetling at buerotiger.de  Mon Dec 16 02:18:46 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Mon, 16 Dec 2019 03:18:46 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <201912142031.09237.chris_hormann@gmx.de>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
Message-ID: <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>

> Von: "Christoph Hormann" <chris_hormann at gmx.de>
>
> The idea that your process of intersecting non-OSM data with OSM based
> admin polygons results in a collective database is not realistic.  To
> me this kind of operation would be a textbook example of something
> generating a derivative database - you combine OSM data with non-OSM
> data to generate something of additional value compared to either of
> these data sets alone.  This is exactly the kind of scenario
> share-alike is meant for and why it was chosen as license for OSM.  But
> there are of course fairly strong economic interests for this not being
> subject to share-alike so people think of ways to interpret the ODbL
> accordingly.

Okay, I'll canceld all plans to use OpenStreetMap for this task.
I've contacted several commercial data providers and hope to get offers tomorrow.

I didn't expected OpenStreetMap to be such non-free and permissive :-(


From dieterdreist at gmail.com  Mon Dec 16 09:23:59 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Mon, 16 Dec 2019 10:23:59 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
Message-ID: <CABPTjTBRa9YU0-YWdkRSq-dfqZ58GwG1Gp5hU1g2eELvu-qbVQ@mail.gmail.com>

Am Mo., 16. Dez. 2019 um 03:19 Uhr schrieb <
matthias.straetling at buerotiger.de>:

> Okay, I'll canceld all plans to use OpenStreetMap for this task.
> I've contacted several commercial data providers and hope to get offers
> tomorrow.
>
> I didn't expected OpenStreetMap to be such non-free and permissive :-(



It depends on your definition of "free". It is free and open in a viral
way, i.e. it also frees the data with which you combine it, at least this
is the conceptual idea behind it (share alike).

Cheers
Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/8c9c7a13/attachment.html>

From nunocapelocaldeira at gmail.com  Mon Dec 16 09:34:05 2019
From: nunocapelocaldeira at gmail.com (Nuno Caldeira)
Date: Mon, 16 Dec 2019 09:34:05 +0000
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
Message-ID: <CAK5cRspp6i2TWnH4YPDOWSrTBz4PLu6wLaUjsfaUsGCVJsQTZw@mail.gmail.com>

that's unfair, it is free, you don't have to pay for it. it just has a
license, or else map companies would use our data

On Mon, 16 Dec 2019, 02:19 , <matthias.straetling at buerotiger.de> wrote:

>
> I didn't expected OpenStreetMap to be such non-free and permissive :-(
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/12e32760/attachment.html>

From chris_hormann at gmx.de  Mon Dec 16 11:03:26 2019
From: chris_hormann at gmx.de (Christoph Hormann)
Date: Mon, 16 Dec 2019 12:03:26 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
Message-ID: <201912161203.26948.chris_hormann@gmx.de>

On Monday 16 December 2019, matthias.straetling at buerotiger.de wrote:
>
> Okay, I'll canceld all plans to use OpenStreetMap for this task.
> I've contacted several commercial data providers and hope to get
> offers tomorrow.

In general (not necessarily specifically in your case - i don't know
enough about it to make that assessment) i think this is a good
approach if you have troubles with the share-alike provisions of the
ODbL.  If you want or need to keep a proprietary data set proprietary
it is natural that you have limitations in using it together with open
data with a viral license.

This is definitely a better approach than trying to find loopholes in
the license with brute force and wishful thinking.  Even if that is
possible and you can present an interpretation of the wording of the
ODbL that supports your use case without share-alike this was clearly
not the intention of the OSM community when adopting the ODbL to do so.

You need to be aware of course that the big corporate data users will
keep looking for loopholes - real or imagined - to achieve a
competitive advantage.  Like in the tale of the frog and the scorpion:
It is in their nature.  So if you respect the spirit of share-alike in
the ODbL you will always be potentially at a competitive disadvantages
to the corporate data users who simply don't give a damn.

The even better approach is of course to adopt the spirit of open data
and use OSM data together with other data sources embracing
share-alike.  Unfortunately so far the OSMF has not provided much
guidance on how to correctly do that, i.e. how to share share-alike
data sets practically.  The LWG unfortunately currently focuses on
guidance on how to avoid share-alike and attribution as much as
possible.

> I didn't expected OpenStreetMap to be such non-free and permissive
> :-(

The usual view is that share-alike provisions do not make something
non-free or non-open because they are meant to protect and extend the
freedom and only constrain users of truly non-free data.  But anyone
can have a different opinion on that of course.

Both share-alike and attribution play an important role in OSM in the
social contract between mappers and data users.  In return for being
able to use the results of the work of the mappers for free, data users
are required to share improvements of the data or the results of
producing something of additional value in combination with other data
under open license terms.

--
Christoph Hormann
http://www.imagico.de/


From simon at poole.ch  Mon Dec 16 12:33:24 2019
From: simon at poole.ch (Simon Poole)
Date: Mon, 16 Dec 2019 13:33:24 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
Message-ID: <f17f51a2-457b-f89b-7341-4428b6f7f798@poole.ch>

Just to be clear: you asked a question on an unmoderated, publicly
accessible mailing list on which everybody can voice their opinions
however unfounded they are or not, and now you are unhappy with that you
got a cacophony of conflicting opinions, which is exactly what you
should have expected.

The official guidance on geo-coding from the OSMF can be found here
https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline


Simon

Am 16.12.2019 um 03:18 schrieb matthias.straetling at buerotiger.de:
>> Von: "Christoph Hormann" <chris_hormann at gmx.de>
>>
>> The idea that your process of intersecting non-OSM data with OSM based
>> admin polygons results in a collective database is not realistic.  To
>> me this kind of operation would be a textbook example of something
>> generating a derivative database - you combine OSM data with non-OSM
>> data to generate something of additional value compared to either of
>> these data sets alone.  This is exactly the kind of scenario
>> share-alike is meant for and why it was chosen as license for OSM.  But
>> there are of course fairly strong economic interests for this not being
>> subject to share-alike so people think of ways to interpret the ODbL
>> accordingly.
> Okay, I'll canceld all plans to use OpenStreetMap for this task.
> I've contacted several commercial data providers and hope to get offers tomorrow.
>
> I didn't expected OpenStreetMap to be such non-free and permissive :-(
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/f5513ee9/attachment.sig>

From matthias.straetling at buerotiger.de  Mon Dec 16 15:02:02 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Mon, 16 Dec 2019 16:02:02 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAK5cRspp6i2TWnH4YPDOWSrTBz4PLu6wLaUjsfaUsGCVJsQTZw@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <CAK5cRspp6i2TWnH4YPDOWSrTBz4PLu6wLaUjsfaUsGCVJsQTZw@mail.gmail.com>
Message-ID: <trinity-39d3509a-e3ef-4e27-aeae-55770e840129-1576508522319@3c-app-gmx-bs57>

An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/ec576b46/attachment.html>

From matthias.straetling at buerotiger.de  Mon Dec 16 15:09:04 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Mon, 16 Dec 2019 16:09:04 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <f17f51a2-457b-f89b-7341-4428b6f7f798@poole.ch>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <f17f51a2-457b-f89b-7341-4428b6f7f798@poole.ch>
Message-ID: <trinity-e4b8f6a5-9c3b-499f-bfee-52ab84361e96-1576508944594@3c-app-gmx-bs57>

Simon,

> Gesendet: Montag, 16. Dezember 2019 um 13:33 Uhr
> Von: "Simon Poole" <simon at poole.ch>
>
> Just to be clear: you asked a question on an unmoderated, publicly
> accessible mailing list on which everybody can voice their opinions
> however unfounded they are or not, and now you are unhappy with that you
> got a cacophony of conflicting opinions, which is exactly what you
> should have expected.

