Licensing datasets

[David Holstius]

Hi all,

I figured there might be some open data expertise on this list. Are any of you thinking about the best way to license (or not license) datasets that you produce using Eggs or similar devices?

I work at a public University in the US, and I'm contacting our IP office to see what our options are. I'm hoping we can apply a CC-BY or similar license so anyone can use them with attribution. I then hope to post them on figshare so they have citable DOIs.

I know some of you work for commercial entities. What do your lawyers say? Also, I know nothing about international copyright law, but this is an international project. I'm operating so far under the impression that datasets are original works, therefore licensable. Does anyone have any good references to share on whether this is or is not the case in any particular jurisdiction(s)?

Thanks all!
David

P.S. What is the plain-English version of the license applicable to data submitted to Cosm? I skimmed the Terms of Use but didn't find what I was looking for ...

[Adam Laskowitz]

I remember a few years ago Usman Haque gave a talk in Copenhagen at Smart Geometry conference. He was talking about (I think) Data Bill of Rights and Pachube.
The closest thing I can find on this on Cosm now is this:
--------------------------------------------------------------------------------
Data ownership
Your data belongs to you and you are free at any time to download an archive of it should you choose to discontinue using the services provided by LogMeIn Cosm Ltd. Additionally, if requested, all personal data linked to you will be removed from the service.

By using the services provided by Cosm Ltd. you agree to allow Cosm Ltd. access to your data. We will at all times respect the privacy settings as set in your account
--------------------------------------------------------------------------------

Not sure if this what you are looking for, or if have seen this already or not.

[Cesar Garcia Saez]

In one of the presentations about Safecast, I heard they were using CC-Zero for the data. Distributing the datasets and reusing them was more important that keeping track of who was the creator of each datapoint in the dataset.

Best,

César

--

[Simone Cortesi]

I vote for CC0 too. It is important that reuse is king here, it is
important that our dataset gets analyzed and put thru algorithms in
order to extract pattern and science from it.

> --
>
>



--
-S

[Usman Haque]

i agree CC0 is the way to go. this can be done on Cosm at the time of feed creation by applying a machine tag to the feed (presumably during provisioning), something like "license:type=cc0" and possibly also "license:url=http://creativecommons.org/publicdomain/zero/1.0/". might even also consider adding a short note to the 'description' attribute of the feed to state in plain english the license.

[Simone Cortesi]

will this be done automatically by cosm upon activation of the device?

otherwise I see big impediments on asking people to add this on their
own. ppl will never do this, licenses are not widely understood.

On Wed, Nov 28, 2012 at 4:10 PM, Usman Haque <us...@pachube.com> wrote:
> i agree CC0 is the way to go. this can be done on Cosm at the time of feed creation by applying a machine tag to the feed (presumably during provisioning), something like "license:type=cc0" and possibly also "license:url=http://creativecommons.org/publicdomain/zero/1.0/". might even also consider adding a short note to the 'description' attribute of the feed to state in plain english the license.
>

> --
>
>



--
-S

[Usman Haque]

yes, for sure -- that's why i said "presumably during provisioning" :)
the AQE device model already has some tags added during the activation (as far as i know) so these would just be others to add.

[Simone Cortesi]

nice!

Thanks,
S.

> --
>
>



--
-S