|Thread title||Replies||Last modified|
|Libre Network services||2||23:55, 29 November 2012|
|Various||0||12:04, 29 June 2012|
|Copyleft/share-alike||14||11:37, 20 April 2011|
|Free file formats||2||09:04, 20 April 2011|
|Do we need yet another license?||2||09:01, 20 April 2011|
|Decoupling uses and abuses from the act of making copies or derived works||0||11:27, 3 April 2009|
|Terminology||2||08:11, 22 July 2008|
|Attribution||1||08:36, 11 February 2008|
Last edit: 23:55, 29 November 2012
An interesting discussion (see wiki and stack) is ensuing on this issue. Again it might lead to new licenses, use of an emblem to indicate "Libre Puro Certified" etc., but for now I offer a diagram which might help the thinking (feel free to improve it). This is a preliminary discussion to brainstorm how far and/or how pure we should be.
In essence, a network service is "libre" if:
- the underlying data is libre (accessible in a free file format which by definition is editable via software libre),
- private data would be accessible, editable and exportable (only) by the party concerned;
- derived data, or information generated using libre data should be released under a libre license;
- the information resources used by the service are libre and are generated with software libre,
- the knowledge resources are libre and all components are generated with software libre, accessing only libre information and libre data.
Where human interaction is involved, it is via a service where users have agreed that their statements and contributions (via text, voice or other media) are libre.
This is all variations on remix, and existing licenses probably deal with most scenarios. Automatic attribution would save a lot of trouble as would automatic checking of license compatibility.
A libre puro service would be all of the above with "libre puro" replacing "libre".
But remember that the Libre Puro License does not require attribution, and that the "puro" idealists would prefer people to keep derived works and services "libre puro" (without stopping others from using and deriving works under CC-BY-SA for example).
What other issues need to be addressed？
Do we allow users to relicense a resource (i.e. a derived work) as CC-BY for example? (a non-copyleft license)
No, because this would open up a way to lock down the knowledge by re-releasing it as (e.g.) copyright all rights reserved.
It might be okay to re-release under a free/libre share-alike license such as CC BY-SA ... but this complicates things (for a start one would need to drop the libre emblem, and successive derived works would have to be exactly CC BY-SA).
So, a "Libre Puro License" would have to be 'copyleftish' (i.e. perpetuate freedoms to have any impact) whereas a Libre Puro Emblem would not (and would probably have a greater impact if it became 'cool').
For a Libre Puro license: people would aspire to produce Libre Puro licensed resources and be assured that the libre nature would be preserved in future derived works. But it is currently difficult to live up to this license and we would be starting from scratch.
To qualify for a Libre Puro emblem all components would need to be appropriately mixed (wrt licenses), comply with the Free Cultural Works/ w:Libre Knowledge definition (the 'Libre Emblem' requires no more than this, Libre Puro requires more: ...), be in free file formats and (where applicable), served via libre services (e.g. via GNU Affero licensed resources), free protocols, etc.. This is easier than meeting the requirements of a Libre Puro License which needs to ensure that the freedoms are perpetuated in derived works.
If I'm reading this correctly, the fundamental difference between a "Libre Puro License" and say CC-BY-SA would be no requirement for attribution under a "Libre Puro License"?
It seems that there is one-way compatibility, i.e. Libre Puro ==> to compatible license, but not the other way around?
You'll need to think about how to resolve the attribution issue. From recollection, the legal code of a CC license through written permission, can remove the requirement for attribution under a CC-License, so the Libre Puro legal code would need to provide for this one-way compatibility.
(I'm speaking under correction here -- so you'll need to check up on this.)
So, should this license be precise wrt share-alike? (i.e. "Libre" only for derived works?). All the share-alike or copyleft licenses I know of specify that derived works should be licensed under exactly the same license as the original work. The designers of those licenses have probably thought about this before ... and I am gravitating towards the same conclusion.
I suspect that in the near future we will see legal compatibility among copyleft licenses. For example the permission to release a derivative work originally licensed under CC-BY-SA to be released as GNU(FDL).
This will go a long way in furthering the aims of the free content movement.
It will be interesting to see how this will be achieved. Probably along the same lines as "share similarly" proposed here.
The Attribution requirement of all the Creative Commons 3.0 licenses undermines the free culture movement. A lot of people seem to ignore it in many contexts ... perhaps as it is so burdensome to keep track of the attribution trees.
In academia there is already a culture of attribution. In other contexts, attribution may not be required - e.g. kids in copy-modify-mix and share mode for fun among their friends.
Abuse of digital resources may be dealt with via legal processes other than copyright.
I'm not sure whether I'd go as far as to say that attribution undermines the free culture movement. I don't see that the ethic of recognising the building blocks of new or reconfigured knowledge is intended to undermine the free knowledge movement.