I was aware of this and just wanted to get a consensus by the data creators: the users.
Looking at the opinion of the mass it shows me, that my approach of using OSM as a source of selecting the data doesn't seem to be fine. More of you are saying "share-alike", so I have to deal with this.

Like I sad before:
I would have been fine at attributing OpenStreetMap as selection. Now, I neither can use OSM data, nor add my dataset to yours.
A lose-lose-situation :-(

> The official guidance on geo-coding from the OSMF can be found here
> https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline

I think that is what Kathleen tried to explain, but got confused by others.
I'm sad that paying a lawyer is more expensive than paying for other datasets :-)

Regards,
Matthias


From matthias.straetling at buerotiger.de  Mon Dec 16 15:13:49 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Mon, 16 Dec 2019 16:13:49 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <201912161203.26948.chris_hormann@gmx.de>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <201912161203.26948.chris_hormann@gmx.de>
Message-ID: <trinity-a4035bc7-d1a5-4c7a-83c3-7d1e7a90879f-1576509229553@3c-app-gmx-bs57>

Christoph.

> Gesendet: Montag, 16. Dezember 2019 um 12:03 Uhr
> Von: "Christoph Hormann" <chris_hormann at gmx.de>
>
> This is definitely a better approach than trying to find loopholes in
> the license with brute force and wishful thinking.  Even if that is
> possible and you can present an interpretation of the wording of the
> ODbL that supports your use case without share-alike this was clearly
> not the intention of the OSM community when adopting the ODbL to do so.

It never was my intention to brute force a hole. I just thought, OSM data can be used, as long I don't mix anything or fill my missing data. I thought, proper attribution like "selected by using OSM data ..." would be fine for your.

> > I didn't expected OpenStreetMap to be such non-free and permissive
> > :-(
>
> The usual view is that share-alike provisions do not make something
> non-free or non-open because they are meant to protect and extend the
> freedom and only constrain users of truly non-free data.  But anyone
> can have a different opinion on that of course.

Sorry to say this, but I don't feel like you want to protect your data. It feels like you want to grab all the data, your data comes into contact with. "Viral" is the right term here - do you know the Borg? :-)

> Both share-alike and attribution play an important role in OSM in the
> social contract between mappers and data users.  In return for being
> able to use the results of the work of the mappers for free, data users
> are required to share improvements of the data or the results of
> producing something of additional value in combination with other data
> under open license terms.

If attribution would pay a role, than "(c) Non-Free data, selected by using OSM data ..." would be possible.
That might be an idea for future license drafts.

Regards,
Matthias


From matthias.straetling at buerotiger.de  Mon Dec 16 15:25:08 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Mon, 16 Dec 2019 16:25:08 +0100
Subject: [OSM-legal-talk] distance calculations
Message-ID: <trinity-9b5221df-1876-4e64-8f2e-f0d6f05b7820-1576509908642@3c-app-gmx-bs57>

Dear IANALs,

I'm sorry to ask an additional question.

A while ago, I've listened to a talk about navigation of pupils from their home to the school - it was used to decide whether the pupil gets a free bus ticket or not.

The distance calculation was done by a land registry office, which didn't have a route-able road & path network, but had trust in the OSM data, since they inspected it for quite a while. For completeness, they've used their own housing and school locations, but didn't use any from OSM.

They routed from the pupil's house to the school on the OSM network. Of course, the results were released in public.
Do such distance calculations also trigger share-alike on the non-free data (here: schools & houses).

Regards,
Matthias



From mikel.maron at gmail.com  Mon Dec 16 15:25:13 2019
From: mikel.maron at gmail.com (Mikel Maron)
Date: Mon, 16 Dec 2019 15:25:13 +0000 (UTC)
Subject: [OSM-legal-talk] update to mailing list description (was use OSM
 data to select proprietary data)
References: <344389541.9286419.1576509913257.ref@mail.yahoo.com>
Message-ID: <344389541.9286419.1576509913257@mail.yahoo.com>

On Monday, December 16, 2019, 07:35:08 AM EST, Simon Poole <simon at poole.ch> wrote:

> Just to be clear: you asked a question on an unmoderated, publicly accessible mailing list on which everybody can voice their opinions however unfounded they are or not, and now you are unhappy with that you got a cacophony of conflicting opinions, which is exactly what you should have expected.

Looks like this was clear to the poster, but I don't think it would necessarily be clear to a random person joining the list.

Suggest we modify the list description on https://lists.openstreetmap.org/listinfo/legal-talk to 

"The list for discussion of all legal matters relating to Openstreetmap, including licensing and copyright. For official information on the license from the OSM Foundation, see https://wiki.osmfoundation.org/wiki/Licence"

* Mikel Maron * +14152835207 @mikel s:mikelmaron


From dieterdreist at gmail.com  Mon Dec 16 16:01:10 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Mon, 16 Dec 2019 17:01:10 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-39d3509a-e3ef-4e27-aeae-55770e840129-1576508522319@3c-app-gmx-bs57>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <CAK5cRspp6i2TWnH4YPDOWSrTBz4PLu6wLaUjsfaUsGCVJsQTZw@mail.gmail.com>
 <trinity-39d3509a-e3ef-4e27-aeae-55770e840129-1576508522319@3c-app-gmx-bs57>
Message-ID: <CABPTjTBBfmTgRHh4q_UMwTLyQnQ=wgWVNnNrkXeVtHJZNNd_1g@mail.gmail.com>

Am Mo., 16. Dez. 2019 um 16:03 Uhr schrieb <
matthias.straetling at buerotiger.de>:

> Now, I neither can use your data, nor add my dataset to yours. A
> lose-lose-situation :-(
>


the problem is that "your dataset" is not yours, otherwise you could add
it, and you could also decide whether to use OSM in combination or not. But
you don't have the right to re-publish / re-license the dataset you have
acquired because they only sold you the right to use it.

It is kind of unfortunate, because OSM as far as I am informed, wouldn't be
interested in the specific dataset (of real estate prices) anyway.

Cheers
Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/e5edadbe/attachment-0001.html>

From chris_hormann at gmx.de  Mon Dec 16 16:02:08 2019
From: chris_hormann at gmx.de (Christoph Hormann)
Date: Mon, 16 Dec 2019 17:02:08 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-a4035bc7-d1a5-4c7a-83c3-7d1e7a90879f-1576509229553@3c-app-gmx-bs57>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912161203.26948.chris_hormann@gmx.de>
 <trinity-a4035bc7-d1a5-4c7a-83c3-7d1e7a90879f-1576509229553@3c-app-gmx-bs57>
Message-ID: <201912161702.08161.chris_hormann@gmx.de>

On Monday 16 December 2019, matthias.straetling at buerotiger.de wrote:
> >
> > The usual view is that share-alike provisions do not make something
> > non-free or non-open because they are meant to protect and extend
> > the freedom and only constrain users of truly non-free data.  But
> > anyone can have a different opinion on that of course.
>
> Sorry to say this, but I don't feel like you want to protect your
> data. It feels like you want to grab all the data, your data comes
> into contact with. "Viral" is the right term here - do you know the
> Borg? :-)

There is a long history of discussion about the benefits of
viral/share-alike licenses in the open data/free software movement.  In
OSM we have had this discussion extensively before the license change.
I tried to provide a bit of insight about why we have share-alike but
people here in general are fairly reluctant to reiterate that
discussion because it rarely brings any new insights.

Apart from the mentioned importance of share-alike for the social
contract between mappers and data users it is also doubtful that OSM
would still exist as a single homogeneous project as we know it today
if in 2012 we would have chosen a non-share-alike license.  It is very
likely that OSM would have split off several proprietary forks with
which corporate data users would have tried to distinguish themselves
from the competition by creating improved versions of the OSM database
adding proprietary data without feeding it back into the openly
licensed public database.

