Re: [OSM-legal-talk] use OSM data to select proprietary data

[Robert Whittaker (OSM lists)]

On Thu, 12 Dec 2019 22:41 Kathleen Lu via legal-talk, <
legal-talk@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.

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

Re: [OSM-legal-talk] Copy information from official business website (WAS: Proposal for a revision of JA:Available Data)

[Robert Whittaker (OSM lists)]

On Fri, 5 Jul 2019 at 21:18, tomoya muramoto  wrote:
> I ask a simple question. May I copy the information of the TESCO Boston 
> Superstore to OSM?
> https://www.tesco.com/store-locator/uk/?bid=2108
>
> This website contains information such as
> - addr=*
> - phone=*
> - opening_hours=*
> - branch=*
>
> The OSM data for this store does not yet contain `phone=*` nor 
> `opening_hours=*`. Can I copy them to OSM?
> https://www.openstreetmap.org/way/61998754

For a single store I believe the answer is yes, since you're
extracting un-copyrightable facts. But if there are a significant
number of stores (as in this case), then the information becomes part
of a database, which is by default protected by database rights (at
least in the EU). You then can't use a significant amount of the
information without an appropriate licence. Moreover, you can't safely
take details from a single store from a chain's website, as there's a
danger that lots of other mappers might do that independently for
different stores, resulting in an infringement for OSM as a whole.

So for practical, use in OSM, I'd say it's ok to take information from
the public website of an independent store, or a group with up to
around half a dozen outlets. But anything larger, and you'd need to
get permission from the company first.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] Licence compatibility: Open Data Licence for The Regional Municipality of Peel (Version 1.0)

[Robert Whittaker (OSM lists)]

On 9 September 2016 at 18:19, Luis Villa  wrote:
> Can you elaborate on the second point, Simon? Are you referring to the
> "third party rights the Information Provider is not authorised to license"
> language? If so, I'm afraid they've merely made explicit what is implicit in
> all licenses - if there is third party material in a work that the open
> licensor isn't authorized to license, then that material isn't licensed to
> you, regardless of what the license says.

Yes, but without that clause, the licensor is implicitly asserting
that they do have the necessary rights to offer the licence, and if
there did turn out to be problems later, the licensee would probably
have some come-back on them. With the clause, it effectively absolves
the licensor of any responsibility for making sure that the licence is
valid for the licencee's use.

There was a case in the UK where (IIRC) house price data was offered
under the UK Open Government Licence (OGL). It turned out later that
the addresses in it had been checked/normalised using a proprietary
address database, and the licensor wasn't able to re-licence them.
According to the OGL, it is still fine to offer the whole dataset
under that licence, and there doesn't seem to be any obligation on the
licensor to point out that some parts of the dataset cannot actually
be re-used under the stated licence.

Normally it's fine as the stuff being licensed is obviously the
licensor's data. But sometimes there are proprietary databases that
get mixed in, so you have to be a bit careful. (Addresses and any GIS
data are particular issues in the UK.) I briefly tried to convince the
UK OGL's authors of the problem with this clause (suggesting they
should add an obligation on the licensor to point out any
un-licensable parts) but I didn't manage to get very far.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] is legal-talk@openstreetmap.org searchable?

[Robert Whittaker (OSM lists)]

On 18 August 2016 at 21:12, Frederik Ramm  wrote:
> use your favourite search engine with something like
> "site:lists.openstreetmap.org legal-talk mykeyword".

Or better still "site:lists.openstreetmap.org/pipermail/legal-talk/
mykeyword", which works on at least Google.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] FYI Collective Database Guideline

[Robert Whittaker (OSM lists)]

On 9 June 2016 at 13:08, Christoph Hormann  wrote:
> On Thursday 09 June 2016, Simon Poole wrote:
>>
>> The LWG has just forwarded the text of
>> http://wiki.openstreetmap.org/wiki/Collective_Database_Guideline to
>> the OSMF board for approval and publishing as definite guidance from
>> the OSMF.
>
> IIRC it was already noted by others that the lack of an example where
> share-alike applies kind of makes the whole thing appear unbalanced and
> endangers meeting the purpose to clarify 'where the line is drawn'.
>
> Independent of the actual content adding a non-trivial counter-example
> would IMO significantly improve practical usefulness and understanding
> of the guideline.

+1

Also (and it may be deliberate) this guideline doesn't address the
question of what filtering / querying you can do with your collective
database. For instance, under the guideline I can take OSM restaurant
data, and add third-party ratings data to each entry, and it will be a
collective database. But what if I then do a query that returns the
locations of restaurants that have >4* ratings in a certain area and
just show those to users? Is this filtered dataset -- including the
ratings used to create it -- subject to share-alike, or is it still a
collective database of OSM restaurant names and locations, together
with independent ratings?

