Archive for December, 2014

Happy UTC+0 New Year

Wednesday, December 31st, 2014
With apologies for the projection.

Smattering of followups on mostly-recent posts, posted at 2015-01-01 00:00:00 UTC. Does anyone celebrate UTC+0 New Year except by coincidence of being in UTC+0 time zone? Yes.

Software Freedom Conservancy released a video with me endorsing them (my recent blog endorsement). I self-recorded the footage and acknowledge total videography incompetence, need of a haircut, and need to be still.

PLOS Biology published a perspective by Daniel Mietchen on The Transformative Nature of Transparency in Research Funding. Riffing on his tweet, that’s early theory; practice is the Wikidata for Research proposal that he is leading creation of in the open (my recent blog endorsement).

Snowdrift.coop’s one-time crowdfunding campaign (my recent blog endorsement and others) is wrapping up very successfully. Looking forward to seeing Snowdrift.coop launch in early 2015.

Free Software Foundation’s call for input on updating its high priority projects list (my blog post) has resulted in over 100 emails to hpp-feedback@gnu.org, most of them very thoughtful and containing numerous suggestions. Some are mirrored in public posts: Antoine Amarilli, Christopher Allan Webber, d3vid seaward, Denver Gingerich, Ingegnue. Please send your feedback! I especially enjoy seeing public posts and explanations of how suggestions are on critical path toward achieving goal of software freedom for everyone.

Speaking of the FSF, they recently released a new video making the case that software freedom is important for everyone. I agree with Christopher Allan Webber’s asseessment of good progress. The video also ties into a free software futurist dinner that Webber said raised money for Software Freedom Conservancy, and some statements I make in the video above: I suspect it’s much easier to take software freedom as a serious issue of top importance if one has a “futurist” bent. This will also figure in a forthcoming post from me casting doubt on everything in this post and the rest from 2014 (last year’s version).

There’s some overlap between the above and OpenHatch’s year-end newsletter (my year-ago blog endorsement).

Finally, check out Don Marti’s below the fold announcement about Aloodo, a project to (if I understand correctly) help sites protect themselves from the long-term damage of being associated with pervasive tracking and door-to-door-like incentives (everything to make immediate conversion, nothing to build trust). I still have not gotten around to blogging other ideas for “fixing” online advertising, but very much look forward to seeing how Marti’s project plays out.

wikidata4research4all

Sunday, December 21st, 2014

Recently I’ve uncritically cheered for Wikidata as “rapidly fulfilling” hopes to “turn the universal encyclopedia into the universal database while simultaneously improving the quality of the encyclopedia.” In April I uncritically cheered for Daniel Mietchen’s open proposal for research on opening research proposals.

Let’s combine the two: an open proposal for work toward establishing Wikidata (including its community, data, ontologies, practices, software, and external tools) as a “collaborative hub around research data” responding to a European Commission call on e-infrastructures. That would be Wikidata for Research (WD4R), instigated by Mietchen, who has already assembled an impressive set of partner institutions and an outline of work packages. The proposal is being drafted in public (you can help) and will be submitted January 14.

4all

The proposal will be strong on its own merits, and very well aligned with the stated desired outcomes from the EC call, and the open proposal dogfood angle is also great. I added for all to this post’s title because I suspect WD4R will be a great for pushing Wikidata toward realizing aforementioned “universal database” hopes (which again means not just the data, but community, tools, etc.; “virtual research environment” is one catch-all term) and will make Wikidata much more useful “research” most broadly construed (e.g., by students, journalists, knowledge workers, anyone), potentially much faster than would happen otherwise.

My suspicion has two bases (please correct me if I’m wrong about either):

  1. A database or virtual environment “for research” might give the impression of someplace to dump data from or perform experiments. Maybe that would be appropriate for Wikidata in some instance, but the overwhelming research-supporting use would seem to be mass collaboration in consolidating, annotating, and correcting data and ontologies which many researchers (and researchers-broadly-construed, everyone) can benefit from, either querying or referencing directly, or extracting and using elsewhere. The pre-existing Gene Wiki project which is beginning to use Wikidata is an example of such useful-to-all collections (as referenced in the WD4R pages).
  2. One of the proposed work packages is to identify and work on features needed for research but not on, or not prioritized on, the Wikidata development plan. I suspect other Wikimedia projects can tremendously benefit from Wikidata integration without Wikidata itself or external tools supporting complex queries and reporting that would be called for by a virtual research environment — and also called for to realize “universal database” hopes. Wikidata’s existing plan looks good to me; here I’m just saying WD4R might help it be even better, faster.

The previously linked Gene Wiki post includes:

For more than a decade many different groups have proposed and many have implemented solutions to this challenge using standards and techniques from the Semantic Web. Yet, today, the vast majority of biological data is still accessed from individual databases such as Entrez Gene that make no attempt to use any component of the Semantic Web or to otherwise participate in the Linked Open Data movement. With a few notable exceptions, the data silos have only gotten larger and problems of fragmentation worse.
[…]
Now, we are working to see if Wikidata can be the bridge between the open community-driven power of Wikipedia and the structured world of semantic data integration. Can the presence of that edit button on a centralized knowledge base associated with Wikipedia help the semantic web break through into everyday use within our community?

I agree that massive centralized commons-oriented resources are needed for decentralization to progress (link analogous but not the same — linked open data : federation :: data silos : messaging silos).

Check out Mietchen’s latest WD4R blog post and the WD4R project page.

Defensive Patent License 1.1 w/diff

Tuesday, December 9th, 2014

13 months ago I wrote about the Defensive Patent License, in particular in relation to free/open source software (followup, 1993 predecessor). Today the DPL project released DPL 1.1 and announced the first licensor; see Internet Archive and EFF posts.

(The EFF post references its earlier guide to alternative patent licensing which I meant to critique, probably along lines of a partially overalpping guide to reform proposals, see patent reform, parts deficient in commons and compare with protect commons from patents. I noticed today that one of the other alternative licensing schemes, License On Transfer, seems to be getting some uptake.)

Most of what I wrote previously about the DPL concept still applies with DPL 1.1 (interesting concept, possibility of substantial good impact in long term). The new version makes one major improvement (especially in relation to FLOSS) — the exclusion of “clone” products or services from the license grant has been removed. Another small (as in a -3 words difference) improvement is that alleging patent invalidity against another DPL user no longer breaches one’s licenses (only alleging infringement does), invalidation being a defensive tactic.

DPL 1.1 also adds the requirement of explicit acceptance, which strikes me as burdensome: one must research licensed patents in order to figure out which DPL users to contact with acceptance, or regularly contact all known DPL users with acceptance of all licensed patents. I understand from the DPL 1.1 announcement telecon that formal acceptance was added because the license grant is more likely to stand up in court with such explicit acceptance, with that more likely assessment based on differences between patent and copyright, and between clubs and public licenses — and further that the “contact all known DPL users” practice will in the future be facilitated by the DPL website.

Finally, a very minor issue: DPL 1.1 reproduces the GPL’s confusing three-option version compatibility scheme (this-version-or-later, only-this-version, or any-version-if-none-specified). If one must have options, I consider less confusing this-version-or-later as default, with option to explicitly mandate only-this-version.