Please keep in mind that the image of a viral license is partly
misleading because everyone has the free choice to not use the data and
not 'be infected' while a biological virus does not typically give you
that freedom.

> > Both share-alike and attribution play an important role in OSM in
> > the social contract between mappers and data users.  In return for
> > being able to use the results of the work of the mappers for free,
> > data users are required to share improvements of the data or the
> > results of producing something of additional value in combination
> > with other data under open license terms.
>
> If attribution would pay a role, than "(c) Non-Free data, selected by
> using OSM data ..." would be possible. That might be an idea for
> future license drafts.

The viewpoint communicated by Kathleen would mean data sets partly
derived from OSM through spatial operations without containing
substantial amounts of the original data in original form (that is
essentially the case we are talking about here in abstract form) would
require neither share-alike nor attribution since they are neither a
Derivative Database, a Collective Database nor a Produced Work.

So while your willingness to attribute is admirable this kind of
attribution for mixed and processed data without share-alike is not
something that the ODbL considers a separate scenario.

--
Christoph Hormann
http://www.imagico.de/


From tlee at mapbox.com  Mon Dec 16 16:03:19 2019
From: tlee at mapbox.com (Tom Lee)
Date: Mon, 16 Dec 2019 11:03:19 -0500
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-e4b8f6a5-9c3b-499f-bfee-52ab84361e96-1576508944594@3c-app-gmx-bs57>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <f17f51a2-457b-f89b-7341-4428b6f7f798@poole.ch>
 <trinity-e4b8f6a5-9c3b-499f-bfee-52ab84361e96-1576508944594@3c-app-gmx-bs57>
Message-ID: <CAEUyxDTqjA1vqMX0VQtOXuYQsdcQzR8Yf=pTx_+x4MYU6UCCbw@mail.gmail.com>

> I was aware of this and just wanted to get a consensus by the data
creators: the users.

This is an admirable impulse, but it is worth emphasizing that those of us
who participate on OSM listservs are a small and unrepresentative fraction
of the project's 5.9 million registered users. Lists like this one are a
great way to find the slice of users who are most interested and passionate
about a particular issue, and who consequently can be expected to have
well-informed (and often strongly held) opinions that reflect the gamut of
possible answers.

But if you are seeking consensus, the closest thing available is the text
of the license itself and guidelines that have been approved by elected
members of the OSMF board. Usually when there is broad agreement on an
issue, the answer is memorialized in a wiki page that people find before
they wind up here :-)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/1e9f62a7/attachment.html>

From nunocapelocaldeira at gmail.com  Mon Dec 16 16:22:15 2019
From: nunocapelocaldeira at gmail.com (Nuno Caldeira)
Date: Mon, 16 Dec 2019 16:22:15 +0000
Subject: [OSM-legal-talk] distance calculations
In-Reply-To: <trinity-9b5221df-1876-4e64-8f2e-f0d6f05b7820-1576509908642@3c-app-gmx-bs57>
References: <trinity-9b5221df-1876-4e64-8f2e-f0d6f05b7820-1576509908642@3c-app-gmx-bs57>
Message-ID: <CAK5cRsohqJsiO_LrTvFoHqmsgkfH+Pvo=_uqVg+6xaXn88uztA@mail.gmail.com>

it's a derivated, therefore share alike. I'm glad they trusted OSM data.

On Mon, 16 Dec 2019, 15:25 , <matthias.straetling at buerotiger.de> wrote:

> Dear IANALs,
>
> I'm sorry to ask an additional question.
>
> A while ago, I've listened to a talk about navigation of pupils from their
> home to the school - it was used to decide whether the pupil gets a free
> bus ticket or not.
>
> The distance calculation was done by a land registry office, which didn't
> have a route-able road & path network, but had trust in the OSM data, since
> they inspected it for quite a while. For completeness, they've used their
> own housing and school locations, but didn't use any from OSM.
>
> They routed from the pupil's house to the school on the OSM network. Of
> course, the results were released in public.
> Do such distance calculations also trigger share-alike on the non-free
> data (here: schools & houses).
>
> Regards,
> Matthias
>
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/66aa934e/attachment.html>

From matthias.straetling at buerotiger.de  Mon Dec 16 18:20:58 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Mon, 16 Dec 2019 19:20:58 +0100
Subject: [OSM-legal-talk] distance calculations
In-Reply-To: <CAK5cRsohqJsiO_LrTvFoHqmsgkfH+Pvo=_uqVg+6xaXn88uztA@mail.gmail.com>
References: <trinity-9b5221df-1876-4e64-8f2e-f0d6f05b7820-1576509908642@3c-app-gmx-bs57>
 <CAK5cRsohqJsiO_LrTvFoHqmsgkfH+Pvo=_uqVg+6xaXn88uztA@mail.gmail.com>
Message-ID: <trinity-700f0cf2-fe53-440b-8265-5e9f87f05129-1576520458056@3c-app-gmx-bap33>

> Von: "Nuno Caldeira" <nunocapelocaldeira at gmail.com>
> it's a derivated, therefore share alike. I'm glad they trusted OSM data. 

So the distance calculations are derivated, of course.
But what about their points of interests? They've interacted with the roads.


From matthias.straetling at buerotiger.de  Mon Dec 16 18:26:58 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Mon, 16 Dec 2019 19:26:58 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAEUyxDTqjA1vqMX0VQtOXuYQsdcQzR8Yf=pTx_+x4MYU6UCCbw@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <f17f51a2-457b-f89b-7341-4428b6f7f798@poole.ch>
 <trinity-e4b8f6a5-9c3b-499f-bfee-52ab84361e96-1576508944594@3c-app-gmx-bs57>
 <CAEUyxDTqjA1vqMX0VQtOXuYQsdcQzR8Yf=pTx_+x4MYU6UCCbw@mail.gmail.com>
Message-ID: <trinity-721caedd-272e-405c-8918-6b00beedd67f-1576520818767@3c-app-gmx-bap33>

> Gesendet: Montag, 16. Dezember 2019 um 17:03 Uhr
> Von: "Tom Lee via legal-talk" <legal-talk at openstreetmap.org>
> 
> This is an admirable impulse, but it is worth emphasizing that those of
> us who participate on OSM listservs are a small and unrepresentative
> fraction of the project's 5.9 million registered users. Lists like this
> one are a great way to find the slice of users who are most interested
> and passionate about a particular issue, and who consequently can be
> expected to have well-informed (and often strongly held) opinions that
> reflect the gamut of possible answers.

I understand this, but the girls and guys here do already have some
knowledge about this topic. I know many OSM mappers, which would never
be able to discuss about this license questions. And many don't even
use a GIS to be able to intersect two different data sources :-)
  
> But if you are seeking consensus, the closest thing available is the
> text of the license itself and guidelines that have been approved by
> elected members of the OSMF board. Usually when there is broad agreement
> on an issue, the answer is memorialized in a wiki page that people find
> before they wind up here :-)_

Why doesn't the OSMF write about fundamental stuff then? I think,
ST_Intersects() is one of the main tools in GIS world. Why don't
give a clear statement on this?

Since the ODbL has never changed, it's fixed. So there could be
something like an FAQ or matrix to look up what triggers share-alike
and what not?


