Archive for November, 2013

Software Patent NATO, 1993

Tuesday, November 26th, 2013

In my thoughts on the Defensive Patent License, I neglected to note in the history section a similar proposal made in 1993 by John Walker, founder of Autodesk, PATO: Collective Security In the Age of Software Patents:

[T]he trend toward increased litigation, constraining innovation in the software industry, is accelerating. The U.S. government is using trade negotiations to force other countries to institute software patents in their own markets.

While eliminating software patents would be the best solution, changing the law takes a long time and is uncertain to succeed. I’ve been trying to puzzle out how the software industry might rescue itself from immolation through litigation and came up with the following proposal.

Could have been written in 2013.

I’ve been thinking about using NATO as a model of a patent defence consortium. Suppose a bunch of big software companies (perhaps led by Oracle, who’s already taken the point on this) were to form PATO–Patent And Technology Organisation–and contribute all their current software patents, and all new software patents they were granted as long as they remained a member of PATO, to its “cross-licensing pool”. To keep the lawyers and shareholders from going nuts, the patents would be licensed through PATO but would remain the property of the member–a member could withdraw with appropriate notice and take the patents back from the pool.

Any member of PATO would be granted an automatic, royalty-free license to use any patent in the cross-licensing pool. Thus, by putting your patents in the pool, you obtain access to all the others automatically (but if you withdraw and pull your patents, of course you then become vulnerable for those you’ve used, which creates a powerful disincentive to quit).

The basic principle of NATO is that an attack on any member is considered an attack on all members. In PATO it works like this–if any member of PATO is alleged with infringement of a software patent by a non-member, then that member may counter-sue the attacker based on infringement of any patent in the PATO cross-licensing pool, regardless of what member contributed it. Once a load of companies and patents are in the pool, this will be a deterrent equivalent to a couple thousand MIRVs in silos–odds are that any potential plaintiff will be more vulnerable to 10 or 20 PATO patents than the PATO member is to one patent from the aggressor. Perhaps the suit will just be dropped and the bad guy will decide to join PATO….

Differences with the DPL, two decades hence:

  • PATO was to cover software patents only; a challenge to define.
  • PATO members could counter-sue attackers with patents from any other member; I have no idea whether this is legally feasible.
  • PATO never moved beyond raw idea stage, as far as I know, while legal work on the DPL has gone on for a few years, DPL 1.0 is complete, and the project is set for a public launch in February.

In 1993, software patents were new, and still opposed by Oracle and Microsoft. Since then both have become software patent aggressors and defend the idea of software patents.

Many companies that claim to dislike software patent aggression in 2013 will become aggressors over the next years, or their patents will be obtained and used by trolls and other aggressors. Becoming a DPL user now may be an effective way for such companies to avoid this fate, and avoid contributing to the stifling of equality, freedom, and innovation.

Addendum 20131202: Another difference between the PATO sketch and the DPL implementation is that the former includes “US$25/year” to be a member, while the latter is gratis. I assume that the nascent DPL Foundation will be able to attract adequate grants and other support, perhaps more than could be obtained through a membership fee, but the choice is at the least an interesting and important one.

Upgrade to CC-BY(-(NC(-(ND|SA))?|ND|SA))?-4\.0

Monday, November 25th, 2013

Today Creative Commons released version 4.0 of six* of its licenses, with many improvements over version 3.0, after more than two years of work. I’ll write more about those details later. But you should skip right past 4.0 and upgrade to CC’s premier legal product, CC0. This is the case whether you’re looking to adopt a CC license for the first time, or to upgrade from version 1.0, 2.0, 2.1, 2.5, or 3.0.

Let’s review the named conditions present in some or all of the CC 4.0 licenses, and why unconditional CC0 is better.

Don’t forget unmitigated © in the basement.

Attribution (BY). Do not take part in the debasement of attribution, and more broadly, provenance, already useful to readers, communities of practice, and publishers, by making them seem mere objects of copyright license compliance. If attribution is useful, it will be provided. If not, robots will find out. Rarely does anyone comply with the exact legal requirements of the attribution term anyway, and as a licensor, you probably won’t provide the information needed by licensees to easily comply. Plus, the corresponding icon looks like a men’s bathroom sign.

NonCommercial (NC). Sounds nice, but nobody knows what it means. Perhaps this goes some way to explaining why NC licensed works are often used by for-profit entities, including with advertising, while NC licensed works are verboten for many community and non-profit projects, most prominently Wikipedia and other Wikimedia projects. (Because commercial entities know there is very low risk of being sued for non-compliance, and can manage risk, while community projects tend to draw and follow bright lines. Perhaps community projects ought to be able to manage risk, and that they can’t is a demonstration of their relative lack of institutional sophistication…but that’s another topic!)