Congratulations and thanks to Jason Schultz, Jennifer Urban, Brewster Kahle, John Gilmore, and others for getting the DPL into production. I hope it is wildly successful; check out the DPL website and help update the Wikipedia article.

Following is a wdiff between DPL 1.0 and 1.1 in two parts (because 1.0 put definitions at the beginning, 1.1 puts them at the end) below, excluding 1.1’s preface, which has no equivalent in 1.0.

DPL 1.0-1.1 wdiff: Grant, conditions, etc.

[-2.-]{+1.+} License Grant

Subject to the conditions and limitations of this [-License and upon-]
[-affirmative assent to the commitments specified in Section 1.7 from an-]
[-individual DPL User,-] {+License,+} Licensor hereby
grants and agrees to grant to [-such-] {+any+} DPL User {+(as defined in Section 7.6) who+}
{+follows the procedures for License Acceptance (as defined in Section 1.1)+}
a worldwide, royalty-free, no-charge, non-exclusive, irrevocable (except
as stated in Sections [-3(e)-] {+2(e)+} and [-3(f))-] {+2(f))+} license, perpetual for the term of
the relevant Licensed Patents, to make, have made, use, sell, offer for
sale, import, and distribute Licensed Products and Services that would
otherwise infringe any claim of Licensed Patents. A Licensee’s sale
of Licensed Products and Services pursuant to this agreement exhausts
the Licensor’s ability to assert infringement [-by-] {+against+} a downstream
purchaser or user of the Licensed Products or Services.

[-2.1-] {+Licensor’s+}
{+obligation to grant Licenses under this provision ceases upon the arrival+}
{+of any applicable Discontinuation Date, unless that Date is followed by+}
{+a subsequent Offering Announcement.+}
{++}
{+1.1+} License Acceptance

In order to accept this License, Licensee must {+qualify as a DPL User+}
{+(as defined in Section 7.6) and must+} contact Licensor via the
[-contact-] information
provided in [-Section 1.16 and-] {+Licensor’s Offering Announcement to+} state affirmatively that
Licensee accepts the terms of this License. Licensee must also {+communicate+}
{+the URL of its own Offering Announcement (as defined in Section 7.13) and+}
specify whether it is accepting the License to all Licensor’s Patents or
only a subset of those Patents. If Licensee is only accepting the License
to a subset of Licensor’s Patents, Licensee must specify each individual
{+Patent’s country of issuance and corresponding+} patent [-by patent number.-]
[--]
[-3.-] {+number for which+}
{+it is accepting a License. There is no requirement that the Licensor+}
{+respond to the Licensee’s affirmative acceptance of this License.+}
{++}
{+2.+} License Restrictions

Notwithstanding the foregoing, this License is expressly subject to and
limited by the following restrictions:

(a) No Sublicensing. This License does not include the right to sublicense
any Licensed Patent of any Licensor.

(b) License Extends Solely to Licensed Patents in Connection with Licensed
Products and Services. For clarity, this License does not purport to
grant any rights in any Licensor’s copyright, trademark, trade dress,
design, trade secret, other intellectual property, or any other rights of
Licensor other than the rights to Licensed Patents granted in Section 2,
nor does the License cover products or services other than the Licensed
Products and Services.  {+For example, this License would not apply to+}
{+any conduct of a Licensee that occurred prior to accepting this License+}
{+under Section 1.1.+}

(c) Scope. This License does not include Patents with a priority date
or Effective Filing Date later than Licensor’s last Discontinuation
Date that has not been followed by a subsequent Offering Announcement
by Licensor.

(d) Future DPL Users. This License does not extend to any DPL User whose
Offering Announcement occurs later than Licensor’s last Discontinuation
Date that has not been followed by a subsequent Offering Announcement
by Licensor.

(e) Revocation and Termination Rights. Licensor reserves the right to
revoke and/or terminate this License with respect to a particular Licensee [-if:-]
{+if, after the date of the Licensee’s most recent Offering Announcement:+}
{++}
{+i.+} Licensee makes any Infringement Claim, not including Defensive Patent
Claims, against a DPL User; or

{+ii.+} Licensee {+assigns, transfers, or+} grants an exclusive [-license,-]
[-    with the right to sue, or assigns or transfers-] {+license for+}
a Patent to an entity or individual other than a DPL User without
conditioning the [-transfer-] {+assignment, transfer, or exclusive license+} on the [-transferee-]
{+recipient+} continuing to abide by the terms of this [-License.-] {+License, including but+}
{+not limited to the revocation and termination rights under this Section.+}

(f) Optional Conversion to FRAND Upon Discontinuation. [-As-] {+Notwithstanding+}
{+any other provision in this License, as+} of any particular Licensee’s
Discontinuation Date, Licensor has the right to convert the License of
that particular Licensee from one that is royalty-free and no-charge to
one that is subject to Fair, Reasonable, And Non-Discriminatory (FRAND) [-terms.-]
[--]
[-4.-]
{+terms going forward. No other terms in the license may be altered in+}
{+any way under this provision.+}
{++}
{+3.+}        Versions of the License

[-4.1-]

{+(a)+} New Versions

The DPL [-Foundation is-] {+Foundation, Jason M. Schultz of New York University, and Jennifer+}
{+M. Urban of the University of California at Berkeley are+} the license [-steward. No-]
{+stewards. Unless otherwise designated by one of the license stewards,+}
{+no+} one other than the license [-steward-] {+stewards+} has the right to modify or publish
new versions of this License. Each version will be given a distinguishing
version number.

[-4.2-]

{+(b)+} Effect of New {+or Revised+} Versions

[-Licensed Products and Services-]

{+Any one of the license stewards+} may {+publish revised and/or new versions+}
{+of the DPL from time to time. Such new versions will+} be [-used, made, sold, offered for sale,-]
[-imported, or distributed under-] {+similar in spirit+}
{+to+} the [-terms-] {+present version, but may differ in detail to address new problems+}
{+or concerns.+}
{++}
{+Each version is given a distinguishing version number. If Licensor+}
{+specifies in her Offering Announcement that she is offering a certain+}
{+numbered version+} of the [-version-] {+DPL “or any later version”, Licensee+}
{+has the option+} of {+following+} the [-License-]
[-originally accepted pursuant to Section 2.1,-] {+terms and conditions either of that+}
{+numbered version+} or [-under-] {+of any later version published by one of+} the [-terms-] {+license+}
{+stewards. If Licensor does not specify a version number+} of {+the DPL in+}
{+her Offering Announcement, Licensee may choose+} any
[-subsequent-] version {+ever+} published
by {+any of+} the license [-steward.-]
[--]
[-5.-] {+stewards.+}
{++}
{+4.+}        Disclaimer of Claims Related to Patent Validity and [-Noninfringement.-]
{+Noninfringement+}

Licensor makes no representations and disclaims any and all warranties
as to the validity of the Licensed Patents or [-that-] {+the+} products or processes
covered by Licensed Patents do not infringe the patent, copyright,
trademark, trade secret, or other intellectual property rights of any
other party.