From kathleen.lu at mapbox.com  Mon Dec 16 21:07:49 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Mon, 16 Dec 2019 13:07:49 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-721caedd-272e-405c-8918-6b00beedd67f-1576520818767@3c-app-gmx-bap33>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <f17f51a2-457b-f89b-7341-4428b6f7f798@poole.ch>
 <trinity-e4b8f6a5-9c3b-499f-bfee-52ab84361e96-1576508944594@3c-app-gmx-bs57>
 <CAEUyxDTqjA1vqMX0VQtOXuYQsdcQzR8Yf=pTx_+x4MYU6UCCbw@mail.gmail.com>
 <trinity-721caedd-272e-405c-8918-6b00beedd67f-1576520818767@3c-app-gmx-bap33>
Message-ID: <CAFekdnCjtduKdqOQW8+2TZ1CiCDViTHbm25fR1XgMv-HfeJeWA@mail.gmail.com>

That's what the guidelines are for!
We can't cover every possible example because there are too many, but as I
already said, I think your usecase is covered by the Geocoding Guideline.
https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline#The_Guideline


> Why doesn't the OSMF write about fundamental stuff then? I think,
> ST_Intersects() is one of the main tools in GIS world. Why don't
> give a clear statement on this?
>
> Since the ODbL has never changed, it's fixed. So there could be
> something like an FAQ or matrix to look up what triggers share-alike
> and what not?
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/d0f5a63c/attachment-0001.html>

From kathleen.lu at mapbox.com  Mon Dec 16 21:13:40 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Mon, 16 Dec 2019 13:13:40 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CABPTjTBBfmTgRHh4q_UMwTLyQnQ=wgWVNnNrkXeVtHJZNNd_1g@mail.gmail.com>
References: <trinity-d4b26a22-15b8-4a35-b720-4b6e02e7051a-1576133968406@3c-app-gmx-bs45>
 <201912132053.32630.chris_hormann@gmx.de>
 <trinity-73efafc6-4e6e-4a5b-af69-b90770aa5a66-1576349626034@3c-app-gmx-bap62>
 <201912142031.09237.chris_hormann@gmx.de>
 <trinity-360b15cf-0cce-4cce-91c0-546dd6c936d0-1576462726676@3c-app-gmx-bs77>
 <CAK5cRspp6i2TWnH4YPDOWSrTBz4PLu6wLaUjsfaUsGCVJsQTZw@mail.gmail.com>
 <trinity-39d3509a-e3ef-4e27-aeae-55770e840129-1576508522319@3c-app-gmx-bs57>
 <CABPTjTBBfmTgRHh4q_UMwTLyQnQ=wgWVNnNrkXeVtHJZNNd_1g@mail.gmail.com>
Message-ID: <CAFekdnAWH67rcV+ZavE70=wuV_nZuRx-EOPVV9jaz-hRxW4t0g@mail.gmail.com>

> It is kind of unfortunate, because OSM as far as I am informed, wouldn't
> be interested in the specific dataset (of real estate prices) anyway.
>
> If it's not the type of data that OSM would be interested in, then why
doesn't it fall under the Collective Database Guideline?
the non-OSM data adds a particular type of geometry or data for a primary
feature that was not already present within a regional cut, and the added
feature data includes no OSM data;
Wasn't a major reason for that guideline to permit nonsharealike usecases
where the data potentially subject to sharealike would not be useful to OSM
anyway?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/624616bc/attachment.html>

From simon at poole.ch  Mon Dec 16 21:36:29 2019
From: simon at poole.ch (Simon Poole)
Date: Mon, 16 Dec 2019 22:36:29 +0100
Subject: [OSM-legal-talk] distance calculations
In-Reply-To: <CAK5cRsohqJsiO_LrTvFoHqmsgkfH+Pvo=_uqVg+6xaXn88uztA@mail.gmail.com>
References: <trinity-9b5221df-1876-4e64-8f2e-f0d6f05b7820-1576509908642@3c-app-gmx-bs57>
 <CAK5cRsohqJsiO_LrTvFoHqmsgkfH+Pvo=_uqVg+6xaXn88uztA@mail.gmail.com>
Message-ID: <c593d319-5589-97fb-44c2-a85125e0be5c@poole.ch>


Am 16.12.2019 um 17:22 schrieb Nuno Caldeira:
> it's a derivated, therefore share alike. I'm glad they trusted OSM data.

I'm not sure what you are referring to here.

Yes the distances are a Produced Work which, if publicly used, implies
that if a Derivative Database was used to produce the distances that
should be made available, but likely this was unmodified OSM data so no
need to actually do that.

The centroids of the locations of the pupils houses and the school are
clearly a separate database and only serve as inputs to the routing
algorithm, per ODbL 4.5.b only the OSM database  component of a
Collective Database used to generate a produced work needs to be
licensed on ODbL terms.

It would seem that this is actually a school book example of when SA
does -not- apply.

Simon

PS: from the description it is not even clear if the output was publicly
used to start with.

>
> On Mon, 16 Dec 2019, 15:25 , <matthias.straetling at buerotiger.de
> <mailto:matthias.straetling at buerotiger.de>> wrote:
>
>     Dear IANALs,
>
>     I'm sorry to ask an additional question.
>
>     A while ago, I've listened to a talk about navigation of pupils
>     from their home to the school - it was used to decide whether the
>     pupil gets a free bus ticket or not.
>
>     The distance calculation was done by a land registry office, which
>     didn't have a route-able road & path network, but had trust in the
>     OSM data, since they inspected it for quite a while. For
>     completeness, they've used their own housing and school locations,
>     but didn't use any from OSM.
>
>     They routed from the pupil's house to the school on the OSM
>     network. Of course, the results were released in public.
>     Do such distance calculations also trigger share-alike on the
>     non-free data (here: schools & houses).
>
>     Regards,
>     Matthias
>
>
>     _______________________________________________
>     legal-talk mailing list
>     legal-talk at openstreetmap.org <mailto:legal-talk at openstreetmap.org>
>     https://lists.openstreetmap.org/listinfo/legal-talk
>
>
> _______________________________________________
> legal-talk mailing list
> legal-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/a718ac3d/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/a718ac3d/attachment.sig>

From dieterdreist at gmail.com  Mon Dec 16 22:48:29 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Mon, 16 Dec 2019 23:48:29 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnCjtduKdqOQW8+2TZ1CiCDViTHbm25fR1XgMv-HfeJeWA@mail.gmail.com>
References: <CAFekdnCjtduKdqOQW8+2TZ1CiCDViTHbm25fR1XgMv-HfeJeWA@mail.gmail.com>
Message-ID: <33985C1C-EA18-4521-8F1A-97B0FCF8779F@gmail.com>



sent from a phone

>> On 16. Dec 2019, at 22:09, Kathleen Lu via legal-talk <legal-talk at openstreetmap.org> wrote:
> That's what the guidelines are for! 
> We can't cover every possible example because there are too many, but as I already said, I think your usecase is covered by the Geocoding Guideline. https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline#The_Guideline


I also believe it is covered by this guideline, but it seems his use would trigger share alike according to this guideline:

> 2. the Geocoding Results are not used to create a new database that contains the whole or a substantial part of the original OSM database


because he wants to create a new database.

Cheers Martin 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/dd2d806d/attachment.html>

From kathleen.lu at mapbox.com  Mon Dec 16 23:04:37 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Mon, 16 Dec 2019 15:04:37 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <33985C1C-EA18-4521-8F1A-97B0FCF8779F@gmail.com>
References: <CAFekdnCjtduKdqOQW8+2TZ1CiCDViTHbm25fR1XgMv-HfeJeWA@mail.gmail.com>
 <33985C1C-EA18-4521-8F1A-97B0FCF8779F@gmail.com>
Message-ID: <CAFekdnBJ+=8CV4v5HdTDYrJnEcW_kZdO1urKfK3J_ECYVGkwyA@mail.gmail.com>

But what that says is not just "create a new database" but one "that
contains the whole or a substantial part of the original OSM database." His
new database will contain very little if any of the original OSM database.

