The Fifth Global Congress on Intellectual Property and the Public Interest: Polis Conversation

The Global Congress brings together an international community of researchers, practitioners, and activists who work on public interest approaches to the regulation of knowledge.  There have been five Global Congresses over the past decade.  Over 400 people participated in the most recent Congress, held at American University in September 2018.

For the Fifth Congress, we reengaged all participants–past and present–in an online conversation about their current work and priorities.

We held this conversation with a tool called Polis, which allows participants to make and respond to each other’s short statements.  The result is a community-directed survey of opinion and—from the participant perspective—also a loose iterative dialogue as participants respond to statements with new statements of their own.

336 members of the Congress community participated over a period of two weeks.  This group submitted 228 statements that received nearly 29,000 total ‘votes’—interactions in which the participants agree, disagree, or pass on a statement.  All participation was anonymous.  The organizers exercised some light moderation to ensure that submitted statements were clear and unique.

We learned from the demographic statements that:

  • A little over half of participants were lawyers (54%)
  • 40% identified as primarily as activists or advocates.
  • Around half were based at universities.
  • 44% were from the US or Canada
  • 17% from Europe
  • 10% from Africa
  • 12% from Latin America
  • 11% from Asia and South Asia.
  •  45% of the respondents were women.
  • A bit over a third indicated that they work primarily on copyright (36%) while just under a quarter work primarily on patents (23%). 12% signaled privacy as a primary focus.

We invited over 1000 past participants and applicants to the Global Congresses to join the conversation. Because respondents then self-selected and were not screened, the survey is not statistically representative of the views of the Global Congress community or of the public interest IP sector.  But it is an interesting window onto some of the dynamics within the several overlapping communities that the Global Congress has tried to mobilize.


Polis clusters participants into larger voting blocs based on their responses to the statements.  In the Global Congress voting community, this process distinguished three loose groups: a large Group A (200 participants) and two smaller groups B (54 participants) and C (45 participants).

These groups were not, for the most part, sharply defined by demographics.  European and US participants, for example, did not have consistent differences in opinion overall.  Rather, the groups organized around broad tendencies or philosophical approaches to intellectual property protection.  We can very roughly characterize these as:

A) Global IP Skeptic

B) Global IP Protectionist

And a partial exception to the demographic mix: the small Group C, which was composed predominantly of US based, academic copyright lawyers.

C) US Law School Copyright Reformist

The Global Congress can be seen as a dialogue between these three positions.

Here is what that grouping looks like to Polis:

If you want to learn more or explore this in more detail, see the full Polis technical report for the conversation


How Should I read the results?

Polis serves up statements in semi-random order, favoring statements that have received the fewest ‘passes’.  In practice, not everyone will see every statement — especially those statements that arrive late in the process.  A typical Polis statement result looks  this this:


Polis shows the overall result, and then the broken-out results for the three voting subgroups that emerged in the survey.


Group A

Group B

Group C

For the foreseeable future, any significant revision to the DMCA is likely to make it worse.

We’ve divided the results into topic areas below.

Platforms and Platform Policy

The US Copyright Office has been discussing possible changes to the Digital Millenium Copyright Act for years.  There is very little appetite for changing the law  – particularly among the US lawyer-heavy group C.


Group A

Group B

Group C


For the foreseeable future, any significant revision to the DMCA is likely to make it worse.

Opinion on the rights of platforms to limit offensive speech are divided overall and across all groups—with stronger opinions both for and against in the US where these issues have become prominent.

Twitter/Facebook, etc. are right to ban Alex Jones (and similar inflammatory voices).

Only a small minority support Google’s plans to launch a censorship-compliant search engine in China.

Google should launch its modified search engine in China, even if that means complying with Chinese government demands on censorship and surveillance.

Did the EU and US have good cases against Google for search bias?  Most passed.  A small plurality overall said yes.  The US-lawyer-heavy Group C was a strong no.

EU and US FTC officials had good cases against Google for search bias.

GDPR and EU Copyright Reform

The responses suggest a great deal of uncertainty about the impact of GDPR, with over 50% of respondents passing on the statement.  Overall, only a third believe that it is, on balance, a positive step for Internet policy.  Among self-identified European Lawyers: 7 agreed, 6 disagreed, and 10 passed.