I wonder if we'd be better having a guideline that's based on rule
that any data used in a query with OSM data has to be shared. Data
that's only used in simple table joins does not. (As in the existing
guideline, it would be a question of whether you can achieve the same
results using such a method. Technical implementations that do things
differently for efficiency reasons don't count against you.)

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] new wiki page ODbL compatibility of common licenses

[Robert Whittaker (OSM lists)]

On 18 January 2016 at 10:53, Martin Koppenhoefer  wrote:
> Following a thread on the OSMF-talk list, I am kindly asking you to review
> and improve a new wiki page that tries to give an overview about the
> compatibility of common licenses with the ODbL and CT:
> http://wiki.openstreetmap.org/wiki/Import/ODbL_Compatibility

This is great to see, and will hopefully become a very useful resource.

Some comments / suggestions:

* In the notes column, it might be better to say "rights holder(s)"
rather than "licensor" since the former is presumably the only
person/body who is able to give such permission.

* For the CC-By notes, I think those giving the permission also need
to be aware that they are (or would need to be) also authorising
downstream use of their data, without necessarily getting any direct
attribution from those downstream uses. I'd suggest adding "including
to cover downstream use in works derived from OSM" to the end of the
note.

* It's not clear from the page whether or not the lack of green in the
"contributor terms" column precludes the use of ODbL data or not.
Presumably not, but more consideration should be taken before using
such data, and with documenting them and attributing it in OSM. If
this is correct, then something to this effect should be added in an
explanatory paragraph.

* It would be good to add the UK's Open Government Licence (OGL) and
Non-Commercial Government Licence (NCGL) to the list. The first should
be the same as the ODbL (as it explicitly states the ODbLs terms are
sufficient to fulfil the obligations under the licence) while the
second is incompatible due to the NC terms.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] NSW LPI permission

[Robert Whittaker (OSM lists)]

On 11 December 2015 at 21:04, Andrew Harvey  wrote:
> Talking with their legal people it was, or at least as far as I
> understood them, their view that the the ODbL style of attribution
> (where downstream don't need to provide attribution for any
> incorporated or derived datasets) is fine within the bounds of the CC
> BY 3.0 AU license already. They mentioned that the CC license has a
> concept of attribution reasonable or appropriate for the medium which
> covered this use case. They also mentioned that when their CC BY 3.0
> AU data is incorporated into a new work (which is more than just a
> trivial transformation) then there is no need for downstream
> attribution within the license.
>
> In a way they are merely letting us know of their interpretation of
> the license that it's terms already meet our requirements.

If their legal people are genuinely happy for the ODbL level of
attribution (particularly with respect to produced works), then it
would make everyone's life much easier if they were able to dual
licence the data under the ODbL in addition to CC-By. Then it's
completely clear to everyone that use under the ODbL is acceptable,
and there wouldn't be any need for lawyery discussions, special
permissions, or user hesitation for particular uses. I guess there may
be political and/or administrative barriers to prevent this, but it
might be worth asking them if you haven't done so already.

As far as use in OSM following the current correspondence goes, I
think the key question is, are they aware that other people could then
use the OSM data, and that with those uses they may only get indirect
attribution -- i.e. the user links back to OSM, which would in turn
link back to them -- even in cases where the data used was dominated
by their data? It would therefore probably be safer if you could get
an explicit statement that they're happy for the data to be used in
OSM, rather than just that they're happy for the attribution that OSM
provides for its own direct use.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] Addresses from Land Registry Price Paid Data

[Robert Whittaker (OSM lists)]

On 1 December 2014 at 21:51, Simon Poole si...@poole.ch wrote:
 Am 01.12.2014 15:08, schrieb Robert Whittaker (OSM lists):
 
 This also raises the question of whether there are any other
 OGL-licensed datasets out there that have been used in OSM, but which
 contain undocumented third-party IP rights that we don't have
 permission to use.
 

 This is, IMHO, not a problem specific to the OGL.

 In general I have yet to see any licence or agreement to include data in
 OSM, that actually states that the licensor has all the necessary rights
 to licence the data on the terms presented and holds the licensee (us)
 harmless for any damages arising out of not having those rights.

That's true, but in both the examples you've given it's a case of
other (non-copyright) rights that are not being licensed. At least an
alert user will be aware of these other rights, and would be able to
conduct their own checks (e.g. searching public patent databases,
checking for people in the images) without needing anything more from
the licensor. Presumably those licences *do* effectively guarantee
that you're ok with the licensed data as far as copyright is concerned
-- which is what the licence is there to license.