On Mon, Dec 16, 2019 at 2:48 PM Martin Koppenhoefer <dieterdreist at gmail.com>
wrote:

>
>
> sent from a phone
>
> On 16. Dec 2019, at 22:09, Kathleen Lu via legal-talk <
> legal-talk at openstreetmap.org> wrote:
>
> That's what the guidelines are for!
> We can't cover every possible example because there are too many, but as I
> already said, I think your usecase is covered by the Geocoding Guideline.
> https://wiki.osmfoundation.org/wiki/Licence/Community_Guidelines/Geocoding_-_Guideline#The_Guideline
>
>
>
> I also believe it is covered by this guideline, but it seems his use would
> trigger share alike according to this guideline:
>
> 2. the Geocoding Results are not used to create a new database that
> contains the whole or a substantial part of the original OSM database
>
>
>
> because he wants to create a new database.
>
> Cheers Martin
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/ea9f0afe/attachment.html>

From dieterdreist at gmail.com  Tue Dec 17 00:00:10 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Tue, 17 Dec 2019 01:00:10 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnBJ+=8CV4v5HdTDYrJnEcW_kZdO1urKfK3J_ECYVGkwyA@mail.gmail.com>
References: <CAFekdnBJ+=8CV4v5HdTDYrJnEcW_kZdO1urKfK3J_ECYVGkwyA@mail.gmail.com>
Message-ID: <A28DEEDA-0E61-4B05-B960-F5082FA4BB0A@gmail.com>



sent from a phone

> On 17. Dec 2019, at 00:04, Kathleen Lu <kathleen.lu at mapbox.com> wrote:
> 
> But what that says is not just "create a new database" but one "that contains the whole or a substantial part of the original OSM database." His new database will contain very little if any of the original OSM database


it will contain a lot of postcode information from the original OpenStreetMap database, in adapted/translated form. Whether the amount is sufficient to be considered substantial will have to be evaluated based on the actual db that is created/the actual numbers.
To create an accurate postcode polygon from point features you will need a lot of them, so probably already a handful of them would be considered substantial.

Cheers Martin 

From matthias.straetling at buerotiger.de  Tue Dec 17 00:10:34 2019
From: matthias.straetling at buerotiger.de (matthias.straetling at buerotiger.de)
Date: Tue, 17 Dec 2019 01:10:34 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <A28DEEDA-0E61-4B05-B960-F5082FA4BB0A@gmail.com>
References: <CAFekdnBJ+=8CV4v5HdTDYrJnEcW_kZdO1urKfK3J_ECYVGkwyA@mail.gmail.com>
 <A28DEEDA-0E61-4B05-B960-F5082FA4BB0A@gmail.com>
Message-ID: <trinity-2dc1fcde-34e2-4aa2-95c6-4c5c973992ae-1576541434042@3c-app-gmx-bs07>

> Gesendet: Dienstag, 17. Dezember 2019 um 01:00 Uhr
> Von: "Martin Koppenhoefer" <dieterdreist at gmail.com>
>
> it will contain a lot of postcode information from the original OpenStreetMap database,
> in adapted/translated form. Whether the amount is sufficient to be considered substantial
> will have to be evaluated based on the actual db that is created/the actual numbers.
> To create an accurate postcode polygon from point features you will need a lot of them,
> so probably already a handful of them would be considered substantial.

There are 5,650,789,072 nodes in OSM database. But the EU database directive wants to
protect the investment (in money). If it was damn hard to collect the nodes belonging
to the postcodes, only a few thousand nodes might be more substantial.

I think, that's a moralistic point of view. I'll neither collect a substantial part
of the whole OSM database, nor you could proof that there was big investment made to
collect the data. Since the users are working for free, the only investment are the
servers.

Like I said, that's a moralistic point of view.

I've got an offer today to get the data for about 3,500 Euro. This allows me to select
the data and even publish the postal code and the merged postal geometries with attribution.
It's another non-free dataset, but it solves my problem.



From kathleen.lu at mapbox.com  Tue Dec 17 00:35:18 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Mon, 16 Dec 2019 16:35:18 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <A28DEEDA-0E61-4B05-B960-F5082FA4BB0A@gmail.com>
References: <CAFekdnBJ+=8CV4v5HdTDYrJnEcW_kZdO1urKfK3J_ECYVGkwyA@mail.gmail.com>
 <A28DEEDA-0E61-4B05-B960-F5082FA4BB0A@gmail.com>
Message-ID: <CAFekdnD-1m5-Oaf1c=x81sJfED9gwvLKJbtUFpiJcf+KqRtaew@mail.gmail.com>

it will contain a lot of postcode information from the original
> OpenStreetMap database, in adapted/translated form.


This doesn't seem correct to me. In the final set, each point will only
tell you yes/no whether it was in a particular postcode. That's not very
much info at all.

>
> To create an accurate postcode polygon from point features you will need a
> lot of them, so probably already a handful of them would be considered
> substantial.
>

 This logic seems backwards. Since it would require a lot of point features
in order to recreate the polygon (and thus something that looks similar to
the original OSM database), it should require a *lot* of points to be
considered substantial.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191216/17c04a9d/attachment.html>

From dieterdreist at gmail.com  Tue Dec 17 00:35:38 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Tue, 17 Dec 2019 01:35:38 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <trinity-2dc1fcde-34e2-4aa2-95c6-4c5c973992ae-1576541434042@3c-app-gmx-bs07>
References: <trinity-2dc1fcde-34e2-4aa2-95c6-4c5c973992ae-1576541434042@3c-app-gmx-bs07>
Message-ID: <9CD55FBC-A7D8-4301-8D52-E49CEEF3938F@gmail.com>



sent from a phone

> On 17. Dec 2019, at 01:11, matthias.straetling at buerotiger.de wrote:
> 
> I think, that's a moralistic point of view. I'll neither collect a substantial part
> of the whole OSM database, nor you could proof that there was big investment made to
> collect the data. Since the users are working for free, the only investment are the
> servers.


you believe the mappers are working for “free” because they do not get paid? One can see their contributions as donations, they are donating their time and knowledge, and the value is what it would cost if they were paid according to the work they are doing. It is out of question that an immense investment had to be made for OpenStreetMap to come to the point where it is now, in survey time, data input, software development, and infrastructure.

“substantial” does not mean it has to be a certain percentage of the whole db, you can see this from the substantial guideline, which has fixed limits that are not growing with the db. “substantial” means it’s more than one or two features (OpenStreetMap-Foundation has declared they see a total of 100 features as substantial, although it is not completely clear what a feature is, for example you could go to an extreme point of view and see the whole border of Germany as a single feature (I am not) while a more credible interpretation would see every border point as a feature, so that the border of Germany would be thousands of features). 