Group A

Group B

Group C

GDPR is, on balance, a positive step for Internet policy.

The right to be forgotten continues to be viewed with uncertainty and concern.  Only 18% of respondents overall feel it has struck a proper balance and almost no one among the North American lawyer heavy Group C.

The Right to Be Forgotten implementation has properly balanced rights to privacy and free expression, which is weighed equally in the EU.

Will GDPR and the new Copyright Act entrench the power of the big platforms?  Opinion split overwhelmingly between ‘Agree’ and ‘Pass’.  Only a few people disagreed.

GDPR and the new Copyright directive are likely to entrench the power of the big platforms.

Is GDPR a threat to business and free trade?  The statement was not heavily engaged but a plurality disagreed.

EU privacy laws, namely GDPR, is a threat to business and is a barrier to free trade.

European efforts to extend the jurisdiction of GDPR and other measures to other countries are very unpopular across the board.

I am generally supportive of EU efforts to extend EU policy jurisdiction on Internet issues globally.

Among Europeans, only 3 out of 21 respondents agreed.

Opinion was split on whether the EU copyright bill could be stopped or improved.  Nearly half of respondents passed on this one.  A related statement about the possibility of activist mobilization similar to the 2011 ACTA and SOPA movement returned a very similar response.  The bill has since passed with its controversial features intact.

There is still a good chance to stop or significantly revise for the better the new EU copyright bill.

The ACTA/SOPA victories can be replicated in the context of the current EU copyright debate.


Copyright Policy

The restoration of copyright formalities (i.e., registration) is generally receives a strong plurality of support overall.  In the US-lawyer-heavy Group C, it commands nearly unanimous support.


Group A

Group B

Group C

Restoring copyright formalities/mandatory registration would make a number of hard problems easier and needs to a top medium-long term priority.

A sharper statement in favor of “mandatory” registration, however, roughly split respondents between agree, disagree, and pass, and loses some support among the US law heavy Group C.

Copyright registration should be made mandatory.

A statement about possible risks of mandatory registration in developing countries further erodes support.  A bare plurality of respondents agree with the statement.

Mandatory registration as a formality for copyright protection could be problematic in developing countries

There is strong support for moral rights, except among the US-law-heavy Group C.  Unlike most other countries, US copyright law does not explicitly recognize moral rights.

Moral rights are important.

A narrow plurality opposed expanding the definition of infringement to include the “making available” of infringing works.  Unexpectedly, Group C  favored the idea.

Uploading unauthorized content to a website where it is available for download (i.e. ‘making available”) should be considered infringement by the uploader.

Only half of participants had an opinion about the role of rights reversion in copyright reform.  Those who did strongly supported it.

Rights reversion is an under-explored and potentially very powerful direction for copyright reform.

An overwhelming majority of those with an opinion about anti-circumvention rules favor their abolition, with unanimous support in Group C.

Anti-circumvention rules have too high a social cost and should be abolished.

A statement related to ‘digital first sale’ principles.  A strong plurality opposes quantitative restrictions on ebook lending by libraries.

The ebook lending model for libraries needs some sort of restriction on the number of copies lent or in use.

On the subject of the famous “monkey selfie” photo from a few years ago, only a small minority (10%) thinks that the monkey holds the copyright.

The monkey holds the copyright.

By the same token, only a small minority felt that AI or algorithmically-generated works should receive copyright protection.

AI generated works should be copyrightable, albeit with excpections regarding moral rights and the duration of the protection.

Around 22% thought that copyright and patents should be abolished—nearly all of them in ‘global skeptic’ Group A.

It is time to make the case for the complete abolition of copyright and patent.

Patents and Medicines

Although there is a large patents and medicines track at the Congress, not many Polis statements addressed these issues.  Among them:


Group A

Group B

Group C

The importance of creating a long-term R&D patent system and the importance of near-term access to medicines are distinct concerns..

Access to medicines is very important, by access to diagnostics is also relevant. We need to eliminate barriers to Diagnostics.

Traditional Knowledge

A majority of respondents is skeptical of strong IP approaches to traditional knowledge.  Sui generis/benefit sharing regimes have more support.

Other than some lightweight measures like marks of origin, the main instruments of IP policy offer no good answers for traditional knowledge systems.

Traditional knowledge (know-how & practices of indigenous people) should be protected through sui generis rights that allow benefit sharing.