With the OGL problem I've flagged up, it's different, in that there's
seemingly no guarantee that the licence applies even to the copyright
in all the copyrightable data that you've been given. If my reading is
correct, then any random third-party could own copyrights in any of
the data, and it then wouldn't be licensed to you to re-use, and
there's no obligation on the licensor to tell you about this. But
there would be no way for a user to spot this, unless they happen to
suspect that some data could only have come from a third-party source
and make enquires of the licensor.

But the most immediate issue, I think, is do we need to do anything
about the Land Registry address use in OSM -- as the addresses are
apparently owned by a third-party, and so not covered by the OGL -- as
a result of this?

Robert.

-- 
Robert Whittaker

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

[OSM-legal-talk] Addresses from Land Registry Price Paid Data

[Robert Whittaker (OSM lists)]

As you may know, the UK's Land Registry makes available historical
Price Paid data for residential property sales, licensed under the
Open Government Licence (OGL). Along with the prices paid, this data
also includes full addresses and postcodes for the properties.

OGL-licensed data is regarded as suitable for use in OSM, and judging
by http://taginfo.openstreetmap.org.uk/search?q=land+registry#values
and 
http://wiki.openstreetmap.org/wiki/Contributors#Land_Registry_-_Price_paid_data
it would appear that people have been making use of the Land Registry
address data in order to help improve OSM's address coverage.

However, I've recently come across this blog post:
http://mapgubbins.tumblr.com/post/103854046790/how-far-can-we-trust-open-data
which points out a potential problem with this data, in its section on
Third Party Intellectual Property Rights.

Briefly, the OGL has an exclusion for third party rights the
Information Provider is not authorised to license, and it appears
that such rights may well exist in the addresses in the Price Paid
data. If this is the case, then the addresses aren't covered by the
OGL licence, and so we're probably unable to use them in OSM.

From 
https://docs.google.com/document/d/1hFQ1I_YfTGxcGiwydVxoMdsJAhDmZaXW30NoDeun4k0/edit
Open Addresses appear to have rejected the Land Registry data as a
source for this reason.

There is one ray of hope from the blog post above though, and that is
that s47 of the Copyright, Designs and Patents Act provides a limited
exemption for some information on public statutory registers. As far
as I can see, under s47(3) the purpose of the copying is limited to
dissemination (is all re-use a form of dissemination?), and you
need authorisation from the organisation originally publishing the
register (and I'm not sure their current OGL Licence would cover
this), and presumably the exemption only applies in the UK. So I'm not
sure whether this would work for us with ODbL.

I think LWG will need to look in to this in more detail, and let us
know whether or not it is ok to continue using the Land Registry's
Price Paid data as a source of addresses in OSM. If not, then do
existing uses also need to be removed from OSM?

This also raises the question of whether there are any other
OGL-licensed datasets out there that have been used in OSM, but which
contain undocumented third-party IP rights that we don't have
permission to use.

Robert.

-- 
Robert Whittaker

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

[OSM-legal-talk] Contents Licence for OSM Data

[Robert Whittaker (OSM lists)]

The ODbL that we now use for OSM data technically only applies to the
database, and not to individual contents contained within it. For
that, the ODbL says you need a separate licence [1]. I was under the
impression that for OSM's data this licence was the ODC's Database
Contents Licence (DbCL) [2].

It therefore surprised me when I read the White Paper at [3], which
said that uncertainty over the content licence was a problem for
downstream users.

When I went to check what the content licence was, I was unable to
find any definitive information where I would expect to find it; i.e.
at http://www.openstreetmap.org/copyright or
http://wiki.openstreetmap.org/wiki/Legal_FAQ or
http://wiki.osmfoundation.org/wiki/License . The latter two seem to
suggest that the OSM Contributor Terms [4] act as a content licence,
but I don't see how that's possible, since the Contributor Terms are
concerned with people giving rights and assurances to OSMF, rather
than OSMF providing rights to data users. The Contributor Terms
themselves mention the DbCL as one of the possible licences OSMF can
use, but don't actually say that OSMF are using it for current data
downloads.

So can I enquire as to exactly what the content licence is for OSM's
geodata, and suggest that it is made clearer on the pages linked
above?

I guess some people may argue that the individual data items in OSM
are facts and so aren't copyrightable anyway. However, it's not
obvious to me that this is necessarily the case for all the data items
(there are certainly some things in OSM that are subject to creative
judgement) and it would seem that uncertainty over the content licence
is a real issue for data users. Even if an explicit content licence
may not be necessary, it would surely be good to soecify one like the
DbCL anyway.

Thanks,

Robert.

