Talk:Open Database License/Contributor Terms/Open Issues

From OpenStreetMap Wiki
Jump to: navigation, search

Scope 1

Problem for current mappers when some previous contributions are incompatible with the new CTs

The new CTs page asks current mappers to agree retrospectively for all their previous contributions too. Many mappers will have made use of sources that were available under CC-By or CC-By-SA licenses in good faith, but that are now not compatible with either the CTs or ODbL. They are therefore legally unable to agree to the new CTs even if they wanted to. A mechanism needs to be found to enable such people to agree to the CTs in such a way that some previous contributions can be exempted. (For instance, I've made use of OS OpenData, but am reasonably confident that I've always attributed the source when using it. So I could agree to the new terms for data I added where there is no mention of OS OpenData as a source. However, it's possible that others may not have been as rigorous in identifying sources.) -- Rjw62 16:25, 18 August 2010 (BST)

  • Where does it say that I have to agree retrospectively to all my previous contributions too? Reading the text I only find wordings like "I agree to only contribute stuff where I own copyright" but stuff that I contributed before agreeing to this contract can certainly not be concerned by this, right? IANAL. --spaetz 15:25, 26 August 2010 (BST)
  • login to your account at then view [1]. You will see the wording "Please read the agreement below and press the agree button to confirm that you accept the terms of this agreement for your existing and future contributions." Dmgroom 16:31, 26 August 2010 (BST)
Well the CT were presumably written with the signup page for new mappers in mind, so in that case there shouldn't be any existing contributions to refer to. But for the relicensing process, it is clearly meant for all previous contributions as well, as it wouldn't make sense otherwise. And I guess the text around the actual CT do state that on the relicensing page. --Amm 19:22, 26 August 2010 (BST)
  • If you have made some edits from sources that aren't compatible with the new CT and license, then I don't think you can agree to them atm. With respect to OS OpenData and other CC-BY or attribution only data, I believe the LWG are still hoping to get clarification from the lawyers (potentially with some changes?) that they are compatible, when at that point you can then agree to the CT. If not, and for those source that are definately not compatible, I guess there will need to be some technical measures to allow users to "split out" those changesets, so that they are at least able to agree to the CTs for their other edits --Amm 19:22, 26 August 2010 (BST)


Because this is a contract, only those of suitable age, mental status and in some countries even gender are able to enter into contracts, minors, those deemed mentally unsound and in some countries women would need a parent or guardian to agree on their behalf or the CTs would not be legally binding.

How is this any different from requiring people to consent to the historical 'contributor terms'. -- Firefishy 01:55, 18 August 2010 (BST)
In principal there is no difference, however the OSM-F is now taking direct ownership of contributions and in turn ownership of problems, the CTs try to minimise the OSM-F's legal standing, but only if the contract is legally binding. -- JohnSmith 02:36, 18 August 2010 (BST)
I call this out as clear FUD. The servers also reside in the UK and are governed by UK law. -- Firefishy 01:55, 18 August 2010 (BST)
You can't contract away anything that is law, it was my understanding that the CTs place more legal requirements on contributors in an attempt to protect OSM-F, however that would only work if the OSM-F had a legally binding contract with the contributor. -- JohnSmith 02:36, 18 August 2010 (BST)

Contradiction within the paragraph no (1)

This appears to me as an alternative, not a contradiction. Rw 14:02, 17 August 2010 (BST)
Really? Where does it mention "or", or "alternative" or any word(s) which could give the impression these are alternatives. Furthermore if it is an alternative is the third sentance an alternative to just the first sentwnce, or to the first and second sentences? Dmgroom 14:54, 17 August 2010 (BST)
It's very poorly written, it should be in point form at the very least showing the options -- JohnSmith 14:14, 17 August 2010 (BST)
* I agree and the below proposal by John Smiths reads much clearer. Think code to see why it currently does not parse well:
       if a:
         if not a:
--spaetz 15:30, 26 August 2010 (BST)
So how would you phrase it? -- Firefishy 01:56, 18 August 2010 (BST)
As I said, at the very least in point form -- JohnSmith 02:39, 18 August 2010 (BST)
1. You agree to only add Contents if:
    a) You are the copyright holder (to the extent the Contents include any copyrightable elements). You represent and warrant that You are legally entitled to grant the license in Section 2 below ; or
    b) If You are not the copyright holder of the Contents, You represent and warrant that You have explicit permission from the rights holder to submit the Contents and grant the license below in Section 2 below.

