As more people and organizations are realizing the importance and benefits of publishing their data under an 'open' license, people are looking for the best license to apply.

Creative Commons CC has provided licenses that have facilitated great changes in the way that people publish their writing and artwork.

I know that there is some controversy over the use of CC licenses for spatial data, what are the technical concerns?

  • btw: It is spelt licence. To give someone a licence is to license them in the same way giving someone advice is to advise them. – JamesRyan Aug 6 '10 at 9:18
  • While both are correct, its probably best to use the American spelling 'license' when dealing with software as its often part of the license name: opensource.org/licenses/category – scw Aug 17 '10 at 23:53

The problem is that in some jurisdictions (in particular the EU) copyright is applied differently for databases than for coherent single works of creation. Therefore, it was necessary for i.e. openstreetmap to work on a new licence that would cover these different ways how copyright law is applied in a way that the ideas of creative commons are also valid for information put in the form of a database.

| improve this answer | |
  • Could you expand on this? Or provide links to more detailed discussion? – fmark Aug 17 '10 at 23:45

There is no problem with CC in general except when people use the optional share alike licence eg. OSM. http://creativecommons.org/licenses/by-sa/2.0/

It is quite common to want to mix copyright data that you can use but have no right to apply the cc licence to. So in not allowing derivatives that are copyright but with attribution, share alike renders such data incompatible. Some people have used layers in mashups to try to get round this but it is very shaky ground legally. Due to the number of people involved it is not usually possible to go back to them and negotiate a different licence for a particular use and in the end it means that companies won't touch it.

The aim is to force people into sharing their own work but in real life people get datasets from a variety of sources and the end result is that it simply reduces the number of people getting involved.

| improve this answer | |
  • I recently encountered this problem with OSM. We wanted to use OSM vector feature data retrieval in our 3D environment generation tools. We cannot use OSM because we would be forcing our users to share the 3D environments they create with our tools using OSM vector feature data. – Jaime Soto Oct 22 '10 at 19:22

Another possible problem is datarot when creative commons data is released as no derivatives allowed.

This means that:

Licensees may copy, distribute, display and perform only verbatim copies of the work, not derivative works based on it.

This becomes problematic when errors are discovered in the dataset, or it needs updating in any way. If the copyright holder is not contactable, or a disinclined/too busy to maintain a dataset, it becomes less and less useful over time.

| improve this answer | |

not a direct answer but I think using an Open Data Commons license would be more appropriate than CC for sharing data


They have everything from a GPL type license to a BSD type license

| improve this answer | |

I would be concerned about correctness, and authority.

The Creative Commons licenses make a lot of sense for spatial data that have key "creative" components - an index of delicious restaurants, or some other subjective topic. The attributes in such a dataset bring more value than the specific location, and a restaurant woudn't require within-a-meter accuracy.

However, for a dataset of gas pipelines, I want that to be as correct as possible! Granted, defining "who may use this" does not deliver a guarantee of correctness, but the nature of an open license makes it harder to place a value on the data.

Consider a published dataset of pipelines, licensed with the Creative Commons license. Someone later makes an update to that dataset to reflect what they feel is the true location of a certain pipe in the ground, perhaps due to construction or erosion or some other event. Now there is a disparity. Is this a technical concern, or a subjective / interpretive concern?

Update: A quick Google search exposed this document:
(I use this as an example)

Under Section V, Use Restrictions and Other Prohibited Activities:

The Licensee is familiar and understands the provisions of the National Map Accuracy Standards and assumes the responsibility and liability for the use of this data at other than the compilation scale (scale at which the digital data was intended to be output in hard copy format).

So, this license does not assume responsibility for the accuracy, but it does transfer responsibility to the licensee clearly. The Creative Commons license assume that the copyright holder has a 'creation', not a 'reference', which is the distinction I'm trying to make. I imagine one could add a clause to your Creative Commons license about accuracy.

I feel that Open Source software is a bad analogy for Spatial Data under a Creative Commons license. Open Source allows the freedom of any user to inspect the insides and validate it for their own satisfaction. Open data does too! Except... I don't want to have to validate it. Validating the position of something is not the same as validating the correctness of some lines of code; if I have to valid something's position on the Earth, why bother using someone else's?

| improve this answer | |
  • 4
    I don't believe that the license that data is published under has any relationship to the accuracy, correctness, or authority of that data. TeleAtlas could make a business decision tomorrow to publish all of their data under an open license, that wouldn't change anything about the accuracy or authority of the data. – DavidF Aug 5 '10 at 17:37
  • Yep, I agree. The analogy to open source software could be made - just because it is open source, it doesn't mean that people have "creatively" inserted bugs. As openstreetmap in my area demonstrates, sometimes openly licensed data can be more accurate that proprietary, purchased data. Due diligence is required whenever using any data set - this is irrespective of its licensing model IMHO. – fmark Aug 17 '10 at 23:48
  • 1
    I don't mean to imply that Creative Commons mean people will be making doodles out of my pipeline network (for whatever reason), I'm more concerned about dilution of authority. Maybe it's just me. – mwalker Aug 18 '10 at 4:06
  • I also don't mean to imply that people who use Creative Commons are mean. Stupid English. – mwalker Aug 18 '10 at 5:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.