[1] http://opendatacommons.org/licenses/odbl/
[2] http://opendatacommons.org/licenses/dbcl/1-0/
[3] 
http://spatiallaw.blogspot.co.uk/2014/10/the-odbl-and-openstreetmap-analysis-and.html
[4] http://wiki.osmfoundation.org/wiki/License/Contributor_Terms

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] YouTube videos

[Robert Whittaker (OSM lists)]

On 8 August 2014 09:48, Simon Poole si...@poole.ch wrote:
 CC-BY is not per se compatible. We need (and I believe this is still the
 case with 4.0) explicit acknowledgement that the way that we provide
 attribution is OK and that we do not provide downstream attribution for
 individual sources.

Getting a bit off-topic, but it seems quite a few people don't realise
this about CC-By (and also some other 'open' licences). I think it
would be a really good idea if there could be some sort of official
lists of known acceptable licences for third-party data and known
unacceptable licences for third-party data either on the wiki or on
the OSMF site, that people could be pointed to. Even if the acceptable
list is extremely short, explicitly listing incompatible licences
would help get across the point that care is needed, and hopefully
help people avoid accidentally making use of incompatible sources.

(At least I don't think such lists exist at the moment -- the closest
I could find is
http://wiki.openstreetmap.org/wiki/Import/Guidelines#Make_sure_data_license_is_OK.
However, the issue doesn't just affect imports, but also other sources
that people may want to use on an ad-hoc or manual basis.)

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] Updated geocoding community guideline proposal

[Robert Whittaker (OSM lists)]

On 11 July 2014 03:52, Alex Barth a...@mapbox.com wrote:
 I just updated the Wiki with a proposed community guideline on geocoding.

 Please review:
 https://wiki.openstreetmap.org/wiki/Open_Data_License/Geocoding_-_Guideline

The whole point of the share-alike aspect of our licence is to stop
people taking OSM, using it to 'improve' their own proprietary
geo-data, and then not sharing the results back with the community.
Whether you are for or against share-alike, that's what the community
has decided to adopt, and that is what is required by the licences
under which various source data has been used. So I think share-alike
is here to stay.

The loop-hole for produced works is to allow people to take
proprietary style instructions / algorithms and run the geo-data
through them to create something artistic, which they then don't have
to share. But the ODbL ensures that the underlying geo-data (and any
additions / modifications made to it) does have to be shared.

So the way I see it, if there's any (substantial) addition of external
geo-data along the way, then that addition creates a derivative
database, before the produced work is created. So if you want to
publicly use this database (or any produced work based on it) then
either the derivative database must be shared-alike, or the algorithm
used to produce it and any additional input data must be shared.

In the case of any substanitial amount of geocoding, you are clearly
having to add additional geographic data to the OSM data in order to
do the geocoding. I would therefore argue that the result must be seen
as a derivative database, and not as a produced work. (In fact I'd go
slightly further, and say that in order to do the geocoding, you have
to create a derivative database comprising the relevant data from OSM
and the relevant address data you want to match against. You then run
a query on that derivative database to produce your geocoded results.)

And I think treating substantial amounts of geocoded results as a
derivative database is certainly within the spirit of the licence, and
something we would want share-alike to apply to. If people are
enriching their own geographic data using OSM, we would like to be
able to use their data to help improve OSM. I don't see why the
specific case of geocoding should be any different to other uses of
OSM data with data companies would like to keep private.

In any case, even without these arguments, I think it would be
impossible to argue that a substantial database of geocoded data
that's been generated using OSM data is anything but a derivative
database of OSM. So I don't think there's any getting around
share-alike if your geocoded results are substantial.

So for those wanting to geocode proprietary datasets using OSM, I
think there are three main options:

1/ Make sure your geocoding only amounts to insubstantial use of OSM.
Then share-alike never kicks in, and it's irrelevant whether the
results are produced works or derivative databases.

2/ Make sure your geocoded database (and any produced works or
derivatives therefore) is kept internal to the company. Hence it is
never publicly used, and share-alike does not apply.

3/ Release the smallest possible derivatve database under the ODbL. As
far as I can see this would need to include whatever input data is
necessary for you to do the geocoding, as you need to include the
address data and the OSM data in the same derivative databse in order
to run your query on them to do the geocoding.

As an example for 3, if you have a databse of company offices with
addresses and other meta-data that you want to obtain lat/lons for,
you'd need to release the address data and the lat-lons you've
obtained. You needn't release the other meta-data, since that could be
kept independently as part of a collective databse, with the two
linked by some unique ID field.