NoDerivatives (ND). This term has no business being in the “Creative Commons” license suite, but sadly still is. If you don’t want to contribute to a creative commons, don’t. If you’d like to, but think copyright (through withholding permission to share adaptations, i.e., the ND term) will prevent people from misrepresenting you, you’re wrong, committing an act of hate toward free speech, and undermining the potential of voluntary license practice to align with and support an obvious baseline objective for copyright reform: noncommercial sharing and remix should always be legal.

ShareAlike (SA). Also sounds nice, and I am a frequent apologist and sometime advocate for the underlying idea, copyleft. But SA is a weak implementation of copyleft. It isn’t “triggered” by the most common use of CC-licensed material (contextual illustration, not full remix), and it has no regulatory condition not present in non-SA CC licenses (cf GPL, which requires sharing source for a work, and is usable for any work; if you care about copyleft, tell CC to finish making CC-BY-SA one-way compatible with GPL). And the SA implementation retains the costs of copyleft: blank stares of incomprehension, even from people who have worked in the “open” world for over a decade, and occasionally intense fear and dislike (the balance is a bit different in the software world, but this is my direct experience among non-software putatively open organizations and people); also, compatibility problems. It’s time to take the unsolicited advice often given to incumbents and others fearful of the internet: ‘obscurity is a greater threat than piracy’ — and apply it: ‘obscurity is a greater threat than proprietarization.’


Upgrade to CC0!

CC0 isn’t perfect, but it is by far the best tool provided by CC. I have zero insight into the future of the CC organization, but I hope it gives ample priority to the public domain, post-4.0 launch.

*CC-BY(-(NC(-(ND|SA))?|ND|SA))?-4\.0 is a regular expression matching all six licenses released today.

Hierarchy of mechanisms for limiting copyright and copyright-like barriers to use of Public Sector Information, or More or Less Universal Government License(s)

Sunday, November 24th, 2013

This sketch is in part motivated by a massive proliferation of copyright and copyright-like licenses for government/public sector information, e.g., sub- and sub-sub-national jurisdiction licenses and sector- and jurisdiction-specific licenses intended to combat license proliferation within a sector within a jurisdiction. Also by longstanding concern about coordination among entities working to limit barriers to use of PSI and knowledge commons governance generally.

Everything following concerns PSI only relative to copyright and copyright-like barriers. There are other pertinent regulations and considerations to follow when publishing or using PSI (e.g., privacy and fraud; as these are pertinent even without copyright, it is silly and unnecessarily complicating to include them in copyright licenses) and other important ways to make PSI more useful technically and politically (e.g., open formats, focusing on PSI that facilitates accountability rather than openwashing).

Eliminate copyright and copyright-like restrictions

No longer barriers to use of PSI, because no longer barriers to use of information. May be modulated down to any general copyright or copyright-like barrier reduction, where the barrier is pertinent to use of PSI. Examples: eliminate sui generis database restrictions where they exist, increase threshold of originality required for information to be subject to copyright restriction, expand exceptions and limitations to copyright restrictions, expand affirmative user rights.

Eliminate copyright and copyright-like restrictions for PSI

For example, works produced by employees of the U.S. federal government are not subject to copyright restrictions in the U.S. Narrower exclusions from copyright restrictions (e.g., of laws, court rulings) are fairly common worldwide. These could be generalized to include eliminate copyright and copyright-like restrictions for PSI, worldwide, and expanded to include PSI produced by contractors or other non-government but publicly funded entities. PSI could be expanded to include any information produced with public funding, e.g., research and culture funded by public grants.

“Standard” international licenses for PSI

Public copyright licenses not specifically intended for only PSI are often used for PSI, and could be more. CC0 is by far the best such license, but other Creative Commons (CC) and Open Data Commons (ODC) licenses are frequently used. Depending on the extent to which the licenses used leave copyright and copyright-like restrictions in place (e.g., CC0: none; CC-BY-NC-ND, lots, thus considered non-open) and how they are applied (from legislative mandate for all PSI to one-off use for individual reports and datasets at discretion of agency), could have effect similar to eliminating copyright and copyright-like restrictions for PSI, or almost zero effect.

Universal Government License

Governments at various levels have chosen to make up their own licenses rather than use a standard international license. Some of the better reasons for doing so will be eliminated by the forthcoming version 4.0 of 6 of the CC licenses (though again, CC0 has been the best choice, since 2009, and will remain so). But some of the less good reasons (uncharitable characterization: vanity) can’t be addressed by a standard international license, and furthermore seem to be driving the proliferation of sub-sub-national licenses, down to licenses specific to an individual town.