I do agree that in many contexts (cultural and otherwise) that attribution may not be required or appropriate.
Regarding Creative Commons - in the early days attribution was an option selected by the copyright holder. I understand that in 98% of the licenses users chose to require attribution and they dropped this as an optional restriction of the license.
Perhaps we should lobby for reintroducing "BY" as an optional restriction in the CC license?
While licensing is important - I think that sometimes we spend too much time grappling with the idiosyncrasies of different licenses at the expense of widening the base of libre knowledge. Licensing is a mechanism or tool to deal with the ludicrous situation of copyright insofar as knowledge is concerned. I view copyleft licenses as a clever hack to deal with the problem of copyright and focus more energy on building capacity and community to achieve a free curriculum.
I did not say "intended" to undermine it. It just does on account of the burden of managing trees of attributions.
In academia we attribute anyway even for old PD works.
The 98% is not a valid reason to drop CC-SA. So, yes, we should lobby for Attribution to be an option and reinstate CC-SA.
Right now CC-SA 3.0 is not an option (retired at 1.0), so the percentage of people using it is not likely to grow (as I believe it would).
Insisting on "BY" reinforces the notion of "IP" which has been contended in many places (e.g. here).
Keep up the good work towards a libre curriculum :-).
Accepting non-share-similarly licenses (i.e. those that do not perpetuate the freedoms in derived works, like CC-BY) would be similar to the GNU Lesser General Public License (LGPL) tactic: - to grow the number of users of libre resources.
Software libraries licensed under the GNU Lesser General Public License (LGPL) may be included with proprietary software.
Typically (and tactically), this is applied where there is already an alternative available to the writers of non-free software.
By allowing writers of non-free software to include LGPL software libraries, we grow awareness and the number of users of free software.
So far, we have not adopted that tactic for libre knowledge. (The Free Culture movement has, in a sense, via CC-BY)
The CC-BY license is the equivalent of LGPL in this sense. Let's keep Attribution there (:-) and reluctantly tolerate those who don't care for freedom in derived works etc. using CC-BY.
On account of the existence of CC-BY, we do not need yet another nuance of "libre licenses" akin to the GNU Lesser General Public License (LGPL).
Currently, most knowledge and cultural works are easier to mimic than software - e.g. if one wants to replace a proprietary image of a house depicting different compartments/facets of some concept, just sketch an equivalent tree with branches (instead of rooms off a passage in the house).
But it is not always that easy, and may become more difficult as intelligent web services become more common among networked knowledge resources. These are provided via software. Currently we recommend the GNU Affero General Public License for such services.
Updates: recently updated the pre-amble to include:
- a definition of "libre license" on the page
- reference to "libre share" - which is not copyleft
- (more like "one more step of freedom" - e.g. a derived work (or even unmodified copies) may be CC-BY if no other components disqualify CC-BY).
Wayne just reminded me of CC0. This would be equivalent to Libre Puro in that attribution is not required and all the freedoms apply. So, perhaps there is no need for the Libre Puro License, and it would be better as an emblem indicating that the resource is in a libre format or is (or is fully editable by) software libre, and not patent-encumbered, etc.
Last edit: 09:04, 20 April 2011
Should we insist on resources actually being in a free file format, or should the license say that you are free to convert it to a free file format (in order to exercise your freedoms)?
Insist on formats convenient for modification by everyone. i.e. make the resources immediately available and localisable - towards digital inclusion.
Several people in the free culture community say that a proliferation of licenses is problematical.
For whom and why?
What if the way someone wants to release their work is not represented in an existing license?
Kim, I'm one of the several people <smile> -- who thinks that a proliferation of licenses is problematic.
- In my view -- we should respect the freedom of individuals to create select and use as many licenses as they want. But I don't think this is practical and we end up focusing on the tool rather than the values underpinning why one license is "libre" and another is not. The essential freedoms are the important bit in the equation -- not the mechanics of how we implement these in practice.
- I think that it's more important to spend time and energy in creating opportunities and resources to educate and inform folk about the essential freedoms rather than spending time and energy creating yet another license. So for example, regarding the attribution issue -- I think its more productive to lobby CC-Learn to bring back attribution as an option.
- The majority of teachers and educators I know already find the licensing to be complicated -- and I not sure they really care about the licenses they use. Typically -- a teacher might go to the CC site and click a couple of options -- eg NC feels good - so I'll select that one, without necessarily understanding the implications.
- I suppose there is a argument that a new "libre" license might simplify things -- for example, if you want to develop free content -- use this license, and avoid all the hassle of choosing from a plethora of options. That said -- I think it will take years before we have a mainstream solution. My view is to work with what we have got -- although not perfect, we can make it better. Already we have WMF thinking about ways to achieve compatibility between FDL and the CC-BY-SA license -- with thanks to a bold move by the FSF.