As an aside, I've yet to actually read a use case where this
interpretation would be particularly problematic for a third party (at
least no more so than any other proprietary data vs share-alike use
case). The only thing I've seen where it might cause difficulties
would be where individual user privacy is at risk. But I would have
thought that either the users can keep their locations private (so not
publicly used) or the locations can be linked to the private meta-data
via an opaque key with the meta-data kept private in another part of a
collective database. A company could always additionally geocode
fictitious points to hide individual users to further increase privacy
if they wanted.

Given what I've written above, my view on
https://wiki.openstreetmap.org/wiki/Open_Data_License/Geocoding_-_Guideline
is that those proposing it need to go back to the drawing board. At
the very least, it needs to start off with proper definitions /
explanations of geocoding, and details included in each example to
say whether or not they include substantial use of OSM. It also
needs to acknowledge that databases of 

Re: [OSM-legal-talk] Community Guidelines - Horizontal Cuts better text

[Robert Whittaker (OSM lists)]

On 21 May 2014 15:08, Frederik Ramm frede...@remote.org wrote:
 I like the message but I am not sure if it really works, license-wise.

 Suppose I have my own data set with restaurant POIs, A.

 Now I take an OSM database with restaurant POIs, B.

 Now I compute the difference, B-A - all restaurants that are in OSM but
 not in my own data set.

 This database, B-A, is clearly derived and needs to be shared. However
 it does not contain anything that is not already in OSM so sharing it
 would be of little use to anyone.

 Now I build a restaurant finder web site that polls both databases, the
 A and the B-A database.

 And you say: Because of this I now need to share A.

 But I don't see how this can ever be possible. At what point has A,
 which has not been modified the slightest in the whole process, been
 tainted with ODbL? The only thing that has any descendance from OSM is
 the B-A database.

One possible argument (and I'm not sure whether it's correct or not)
would be that while initially A and B are independent elements of a
collective database, in order to run the query that works out B-A,
then A and B (or at least the information required to run the query)
are no longer independent. Therefore you've implicitly created a
derivative database of (at least parts of) A and B, in order to run
the query. If that's the case then either the (parts of) A+B
derivative database must be shared under ODbL, or the parts of A used
in the query and the details of the query must be made available.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] OpenData attributes from closed vector data

[Robert Whittaker (OSM lists)]

On 7 March 2014 22:40, Rob Nickerson rob.j.nicker...@gmail.com wrote:
 I have been provided (i) original vector data and (ii) a printed map leaflet
 both of which include attribute data about roads - for example, whether the
 road is lit.

 The owner of the attribute data (whether the road is lit) has explicitly
 stated that their data is available as OpenData and are happy for it to be
 added to OSM. However, I know that the underlying vector data is most likely
 derived from a closed source (national mapping agency).

 Given that I am only interested in the attribute data (we already have our
 own version of the road vector data) can I go ahead and add it to OSM using
 (i), (ii) or both?

I think it will depend on the attribute data, what additional
information you need to use/interpret it, and how it got to be tangled
up with the underlying vector data in the first place.

My first test would be a sort of clean-room test: Can you manipulate
the vector data-set in such a way that all the tainted information is
removed, and your new dataset contains just the attributes, and some
non-copyrightable identifiers (e.g. street names -- though the lack of
IP rights in a list of street names may be debatable)? Then can you
get the information you want to add to OSM from this new dataset? If
not, then arguably you would have to be making some use of the tainted
data in order to add the attributes to OSM, and hence it wouldn't be
allowed.

However, I don't think this is sufficient, since the tying of the
attributes to the identifiers may have made use of some of the tainted
data (e.g. the geometry). If this is the case, then arguably the
attribute-identifier relationships have been derived from the tainted
data in some way, and so are tainted also.

So I think, strictly speaking, could be on quite shaky ground here.
And if the underlying vector data has come from OSGB, they're likely
to claim IP rights in the derived dataset even if they don't legally
or morally have any. Rather than people playing arm-chair lawyers
here, I'd suggest that you get LWG to review the situation to give a
definitive answer of whether they're happy for the data to be used in
the way you're suggesting.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] Licensing advice for a potential data source?

[Robert Whittaker (OSM lists)]

On 21 January 2014 18:18, Adam Williamson ad...@happyassassin.net wrote:
 Hi, folks! I'm a new OSM contributor in Vancouver, BC, Canada. I'm doing
 some manual, on-the-ground, local knowledge mapping, but I'm also
 looking for importable sources of important data types we're currently
 missing locally.

 There is a guy running a project at http://wherepost.ca/ to produce a
 crowdsourced database of post box locations in Canada. There's a
 feedback page at http://wherepost.ca/about/ where I've been interacting
 with him. I believe his intent is for this information to be free, but I
 don't think he has the necessary framework in place for this:

IANAL, but I think there are three considerations here:

1/ That those contributing give away / license their own rights in the
data they submit

2/ That those contributing don't use sources that are
copyright-encumbered in some way

3/ That there is no over-riding encumbrance on the data-set, allowing
someone to claim rights over it even though it's been collected
independently rather than copied.