Furthermore You represent and warrant that such license does not violate any law, breach any contract, or, to the best of Your knowledge, infringe any third party’s rights.
There should probably be a third option too, to deal with the possibility that you are a joint copyright holder with one or more third parties. In this case, neither (a) or (b) strictly applies, but presumably the intention is that you grant the license to OSMF and you have have explicit permission from the other copyright holders. -- Rjw62 14:42, 18 August 2010 (BST)
I don't think a third option is needed, as part 'a' covers direct ownership and part 'b' covers ownership by others, although the 'or' might need to be turned into 'and/or' to satisfy the situation that both are needed, although I'm not familiar enough with legalese to know how to write it in a suitable manner for contracts. --JohnSmith 15:04, 18 August 2010 (BST)

Explicit permission

CC-zero is explicit about waiving "Affirmer's Copyright" (s2) and is explicit that affirmer's copyright and related rights include "database rights"[s1,iv]. CC-zero then explicitly grants those rights to "each member of the public" in the waiver of affirmer's rights. In the public license fallback (s3) those rights are granted to "each affected person". If you are the one using CC-zero works then the rights are explicitly granted to you.
PDDL is explicit about "Rightsholder ... relinquishes all rights in Copyright and Database Rights". PDDL dedication (s3.1 a) is explicit that "the Work is free and open for others to Use" and in the license (s3.3) "The Licensor grants to You ... any applicable Copyright and Database Rights." If you are the one using PDDL works then the rights are explicitly granted to you.
Both ODBL and CC-zero are explicit and verbose about the permissions they grant and to whom but they are not specific. Neither license states that $Specific_OSM_User is permitted. I presume that listing all specific permitted users and all specific permitted projects would require too much space when "the public", "each affected person", "others" and "You" covers all users explicitly. Rw 13:52, 17 August 2010 (BST)
User RW writes "I presume that listing all specific permitted users and all specific permitted projects would require too much space"
Exactly my point - the permission granted is not explicit. I'm not saying that PD or CC-zero should give explicit permission, I agree it would be unreasonable for them to do so, but since they don't then you can't say they are compatible with the CT's. Dmgroom 15:19, 17 August 2010 (BST)
I am not a lawyer, but how much more explicit can one be than to state in a written legal document (i.e. license) that everyone and therefore definitely include You has the permission to do what the license sais? To me, that seems as explicit as one can possibly be, unless you can claim that everyone does not apply to You. Amm 20:03, 18 August 2010 (BST)
The problem is most people confuse legal terms, which like scientific terms may have a completely different meaning to every day use. In this case the above licenses give implied rights, not explicit rights, as Dmgroom points out above, since the contributor isn't explicitly mentioned in the license and there is no contract explicitly giving the contributor permission -- JohnSmith 20:59, 18 August 2010 (BST)
Thank you for clarifying it better than I have been able to. Furthermore the right to use in OSM is not explicitly mentioned. So the rights given are neither explicit with regards who can use them, nor explicit about the fact they can be used in OSM. -- Dmgroom 22:10, 18 August 2010 (BST)
Do you have a link to the legal definition of "explicit" in that case? Where it explicitly ;-) -- Amm
Which jurisdiction do you want it for, I'm sure it would vary between them, here is one definition -- JohnSmith 19:41, 19 August 2010 (BST)
Any jurisdiction where an explicit licence to everyone would not include an explicit licence to OpenStreetMap would do. Unfortunately, I fail to see how your definition you linked to would cover that. It states as synonyms words like "absolute", "crystal-clear", "definite", "clearly stated". I still don't see, how a formal, written, legal document giving permission to everyone would not meet those definitions.
But you ignore the synonyms "precise", "strict", "specfic", which to my mind can not apply to the general permissions given by these licences. Dmgroom 15:01, 22 August 2010 (BST)
sais that when it in every day terms the license explicitly sais everyone has the right that it then does not mean they have explicit rights in legal terms? I mean, if the license says that "OpenStreetMap" or company X has the right to do this, would that be legally explicit? It does not mention specifically JohnSmith in the license, so how can you say you have the explicit right? Same if you are the employer of Company X, it does not specifically mention your name, so you can't have explicit legal rights? Or why would the group "OpenStreetMap" or company X of which one is (implicitly) a member of work any better than the group "everyone" which one definitely is a member of? -- Amm
Well Nearmap until recently explicitly allowed use of their map data in OSM under CC-by-SA, now they don't... -- JohnSmith 19:41, 19 August 2010 (BST)
One way to clarify things would be to require either "explicit permission to submit the Contents under these terms" or that "the Contents have been released under a license and/or terms and conditions that is/are compatible with these terms." Then add a statement that "For the purposes of this clause, OSMF will maintain the definitive list of licenses and terms that are deemed compatible." This would remove any need for (and risks associated with) individual mappers to have to worry about interpreting any licenses / T&C's, or how to interpret explicit -- they can either ask the copyright owner for explicit permission to submit to OSM, or rely on a pre-vetted license that OSMF says is ok. -- Rjw62 14:49, 18 August 2010 (BST)
That would seem an excellent solution . Dmgroom 15:45, 18 August 2010 (BST)