[-6.-]

{+5.+}        Disclaimer of [-Warranties.-] {+Warranties+}

UNLESS OTHERWISE MUTUALLY AGREED TO BY THE PARTIES IN WRITING,
LICENSOR OFFERS THE PATENT LICENSE GRANTED HEREIN “AS IS” AND
MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND CONCERNING THE
LICENSED PATENTS OR ANY PRODUCT EMBODYING ANY LICENSED PATENT, EXPRESS
OR IMPLIED, STATUTORY OR OTHERWISE, INCLUDING, WITHOUT LIMITATION,
WARRANTIES OF TITLE, [-MERCHANTIBILITY,-] {+MERCHANTABILITY,+} FITNESS FOR A PARTICULAR PURPOSE,
NONINFRINGEMENT, OR THE PRESENCE OR ABSENCE OF ERRORS, REGARDLESS OF THEIR
DISCOVERABILITY. SOME JURISDICTIONS DO NOT ALLOW THE EXCLUSION OF IMPLIED
WARRANTIES, IN WHICH CASE SUCH EXCLUSION MAY NOT APPLY TO LICENSEE.

[-7.-]

{+6.+}        Limitation of [-Liability.-] {+Liability+}

LICENSOR SHALL NOT BE LIABLE FOR ANY DAMAGES ARISING FROM OR RELATED TO
THIS LICENSE, INCLUDING INDIRECT, INCIDENTAL, CONSEQUENTIAL, PUNITIVE OR
SPECIAL DAMAGES, WHETHER ON WARRANTY, CONTRACT, NEGLIGENCE, OR OTHERWISE,
EVEN IF LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES
PRIOR TO SUCH AN OCCURRENCE.

DPL 1.0-1.1 wdiff: Definitions

[-1.-]{+7.+} Definitions

[-1.1-]

{+7.1+} “Affiliate” means a corporation, partnership, or other entity in
which the Licensor or Licensee possesses more than fifty percent (50%) of
the ownership interest, representing the right to make the decisions for
such corporation, partnership or other entity which is now or hereafter,
owned or controlled, directly or indirectly, by Licensor or Licensee.

[-1.2  “Clone Products or Services” means products or services of-]
[-Licensee that include the same or substantially identical functionality of-]
[-all or a commercially substantial portion of a prior released product or-]
[-service of a Licensor and implement the same or a substantially identical-]
[-proprietary user interface of the prior product or service.-]
[--]
[-1.3-]

{+7.2+} “Defensive Patent Claim” means an Infringement Claim against a
DPL User made in response to a pending prior Infringement Claim by said
DPL User against the asserter of the Defensive Patent Claim.

[-1.4-]

{+7.3+} “Discontinuation Announcement” means a DPL User’s announcement
that:

{+(a)+} declares the DPL User’s intent to discontinue offering to license
its Licensed Patents under the DPL, effective as of the Discontinuation
Date; and

{+(b)+} contains the DPL [-User's-] {+User’s+} contact information for licensing purposes;
and [-is submitted to the DPL-]
[-    Website via the Websites's official email address-]

{+(c)+} at least 180 days prior to [-a-] {+the+} Discontinuation [-Date;-] {+Date is posted to a+}
{+publicly accessible website;+} and

{+(d)+} at least 180 days prior to the Discontinuation Date is [-posted to a publicly accessible-]
[-    indexed-] {+communicated+}
{+reasonably and promptly, along with the URL of the+} website [-controlled-] {+mentioned in+}
{+subsection (c) of this provision,+} by the {+discontinuing+} DPL User [-using a URL accessible-]
[-    via at least the following syntax: "http://www.NAME.com/DPL" or-]
[-    "http://www.NAME.com/defensivepatentlicense" where "NAME" is-] {+to every+}
{+Licensor of+} a [-name-]
[-    commonly associated with-] {+Patent to which+} the {+discontinuing+} DPL [-user, such as-] {+User is+} a [-company name.-]
[--]
[-1.5-] {+Licensee.+}
{++}
{+7.4+} “Discontinuation Date” means the date a DPL User specifies in
[-their-]
{+its+} Discontinuation Announcement to discontinue offering to license its
Licensed Patents under the DPL, which must be at least 180 days after
the date of an individual or entity’s most recent Discontinuation
Announcement.

[-1.6-]

{+7.5+} “DPL” and “License” mean the grant, conditions, and
limitations herein.

[-1.7-]

{+7.6+} “DPL User” means an entity or individual that:

{+(a)+} has committed to offer a license to each of its Patents under the [-DPL, or, if such entity or individual has no Patents, has-]
[-    committed to offer a license to any Patents it may obtain in the-]
[-    future under the-]
DPL; and

{+(b)+} has declared such commitment by means of an Offering Announcement; and

{+(c)+} if the entity or individual has made a Discontinuation Announcement,
the Discontinuation Date has not yet occurred; and

{+(d)+} has not engaged in the conduct described in either Sections [-3(e)(i)-] {+2(e)(i)+}
or [-3(e)(ii).-]
[--]
[-1.8 “DPL Website” means the website-]
[-at http://www.defensivepatentlicense.org,-]
[-http://www.defensivepatentlicense.com, or any future site designated by-]
[-the DPL Foundation.-]
[--]
[-1.9-] {+2(e)(ii).+}
{++}
{+7.7+} “Effective Filing Date” is the effective filing date determined
by the applicable patent office that issued the relevant Licensed Patent.

[-1.10 “Foundry Services or Products” means services provided by-]
[-Licensee to, or products manufactured by Licensee for or on behalf of,-]
[-a specific third party, using designs or specifications received in-]
[-a substantially completed form from that third party, for resale or-]
[-relicense to or on behalf of that third party. This definition will not-]
[-apply when:-]
[--]
[-    Licensee or its Affiliate owns the design or specification of such-]
[-    service or product and the service or product is not specifically-]
[-    designed for commercial exploitation substantially only by such third-]
[-    party; or such design or specification resulted from a bona fide joint-]
[-    development or joint participation between Licensee or its Affiliate-]
[-    and such third party, including but not limited to a standards body-]
[-    or community organization and the resulting products, services or-]
[-    components provided by Licensee or its Affiliate meet the definition-]
[-    of Licensed Services Product or Products as set forth herein; or-]
[-    the third party recipient of the products or services is a DPL User.-]
[--]
[-1.11-]

{+7.8+} “Infringement Claim” means any legal action, proceeding or
procedure for the resolution of a controversy in any jurisdiction in
the world, whether created by a claim, counterclaim, or cross-claim,
alleging patent [-infringement or patent invalidity.-] {+infringement.+} Such actions, proceedings, or procedures
shall include, but not be limited to, lawsuits brought in state or
federal court, binding arbitrations, and administrative actions such as
a proceeding before the International Trade Commission.

[-1.12-]

 {+7.9+} “Licensed Patents” means any and all Patents (a) owned or
 controlled by Licensor; or (b) under which Licensor has the right
 to grant licenses without the consent of or payment to a third party
 (other than an employee inventor).