Point 1 is easy to address, you just have any contributors agree to
either place their contributions in the public domain, or licence them
under a suitable free licence. If this hasn't been done so far, then
the current data may not be ok to use. Point 2 may require some user
education as to what is and isn't acceptable. But for this particular
case, there is the more important issue that the background map is
from Google. As far as I'm aware, Google claims rights in any
coordinates obtained by clicking on a Google map. This could well be a
show-stopper for the existing data, but could be fixed for new
submissions by switching to an open base map such as OSM.

As for Point 3, I would say that anyone would have a rather hard job
claiming any rights over the locations of publicly viewable bright red
boxes that someone else has collected individually. But that's not to
say they wouldn't try. (This could be a problem for all sorts of data
in OSM though, including any Canadian postboxes already present.) The
case is significantly weaker than for postcodes though, since we're
talking about coordinates of physical objects that anyone could obtain
independently, rather than crowd-sourced copying of data that could
only have come initially from Canada Post's official list.

I don't know what the current coverage of post boxes in Canada is like
in OSM. One option that might be worth considering is for him to run a
site that allows contributors to directly add a postbox node in OSM,
along the lines of http://onosm.org/ but adding the data live through
a dedicated account. (You'd need to watch for vandalism though.) And
then for the database his site uses for showing the locations to come
directly from a regularly updated OSM data export. That way we don't
have to worry about importing data, and his site gets the benefit of
any boxes already in OSM or added to OSM in the future by other means.

Hope that helps,

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] [Talk-GB] Using store locator as source

[Robert Whittaker (OSM lists)]

On 17 September 2013 08:38, OpenStreetmap HADW osmh...@gmail.com wrote:
 However, basic postcode centre locations are part of the OS OpenData releases.

Unfortunately, CodePoint Open is the one dataset in the OS OpenData
collection that hasn't been cleared for use in OSM. See
https://lists.openstreetmap.org/pipermail/talk-gb/2013-July/015028.html

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] Clarifying Geocoding and ODbL

[Robert Whittaker (OSM lists)]

On 13 June 2013 14:58, Olov McKie o...@mckie.se wrote:
 Manual geocoding
 A person using an OSM map to find the latitude and longitude coordinates 
 associated with a point or an area, normally by clicking, drawing or 
 similarly marking where that point or area is on a map. As an example, the 
 process of marking the point or area could be very similar to adding a new 
 POI or area to OSM using the iD in-browser editor, but instead of uploading 
 the additions to OSM, the coordinates and metadata are stored somewhere else. 
 Manual geocoding can also be done completely whitout computer support for 
 instance by reading coordinates of a printed paper map based on OSM data.

I think you may be mixing up two different things here:

* The method of doing the geocoding, which can be either automated or by hand.

* The type of input data used, which could be details of the object
itself (if the object exists in OSM, or can be interpolated from data
that does), or local knowledge of where an object is in relation to
other mapped data (e.g. knowing that a certain house is on the corner
of two streets). (The second type here would be rather hard to
automate, but the first type could be done automatically or by hand.)

With either interpretation of manual, the insubstantial exclusions
in ODbL will allow you do do a few such geocodings and retain all
rights anyway. However, if you were to do a lot -- to the point at
which is becomes substantial I don't see any reason for the
share-alike provisions not to apply. You are making use of
contributors' works to derive some additional data. The whole point of
having a share-alike license is to ensure that such derivatives are
shared back with the community.

 Geocoding and license implications
 Manual geocoding of an entity that a person has prior local knowledge of, is 
 the same process as adding a new entity to the OSM, and as such the person 
 geocoding the entity retains their full copyright over the geocoded entity. 
 All other geocoding results in a Produced Work, as that term is defined in 
 the ODbL. Section 4.5 provides that a Produced Work is not subject to the 
 share-alike provisions of Section 4.4 of the ODbL.

I think it's quite a stretch to claim that something that is clearly a
derived dataset is actually a produced work. The point of the Produced
Work provisions is to allow artistic renditions of the data to be
licensed differently, as long as the underlying data is still shared.
It is not supposed to be used as a get-out clause for avoiding sharing
additional data.