Ideally this extreme license proliferation trend would terminate with mass implementation of one of the above options, though this seems unlikely in the short term. Maybe yet another standard license would help! The idea of an “open government license” which various governments would have a direct role in creating and stewarding has been casually discussed in the past, particularly several years ago when the current proliferation was just beginning, the CC 4.0 effort had not begun, and CC and ODC were not on the same page. Nobody is particularly incented to make this unwieldy project happen, but nor is it an impossibility — due to the relatively small world of NGOs (such as CC and the Open Knowledge Foundation, of which ODC is a project) and government people who really care and know about public licenses, and the possibility their collective exhaustion and exasperation over license details, incompatibility, and proliferation could reach a tipping point into collective action. There’s a lot to start from, including the research that went into CC-BY-4.0, and the OGL UK 2.0, which is a pretty good open license.

But why think small? How many other problems could be addressed simultaneously?

  • Defend the traditional meaning of ‘open government’ by calling the license something else, e.g., Universal/Uniform/Unified Government License.
  • Rallying point for public sector worldwide to commit more firmly and broadly to limiting copyright and copyright-like barriers to use of PSI, more rapidly establishing global norm, and leading to mandates. The one thing to be said for massive PSI license proliferation could be increased commitment from proliferating jurisdictions to use their custom licenses (I know of no data on this). A successful UGL would swamp any increased local commitment due to local vanity licenses through much higher level expectation and mandate.
  • Make the license work well for software (including being approved by the Open Source Initiative), as:
    • Generically “open” licenses are inevitably used for software, whether the steward apparently intends this (OGL UK 2.0) or does not (CC).
    • The best modern permissive license for software (Apache 2.0) is relatively long and unreadable for what it does, and has an discomfiting name (not nearly as bad as certain pro sports organizations, but still); it ought be superseded.
  • Ensure the license works for other domains, e.g., open hardware, which don’t really require domain-specific licenses, are headed down the path of proliferation and incompatibility, and that governments have obvious efficiency, regulatory, security, and welfare interests in.
  • Foster broader “open innovation community” engagement with government and public policy and vice versa, and more knowledge transfer across OIC domains, on legal instruments at the least.
  • Uniform Public License may be a better name than UGL in some respects (whatever the name, it ought be usable by the public sector, and the general public), but Government may be best overall, a tip of the hat to both the vision within governments that would be necessary to make the license succeed, and to the nature of copyright and copyright-like barriers as government regulatory regimes.

National jurisdiction licenses for PSI

A more likely mechanism for license proliferation deceleration and harm reduction in the near term is for governments within a national jurisdiction to use a single license, and follow various license stewardship and use best practices. Leigh Dodds recently blogged about the problem and highlighted this mechanism in a post titled The Proliferation of Open Government Licences.

Sub-national jurisdiction licenses for PSI

Each province/state and sub-jurisdiction thereof, down to towns and local districts, could use its own vanity license. This appears to be the trend in Canada. It would be possible to push further in this direction with multiple vanity licenses per jurisdiction, e.g., various licenses for various kinds of data, reports, and other materials.

Licenses for each PSI dataset or other work

Each and every government dataset or other publication could come with its own bespoke license. Though these licenses would grant permissions around some copyright and copyright-like restrictions, I suspect their net effect would be to heighten copyright and copyright-like restrictions as a barrier to both the use and publication of PSI, on an increased cost basis alone. This extreme highlights one of the downsides of copyright licenses, even unambiguously open ones — implementing, understanding, and using them can be seen as significant cost centers, creating an additional excuse for not opening materials, and encouraging the small number of people who really understand the mechanisms to be jealous and wary of any other reform.

None

Included for completeness.

Privatization of PSI copyright

Until now, I’ve assumed that copyright and copyright-like restrictions are barriers to use of PSI. But maybe there aren’t enough restrictions, or they aren’t allocated to the right entities, such that maximum value is realized from use of PSI. Control of copyright and copyright-like restrictions in PSI could be auctioned off to entities with the highest ability to extract rents from PSI users. These businesses could be government-owned, with various public-private partnerships in between. This would increase the direct contribution of PSI to GDP, incent the creation and publication of more PSI, ensure PSI is maintained and marketed, reaching citizens that can affordneed it, and provide a solid business model for Government 2.0, academia, cultural heritage, and all other publicly funded and publicly interested sectors, which would otherwise fail to produce an optimal level of PSI and related materials and innovations.

Do not let any of the above trick you into paying more attention to possible copyright and copyright-like barriers and licenses than actually doing stuff, especially with PSI, especially with “data”, doubly with “government data”.

I agree with Denny Vrandečić’s paradoxical sounding but correct directive:

Data is free. Free the data!

I tried to communicate the same in a chapter of the Data Journalism Handbook, but lacked the slogan.