[-1.13-]

{+7.10+} “Licensed Products and Services” means any products, services
or other activities of a Licensee that practice one or more claims of
one or more Licensed Patents of a [-Licensor, but excluding Foundry Services-]
[-or Products and Clone Products or Services.-]
[--]
[-1.14-] {+Licensor.+}
{++}
{+7.11+} “Licensee” means any individual, corporation, partnership or
other entity exercising rights granted by the Licensor under this License
including all Affiliates of such entity.

[-1.15-]

{+7.12+} “Licensor” means any individual, corporation, partnership or
other entity with the right to grant licenses in Licensed Patents under
this License, including any Affiliates of such entity.

[-1.16-]

 {+7.13+} “Offering Announcement” means a Licensor’s announcement that:

{+(a)+} declares the Licensor’s commitment to offer a [-license to-] {+Defensive Patent+}
{+License for any of+} its Patents
    [-under the DPL, or, if such Licensor has no Patents, the commitment to-]
[-    offer a license-] to any [-Patents it may obtain in the future under the-]
[-    DPL;-] {+DPL User;+} and

{+(b)+} contains the Licensor’s contact information for licensing purposes;
and [-is submitted to the DPL Website via the Website’s-]
[-    official email address; and-]

{+(c)+} is posted to a publicly accessible
    [-indexed website controlled by Licensor using a URL accessible-]
[-    via at least-] {+website.+}
{++}
{+An Offering Announcement may, but is not required to, specify the+}
{+particular version of the DPL that+} the [-following syntax: "http://www.NAME.com/DPL" or-]
[-    "http://www.NAME.com/defensivepatentlicense" where "NAME"-] {+Licensor+} is {+committed to+}
{+offering. It may also specify+} a [-name-]
[-    commonly associated with Licensor, such as a company name.-]
[--]
[-1.17-] {+particular version of the DPL “or any+}
{+later version” to allow Licensees to accept subsequent new or revised+}
{+versions of the DPL.+}
{++}
{+7.14+} “Patent” means any right, whether now or later acquired,
under any national or international patent law issued by a governmental
body authorized to issue such rights. For clarity, this definition
includes any rights that may arise in patent applications, utility
models, granted patents, including, but not limited to, continuations,
continuations-in-part, divisionals, provisionals, results of any patent
reexaminations, and reissues, but excluding design patents or design
registrations.

prioritize(projects, freedom_for_all_computer_users)

Monday, December 8th, 2014

Last week the Free Software Foundation published its annual appeal, which includes the following:

In another 30 years, we believe that we can achieve our goal. We believe that free software can be everywhere, and that proprietary software can go the way of the dinosaur. With the experience we’ve gained, and our community surrounding us, we can win this.

My immediate reaction: I’d love to see the last sentence expanded. How exactly?

Sadly I do not live in a world that laughs at any fundraising appeal lacking an explicit theory of change and only esteems those that one can bet on. At least the FSF has a goal. Perhaps its surrounding community can figure out what it will take to achieve that goal.

Helping “the FSF stay strong for 30 more years” is plainly insufficient, though of course I hope the FSF does stay strong for decades and encourage helping financially. The entire free software movement on its current trajectory is insufficient; some of its staunchest advocates predict a “dark ages” of software freedom (e.g., Bradley Kuhn, Stefano Zacchiroli).

Since 2005 the FSF has published a list of high priority free software projects in order “to foster work on projects that are important for increasing the adoption and use of free software and free software operating systems.”

Today the FSF announced a review of this list. Excerpt:

Undoubtedly there are thousands of free software projects that are high priority, each having potential to displace non-free programs for many users, substantially increasing the freedom of those users. But the potential value of a list of High Priority Free Software Projects maintained by the Free Software Foundation is its ability to bring attention to a relatively small number of projects of great strategic importance to the goal of freedom for all computer users.

[…]

Keep in mind that not every project of great strategic importance to the goal of freedom for all computer users will be a software development project. If you believe other forms of activism, internal or external (e.g., making free software communities safe for diverse participants, mandating use of free software in the public sector), are most crucial, please make the case and suggest such a project!

I hope the announcement text indicates the possibility of exploiting the review and list to encourage debate about how to achieve the FSF’s goal of software freedom for all over the next decades, and that the how might (must, in my view) go far beyond hacking of code (and secondarily, copyright). How can demand for software freedom be both increased and made more effective? Same for supply, inclusive of distribution and marketing?

Send your suggestions to hpp-feedback@gnu.org or better yet post publicly. (I’m on the review committee.)

Because it is undoubtedly out of scope for above activity, I’ll note here a project I consider necessary for FSF’s goal to become plausible: question software freedom.

The “dark ages” links above largely concern “the cloud”, the topic of the other FSF-related committee I’ve participated in, over 6 years ago, correctly implying that effort was not very influential. I hope to post an assessment and summary of my current take on the topic in the near future.

Friends of Software Freedom, Conservationists, Rangers: Support!

Wednesday, December 3rd, 2014

Software Freedom Conservancy, a nonprofit I serve on the board of, today announced a supporter program.

There’s plenty of info at the previous links, so I’ll just add four notes:

  1. Since Karen Sandler started as executive director earlier this year, Conservancy has been doing especially great work for its member projects (and some exciting new ones will be coming on board in the next months; if you lead a free software project, look at info on becoming a member) and is becoming even more the thought leader in both reactionary and progressive (I’m not using those terms to mark ideology here) strategies for software freedom (the italicized modifiers are there because Conservancy was already doing great work, and Sandler had been involved from the very beginning has a volunteer).
  2. Conservancy’s free software non-profit accounting project did not meet its funding goal last year, so progress this year has been slow. It’s still an important project and a successful supporters program will help free up and add resources to move it forward.
  3. Because this is the time of year end appeals (I’ll probably mention a few others in passing shortly), it’s worth noting that Bradley Kuhn (Conservancy’s president and distinguished technologist, but this is one of his many volunteer efforts) is the primary maintainer of a repository of tax filings and other data for free/open organizations. Most of this information can be found somewhere on the web, but not conveniently. Patches welcome (I’d enjoy a dedicated website, and worldwide coverage). Conservancy itself has a page with its filings and a couple explanatory posts.
  4. Didn’t I just a few days ago promote a sort of meta service for free/libre/open projects, Snowdrift.coop? Yes, but it and Conservancy are rather different on multiple dimensions. Snowdrift.coop is a crazy speculative crowdfunding platform. Conservancy is a fiscal sponsor (plus), an old and well understood model that emphasizes administration and stewardship rather than (confusingly to many people, given “fiscal” and “sponsor”) fundraising. Conservancy is able to process donations for member projects (taking a very reasonable 10% cut). If Snowdrift.coop were to become a well-established mechanism for collecting donations to free/libre/open projects such that Conservancy member projects demanded to use the Snowdrift.coop mechanism, I imagine it would be natural for Conservancy to process such funds for its projects. But, this is speculative.

Christopher Allan Webber also says that donating to Conservancy is a real bang for your buck.

copyleft.org

Monday, December 1st, 2014

Last month the Free Software Foundation and Software Freedom Conservancy launched copyleft.org, “a collaborative project to create and disseminate useful information, tutorial material, and new policy ideas regarding all forms of copyleft licensing.” The main feature of the project now is a 157 page tutorial on the GPL which assembles material developed over the past 10 years and a new case study. I agreed to write a first draft of material covering CC-BY-SA, the copyleft license most widely used for non-software works. My quote in the announcement: “I’m glad to bring my knowledge about the Creative Commons copyleft licenses as a contribution to improve further this excellent tutorial text, and I hope that copyleft.org as a whole can more generally become a central location to collect interesting ideas about copyleft policy.”