CC-licensing has room to grow, even in the Creative Commons-friendly territory of the Global Congress.


Group A

Group B

Group C

All or nearly all of my own work is published under a CC license.

Around a third of participants had used ‘shadow libraries’ like Sci-Hub for their work—with lower numbers only among the strong IP Group B.

I’ve used Sci-Hub or other shadow libraries for my work.

Even in a group of Internet policy experts, only around a quarter say they understand how the Internet advertising ecosystem works.

I definitely understand how advertising works on Google and Facebook.

Fewer (though a plurality of Group C) claim to understand how Content ID works.

I definitely know how ContentID works.

Framing and Strategy

There is strong consensus that “Access to Knowledge” is not (or is no longer) a sufficient way to frame copyright reform.


Group A

Group B

Group C

A reframing for copyright advocacy narratives is necessary. Relying on “access to knowledge” is not enough to deal with current problems.

This skepticism extends to claims that bad copyright policy will “break” the Internet.

The advocacy position that bad (c) or other policy will ‘break’ the Internet has become counterproductive. There has been a lot of site blocking, over-filtering, violations of net neutrality, and so on. But very few people experience it as breaking anything.

There is strong support for organizing around shorter copyright terms in this community but somewhat more disagreement—especially in American Law heavy Group C, about whether this is a productive issue to campaign around.

It would be productive to try to launch a public debate about shortening copyright terms.

The US law wing sees a  relationship between the strong funding environment of the early 2000s and policy wins.  In the other groups (which didn’t benefit as much from US philanthropic attention) there is more uncertainty.

The wins in 2011/2012 were built on a decade of funder investment in civil society and research in this area. The losses since then on a decade of funder retreat.

A large part of the community sees a deterioration in the copyright policy environment since 2011—and a strong majority in the US law pool.

Most of the SOPA/ACTA victories (e.g, on surveillance, ISP liability, overcriminalization) have been whittled away since 2011–even prior to the EU (c) disaster.

Most of the community puts high value on copyright advocacy at the World Intellectual Property Organization.

We should be prioritizing advocacy work at WIPO focused on protecting and promoting copyright user rights.

‘Agree’ sharply outweighed ‘disagree’ on the topic of whether US withdrawal from the TPP was a good thing—especially in Group C.  But a large number of participants passed.

US withdrawal from the TPP was, on balance, a good thing.

There is a mix of worry and uncertainty about whether the OER community is prepared for the shift to big-data-driven learning analytics.

The Open/OER community isn’t well prepared for the next generation of learning tools, which will be big-data learning analytics driven.

Not many people have a clear picture of the ‘technological sovereignty’ issue, which refers to national control of data and infrastructure.  But those that do think it will be a major concern.

Technological sovereignty – for example with regard to public data – will in 10 years be more important than the idea of open data.

Most respondents don’t put much hope in a generational shift in attitudes and politics on copyright—though here too 40% passed.

Belief that generational turnover would be a major positive factor in (c) policy advocacy is proving misplaced. The kids are growing up in walled gardens, not DIY sharing utopias.

Algorithmic transparency is a very strong priority of this community.

Algorithmic transparency is crucial to defending human rights and democracy online.


By a small margin, respondents agreed that research on copyright had improved since 2011.  In the US law group, the margin was larger.  A third of participants passed.


Group A

Group B

Group C

The research environment around copyright has gotten a lot more robust since 2011. Industry groups don’t have as much latitude to make up nonsense.

By a small margin, respondents agreed that critiques of industry accounts of losses to piracy had changed the politics around enforcement.

Repeatedly debunking industry accounts of losses to piracy has meaningfully changed the politics around copyright enforcement.

A strong majority across all groups sees a need for more systematic and empirical work on the relationship between speech rights and copyright.

We badly lack a framework for measuring and discussing tradeoffs between copyright and free speech. Hypotheticals and anecdotal abuses are not enough.

For trade, intellectual property and access to medicines, we lack a strong peer-reviewed journal and should consider starting one together.

What else?


Africa needs a preconference meeting in order to have more impact in future GC’s especially in A2M.

Apple Watch tracks when we have sex!

Frankenstein is a warning against the hubris of scientific communities.

Counterpoint: Frankenstein is a tragedy about a parent that cannot love its child.