Cheers Martin 

From dieterdreist at gmail.com  Tue Dec 17 00:39:13 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Tue, 17 Dec 2019 01:39:13 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnD-1m5-Oaf1c=x81sJfED9gwvLKJbtUFpiJcf+KqRtaew@mail.gmail.com>
References: <CAFekdnD-1m5-Oaf1c=x81sJfED9gwvLKJbtUFpiJcf+KqRtaew@mail.gmail.com>
Message-ID: <75FE856F-1EED-47DF-B38D-FE1AC09544AE@gmail.com>



sent from a phone

> On 17. Dec 2019, at 01:35, Kathleen Lu <kathleen.lu at mapbox.com> wrote:
> 
>  
>> 
>> To create an accurate postcode polygon from point features you will need a lot of them, so probably already a handful of them would be considered substantial.
> 
>  This logic seems backwards. Since it would require a lot of point features in order to recreate the polygon (and thus something that looks similar to the original OSM database), it should require a *lot* of points to be considered substantial.


it _took_ a lot of address points to create the aggregate postcode polygon. The Germans did not survey the polygon, they surveyed addresses

Cheers Martin 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191217/bdee108f/attachment.html>

From kathleen.lu at mapbox.com  Thu Dec 19 23:15:20 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Thu, 19 Dec 2019 15:15:20 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <9CD55FBC-A7D8-4301-8D52-E49CEEF3938F@gmail.com>
References: <trinity-2dc1fcde-34e2-4aa2-95c6-4c5c973992ae-1576541434042@3c-app-gmx-bs07>
 <9CD55FBC-A7D8-4301-8D52-E49CEEF3938F@gmail.com>
Message-ID: <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>

>
> “substantial” does not mean it has to be a certain percentage of the whole
> db, you can see this from the substantial guideline, which has fixed limits
> that are not growing with the db. “substantial” means it’s more than one or
> two features (OpenStreetMap-Foundation has declared they see a total of 100
> features as substantial, although it is not completely clear what a feature
> is, for example you could go to an extreme point of view and see the whole
> border of Germany as a single feature (I am not) while a more credible
> interpretation would see every border point as a feature, so that the
> border of Germany would be thousands of features).
>
> This is not what the Substantial Guideline says. It says that fewer than
100 features is "not Substantial". It also gives as an example "More that
100 Features only if the extraction is non-systematic and clearly based on
your own qualitative criteria for example an extract of all the the
locations of restaurants you have visited for a personal map to share with
friends or use the locations of a selection of historic buildings as an
adjunct in a book you are writing, we would regard that as non
Substantial." BTW, a list of flats/houses for sale in the current time
period is not too far off from these examples.
And that is for extracting the entirety of each feature. The Geocoding
Guideline states: "Furthermore, if only names are provided in Geocoding
Results from OSM -- in particular, latitude/longitude information from OSM
is not included in the Geocoding Results -- a collection of such results is
not a substantial extract." There's no 100 feature limit at all for
geocoding.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191219/95f4f1cb/attachment.html>

From dieterdreist at gmail.com  Fri Dec 20 00:13:19 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Fri, 20 Dec 2019 01:13:19 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
References: <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
Message-ID: <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com>



sent from a phone

> On 20. Dec 2019, at 00:16, Kathleen Lu via legal-talk <legal-talk at openstreetmap.org> wrote:
> 
> This is not what the Substantial Guideline says. It says that fewer than 100 features is "not Substantial". It also gives as an example "More that 100 Features only if the extraction is non-systematic and clearly based on your own qualitative criteria for example an extract of all the the locations of restaurants you have visited for a personal map to share with friends or use the locations of a selection of historic buildings as an adjunct in a book you are writing, we would regard that as non Substantial."


If I recall correctly there is no definition what a feature is. Nobody has yet commented how they would interpret this for a border: is it about the border way or about the individual border points from which the border is made of? 

100 features may be a reasonable limit for point features, but for complex ways and relations, already very few (maybe even a single one) may be substantial?

Kathleen, I would be interested in your thoughts what a feature is in the context of ways.

Cheers Martin 

From dieterdreist at gmail.com  Fri Dec 20 00:21:17 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Fri, 20 Dec 2019 01:21:17 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
References: <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
Message-ID: <DA774FF5-0C41-4ECF-9A3D-C19958363D01@gmail.com>



sent from a phone

> On 20. Dec 2019, at 00:16, Kathleen Lu via legal-talk <legal-talk at openstreetmap.org> wrote:
> 
> And that is for extracting the entirety of each feature. The Geocoding Guideline states: "Furthermore, if only names are provided in Geocoding Results from OSM -- in particular, latitude/longitude information from OSM is not included in the Geocoding Results -- a collection of such results is not a substantial extract."


it should be noted that in the case at hand the intention was to create a new database with translated/adapted OpenStreetMap data. The guideline says: 
“ Since individual Geocoding Results are insubstantial extracts, they may be stored and used together with other proprietary or third party data without having a share-alike impact on such other data, provided the Geocoding Results have not been aggregated to create a new database that contains the whole or a substantial part of the OSM database.
If Geocoding Results are used to create a new database that contains the whole or a substantial part of the contents of the OSM database, this new database would be considered a Derivative Database and would trigger share-alike obligations under section 4.4.b of the ODbL. ”



the guideline is about individual results, not about aggregations, for which the share alike provisions persist. From my interpretation this also implies that the attribution requirements persist for individual results, because otherwise it would not be clear that you cannot aggregate them. Do you agree?



Cheers Martin 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191220/cfdcf8a0/attachment.html>

From matkoniecz at tutanota.com  Fri Dec 20 07:02:25 2019
From: matkoniecz at tutanota.com (Mateusz Konieczny)
Date: Fri, 20 Dec 2019 08:02:25 +0100 (CET)
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com>
References: <CAEUyxDTqjA1vqMX0VQtOXuYQsdcQzR8Yf=pTx_+x4MYU6UCCbw@mail.gmail.com>
 <trinity-721caedd-272e-405c-8918-6b00beedd67f-1576520818767@3c-app-gmx-bap33>
 <CAFekdnCjtduKdqOQW8+2TZ1CiCDViTHbm25fR1XgMv-HfeJeWA@mail.gmail.com>
 <33985C1C-EA18-4521-8F1A-97B0FCF8779F@gmail.com>
 <CAFekdnBJ+=8CV4v5HdTDYrJnEcW_kZdO1urKfK3J_ECYVGkwyA@mail.gmail.com>
 <A28DEEDA-0E61-4B05-B960-F5082FA4BB0A@gmail.com>
 <trinity-2dc1fcde-34e2-4aa2-95c6-4c5c973992ae-1576541434042@3c-app-gmx-bs07>
 <9CD55FBC-A7D8-4301-8D52-E49CEEF3938F@gmail.com>
 <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
 <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com>
Message-ID: <LwX6KOn--3-1@tutanota.com>




20 Dec 2019, 01:13 by dieterdreist at gmail.com:

>
>
> sent from a phone
>
>> On 20. Dec 2019, at 00:16, Kathleen Lu via legal-talk <legal-talk at openstreetmap.org> wrote:
>>
>> This is not what the Substantial Guideline says. It says that fewer than 100 features is "not Substantial". It also gives as an example "More that 100 Features only if the extraction is non-systematic and clearly based on your own qualitative criteria for example an extract of all the the locations of restaurants you have visited for a personal map to share with friends or use the locations of a selection of historic buildings as an adjunct in a book you are writing, we would regard that as non Substantial."
>>
>
>
> If I recall correctly there is no definition what a feature is. Nobody has yet commented how they would interpret this for a border: is it about the border way or about the individual border points from which the border is made of? 
>
Obviously, both nodes, ways and
relations should be counted.

Otherwise one would be able to
temporarily create one relation,
that would include all data (s)he
wish to use and export this.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191220/f9d024a4/attachment.html>

From dieterdreist at gmail.com  Fri Dec 20 09:35:36 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Fri, 20 Dec 2019 10:35:36 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <LwX6KOn--3-1@tutanota.com>
References: <LwX6KOn--3-1@tutanota.com>
Message-ID: <4235C572-88EA-4F55-A6B1-1F7A4506CA72@gmail.com>



sent from a phone

> On 20. Dec 2019, at 08:04, Mateusz Konieczny <matkoniecz at tutanota.com> wrote:
> 
> Obviously, both nodes, ways and
> relations should be counted.
> 
> Otherwise one would be able to
> temporarily create one relation,
> that would include all data (s)he
> wish to use and export this.


and if you count both, one could split all ways in 2-node segments so that the same data becomes lots of features. ;-)
I agree that (significant) nodes are relevant and just counting ways does not fit with the concept. Significant nodes could be those that define an angle of more than 1-2 degrees compared to the previously counted node or that define an intersection  (i.e. additional, otherwise unconnected nodes in a straight way would not count).