I tend to offer apologia to copyleft detractors and criticism to copyleft advocates, and cheer whatever improvements to copyleft licenses can be mustered (I hope to eventually write a cheery post about the recent compatibility of CC-BY-SA and the Free Art License), but I’m far more interested in copyleft licenses as prototypes for non-copyright policy.

For now, below is that first draft. It mostly stands alone, but might be merged in pieces as the tutorial is restructured to integrate material about non-GPL and non-software copyleft licenses. Your patches and total rewrites welcome!


Detailed Analysis of the Creative Commons Attribution-ShareAlike Licenses

This tutorial gives a comprehensive explanation of the most popular free-as-in-freedom copyright licenses for non-software works, the Creative Commons Attribution-ShareAlike (“CC-BY-SA”, or sometimes just “BY-SA”) – with an emphasis on the current version 4.0 (“CC-BY-SA-4.0”).

Upon completion of this part of the tutorial, readers can expect to have learned the following:

  • The history and role of copyleft licenses for non-software works.
  • The differences between the GPL and CC-BY-SA, especially with respect to copyleft policy.
  • The basic differences between CC-BY-SA versions 1.0, 2.0, 2.5, and 4.0.
  • An understanding of how CC-BY-SA-4.0 implements copyleft.
  • Where to find more resources about CC-BY-SA compliance.

FIXME this list should be more aggressive, but material is not yet present

WARNING: As of November 2014 this part is brand new, and badly needs review, referencing, expansion, error correction, and more.

Freedom as in Free Culture, Documentation, Education…

Critiques of copyright’s role in concentrating power over and making culture inaccessible have existed throughout the history of copyright. Few contemporary arguments about “copyright in the digital age” have not already been made in the 1800s or before. Though one can find the occasional ad hoc “anti-copyright”, “no rights reserved”, or pro-sharing statement accompanying a publication, use of formalized public licenses for non-software works seems to have begun only after the birth of the free software movement and of widespread internet access among elite populations.

Although they have much older antecedents, contemporary movements to create, share, and develop policy encouraging “cultural commons”, “open educational resources”, “open access scientific publication” and more, have all come of age in the last 10-15 years – after the huge impact of free software was unmistakable. Additionally, these movements have tended to emphasize access, with permissions corresponding to the four freedoms of free software and the use of fully free public licenses as good but optional.

It’s hard not to observe that it seems the free software movement arose more or less shortly after as it became desirable (due to changes in the computing industry and software becoming unambiguously subject to copyright in the United States by 1983), but non-software movements for free-as-in-freedom knowledge only arose after they became more or less inevitable, and only begrudgingly at that. Had a free culture “constructed commons” movement been successful prior to the birth of free software, the benefits to computing would have been great – consider the burdens of privileged access to proprietary culture for proprietary software through DRM and other mechanisms, toll access to computer science literature, and development of legal mechanisms and policy through pioneering trial-and-error.

Alas, counterfactual optimism does not change the present – but might embolden our visions of what freedom can be obtained and defended going forward. Copyleft policy will surely continue to be an important and controversial factor, so it’s worth exploring the current version of the most popular copyleft license intended for use with non-software works, Creative Commons Attribution-ShareAlike 4.0 International (CC-BY-SA-4.0), the focus of this tutorial.

Free Definitions

When used to filter licenses, the Free Software Definition and Open Source Definition have nearly identical results. For licenses primarily intended for non-software works, the Definition of Free Cultural Works and Open Definition similarly have identical results, both with each other and with the software definitions which they imitate. All copyleft licenses for non-software works must be “free” and “open” per these definitions.

There are various other definitions of “open access”, “open content”, and “open educational resources” which are more subject to interpretation or do not firmly require the equivalent of all four freedoms of the free software definition. While these definitions are not pertinent to circumscribing the concept of copyleft – which is about enforcing all four freedoms, for everyone. But copyleft licenses for non-software works are usually considered “open” per these other definitions, if they are considered at all.

The open access to scientific literature movement, for example, seems to have settled into advocacy for non-copyleft free licenses (CC-BY) on one hand, and acceptance of highly restrictive licenses or access without other permissions on the other. This creates practical problems: for example, nearly all scientific literature either may not be incorporated into Wikipedia (which uses CC-BY-SA) or may not incorporate material developed on Wikipedia – both of which do happen, when the licenses allow it. This tutorial is not the place to propose solutions, but let this problem be a motivator for encouraging more widespread understanding of copyleft policy.

Non-software Copylefts

Copyleft is a compelling concept, so unsurprisingly there have been many attempts to apply it to non-software works – starting with use of GPLv2 for documentation, then occasionally for other texts, and art in various media. Although the GPL was and is perfectly usable for any work subject to copyright, several factors were probably important in preventing it from being the dominant copyleft outside of software:

  • the GPL is clearly intended first as a software license, thus requiring some perspective to think of applying to non-software works;
  • the FSF’s concern is software, and the organization has not strongly advocated for using the GPL for non-software works;
  • further due to the (now previous) importance of its hardcopy publishing business and desire to retain the ability to take legal action against people who might modify its statements of opinion, FSF even developed a non-GPL copyleft license specifically for documentation, the Free Documentation License (FDL; which ceases to be free and thus is not a copyleft if its “invariant sections” and similar features are used);
  • a large cultural gap and lack of population overlap between free software and other movements has limited knowledge transfer and abetted reinvention and relearning;
  • the question of what constitutes source (“preferred form of the work for making modifications”) for many non-software works.

As a result, several copyleft licenses for non-software works were developed, even prior to the existence of Creative Commons. These include the aforementioned FDL (1998), Design Science License (1999), Open Publication License (1999; like the FDL it has non-free options), Free Art License (2000), Open Game License (2000; non-free options), EFF Open Audio License (2001), LinuxTag Green OpenMusic License (2001; non-free options) and the QING Public License (2002). Additionally several copyleft licenses intended for hardware designs were proposed starting in the late 1990s if not sooner (the GPL was then and is now also commonly used for hardware designs, as is now CC-BY-SA).1

At the end of 2002 Creative Commons launched with 11 1.0 licenses and a public domain dedication. The 11 licenses consisted of every non-mutually exclusive combination of at least one of the Attribution (BY), NoDerivatives (ND), NonCommercial (NC), and ShareAlike (SA) conditions (ND and SA are mutually exclusive; NC and ND are non-free). Three of those licenses were free (as was the public domain dedication), two of them copyleft: CC-SA-1.0 and CC-BY-SA-1.0.

Creative Commons licenses with the BY condition were more popular, so the 5 without (including CC-SA) were not included in version 2.0 of the licenses. Although CC-SA had some advocates, all who felt very strongly in favor of free-as-in-freedom, its incompatibility with CC-BY-SA (meaning had CC-SA been widely used, the copyleft pool of works would have been further fragmented) and general feeling that Creative Commons had created too many licenses led copyleft advocates who hoped to leverage Creative Commons to focus on CC-BY-SA.