Data is free. Free the data!

And what is not data? ☻

Addendum: Entirely by coincidence (in response to a European Commission consultation on PSI, which I had already forgotten about), today posts by Timothy Vollmer for the Communia Association and Creative Commons call out the license proliferation problem and endorse public domain as the default for PSI.

Innovation Pending

Wednesday, November 20th, 2013

Does the U.S. Patent System Stifle Innovation? Pro: Christopher Kelty, Laura Sydell. Con: Jaz Banga, Scott Snibbe. Moderator: Eric Goldman. Video:

The moderator was by far the best performer. Watch above, or read his introduction and audience voting instructions.

The pro side’s opening statement was funny, involving the definition of “stifle”, freedom as the oxygen of innovation, and innovation occurring within the iron lungs of large corporations, due to the patent system. Otherwise they stuck to a narrow argument: the current U.S. patent system is beset by trolls (Sydell was a reporter for When Patents Attack and II) and lawsuits and some would-be inventors do give up after realizing they are in a heavily patented field, ergo, the U.S. patent system stifles innovation.

The con side often seemed to make contradictory arguments that didn’t support their side. At one point the moderator interrupted to ask if they were really making a claim they seemed to be; nobody was phased, though I could swear at various points the pro side was looking incredulously at the con side (the recording is at the wrong angle to really see). But their fundamental argument was that there’s lots of innovation happening, patents and IP generally are American as apple pie, and trolls, while bad, aren’t a big deal for companies like Apple with many billions of dollars, ergo, the U.S. patent system does not stifle innovation.

The audience voted for the con side.

In my previous post noting that this debate was coming up, I concluded with “I hope they also consider equality and freedom.” They did a bit with regard to innovators — “freedom to innovate” and how “small” and “large” innovators fare in the system. But I had in mind expanding the discourse to include the effects of innovation policy on the freedom and equality of all humans.

“Patent” and “stifle” were expertly and humorously defined by Goldman and Kelty, but “innovation” remained undefined. The closest the debate came to exploring the contours of what innovation means, or ought mean, may have been in points made about the triviality of some patents, and the contrast between “small” and “large” innovators. Is innovation ‘done in a fashion that has served to maximize the patent encumbrances’ so it can be controlled by Apple, Microsoft, IBM, Monsanto, et al, the innovation we want?

Both the pro and con sides seemed to dislike patent trolls (while disagreeing on their importance). I wonder if any of the participants (particularly the con side) will endorse, or better yet, sign up for the Defensive Patent License (my discussion)? Or any of the other reforms reviewed by Goldman in Fixing Software Patents?

The debate was part of ZERO1 Garage’s Patent Pending exhibition, open through December 20. Each of the exhibited works is somehow related to a patent held or filed for by the artist.

One patent related to a work is pending, thus the work required an NDA for viewing:

nda

The handful of people I showed this image to were each appalled. But, in the context of the show, I have to admit it is cute. And, perhaps unintended, a critique of patent theory — which claims that patents encourage revelation.

Each of the pieces is interesting to experience. I particularly enjoyed the sounds made and shadows cast by (con side debater) Snibbe’s fan work (controlled by blowing through a smaller fan):

fans

My only disappointment from the exhibition is that there wasn’t a touching sample of these bricks, apparently made in part from fungus:

fungus brick

Bonus link: Discussions On The Abolition Of Patents In The UK, France, Germany And The Netherlands, From 1869. As I’ve mentioned before, these debates are nothing new, though it’s popular even for “reformers” to claim that current innovation policy is somehow mismatched with the “digital age”. The only difference between old and current debates is that the public interest is far more buried in the current ones.

Defensive Patent License 1.0 birthday

Saturday, November 16th, 2013

Defensive Patent License version 1.0 turned 0 yesterday. The Internet Archive held a small celebration. The FAQ says the license may be used now:

Sign up and start using the DPL by emailing defensivepatent@gmail.com.

There will be a launch conference 2014-02-2811-07 in Berkeley: gratis registration. By that time I gather there should be a list of launch DPL users, a website for registering and tracking DPL users, and a non-profit organization to steward the license, for which the Internet Archive will serve as a 501(c)3 fiscal sponsor.