What if the way someone wants to release their work is not represented in an existing license?
I'd say that we should lobby existing licenses to improve and meet their requirements <smile>. I don't think its on the best interests of the freedom culture to create another license fork. However -- if a fork will take us forward in the long run -- it will. These things are dependent on the eco-systems that support them.
Last edit: 09:01, 20 April 2011
What should we call such a Libre Ideal License? Is there a Spanish, English or other language word which somehow captures the essence of the following?
- Licencia Libre Absoluto (or absolut ?)/Libre Absolute License
- Licencia Libre Base/ Libre Base License
- Licencia Libre Blanca/Clara - Libre Clear License
- Licencia Libre Cabal/Absolutely Libre License
- Licencia Libre Centro or Licencia Libre Núcleo/ Libre Core License
- Licencia Libre Cero/Libre Zero License
- Licencia Libre Completo/Libre Complete License or Completely Libre License
- Licencia Libre Fundación/ Libre Foundation License
- Licencia Libre Ideal/ Libre Ideals License
- Licencia Libre Inclusivo/ Libre Inclusive License
- Licencia Libre Perpetuo/Libre Perpetual License
- Licencia Libre Conservar/Libre Preserved License
- Licencia Libre Principio/ Libre Principle License
- Licencia Libre Prístino/Libre Pristine License
- Licencia Libre Puro/Libre Pure License
- Licencia Libre sin procesar/Libre Raw License
- Licencia Libre Universal/ Libre Universal License
- Permeso Libera Eterna (Esperanto)
A possible source of confusion: do we call "libre compatible" licenses (e.g. CC-BY-SA) "libre"? What about non-libre compatible licenses which do not violate the 4 freedoms (e.g. CC-BY)?
I'm not sure that I buy into the argument that CC-BY is a "non-libre" license.
It doesn't violate or restrict the essential freedoms and I would prefer to label CC-BY as a "libre" license.
Clearly NC and ND are "non-libre" licenses because they violate the essential freedoms.
There is strong opinion on both sides of the SA fence (within the libre friendly community). I don't think that its productive to have the SA versus non-SA teams at loggerheads -- we should respect the core freedoms of choice. I see the CC-BY and CC-BY-SA proponents as two sides of the same "libre" coin. They choose to emphasise freedom from different nuances -- and as an educationalist , I think this is healthy.
I prefer an approach like the Free Cultural Works Definition (not unlike the list of FSF compatible licenses) where we say "the following licenses meet the requirements of the essential freedoms". Incidentally -- Benjamin Mako Hill, who co-author of the Free Cultural Works Definition is a member of the FSF board (I'm sure you know this -- but readers of the list may not). WMF and WE have subscribed to the Free Cultural Works Definition -- the point being that the core freedoms are the foundations for deciding what is libre and what is not. The licenses are arbitrary.
Agreed. Though I think there is a missing nuance among the existing libre licenses.
Benjamin Mako Hill has suggested finding an alternative name for this nuance such as "Libre Ideal License" (Our aim is to produce a definition and license which represent "pure" freedom - another option for those who aspire to such ideals).
So, if we agree on a name for the license, let's say "Libre Puro License" for this discussion, then CC-BY and CC-BY-SA would both be "libre licenses" in the sense that they do not violate the 4 freedoms.
CC-BY-SA would be "Libre Puro compatible" - meaning that "Libre Puro" works could be modified or incorporated into another work and the result licensed as CC-BY-SA because CC-BY-SA perpetuates the freedoms (via SA).
CC-BY would be a "libre license" but not "Libre Puro compatible" since it does not ensure freedom for future derived works (or uses of the work in restrictively-licensed collections, etc.).
Attribution, which is part of each and every CC 3.0 license, might be in conflict with deep values in some cultures.
The draft libre license does not require attribution, leaving people free to decide whether to attribute or not. In academia we reference even public domain works - it is part of the culture. In other contexts it might be less important. e.g. kids in rip-mix-share activities for fun (and social construction). In those contexts we should relieve the sharers of the burden of keeping track of trees of attributions in derived/mixed works. Where Attribution does matter, we can develop tools to make it easier.
Personally I would like to see license options which don't force the requirement for attribution. In fact earlier versions of the CC had attribution as an option.
As an academic -- I will always reference my sources through attribution -- even if there is no legal requirement to do so. That said I can also see why many academics may want to ensure attribution. Either way -- I think its more important to widen the knowledge base of the intellectual commons -- and if this means a "thousand flowers" of licenses -- sobeit as long as there are reciprocal compatability clauses among all the free content licenses.