Cheers Martin 

From chris_hormann at gmx.de  Fri Dec 20 10:34:12 2019
From: chris_hormann at gmx.de (Christoph Hormann)
Date: Fri, 20 Dec 2019 11:34:12 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <LwX6KOn--3-1@tutanota.com>
References: <CAEUyxDTqjA1vqMX0VQtOXuYQsdcQzR8Yf=pTx_+x4MYU6UCCbw@mail.gmail.com>
 <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com> <LwX6KOn--3-1@tutanota.com>
Message-ID: <201912201134.12333.chris_hormann@gmx.de>

On Friday 20 December 2019, Mateusz Konieczny wrote:
>
> Obviously, both nodes, ways and
> relations should be counted.
>
> Otherwise one would be able to
> temporarily create one relation,
> that would include all data (s)he
> wish to use and export this.

The "100 Features" limit as a rule of thumb for substantiality was not
really well thought through.  IMO a data volume limit would make more
sense - which would probably make sense to position somewhere between
1kB (approximately equivalent to a hundred untagged nodes) and 5kB
(addresses, buildings etc.)  Talking about compact binary data
representation here of course, not raw OSM XML and no lossy
compression.

--
Christoph Hormann
http://www.imagico.de/


From michal.palenik at freemap.sk  Fri Dec 20 16:03:00 2019
From: michal.palenik at freemap.sk (Michal Palenik)
Date: Fri, 20 Dec 2019 17:03:00 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <201912201134.12333.chris_hormann@gmx.de>
References: <CAEUyxDTqjA1vqMX0VQtOXuYQsdcQzR8Yf=pTx_+x4MYU6UCCbw@mail.gmail.com>
 <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com>
 <LwX6KOn--3-1@tutanota.com>
 <201912201134.12333.chris_hormann@gmx.de>
Message-ID: <20191220160300.GA2154192@tanicka.iz.sk>

On Fri, Dec 20, 2019 at 11:34:12AM +0100, Christoph Hormann wrote:
> On Friday 20 December 2019, Mateusz Konieczny wrote:
> >
> > Obviously, both nodes, ways and
> > relations should be counted.
> >
> > Otherwise one would be able to
> > temporarily create one relation,
> > that would include all data (s)he
> > wish to use and export this.
> 
> The "100 Features" limit as a rule of thumb for substantiality was not
> really well thought through.  IMO a data volume limit would make more
> sense - which would probably make sense to position somewhere between
> 1kB (approximately equivalent to a hundred untagged nodes) and 5kB
> (addresses, buildings etc.)  Talking about compact binary data
> representation here of course, not raw OSM XML and no lossy
> compression.

my rule of thumb proposal (from a long time age) was 
"1 day work of a semi experienced mapper"

(which would take into account availability of aerial photos or other
sources to import)


m



-- 
michal palenik
www.freemap.sk
www.oma.sk


From list-osm-legal-talk at cyclestreets.net  Fri Dec 20 23:10:12 2019
From: list-osm-legal-talk at cyclestreets.net (Martin - CycleStreets)
Date: Fri, 20 Dec 2019 23:10:12 +0000 (GMT)
Subject: [OSM-legal-talk] Use of OSM as a background - derivative data
Message-ID: <alpine.DEB.2.20.1912201236420.9434@pling.qwghlm.org>



I'm e-mailing on behalf of London Cycling Campaign, LCC.

LCC is currently undertaking a project to test the viability of 
crowdsourcing updates to Transport for London’s Cycling Infrastructure 
Database. A website has been built that overlays the TfL CID data on top of 
a map base, currently OpenStreetMap. Volunteers are visiting locations to 
verify that the infrastructure has been correctly entered and still exists. 
Where new infrastructure has been newly built or altered they are entereing 
the location, asset details and taking photographs.

We are looking to confirm that our use of OpenStreetMap in this way does 
not trigger share-alike provisions.  While TfL is making its data freely 
available (and we hope that it will be used to enhance OSM), TfL is doing 
this under its own licence terms. As such we would be grateful for an 
opinion from those here or ultimately from Licence Working Group.

Looking at section 3g of the Legal FAQ we would argue our data is 
independent – we are not taking OSM data, altering and republishing it. We 
are using OSM as a locational reference and physically visiting locations 
to verify the data in our dataset – not simply copying or tracing OSM.  We 
may notice something on OSM – e.g. some cycle parking – that we don’t have 
in our database but again, we are visiting the location to check and to 
take photographs.


https://opendatacommons.org/licenses/odbl/1.0/
https://wiki.openstreetmap.org/wiki/Legal_FAQ

3g. Can I use OSM data and OpenStreetMap-derived maps to verify my own data 
without triggering share-alike?
Yes, provided that you are only comparing and do not copy any OpenStreetMap 
data. If you make any changes to your data after making the comparison, you 
should be able to reasonably demonstrate that any such change was made 
either from your own physical observation or comes from a non-OpenStreetMap 
source accessed directly by you. I.e you can compare but not take!
- Example 1: You notice that a street is called one name on your map and 
another in OpenStreetMap 
[1]<https://docs.google.com/document/d/1PI83Tx48iQn7qFJgEUAG8U7lUb1jMzu8VS4F4famseQ/pub>. 
You should visit the street and check the name, then you are free to put 
that name in your data as it is your own observation.
- Example 2: You notice that a boundary is different in your data and 
OpenStreetMap. You should check back to original authoritative sources and 
make any correction required.


(Note that much of the data being maintained is not of relevance to OSM - 
it includes things like signage pole locations, painted symbols, which 
traditionally OSM has not covered. There is a separate project related to 
conflation of the relevants parts of the data into OSM:
https://wiki.openstreetmap.org/wiki/TfL_Cycling_Infrastructure_Database
but in the meanwhile it is considered important to keep the current dataset 
maintained.)




Martin,                     **  CycleStreets - For Cyclists, By Cyclists
Developer, CycleStreets     **  https://www.cyclestreets.net/

From kathleen.lu at mapbox.com  Sat Dec 21 02:01:37 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Fri, 20 Dec 2019 18:01:37 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <DA774FF5-0C41-4ECF-9A3D-C19958363D01@gmail.com>
References: <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
 <DA774FF5-0C41-4ECF-9A3D-C19958363D01@gmail.com>
Message-ID: <CAFekdnA3CktGXgyGj0+7fLVh+dKM_0-Qyoe9gXRWS2XpF2japw@mail.gmail.com>

the guideline is about individual results, not about aggregations, for
> which the share alike provisions persist. From my interpretation this also
> implies that the attribution requirements persist for individual results,
> because otherwise it would not be clear that you cannot aggregate them. Do
> you agree?
>

No, the guideline was explicitly about both individual results and
aggregations. Individual results are insubstantial, so no ODbL obligations
attach at all (attribution is one of the ODbL obligations). A collection of
results is not a Derivative Database *unless* the collection is used as a
general geodatabase (basically, if you tried to reverse engineer OSM by
mass geocoding). That means that in normal circumstances, for a collection
of results (an aggregation), the sharealike provisions *do not* persist.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191220/967b07c3/attachment.html>

From kathleen.lu at mapbox.com  Sat Dec 21 02:22:03 2019
From: kathleen.lu at mapbox.com (Kathleen Lu)
Date: Fri, 20 Dec 2019 18:22:03 -0800
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com>
References: <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
 <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com>
Message-ID: <CAFekdnD=Zc2eW+hfAHcuxoYb6m8eod+ur+q+goH2o0rsDOChwg@mail.gmail.com>