Creative Commons began with a small amount of funding and notoriety, but its predecessors had almost none (FSF and EFF had both, but their entries were not major focuses of those organizations), so Creative Commons licenses (copyleft and non-copyleft, free and non-free) quickly came to dominate the non-software public licensing space. The author of the Open Publication License came to recommend using Creative Commons licenses, and the EFF declared version 2.0 of the Open Audio License compatible with CC-BY-SA and suggested using the latter. Still, at least one copyleft license for “creative” works was released after Creative Commons launched: the Against DRM License (2006), though it did not achieve wide adoption. Finally a font-specific copyleft license (SIL Open Font License) was introduced in 2005 (again the GPL, with a “font exception”, was and is now also used for fonts).

Although CC-BY-SA was used for licensing “databases” almost from its launch, and still is, copyleft licenses specifically intended to be used for databases were proposed starting from the mid-2000s. The most prominent of those is the Open Database License (ODbL; 2009). As we can see public software licenses following the subjection of software to copyright, interest in public licenses for databases followed the EU database directive mandating “sui generis database rights”, which began to be implemented in member state law starting from 1998. How CC-BY-SA versions address databases is covered below.

Aside on share-alike non-free therefore non-copylefts

Many licenses intended for use with non-software works include the “share-alike” aspect of copyleft: if adaptations are distributed, to comply with the license they must be offered under the same terms. But some (excluding those discussed above) do not grant users the equivalent of all four software freedoms. Such licenses aren’t true copylefts, as they retain a prominent exclusive property right aspect for purposes other than enforcing all four freedoms for everyone. What these licenses create are “semicommons” or mixed private property/commons regimes, as opposed to the commons created by all free licenses, and protected by copyleft licenses. One reason non-free public licenses might be common outside software, but rare for software, is that software more obviously requires ongoing maintenance.2 Without control concentrated through copyright assignment or highly asymmetric contributor license agreements, multi-contributor maintenance quickly creates an “anticommons” – e.g., nobody has adequate rights to use commercially.

These non-free share-alike licenses often aggravate freedom and copyleft advocates as the licenses sound attractive, but typically are confusing, probably do not help and perhaps stymie the cause of freedom. There is an argument that non-free licenses offer conservative artists, publishers, and others the opportunity to take baby steps, and perhaps support better policy when they realize total control is not optimal, or to eventually migrate to free licenses. Unfortunately no rigorous analysis of any of these conjectures exists. The best that can be done might be to promote education about and effective use of free copyleft licenses (as this tutorial aims to do) such that conjectures about the impact of non-free licenses become about as interesting as the precise terms of proprietary software EULAs – demand freedom instead.

In any case, some of these non-free share-alike licenses (also watch out for aforementioned copyleft licenses with non-free and thus non-copyleft options) include: Open Content License (1998), Free Music Public License (2001), LinuxTag Yellow, Red, and Rainbow OpenMusic Licenses (2001), Open Source Music License (2002), Creative Commons NonCommercial-ShareAlike and Attribution-NonCommercial-ShareAlike Licenses (2002), Common Good Public License (2003), and Peer Production License (2013). CC-BY-NC-SA is by far the most widespread of these, and has been versioned with the other Creative Commons licenses, through the current version 4.0 (2013).

Creative Commons Attribution-ShareAlike

The remainder of this tutorial exclusively concerns the most widespread copyleft license intended for non-software works, Creative Commons Attribution-ShareAlike(CC-BY-SA). But, there are actually many CC-BY-SA licenses – 5 versions (6 if you count version 2.1, a bugfix for a few jurisdiction “porting” mistakes), ports to 60 jurisdictions – 96 distinct CC-BY-SA licenses in total. After describing CC-BY-SA and how it differs from the GPL at a high level, we’ll have an overview of the various CC-BY-SA licenses, then a section-by-section walkthrough of the most current and most clear of them – CC-BY-SA-4.0.

CC-BY-SA allows anyone to share and adapt licensed material, for any purpose, subject to providing credit and releasing adaptations under the same terms. The preceding sentence is a severe abridgement of the “human readable” license summary or “deed” provided by Creative Commons at the canonical URL for one of the CC-BY-SA licenses – the actual license or “legalcode” is a click away. But this abridgement, and the longer the summary provided by Creative Commons are accurate in that they convey CC-BY-SA is a free, copyleft license.

GPL and CC-BY-SA differences

FIXME this section ought refernence GPL portion of tutorial extensively

There are several differences between the GPL and CC-BY-SA that are particularly pertinent to their analysis as copyleft licenses.

The most obvious such difference is that CC-BY-SA does not require offering works in source form, that is their preferred form for making modifications. Thus CC-BY-SA makes a huge tradeoff relative to the GPL – CC-BY-SA dispenses with a whole class of compliance questions which are more ambiguous for some creative works than they are for most software – but in so doing it can be seen as a much weaker copyleft.

Copyleft is sometimes described as a “hack” or “judo move” on copyright, but the GPL makes two moves, though it can be hard to notice they are conceptually different moves, without the contrast provided by a license like CC-BY-SA, which only substantially makes one move. The first move is to neutralize copyright restrictions – adaptations, like the originally licensed work, will effectively not be private property (of course they are subject to copyright, but nobody can exercise that copyright to prevent others’ use). If copyright is a privatized regulatory system (it is), the first move is deregulatory. The second move is regulatory – the GPL requires offer of source form, a requirement that would not hold if copyright disappeared, absent a different regulatory regime which mandated source revelation (one can imagine such a regime on either “pragmatic” grounds, e.g., in the interest of consumer protection, or on the grounds of enforcing software freedom as a universal human right).

FIXME analysis of differences in copyleft scope (eg interplay of derivative works, modified copies, collections, aggregations, containers) would be good here but might be difficult to avoid novel research

CC-BY-SA makes the first move3 but adds the second in a limited fashion. It does not require offer of preferred form for modification nor any variation thereof (e.g., the FDL requires access to a “transparent copy”). CC-BY-SA does prohibit distribution with “effective technical measures” (i.e., digital restrictions management or DRM) if doing so limits the freedoms granted by the license. We can see that this is regulatory because absent copyright and any regime specifically limiting DRM, such distribution would be perfectly legal. Note the GPL does not prohibit distribution with DRM, although its source requirement makes DRM superfluous, and somewhat analogously, of course GPLv3 carefully regulates distribution of GPL’d software with locked-down devices – to put it simply, it requires keys rather than prohibiting locks. Occasionally a freedom advocate will question whether CC-BY-SA’s DRM prohibition makes CC-BY-SA a non-free license. Few if any questioners come down on the side of CC-BY-SA being non-free, perhaps for two reasons: first, overwhelming dislike of DRM, thus granting the possibility that CC-BY-SA’s approach could be appropriate for a license largely used for cultural works; second, the DRM prohibition in CC-BY-SA (and all CC licenses) seems to be mainly expressive – there are no known enforcements, despite the ubiquity of DRM in games, apps, and media which utilize assets under various CC licenses.