In any case, I'm not sure that calling the geocoded dataset a produced
work will necessarily help. For if it were to be publicly used then
you would be required to share the derived database behind it and/or
the algorithm to create it, including AFAIK any external data you you
added (i.e. the data associated with each object that you used in the
search). So I don't think this would actually get you out of the
obligation to share-alike the search data employed. If you don't
publicly use the geocoded data, then there would be no-one external
getting a copy that would need to be given share-alike rights to it
anyway. (Share-alike only applies to those receiving a copy of the
data/work.) So it then wouldn't matter if the data what the
share-alike rules were.

I'd still very much like to hear of potential use cases, where
regarding the inputted search data plus returned coordinates as a
derivative database (which may be part of a collective database with
other proprietary data in it) would actually cause problems.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] Clarifying Geocoding and ODbL

[Robert Whittaker (OSM lists)]

On 7 June 2013 01:56, Alex Barth a...@mapbox.com wrote:
 With two State of the Map conferences coming up now and plenty of
 opportunities for face time, I'd like to restart our conversation around
 clarifying the ODbL's implications for geocoding and get to a result. Over
 here at MapBox we're hoping to use OpenStreetMap soon as much as possible
 for geocoding (right now we don't) and we'd like to do this on firm legal
 ground. I know that others have raised similar questions in the past [1].

For the avoidance of doubt, could you clarify exactly what you mean by
geocoding here? I presume it would be something along the lines of
taking some form of location description (e.g. a typical written
address), searching for an appropriate match in an OSM-derived
database, and then returning the latitude and longitude coordinates
associated with the matching OSM object. This process could possibly
be repeated many times, once for each record with a location
description in an external database.

In that case, then from a philosophical point of view, I think I'd
agree that other data about the location found in the database
shouldn't be tainted by share-alike. But at the same time, I don't
think that the coordinates should be available to be completely freely
used by the person obtaining them. As for the actual location data
used in the search, I think that's a more difficult question. On the
one hand, it's sort of necessary to do the search and interpret the
coordinates, so we'd want it to be shared). On the other hand, while
individual location data items aren't really proprietary, the
collection of them could be (e.g. identifying a set of customers), so
there may be reasons why it wouldn't be appropriate to share it.

My reasoning behind not wanting to allow the coordinates to be used
freely would be that I could, for example, produce a list of all
possible post box reference numbers in the UK since they're always a
postal district plus a 1-4 digit number. Then I could use OSM to get
coordinates for each reference number where it existred. If I was able
to freely use the resulting data without any restrictions, I'd then
have a public-domain dataset of all the post box locations that were
in OSM, sidestepping the share-alike provisions of ODbL. I think this
would be unacceptable.

It seems to me, that each location description (whatever was used to
search in OSM) plus the returned coordinates should probably be
regarded as a derivative database, which then forms part of a
collective database with any other (possibly private) data associated
with the location description. It's only if you publicly use the
data that the share-alike provisions kick in, and then you'd only need
to share the location descriptions and coordinates for points that are
shown to a user (which would probably be visible to them anyway).
Remember that the share alike provisions only apply to those receiving
the public use of the work. So if you only provide something based
on the data to a particular customer, that's the only person you have
to allow share-alike use to. (Of course there's nothing to stop the
customer sharing that data further, but that's up to them.)

Would something like this be a problem for any of the use cases that
you have in mind?

Of course, if you can argue that your geocoding results are
insubstantial under the ODbL then you can do what you want with
them. The above would only apply to substantial uses.

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] Current status on UK Council footpath data

[Robert Whittaker (OSM lists)]

On 6 June 2013 08:11, Nick Whitelegg nick.whitel...@solent.ac.uk wrote:
 Just wondering what the current state of what we can do with the UK council
 footpath open data is?

It will depend what data you are referring to. But the general rule
will apply: you can only use data/information that is subject to
someone else's copyright if you either have explicit permission to use
it in OSM, or permission to use it under a license that's compatible
with OSM's license.

* In the case of current OS Landranger and Explorer maps showing
Public Rights of Way, these are copyright Ordnance Survey. I'm not
aware of OS giving any permissions to re-use these maps, and so are
not usable in OSM.

* In the case of the Definitive Maps maintained by each council, then
these contain IP rights belonging to both the Council and Ordnance
Survey. I'm not aware of OS giving any permissions to re-use these
maps themselves. Hence they're not usable for OSM.

* In the case of GIS data for PRoW routes derived from Definitive
Maps, these also contain IP rights belonging to both the Council and
Ordnance Survey. However, under the Public Sector Mapping Agreement,
the councils can apply for permission from OS to release them under
the OS OpenData License. However, there are question marks over
whether this license is compatible with the ODbL+DbCL used by OSM. The
most recent statement I'm aware of from OS maintains that their
license is not compatible, and hence we shouldn't make use of this
data in OSM (unless we can obtain separate explicit permission from
both the council and OS). See
http://robert.mathmos.net/osm/prow/council-gis.html and
http://robert.mathmos.net/osm/os-open-data.html for more discussion.