Loosely organized thoughts follow. But in short:

  • DPL users grant a royalty free license (except for the purpose of cloning products) for their entire patent portfolio, to all other DPL users. This grant is irrevocable, unless the licensee (another DPL user) withdraws from the DPL or initiates patent litigation against any DPL user — but note that the withdrawing or aggressing entity’s grant of patents to date to all other DPL users remains in force forever.
  • Participation is on an entity basis, i.e., a DPL user is an organization or individual. All patents held or gained while a DPL user are included. But the irrevocable license to other DPL users then travels with individual patents, even when transferred to a non-DPL user entity.
  • An entity doesn’t need any patents to become a DPL user.
  • DPL doesn’t replace or conflict with patent peace provisions in modern free/open source licenses (e.g., Apache2, GPLv3, MPL2); it’s a different, complementary approach.
  • It may take years for the pool of DPL users’ patents to be significant enough to gain strong network effects and become a no-brainer for businesses in some industries to join. It may never. But it seems possible, and well worth trying.
  • Immediately, DPL seems like something for organizations that want to make a strong commitment, but a narrow one (only to others making the commitment), to patent non-aggression, ought to get on board with. Entities that want to make a broader commitment, including those that have already made complementary commitments through free/open source licenses or non-aggression pledges for certain uses (e.g., implementing a standard), should also get on board.

History

Last year I’d read Protecting Open Innovation: The Defensive Patent License as a New Approach to Patent Threats, Transaction Costs, and Tactical Disarmament (by Jennifer Urban and Jason Schultz, also main authors of the DPL 1.0) with interest and skepticism, and sent some small comments to the authors. The DPL 1.0, available for use now, incorporates some changes suggested in A Response to a Proposal for a Defensive Patent License (DPL) (and probably elsewhere; quite a few people worked on the license). Both papers are pretty good reads for understanding the idea and some of the choices made in DPL 1.0.

Two new things I learned yesterday are that the DPL was Internet Archive founder Brewster Kahle’s idea, and work on the license started in 2009. Kahle had been disturbed that patents with his name on them that he had been told were obtained for defensive purposes while an engineer at Thinking Machines, were later used offensively by an entity that had acquired the patents. This made him wonder if there could be a way for an entity to commit to using patents only defensively. Kahle acknowledged that others have had similar ideas, but the DPL is now born, and it just may be the idea that works.

(No specific previous ideas were mentioned, but a recent one that comes to mind is Paul Graham’s 2011 suggestion of a pledge to not initiate patent litigation against organizations with fewer that 25 employees. Intentionally imprecise, not legally binding, and offering no benefit other than appearing on a web page, probably not surprising it didn’t take off. Another is Twitter’s Innovator’s Patent Agreement (2012), in which a company promises an employee to seek their permission for any non-defensive uses of patents in the employee’s name; unclear uptake. Additional concepts are covered at End Soft Patents.)

Kahle, Urban, and Schultz acknowledged inspiration from the private ordering/carving out of free spaces (for what Urban and Schulz call “open innovation communities” to practice) through public licenses such as the GPL and various Creative Commons licenses. But the DPL is rather different in a few big ways (and details which fall out of these):

  1. Subject of grant: patent vs. copyright
  2. Scope of grant: all subject rights controlled by an entity vs individual things (patents or works subject to copyright)
  3. Offered to: club participants vs. general public

I guess there will be a tendency to assume the second and third follow strictly from the first. I’m not so sure — I can imagine free/open source software and/or free culture/open content/data worlds which took the entity and club paths (still occasionally suggested) — and I think the assumption would under-appreciate the creativity of the DPL.

DPL and free/open source software

The DPL is not replacement for patent clauses in free/open source licenses, which are conditions of public copyright licenses with different subject, scope, and audience (see previous). Additionally, the DPL’s non-grant for cloning products, which I do not understand the scope of, probably further reduces any overlap between modern FLOSS license patent provisions and the DPL that may exist. But, I see no conflict, and some complementarity.

A curiosity would be DPL users releasing software under free software licenses without patent provisions, or even with explicit patent non-grants, like CC0. A complementary curiosity would be free/source projects which only accept contributions from DPL users. Yet another would be a new software license only granting copyright permissions to DPL users (this would almost certainly not be considered free/open source), or releasing DPL users from some license conditions (this could be done as an exception to an existing license).

The DPL isn’t going to directly solve any patent problems faced by free/open source software (e.g., encumbered ‘standards’) any time soon. But, to the extent the DPL decreases the private value (expected rents) of patents and encourages more entities to not see patents as useful for collecting rents, this ought push the problems faced away, just a bit. Even if software patents were to evaporate tomorrow (as they should!), users of free/open source software would encounter patents impacted all sorts of devices running said software; patents would still be a problem for software freedom.

I hope that many free/open source software entities become DPL users, for the possible slowly accruing benefits above, but also to make common cause with others fighting for (or reforming slightly towards) intellectual freedom. Participation in broader discourse by free/open source software entities is a must, for the health of free software, and the health of free societies.

End Soft Patents’ entry on the DPL will probably be a good place to check years hence on how the DPL is viewed from the perspective of free/open source software.