Incompatibility with CC-BY / Attribution Data

OSM currently provides attribution pages. These attribution pages could be made permanent to satisfy donors with attribution requirements. Rw 13:59, 17 August 2010 (BST)
Section 4 *may* cover attribution issues in section 3, it does not fix section 1 being in conflict with section 3 according to Francis Davey.
While Section 4 covers attribution by OSM, it doesn't enforce that after a future relicensing that people are required to attribute OSM itself, particular if the future license does not require attribution. It's no point OSM having an attribution page if end users of the data don't have to mention that the data is from OSM, which would be possible if a vote to relicense as CC-0, for example, got through --DavidDean 13:19, 19 August 2010 (BST)

Incompatibility with ODBL Data

citation needed? ;-) Please expand this "open issue". If the "Same as above" refers to the ccby attribution issue above, the resolution is just as simple. Include attribution and disclaimer in the wiki and protect those pages from editing if required. Rw 14:07, 17 August 2010 (BST)
The CTs don't guarantee attribution or share-a-like, ODBL requires both... -- JohnSmith 14:16, 17 August 2010 (BST)
The CTs require you allow OSM-F to be able to relicense the data in future, while ODBL would allow you to use the data, you explicitly don't have the right to allow it to be relicensed... -- JohnSmith 02:51, 18 August 2010 (BST)
I really wish you would stop repeating this mis-information. OSMF does not have the automatic right to relicense anyone's data. -- Firefishy 04:08, 18 August 2010 (BST)
I never said anything about 'automatic', are you intentionally trying to confuse the issue? Section 2 requires you to grant OSM a right to relicense in future, which is all I was commenting about. If I'm mistaken please explain how I'm mistaken. -- JohnSmith 04:55, 18 August 2010 (BST)
I believe that the problem here is that the CTs don't allow ODbL-licensed data to be used / imported without additional rights grants from the copyright holders. Being ODbL-licensed does not give you sufficient rights to grant OSMF the rights required under CT clause 2. This means a third-party can take some OSM data, combine it with their own, and release the results under ODbL to fulfil their obligations, but there's nothing to force them to agree to the CTs to allow the derived database to be re-imported back into OSM. One of the main purposes of having a share-alike license for the OSM data is to allow OSM to be able to utilise any derived datasets produced by third parties. But with these CTs, there is no guarantee that we will be able to do so. -- Rjw62 14:59, 18 August 2010 (BST)

Action in Event of Breach of CT

Current action on inappropriate sources, suspicious data, edits or accounts is handled by the community. Contributors can discuss issues with other contributors, revert mistakes, correct errors, and advise on community norms. The overwhelming majority of issues that I run into are simple misunderstandings and are easily fixed with a friendly email exchange. YMMV. Contributors can also request the assistance of the Data Working Group, who are also OpenStreetMap contributors. The DWG, will also engage in discussion and advice but also are able to draw on the Technical Working Group for additional technical measures in severe cases, like account suspensions. I think this works. The community largely manages itself. Should this change? Does the community want the OSMF to take the responsibility for enforcing community norms, or to continue as a fall back when contributors request assistance? Rw 23:07, 3 September 2010 (BST)
By taking ownership of the data to an extent, OSM-F will be more directly responsible for the contents, rather than just another end user as the current situation exists. While the OSM-F are attempting to offload some of the problems onto the contributor by way of CTs, they haven't specifically listed actions that will be taken against the user, at the very least they should have a clause about removing the account and/or banning the person if it's determined that the breach of contract is significant enough. -- JohnSmith 00:44, 22 September 2010 (BST)

Contradiction between CT text and FAQ

The contributor terms FAQ states that accepting the contributor terms is per account, not per person, but the CT themselves say in section 3 An 'active contributor' is defined as: a natural person (whether using a single or multiple accounts).

While this may not strictly be a contradiction, because CT and license are two different things, the consequences should be thoroughly considered and clarified, e.g.

  • what if a contributor wants to accept a future license change only for some of its data? With the current terms this is not possible.
  • what other implications does that have, whether intended or not?

--Schuetzm 18:21, 8 October 2010 (BST)

Very wide range of paragraph 2

I second the concerns of NearMap stated on the article page.

Why does the OSMF need a blanket license for doing anything they want with the data, when their only purpose is to publish the data under a handful of licenses (and in addition enforce these licenses)? Note that section 3 does not restrict the OSMFs rights in any way: it merely requires them to sub-license the data under ODBL or the other licenses mentioned, but does not forbid doing anything else.

OSMF agrees to use or sub-license Your Contents as part of a database and only under the terms of one of the following licences: ...

This was probably intended to mean: OSMF promises to sub-license your contents ..., and will never do anything else with them., but as it reads now, I cannot see how this could be interpreted in any way to restrict the rights granted in section 2.

I see this as a very serious issue, because the OSMF could relicense the data in any way they like without breaking their part of the agreement, as long as they also publish it under the ODBL.

--Schuetzm 18:42, 8 October 2010 (BST)