Another obvious difference between the GPL and CC-BY-SA is that the former is primarily intended to be used for software, and the latter for cultural works (and, with version 4.0, databases). Although those are the overwhelming majority of uses of each license, there are areas in which both are used, e.g., for hardware design and interactive cultural works, where there is not a dominant copyleft practice or the line between software and non-software is not absolutely clear.

This brings us to the third obvious difference, and provides a reason to mitigate it: the GPL and CC-BY-SA are not compatible, and have slightly different compatibility mechanisms. One cannot mix GPL and CC-BY-SA works in a way that creates a derivative work and comply with either of them. This could change – CC-BY-SA-4.0 introduced4 the possibility of Creative Commons declaring CC-BY-SA-4.0 one-way (as a donor) compatible with another copyleft license – the GPL is obvious candidate for such compatibility. Discussion is expected to begin in late 2014, with a decision sometime in 2015. If this one-way compatibility were to be enacted, one could create an adaptation of a CC-BY-SA work and release the adaptation under the GPL, but not vice-versa – which makes sense given that the GPL is the stronger copyleft.

The GPL has no externally declared compatibility with other licenses mechanism (and note no action from the FSF would be required for CC-BY-SA-4.0 to be made one-way compatible with the GPL). The GPL’s compatibility mechanism for later versions of itself differs from CC-BY-SA’s in two ways: the GPL’s is optional, and allows for use of the licensed work and adaptations under later versions; CC-BY-SA’s is non-optional, but only allows for adaptations under later versions.

Fourth, using slightly different language, the GPL and CC-BY-SA’s coverage of copyright and similar restrictions should be identical for all intents and purposes (GPL explicitly notes “semiconductor mask rights” and CC-BY-SA-4.0 “database rights” but neither excludes any copyright-like restrictions). But on patents, the licenses are rather different. CC-BY-SA-4.0 explicitly does not grant any patent license, while previous versions were silent. GPLv3 has an explicit patent license, while GPLv2’s patent license is implied (see [gpl-implied-patent-grant] and [GPLv3-drm] for details). This difference ought give serious pause to anyone considering use of CC-BY-SA for works potentially subject to patents, especially any potential licensee if CC-BY-SA licensor holds such patents. Fortunately Creative Commons has always strongly advised against using any of its licenses for software, and that advice is usually heeded; but in the space of hardware designs Creative Commons has been silent, and unfortunately from a copyleft (i.e., use mechanisms at disposal to enforce user freedom) perspective, CC-BY-SA is commonly used (all the more reason to enable one-way compatibility, allowing such projects to migrate to the stronger copyleft).

The final obvious difference pertinent to copyleft policy between the GPL and CC-BY-SA is purpose. The GPL’s preamble makes it clear its goal is to guarantee software freedom for all users, and even without the preamble, it is clear that this is the Free Software Foundation’s driving goal. CC-BY-SA (and other CC licenses) state no purpose, and (depending on version) are preceded with a disclaimer and neutral “considerations for” licensors and licensees to think about (the CC0 public domain dedication is somewhat of an exception; it does have a statement of purpose, but even that has more of a feel of expressing yes-I-really-mean-to-do-this than a social mission). Creative Commons has always included elements of merely offering copyright holders additional choices and of purposefully creating a commons. While CC-BY-SA (and initially CC-SA) were just among the 11 non-mutually exclusive combinations of “BY”, “NC”, “ND”, and “SA”, freedom advocates quickly adopted CC-BY-SA as “the” copyleft for non-software works (surpassing previously existing non-software copylefts mentioned above). Creative Commons has at times recognized the special role of CC-BY-SA among its licenses, e.g., in a statement of intent regarding the license made in order to assure Wikimedians considering changing their default license from the FDL to CC-BY-SA that the latter, including its steward, was acceptably aligned with the Wikimedia movement (itself probably more directly aligned with software freedom than any other major non-software commons).

FIXME possibly explain why purpose might be relevant, eg copyleft instrument as totemic expression, norm-setting, idea-spreading

FIXME possibly mention that CC-BY-SA license text is free (CC0)

There are numerous other differences between the GPL and CC-BY-SA that are not particularly interesting for copyleft policy, such as the exact form of attribution and notice, and how license translations are handled. Many of these have changed over the course of CC-BY-SA versioning.

CC-BY-SA versions

FIXME section ought explain jurisdiction ports

This section gives a brief overview of changes across the main versions (1.0, 2.0, 2.5, 3.0, and 4.0) of CC-BY-SA, again focused on changes pertinent to copyleft policy. Creative Commons maintains a page detailing all significant changes across versions of all of its CC-BY* licenses, in many cases linking to detailed discussion of individual changes.

As of late 2014, versions 2.0 (the one called “Generic”; there are also 18 jurisdiction ports) and 3.0 (called “Unported”; there are also 39 ports) are by far the most widely used. 2.0 solely because it is the only version that the proprietary web image publishing service Flickr has ever supported. It hosts 27 million CC-BY-SA-2.0 photos 5 and remains the go-to general source for free images (though it may eventually be supplanted by Wikimedia Commons, some new proprietary service, or a federation of free image sharing sites, perhaps powered by GNU MediaGlobin). 3.0 both because it was the current version far longer (2007-2013) than any other and because it has been adopted as the default license for most Wikimedia projects.

However apart from the brief notes on each version, we will focus on 4.0 for a section-by-section walkthrough in the next section, as 4.0 is improved in several ways, including understandability, and should eventually become the most widespread version, both because 4.0 is intended to remain the current version for the indefinite and long future, and it would be reasonable to predict that Wikimedia projects will make CC-BY-SA-4.0 their default license in 2015 or 2016.

FIXME subsections might not be the right strcuture or formatting here

1.0 (2002-12-16)

CC-BY-SA-1.0 set the expectation for future versions. But the most notable copyleft policy feature (apart from the high level differences with GPLv2, such as not requiring source) was no measure for compatibility with future versions (nor with the CC-SA-1.0, also a copyleft license, nor with pre-existing copyleft licenses such as GPL, FDL, FAL, and others, nor with CC jurisdiction ports, of which there were 3 for 1.0).

2.0 (2004-05-25)

CC-BY-SA-2.0 made itself compatible with future versions and CC jurisdiction ports of the same version. Creative Commons did not version CC-SA, leaving CC-BY-SA-2.0 as “the” CC copyleft license. CC-BY-SA-2.0 also adds the only clarification of what constitutes a derivative work, making “synchronization of the Work in timed-relation with a moving image” subject to copyleft.

2.5 (2005-06-09)

CC-BY-SA-2.5 makes only one change, to allow licensor to designate another party to receive attribution. This does not seem interesting for copyleft policy, but the context of the change is: it was promoted by the desire to make attribution of mass collaborations easy (and on the other end of the spectrum, to make it possible to clearly require giving attribution to a publisher, e.g., of a journal). There was a brief experiment in branding CC-BY-SA-2.5 as the “CC-wiki” license. This was an early step toward Wikimedia adopting CC-BY-SA-3.0, four years later.