This is an interesting question. I'm not sure what exactly one feature is.
But I would find it very hard to claim that a single way, even a complex
one, was "substantial" by itself. Remember that it's a "substantial
part...of the contents of a database" (in this case OSM), and one way would
be a very very small part of OSM.
But I think this is why the guideline defined mostly "insubstantial"
instead of substantial, because defining substantial is much more
difficult, context dependent, and would depend on case law (of which there
is very little)

On Thu, Dec 19, 2019 at 4:13 PM Martin Koppenhoefer <dieterdreist at gmail.com>
wrote:

>
>
> sent from a phone
>
> > On 20. Dec 2019, at 00:16, Kathleen Lu via legal-talk <
> legal-talk at openstreetmap.org> wrote:
> >
> > This is not what the Substantial Guideline says. It says that fewer than
> 100 features is "not Substantial". It also gives as an example "More that
> 100 Features only if the extraction is non-systematic and clearly based on
> your own qualitative criteria for example an extract of all the the
> locations of restaurants you have visited for a personal map to share with
> friends or use the locations of a selection of historic buildings as an
> adjunct in a book you are writing, we would regard that as non Substantial."
>
>
> If I recall correctly there is no definition what a feature is. Nobody has
> yet commented how they would interpret this for a border: is it about the
> border way or about the individual border points from which the border is
> made of?
>
> 100 features may be a reasonable limit for point features, but for complex
> ways and relations, already very few (maybe even a single one) may be
> substantial?
>
> Kathleen, I would be interested in your thoughts what a feature is in the
> context of ways.
>
> Cheers Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191220/60e9fc34/attachment.html>

From matkoniecz at tutanota.com  Sat Dec 21 10:19:16 2019
From: matkoniecz at tutanota.com (Mateusz Konieczny)
Date: Sat, 21 Dec 2019 11:19:16 +0100 (CET)
Subject: [OSM-legal-talk] Use of OSM as a background - derivative data
In-Reply-To: <alpine.DEB.2.20.1912201236420.9434@pling.qwghlm.org>
References: <alpine.DEB.2.20.1912201236420.9434@pling.qwghlm.org>
Message-ID: <LwbxyUg--3-2@tutanota.com>

21 Dec 2019, 00:10 by list-osm-legal-talk at cyclestreets.net:

> A website has been built that overlays the TfL CID data on top of a map base, currently OpenStreetMap. Volunteers are visiting locations to verify that the infrastructure has been correctly entered and still exists. Where new infrastructure has been newly built or altered they are entereing the location, asset details and taking photographs.
>
> We are looking to confirm that our use of OpenStreetMap in this way does not trigger share-alike provisions.
>
Not a lawyer, but AFAIK this is perfectly fine.

I used copyrighted sources to plan
my trips/mapping in OSM on exactly the same way,
sometimes with addition of an automatic
data comparison.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191221/e8d03949/attachment.html>

From matkoniecz at tutanota.com  Sat Dec 21 20:14:42 2019
From: matkoniecz at tutanota.com (Mateusz Konieczny)
Date: Sat, 21 Dec 2019 21:14:42 +0100 (CET)
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <20191220160300.GA2154192@tanicka.iz.sk>
References: <33985C1C-EA18-4521-8F1A-97B0FCF8779F@gmail.com>
 <CAFekdnBJ+=8CV4v5HdTDYrJnEcW_kZdO1urKfK3J_ECYVGkwyA@mail.gmail.com>
 <A28DEEDA-0E61-4B05-B960-F5082FA4BB0A@gmail.com>
 <trinity-2dc1fcde-34e2-4aa2-95c6-4c5c973992ae-1576541434042@3c-app-gmx-bs07>
 <9CD55FBC-A7D8-4301-8D52-E49CEEF3938F@gmail.com>
 <CAFekdnDyn-LfBBMeNPCOxBz7Q=E8Hgpeyq2O0nkWE94Fczcb4w@mail.gmail.com>
 <A4CFDC16-52C6-4A9B-A7B9-C87CC88D6484@gmail.com> <LwX6KOn--3-1@tutanota.com>
 <201912201134.12333.chris_hormann@gmx.de>
 <20191220160300.GA2154192@tanicka.iz.sk>
Message-ID: <Lwe5Fgu--3-2@tutanota.com>




20 Dec 2019, 17:03 by michal.palenik at freemap.sk:
> my rule of thumb proposal (from a long time age) was 
> "1 day work of a semi experienced mapper"
>
> (which would take into account availability of aerial photos or other
> sources to import)
>
As in "output of mapping for entire day"
or "average daily output of mapper'?

First one would be unreasonably large
(far more than LWG proposed rule),
second would be hard to compute.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20191221/fd2c23b8/attachment.html>

From dieterdreist at gmail.com  Sat Dec 21 23:43:48 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Sun, 22 Dec 2019 00:43:48 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnA3CktGXgyGj0+7fLVh+dKM_0-Qyoe9gXRWS2XpF2japw@mail.gmail.com>
References: <CAFekdnA3CktGXgyGj0+7fLVh+dKM_0-Qyoe9gXRWS2XpF2japw@mail.gmail.com>
Message-ID: <44283734-11C5-4BC5-8552-D6C3F09EE7E4@gmail.com>



sent from a phone

> On 21. Dec 2019, at 03:01, Kathleen Lu <kathleen.lu at mapbox.com> wrote:
> 
> No, the guideline was explicitly about both individual results and aggregations. Individual results are insubstantial, so no ODbL obligations attach at all (attribution is one of the ODbL obligations).


while an individual result is probably insubstantial, a collection of individual results is not, so even if the individual result would not require attribution, recollecting them will, so somehow these strings must be attached to make sure the recipients are aware that collecting them would trigger share alike and attribution requirements.


> A collection of results is not a Derivative Database *unless* the collection is used as a general geodatabase


this is also something I’ve always struggled to understand, what is a “general” geodatabase? If I take all power network related information from OpenStreetMap, is this a general database? Or a list of all streets in France and the municipality they are in? Or all gas stations in the US with their postcode (but no location)?



> basically, if you tried to reverse engineer OSM by mass geocoding). That means that in normal circumstances, for a collection of results (an aggregation), the sharealike provisions *do not* persist.  


Isn’t the guideline going a big beyond what ODbL permits? I could reverse engineer a list of postcodes for streets from geocoding results, something that would create  a derived db under the ODbL (adaptation), and according to this reading of the guideline the result could be any license I want?

Cheers Martin 



From dieterdreist at gmail.com  Sat Dec 21 23:54:01 2019
From: dieterdreist at gmail.com (Martin Koppenhoefer)
Date: Sun, 22 Dec 2019 00:54:01 +0100
Subject: [OSM-legal-talk] use OSM data to select proprietary data
In-Reply-To: <CAFekdnD=Zc2eW+hfAHcuxoYb6m8eod+ur+q+goH2o0rsDOChwg@mail.gmail.com>
References: <CAFekdnD=Zc2eW+hfAHcuxoYb6m8eod+ur+q+goH2o0rsDOChwg@mail.gmail.com>
Message-ID: <8BF39720-EFE3-499D-BB2B-8D81F353513D@gmail.com>



sent from a phone

> On 21. Dec 2019, at 03:22, Kathleen Lu <kathleen.lu at mapbox.com> wrote:
> 
> Remember that it's a "substantial part...of the contents of a database" (in this case OSM), and one way would be a very very small part of OSM.


If “substantial“ has to be seen in relation to the size of the database it would imply that the more data we collect the more one could take without it being protected. It’s hard to believe that this is the intention of the wording, but if it was confirmed we could protect by splitting the planet db in several parts, eg. by tenth of degrees, and distribute it as collection of databases rather than a single one.

A single feature would be more of an exception (on average), but when you look at  relations rather than ways they can become easily substantial

Cheers Martin