DPL “enforcement”

In one sense, the DPL requires no enforcement — it is a grant of permission, which one either takes or not by also becoming a DPL user. But, although it contains provisions to limit obvious gaming, if it becomes significant, doubtless some entities will try to push its boundaries, perhaps by obfuscating patent ownership, or interpreting “cloning” expansively. Or, the ability to leave with 180 days notice could prove to be a gaping hole, with entities taking advantage of the pool until they are ready to file a bunch of patents. Or, the lack of immediate termination of licenses from all DPL users and the costliness of litigation may mean the DPL pool does little to restrain DPL users from leaving, or worse, initiating litigation (or threatening to do so, or some other extortion) against other DPL users.

Perhaps the DPL Foundation with a public database of DPL users will play a strong coordinating function, facilitating uncovering obfuscated ownership, disseminating notice of bad behavior, and revocation of licenses to litigators and leavers.

DPL copyleft?

In any discussion of X remotely similar to free/open source software, the question of “what is copyleft for X?” comes up — and one of the birthday presenters mentioned that the name DPL is a hat tip to the GPL — is the DPL “copyleft for patents”?

It does have reciprocality — only DPL users get DPL grants from other DPL users. I will be surprised if at some point someone doesn’t pejoratively say the DPL is “viral” — because the license to DPL users stays with patents even if they are transferred to a non-DPL user entity. A hereditary effect more directly analogous to the GPL might involve a grant conditioned on an licensee’s other patents which read on the licensed patent being similarly licensed, but this seems ineffective at first blush (and has been thought of and discarded innumerable times).

The DPL doesn’t have a regulatory side. Forced revelation, directly analogous to the GPL’s primary regulatory side, would be the obvious thing to investigate for a DPL flavor, but the most naive requirement (entity must reveal all patentable inventions in order to remain a DPL user in good standing) would be nearly impossible to comply with, or enforce. It may be more feasible to require revelation of designs and documentation for products or services (presumably source code, for software) that read on any patents in the DPL pool. This would constitute a huge compliance and enforcement challenge, and probably very difficult to bootstrap a significant pool, but would be an extremely interesting regulatory experiment if it gained any traction.

DPL “Troll-proof”?

The slogan must be taken with a mountain of salt. Still, the DPL, if widely adopted, would mitigate the troll problem. Because grants to DPL users are irrevocable, and follow a patent upon changes of ownership, any patent with a grant to DPL users will be less valuable for a troll to acquire, because there are fewer entities for the troll to sue. To the extent DPL adoption reduces patenting in an industry, or overall, there will be less ammunition available for trolls to buy and use to hold anyone up. In the extreme of success, all practicing entities become DPL users. Over a couple decades, the swamp is drained.

Patents are still bad

The only worrisome thing I heard yesterday (and I may have missed some nuance) was the idea that it is unfortunate that many engineers, and participants in open innovation communities in particular, see patents as unethical, and that as free/open source software people learned to use public copyright licenses (software was not subject to copyright until 30-40 years ago), they and others should learn to use appropriate patent tools, i.e., the DPL.

First, the engagement of what has become free/open source software, open access, open data, etc., with copyright tools, has not gone swimmingly. Yes, much success is apparent, but compared to what? The costs beg to be analyzed: isolation, conservatism, internal fighting, gaming of tools used, disengagement from policy and boundary-pushing, reduction (and stunting) of ethics to license choice. My ideal, as hinted above, would be for engagement with the DPL to help open innovation communities escape this trap, rather than adding to its weight.

Second, in part because extreme “drain the swamp” level of success is almost certainly not going to be achieved, abolition (of software patents) is the only solution. And beyond software, the whole system should be axed. Of course this means not merely defending innovators, including open innovation communities, from some expense and litigation, but moving freedom and equality to the top of our innovation policy ordering.

DPL open infrastructure?

I hope, in part to make the DPL attractive to existing open innovation communities, I really hope the DPL Foundation will make everything it does free and open with traditional public copyright and publishing tools;

  • Open content: the website and all documentation ought be licensed under CC0 (though CC-BY or CC-BY-SA would be acceptable).
  • Open source/open service: source code of the eventual website, including applications for tracking DPL users, should be developed in a public repository, and licensed under either Apache2 or AGPLv3 (latter if the Foundation wishes to force those using the software elsewhere to reveal their modifications).
  • Open data: all data concerning DPL users, licensed patents, etc., should be machine-readable, downloadable in bulk, and released under CC0.