3.0 (2007-02-23)

CC-BY-SA-3.0 introduced a mechanism for externally declaring bilateral compatibility with other licenses. This mechanism to date has not been used for CC-BY-SA-3.0, in part because another way was found for Wikimedia projects to change their default license from FDL to CC-BY-SA: the Free Software Foundation released FDL 1.3, which gave a time-bound permission for mass collaboration sites to migrate to CC-BY-SA. While not particularly pertinent to copyleft policy, it’s worth noting for anyone wishing to study old versions in depth that 3.0 is the first version to substantially alter the text of most of the license, motivated largely by making the text use less U.S.-centric legal language. The 3.0 text is also considerably longer than previous versions.

4.0 (2013-11-25)

CC-BY-SA-4.0 added to 3.0’s external compatibility declaration mechanism by allowing one-way compatibility. After release of CC-BY-SA-4.0 bilateral compatibility was reached with FAL-1.3. As previously mentioned, one-way compatibility with GPLv3 will soon be discussed.

4.0 also made a subtle change in that an adaptation may be considered to be licensed solely under the adapter’s license (currently CC-BY-SA-4.0 or FAL-1.3, in the future potentially GPLv3 or or a hypothetical CC-BY-SA-5.0). In previous versions licenses were deemed to “stack” – if a work under CC-BY-SA-2.0 were adapted and released under CC-BY-SA-3.0, users of the adaptation would need to comply with both licenses. In practice this is an academic distinction, as compliance with any compatible license would tend to mean compliance with the original license. But for a licensee using a large number of works that wished to be extremely rigorous, this would be a large burden, for it would mean understanding every license (including those of jurisdiction ports not in English) in detail.

The new version is also an even more complete rewrite of 3.0 than 3.0 was of previous versions, completing the “internationalization” of the license, and actually decreasing in length and increasing in readability.

Additionally, 4.0 consistently treats database (licensing them like other copyright-like rights) and moral rights (waiving them to the extent necessary to exercise granted freedoms) – in previous versions some jurisdiction ports treated these differently – and tentatively eliminates the need for jurisdiction ports. Official linguistic translations are underway (Finnish is the first completed) and no legal ports are planned for.

4.0 is the first version to explicitly exclude a patent (and less problematically, trademark) license. It also adds two features akin to those found in GPLv3: waiver of any right licensor may have to enforce anti-circumvention if DRM is applied to the work, and reinstatement of rights after termination if non-compliance corrected within 30 days.

Finally, 4.0 streamlines the attribution requirement, possibly of some advantage to massive long-term collaborations which historically have found copyleft licenses a good fit.

The 4.0 versioning process was much more extensively researched, public, and documented than previous CC-BY-SA versionings; see https://wiki.creativecommons.org/4.0 for the record and https://wiki.creativecommons.org/Version_4 for a summary of final decisions.

CC-BY-SA-4.0 International section-by-section

FIXME arguably this section ought be the substance of the tutorial, but is very thin and weak now

FIXME formatted/section-referenced copy of license should be added to license-texts.tex and referenced throughout

The best course of action at this juncture would be to read http://creativecommons.org/licenses/by-sa/4.0/legalcode – the entire text is fairly easy to read, and should be quickly understood if one has the benefit of study of other public licenses and of copyleft policy.

The following walk-through will simply call out portions of each section one may wish to study especially closely due to their pertinence to copyleft policy issues mentioned above.

FIXME subsections might not be the right structure or formatting here

1 – Definitions

The first three definitions – “Adapted Material”, “Adapter’s License”, and “BY-SA Compatible License” are crucial to understanding copyleft scope and compatibility.

2 – Scope

The license grant is what makes all four freedoms available to licensees. This section is also where waiver of DRM anti-circumvention is to be found, also patent and trademark exclusions.

3 – License Conditions

This section contains the details of the attribution and share-alike requirements; the latter read closely with aforementioned definitions describe the copyleft aspect of CC-BY-SA-4.0.

4 – Sui Generis Database Rights

This section describes how the previous grant and condition sections apply in the case of a database subject to sui generis database rights. This is an opportunity to go down a rabbit-hole of trying to understand sui generis database rights. Generally, this is a pointless exercise. You can comply with the license in the same way you would if the work were subject only to copyright – and determining whether a database is subject to copyright and/or sui generis database rights is another pit of futility. You can license databases under CC-BY-SA-4.0 and use databases subject to the same license as if they were any other sort of work.

5 – Disclaimer of Warranties and Limitation of Liability

Unsurprisingly, this section does its best to serve as an “absolute disclaimer and waiver of all liability.”

6 – Term and Termination

This section is similar to GPLv3, but without special provision for cases in which the licensor wishes to terminate even cured violations.

7 – Other Terms and Conditions

Though it uses different language, like the GPL, CC-BY-SA-4.0 does not allow additional restrictions not contained in the license. Unlike the GPL, CC-BY-SA-4.0 does not have an explicit additional permissions framework, although effectively a licensor can offer any other terms if they are the sole copyright holder (the license is non-exclusive), including the sorts of permissions that would be structured as additional permissions with the GPL. Creative Commons has sometimes called offering of separate terms (whether additional permissions or “proprietary relicensing”) the confusing name “CC+”; however where this is encountered at all it is usually in conjunction with one of the non-free CC licenses. Perhaps CC-BY-SA is not a strong enough copyleft to sometimes require additional permissions, or be used to gain commercially valuable asymmetric rights, in contrast with the GPL.

8 – Interpretation

Nothing surprising here. Note that CC-BY-SA does not “reduce, limit, restrict, or impose conditions on any use of the Licensed Material that could lawfully be made without permission under this Public License.” This is a point that Creative Commons has always been eager to make about all of its licenses. GPLv3 also “acknowledges your rights of fair use or other equivalent”. This may be a wise strategy, but should not be viewed as mandatory for any copyleft license – indeed, the ODbL attempts (somewhat self-contradictorily; it also acknowledges fair use or other rights to use) make its conditions apply even for works potentially subject to neither copyright nor sui generis database rights.

Enforcement

There are only a small number of court cases involving any Creative Commons license. Creative Commons lists these and some related cases at https://wiki.creativecommons.org/Case_Law.

Only two of those cases concern enforcing the terms of a CC-BY-SA license (Gerlach v. DVU in Germany, and No. 71036 N. v. Newspaper in a private Rabbinical tribunal) each hinged on attribution, not share-alike.

Further research could uncover out of compliance uses being brought into compliance without lawsuit, however no such research, nor any hub for conducting such compliance work, is known. Editors of Wikimedia Commons document some external uses of Commons-hosted media, including whether user are compliant with the relevant license for the media (often CC-BY-SA), resulting in a category listing non-compliant uses (which seem to almost exclusively concern attribution).

Compliance Resources

FIXME this section is just a stub; ideally there would also be an additional section or chapter on CC-BY-SA compliance

Creative Commons has a page on ShareAlike interpretation as well as an extensive Frequently Asked Questions for licensees which addresses compliance with the attribution condition.

English Wikipedia’s and Wikimedia Commons’ pages on using material outside of Wikimedia projects provide valuable information, as the majority of material on those sites is CC-BY-SA licensed, and their practices are high-profile.

FIXME there is no section on business use of CC-BY-SA; there probably ought to be as there is one for GPL, though there’d be much less to put.