Re: [OSM-legal-talk] attributive data enrichment using OSM

2016-07-25 Thread Stefan Jäger
Thanks you Simon for your views.

Unfortunately disclosing any of the data is not an option, because it's clearly 
not allowed, be it a centroid or the polygon.
Actually I was using  this discussion list because I want to openly discuss 
such problems.

However, I have the impression it is rather difficult to really build a 
business case based on OSM data.
Being in a gray area doesn't feel comfortable. When in doubt, leave it out.
I am in doubt and I will leave it out.

In my case, the OSM data would be a nice add on for which I would also be 
willing to contribute something, but the contribution of data is not an option
The data would however not be a very big improvement, and the product can very 
well exist without it.
But it is , as I said, no option at all to disclose any other of the data to 
the public.

Stefan




> -Ursprüngliche Nachricht-
> Von: Simon Poole [mailto:si...@poole.ch]
> Gesendet: Montag, 25. Juli 2016 16:22
> An: legal-talk@openstreetmap.org
> Betreff: Re: [OSM-legal-talk] attributive data enrichment using OSM
> 
> IMHO it is rather unclear
> 
> a) who is contracting the work you are supposed to do
> 
> b) who is going to receive the product(s ?) in the end
> 
> and yes you may not be able to disclose details in public, which is why I'm
> always very sceptical about having such discussions on this list.
> 
> In the end it depends on what kind of operations you want to do with the
> result, for example you could generate your building type data as described,
> however not apply it to the building outline, but to the building centroid. 
> The
> resulting dataset would clearly be a Derivative Database and subject to share
> alike, however it might be that making that available would not be so
> problematic for the original data owners.
> 
> Simon
> 
> 
> Am 25.07.2016 um 12:46 schrieb Stefan Jäger:
> > OK,
> > I am afraid that this renders impossible using OSM data for my use case.
> >
> > I appreciate the comments from everybody.
> >
> > Stefan
> >
> >> -Ursprüngliche Nachricht-
> >> Von: Christoph Hormann [mailto:chris_horm...@gmx.de]
> >> Gesendet: Montag, 25. Juli 2016 09:43
> >> An: Licensing and other legal discussions.
> >> <legal-talk@openstreetmap.org>
> >> Betreff: Re: [OSM-legal-talk] attributive data enrichment using OSM
> >>
> >> On Monday 25 July 2016, Stefan Jäger wrote:
> >>> "Viral aspects" like Martin puts it, sound a bit like a disease :-)
> >> The connotation of the term 'viral' has changed quite a bit.  Viral
> >> marketing for example has - at least for those initiating it - a quite 
> >> positive
> appeal.
> >>
> >>> However, If  I  understand Ilya correctly, the Odbl only applies to
> >>> data that is made publicly available. That, in turn, means, any data
> >>> put together using also OSM data, that is not publicly accessible,
> >>> but only  in internal networks, can be produced?
> >> The ODbL applies to all OSM data everywhere - however the share-alike
> >> requirements of the ODbL only apply if the data is publicly 'conveyed'.
> >> Keep in mind though that this includes any produced works (like map
> >> renderings or other illustrations created using the data) - if a
> >> derivative database is used in producing such illustrations this
> >> database is subject to the share-alike requirements.
> >>
> >> And note if you (as indicated in your initial inquiry) intend to
> >> license this derivative database or a produced work to a third party
> >> that is already a public process under the ODbL.
> >>
> >> --
> >> Christoph Hormann
> >> http://www.imagico.de/
> >>
> >> ___
> >> legal-talk mailing list
> >> legal-talk@openstreetmap.org
> >> https://lists.openstreetmap.org/listinfo/legal-talk
> > ___
> > legal-talk mailing list
> > legal-talk@openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/legal-talk
> 

___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] attributive data enrichment using OSM

2016-07-25 Thread Stefan Jäger
OK,
I am afraid that this renders impossible using OSM data for my use case.

I appreciate the comments from everybody.

Stefan

> -Ursprüngliche Nachricht-
> Von: Christoph Hormann [mailto:chris_horm...@gmx.de]
> Gesendet: Montag, 25. Juli 2016 09:43
> An: Licensing and other legal discussions. <legal-talk@openstreetmap.org>
> Betreff: Re: [OSM-legal-talk] attributive data enrichment using OSM
> 
> On Monday 25 July 2016, Stefan Jäger wrote:
> >
> > "Viral aspects" like Martin puts it, sound a bit like a disease :-)
> 
> The connotation of the term 'viral' has changed quite a bit.  Viral marketing
> for example has - at least for those initiating it - a quite positive appeal.
> 
> > However, If  I  understand Ilya correctly, the Odbl only applies to
> > data that is made publicly available. That, in turn, means, any data
> > put together using also OSM data, that is not publicly accessible, but
> > only  in internal networks, can be produced?
> 
> The ODbL applies to all OSM data everywhere - however the share-alike
> requirements of the ODbL only apply if the data is publicly 'conveyed'.
> Keep in mind though that this includes any produced works (like map
> renderings or other illustrations created using the data) - if a derivative
> database is used in producing such illustrations this database is subject to 
> the
> share-alike requirements.
> 
> And note if you (as indicated in your initial inquiry) intend to license this
> derivative database or a produced work to a third party that is already a
> public process under the ODbL.
> 
> --
> Christoph Hormann
> http://www.imagico.de/
> 
> ___
> legal-talk mailing list
> legal-talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk
___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] attributive data enrichment using OSM

2016-07-25 Thread Christoph Hormann
On Monday 25 July 2016, Stefan Jäger wrote:
>
> "Viral aspects" like Martin puts it, sound a bit like a disease :-)

The connotation of the term 'viral' has changed quite a bit.  Viral 
marketing for example has - at least for those initiating it - a quite 
positive appeal.