* In the case of the Definitive Statements that each council must
maintain, OS has publicly stated that they don't claim any rights in
them, so the only IP rights rest with the council. Hence if you can
obtain permission from the Council (either explicitly for OSM, or
under a suitable licence), then it is ok to use them for OSM. FOr some
advice about how to obtain  Defintiive Statements and ask for
permission to use them in OSM, see
http://robert.mathmos.net/osm/prow/council-docs.html

Having said that, while various sources listed above are not usable
directly in OSM, there's nothing to stop you using such a source to
look for discrepancies in the current OSM data, and then using that
information to choose where to survey or search other sources for
information that can be used for OSM mapping. However, doing this,
you'd have to be careful that whatever you map in OSM comes only from
the sources you can use, and isn't tainted by the sources that you
can't.

Hope that helps,

Robert.

-- 
Robert Whittaker

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

[OSM-legal-talk] Content Licence for OSM Data

[Robert Whittaker (OSM lists)]

My understanding of the ODbL is that it covers an overall database,
but not individual contents within it. So in order to use an ODbL
database you also need a license (or other permission) to use the
contents. Conversely, when offering a database to others under the
ODbL, if you actually want them to be able to use it, you also need to
provide a suitable licence for the contents. See the ODC FAQ at
http://opendatacommons.org/faq/licenses/#db-versus-contents

In particular, the licensing instructions at
http://opendatacommons.org/licenses/odbl/ suggest you need to include
both the ODbL for the database, and a licence for the contents. The
suggested form is:

This {DATA(BASE)-NAME} is made available under the Open Database
License: http://opendatacommons.org/licenses/odbl/1.0/. Any rights in
individual contents of the database are licensed under the Database
Contents License: http://opendatacommons.org/licenses/dbcl/1.0/ 

However, on the OSM license page at
http://www.openstreetmap.org/copyright only the Open Database Licence
is mentioned. There is no mention of any licence for the contents.
Should we be specifying a content license for the OSM data on that
page? If so, should this be the Database Contents License (DbCL)?

(The DbCL is mentioned in the contributor terms, and there is a
reference at 
http://wiki.osmfoundation.org/wiki/License/About_The_License_Change#The_documents
, but neither explicitly says that the OSM contents are indeed
licensed under this licence. I also guess you could take the view that
there are no rights in the OSM contents since each is individually an
un-copyrightable fact. But to be on the safe side, to create an level
playing field in all jurisdictions, and to re-assure potential users,
I'd have thought it would be better to provide an explicit license for
the contents anyway.)

Any thoughts?

Robert.

-- 
Robert Whittaker

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

Re: [OSM-legal-talk] tesco store location data

[Robert Whittaker (OSM lists)]

On 5 November 2012 17:56, Richard Fairhurst rich...@systemed.net wrote:
 Chris Hill wrote:
 So the answer, as always with this sort of question, is no we cannot
 use that data without written permission of the copyright holder to
 use this data in OSM for any purpose. I don't think that is likely to
 be forthcoming.

 Don't forget, too, that Tesco probably didn't create the data themselves -
 they might have sent a guy out with a GPS to each of their stores, but I
 doubt it. More likely it's referenced against some external, commercial,
 protected-up-to-the-eyeballs dataset. Even if someone in Tesco's Customer
 Contact Centre (or whatever) says yes, they likely aren't aware of upstream
 copyright issues.

I'd have thought that Tesco would only benefit from having their
stores shown on OSM maps, so I don't see why they wouldn't give
permission (to the extent that they can) for us to use their data --
if we can manage to persuade the right person to consider the
question. The coordinates are indeed problematic, but even without
them the data could be useful.

I agree that an automated import isn't the way to go here, but some
sort of keep-right-esq tool that highlights places where a store is
expected but is not present in OSM would be really useful. Given
aerial imagery and an approximate store location, it will often be
possible to identify the right building and map it out, adding the
store details from the Tesco data.

As for finding the locations, the address may give enough of a clue to
local mappers. Maybe this could even be converted to approximate
coordinates by Nominatim?

More useful might be the postcodes. Since it seems we now have
centroid coordinates available via an ONS dataset (see
http://lists.openstreetmap.org/pipermail/talk-gb/2012-October/014055.html
) we could presumably use them to locate each store with reasonable
accuracy.

Would there be any copyright / database right issues with doing this,
assuming Tesco give their permission to use the addresses they hold,
and we're able to use the postcode data from OSN under the Open
Government License?

Robert.

-- 
Robert Whittaker

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