DPL readability

I found the DPL surprisingly brief and readable. My naive guess, given a description of how it works, would have been something far longer and more inscrutable. But the DPL actually compares to public licenses very favorably on automated readability metrics. Table below shows these for DPL 1.0 and some well known public copyright licenses (lower numbers indicate better readability, except in the case of Flesch; Chars/(Flesch>=1) is my gross metric for how painful it is to read a document; see license automated readability metrics for an explanation):

SHA1 License Characters Kincaid ARI Coleman-Liau Fog Lix SMOG Flesch Chars/(Flesch>=1)
8ffe2c5c25b85e52f42fcde68c2cf6a88b7abd69 Apache-2.0 8310 16.8 19.8 15.1 20.7 64.6 16.6 33.6 247
20dc61b94cfe1f4ba5814b340095b4c3fa23e801 CC-BY-3.0 14956 16.1 19.4 14.1 20.4 66.1 16.2 40.0 373
bbf850220781d9423be9e478fbc07098bfd2b5ad DPL-1.0 8256 15.1 18.9 15.7 18.4 65.9 15.0 40.6 203
0473f7b5cf37740d7170f29232a0bd088d0b16f0 GPL-2.0 13664 13.3 16.2 12.5 16.2 57.0 12.7 52.9 258
d4ec7d0b46077b89870c66cb829457041cd03e8d GPL-3.0 27588 13.7 16.0 13.3 16.8 57.5 13.8 47.2 584
78fe0ed5d283fd1df26be9b4afe8a82124624180 MPL-2.0 11766 14.7 16.9 14.5 17.9 60.5 14.9 40.1 293

Automated readability metrics are probably at best an indicator for license drafters, but offer no guidance on actually improving readability. Last month Luis Villa (incidentally, on the DPL’s advisory board) reviewed a manual of style for contract drafting by editing Twitter’s Innovator’s Patent Agreement per the manual’s advice. I enjoyed Villa’s post, but have not attempted to discern (and discernment may be beyond my capability) how closely DPL 1.0 follows the manual’s advice. By the way, Villa’s edit of the IPA per the manual did improve its automated readability metrics:

SHA1 License Characters Kincaid ARI Coleman-Liau Fog Lix SMOG Flesch Chars/(Flesch>=1)
8774cfcefbc3b008188efc141256b0a8dbe89296 IPA 4778 19.6 24.0 15.5 22.7 75.8 17.0 27.1 176
b7a39883743c7b1738aca355c217d1d14c511de6 IPA-MSCD 4665 17.4 21.2 15.6 20.4 70.2 16.0 32.8 142

Net

Go back to the top, read the DPL, get your and other entities in the queue to be DPL users at its launch! Or, explain to me why this is a bad idea.

Social mobilization for the Internet post-epochals grew up with

Thursday, November 14th, 2013

Puneet Kishor has organized a book talk tomorrow (2013-11-15) evening in San Francisco by Edward Lee, author of The Fight for the Future: How People Defeated Hollywood and Saved the Internet–For Now (pdf).

I can’t attend, so I watched a recording of a recent talk by Lee and skimmed the book.

The book gives a narrative of the SOPA/PIPA and ACTA protests, nicely complementing Social Mobilization and the Networked Public Sphere: Mapping the SOPA-PIPA Debate, which does what the title says by analyzing relevant posts and links among them.

Lee in the talk and book, and the authors of the mapping report, paint a picture of a networked, distributed, and dynamic set of activists and organizations, culminating in a day of website blackouts and millions of people contacting legislators, and street protests in the case of ACTA.

The mapping report puts the protests and online activity leading up to them in the context of debate over whether the net breeds conversations that are inane and silo’d, or substantive and boundary-crossing: data point for the latter. What does this portend for social mobilization and politics in the future? Unknown: (1) state or corporate interests could figure out how to leverage social mobilization as or more effectively than public interest actors (vague categories yes), (2) the medium itself (which now, a few generations have grown up with, if we allow for “growing up” to extend beyond high school) being perceived at risk may have made these protests uniquely well positioned to mobilize via the medium, or (3) this kind of social mobilization could tilt power in a significant and long-term way.

Lots of people seem to be invested in a version of (3). They may be right, but the immediate outcome makes me sad: the perceived cutting edge of activism amounts to repeated communications optimization, i.e., spam science. Must be the civil society version of “The best minds of my generation are thinking about how to make people click ads. That sucks.” This seems eminently gameable toward (1), in addition to being ugly. We may be lucky if (2) is most true.

On the future of “internet freedoms” and social mobilization, Lee doesn’t really speculate. In the talk Q&A, lack of mass protest concerning mass surveillance is noted. The book’s closing words:

“We tried not to celebrate too much because it was just a battle. We won a battle, not the war. We’re still fighting other free trade agreements and intellectual property enforcement that affect individual rights.”

In a way, the fight for digital rights had only just begun.