> However, If  I  understand Ilya correctly, the Odbl only applies to
> data that is made publicly available. That, in turn, means, any data
> put together using also OSM data, that is not publicly accessible,
> but only  in internal networks, can be produced?

The ODbL applies to all OSM data everywhere - however the share-alike 
requirements of the ODbL only apply if the data is publicly 'conveyed'.  
Keep in mind though that this includes any produced works (like map 
renderings or other illustrations created using the data) - if a 
derivative database is used in producing such illustrations this 
database is subject to the share-alike requirements.

And note if you (as indicated in your initial inquiry) intend to license 
this derivative database or a produced work to a third party that is 
already a public process under the ODbL.

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

___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] attributive data enrichment using OSM

2016-07-25 Thread Frederik Ramm
Hi,

On 07/25/2016 09:05 AM, Stefan Jäger wrote:
> That, in turn, means, any data put together using also OSM data, that is not 
> publicly accessible, but only  in internal networks, can be produced?

In the terms of ODbL, "publicly" means (quote)

"to Persons other than You or under Your control by
either more than 50% ownership or by the power to direct their
activities (such as contracting with an independent consultant)."

Bye
Frederik

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

___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] attributive data enrichment using OSM

2016-07-22 Thread Ilya Zverev
Hi Stefan, and thanks for writing to this mailing list.

Your case is not much different from geocoding, when you borrow some attributes 
(addresses, or in your case, POI or landuse tags) from OSM and put it into a 
proprietary database. That would clearly make a derived database out of your 
proprietary one, so you would have to provide it under an open license.

Now, if you are not giving any access to the resulting database outside a 
private network, that may be considered not using it publicly, in which case 
the whole license does not apply. Section 4.2 starts with "If You Publicly 
Convey this Database, any Derivative Database, or the Database as part of a 
Collective Database, then You must", so this depends on how non-public your 
system is. I hope somebody else here can elaborate on this.

IZ

> 22 июля 2016 г., в 5:46, Stefan Jäger  написал(а):
> 
> Dear all from legal-talk,
>  
> We have the idea of using OSM data for an enrichment process in order to 
> improve the attributive information of proprietary building footprints.
>  
> Let me briefly explain, what the purpose of using OSM for our enrichment 
> process is.
>  
> We are developing a data product for building footprints  that is based on 
> 3D-building data from the official German surveying authorities (there 16 of 
> them) .
> The license of these data is not open, neither for the original data nor the 
> data product we plan to produce.
>  
> These original data are rather heterogeneous with respect to quality , 
> quality not in the sense of geometrical correctness or completeness but in 
> the sense of building function attributes (residential, official, hospital, 
> etc.).
> Our Idea is to use OSM data in a quality improvement process for the building 
> types, in combination with other processes like analyzing the shape and/or 
> size of buildings.
>  
> An example:
> Suppose we  have a building footprint from our data which has no information 
> on the building type.
> We now take the centroid of that building and analyze the underlying osm area 
> information , which is, let’s say an industrial complex.
> We would then assign the official code, let’s say ‘4711 (industrial)’ to that 
> building.
>  
> Another example:
> Again, suppose we  have a building footprint of our data which has no 
> information on the building type.
> We would than take OSM amenity (point) information and check whether there 
> are many shops (points) inside that building polygon.
> We would then designate the building a shopping mall (provided a certain size 
> criterion derived from the original data source is met) and assign it the 
> code 1234 for shopping mall.
>  
> My question now is: if we enrich our data (with only underlying attributes, 
> no geometry from OSM at all) with such a process using OSM data, is this then 
> a produced work (or a collective database) or would I have to license my 
> enriched data product under the odbl, which I would not be allowed to do, 
> because it would conflict with the license of the main data source?
>  
> The final data product itself is not intended for public use (e.g. on 
> publicly accessible websites) or display, which is forbidden by the original 
> license in the first place.
> It will not have any osm information or reference to osm  entries in the 
> final data product.
>  
> I have read the text here:
> https://wiki.openstreetmap.org/wiki/Open_Data_License/Horizontal_Layers_-_Guideline
> in particular the last paragraph “Combining OSM data with proprietary data?”
> …
> But what happens in case a legal entity wants to combine OSM data with 
> third-party data? Let's assume the third-party data is proprietary, e.g. a 
> list of restaurants that was bought by the legal person with the right to 
> publicly use it, but of course not publicly release it.
> ….
>  
> Unfortunately this question has not been answered yet.
>  
> Of course I am more than happy with acknowledging OSM as an additional data 
> enrichment source.
>  
> I would also be happy to support the OSM project.
>  
> Most importantly, I need to be on the safe side.
>  
> Thanks for feedback!
>  
> Stefan Jäger
> 
>  
> ___
> legal-talk mailing list
> legal-talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/legal-talk


___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] attributive data enrichment using OSM

2016-07-22 Thread Martin Koppenhoefer


sent from a phone

> Il giorno 22 lug 2016, alle ore 14:46, Stefan Jäger  ha 
> scritto:
> 
> My question now is: if we enrich our data (with only underlying attributes, 
> no geometry from OSM at all) with such a process using OSM data, is this then 
> a produced work (or a collective database) or would I have to license my 
> enriched data product under the odbl,


I think it's the latter (derivative database). You are mixing up databases and 
they are not independent, so the viral aspect of ODbL comes into play 

cheers,
Martin 
___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk


Re: [OSM-legal-talk] attributive data enrichment using OSM

2016-07-22 Thread Stefan Jäger
Thanks fort he quick answer,

Actually I did consult the mentioned guideline , however, I could not really 
relate it 100% to my use case.
That's what I hope for some clarification here.

Stefan

___
legal-talk mailing list
legal-talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/legal-talk