Of course my standard complaint about this fight, which is decades old at least, is that it does not consist merely of a series of rearguard battles, but also altering the ecosystem.

“I would love it if all patents evaporated” (WebRTC)

Monday, November 11th, 2013

I’ve been following WebRTC (Real Time Communications) because (1) it is probably the most significant addition to the web in terms of enabling a new class of applications at least since the introduction of Ajax (1998, standardized by 2006), and perhaps since the introduction of Javascript (1995, standardized by 1997). The IETF working group charter puts it well (another part of the work is at W3C):

There are a number of proprietary implementations that provide direct interactive rich communication using audio, video, collaboration, games, etc. between two peers’ web-browsers. These are not interoperable, as they require non-standard extensions or plugins to work. There is a desire to standardize the basis for such communication so that interoperable communication can be established between any compatible browsers. The goal is to enable innovation on top of a set of basic components. One core component is to enable real-time media like audio and video, a second is to enable data transfer directly between clients.

(See pad.textb.org (source) for one simple application; simpleWebRTC seems to be a popular library for building WebRTC applications.)

And (2) because WebRTC is the scene of the latest fight to protect open web standards from rent seekers.

The IETF working group is choosing between H.264 Constrained Baseline Profile Level 1.2 and VP8 as the Mandatory To Implement (MTI) video codec (meaning all applications can count on that codec being available) for WebRTC. H.264 cannot be included in free and open source software, VP8 can, due to their respective patent situations. (For audio-only WebRTC applications, the free Opus codec seems to be a non-controversial requirement.)

Cisco has recently promised that in 2014 they will make available a binary implementation of H.264 for which they will pay license fees for all comers (there is an annual cap on fees, allowing them to do this). That’s nice of them, but the offer is far from ideal for any software (a binary must be downloaded from Cisco servers for each user), and a nonstarter for applications without some kind of plugin system, and for free and open source software distributions, which must be able to modify source code.

Last week I remotely attended a meeting on the MTI video codec choice. No consensus was reached; discussion continues on the mailing list. One interesting thing about the non-consensus was the split between physical attendees (50% for H.264 and 30% for VP8) and remote attendees (20% for H.264, 80% for VP8). A point mentioned several times was the interest of “big players” (mostly fine with paying H.264 fees, and are using it in various other products) and “little players” (fees are significant, eg startups, or impossible, eg free and open source projects); depending on one’s perspective, the difference shows how venue biases participation in one or both directions.

Jonathan Rosenberg, the main presenter for H.264, at about 22 minutes into a recording segment:

I would love it if all patents evaporated, if all the stuff was open source in ways that we could use, and we didn’t have to deal with any of this mess.

The argument for why H.264 is the best choice for dealing with “this mess” boils down to H.264 having a longer history and broader adoption than VP8 (in other applications; the two implementation of WebRTC so far, in recent versions of Chrome and Firefox, so far exclusively use VP8).

Harald Alvestrand, the main presenter for VP8, at about 48 minutes into another recording segment:

Development of codecs has been massively hampered and held back by the fact that it has been done in a fashion that has served to maximize the patent encumbrances on codecs. Sooner or later, we should see a way forward to abandon the dependence on encumbered codecs also for video software. My question, at this juncture, is if not now, when?

Unsurprisingly, I find this (along with the unworkability of H.264 for free and open source software) a much more compelling argument. The first step toward making patents evaporate (or at least irrelevant for digital video) is to select a codec which has been developed to maximize freedom, rather than developed to maximize encumbrances and rent collection.

What are individuals and entities pushing H.264 as the best codec for now, given the mess, doing for the longer term? Are they working on H.265, in order to bake in rents for the next generation? Or are they contributing to VP9, the next-next generation Daala, and the elimination of software patents?

Addendum: Version of this post sent to rtcweb@ietf.org (and any followups).

NFL IP II

Friday, November 8th, 2013

In an imperial capital city, expect to see the heads of conquered people on display.

Second time in about a month, someone has suggested US professional football businesses’ ability to censor be modulated if they continue to act against the public interest. First, copyright and civic extortion, now trademark and display of the heads of conquered people.

The first is fantasy at this point. The second may well happen, and not soon enough.

That modulating pro sports businesses’ ability to censor is deemed a potentially powerful incentive to stop bad behavior highlights the role of copyright and trademark in fostering a culture of spectacle and inequality — without these rents, team owners’ wealth and power would decrease significantly.

If professional sport is one of the things that brings classes and cultures in a community together, let’s enhance that by allowing everyone to view, share, make, and vend bits and atoms featuring elements of this togetherness, in their own way, without legal threat from ultra rich business owners.

Yes, let’s bring the heads down; that’ll get us some distance into modernity. But the empire, and its killing and torture, goes on. End that.