del-fi

SCOTUS Plays Solomon on Gene Patents

Thu, 13 Jun 2013 08:16:00 -0700

Big news today: the Supreme Court has ruled that naturally occurring genes are not patentable.

This is a Big Fucking Deal, as Joe Biden might say. But it’s not as complete a victory as it may seem on the surface. The ruling explicitly excludes cDNA, and notes that the case didn’t address methods, applications of knowledge, synthetic DNA, or alterations of gene sequences that do occur in nature*. Lots of room to negotiate privatization there - indeed, that sentence could be the next “by means of a computer program” in patent law, though it doesn’t have to be so.

This is a ruling that resets the default to unpatentable, as Mike Eisen rightly pointed out on twitter. That’s why it’s big. But it doesn’t close the door, at all, on lots of patents in and around DNA*. That’s why the biotech stock index is up today*.

I’m actually not nearly as interested in those parts of the ruling though. What this means for me is more that the biggest barrier to building a commons of mutations with diagnostic potential is gone: the inability of DTC sequencing companies like 23andme to reveal the status of its customers to its customers because of patents. The companies that rely on these patents now have to move to trade secret approaches, as Myriad already has done, and the thing about trade secrets is…we can compete with them.

What the patents did was make it impossible, illegal, for us to build commons-based competition. They were enforcers on trade secrecy. Those enforcers are gone. We can now go straight to the citizen and say, get yourself genotyped, and donate your data to science.

At Sage Bionetworks, my non-profit employer, we’ve built a system that allows precisely that. It’s called Portable Consent, and we’ve got a study called the Self-Contributed Cohort for Common Genomics Research. You can enroll and donate your data in less than ten minutes, start to finish.

So go get yourself genotyped. Download your data. And donate it to science. Let’s stop fighting companies that privatize, and start competing with them.

Note: edited post at 12:15PM EST for clarification of a few points. Those sentences carry asterisks at the end.

Speaking Notes - Health Datapalooza

Mon, 03 Jun 2013 11:30:20 -0700

My notes for the panel today at Health Datapalooza. I’ll come back later and add links, fix typos, and so forth…

Why consumer access to data is important

- because data is a digital representation of us. and it’s increasingly being used to affect our real-world services, their costs, their benefits.
- we’re increasingly able to generate data that used to be the exclusive space of the clinical system. genomes are just part of it. health data is in everything. Facebook cups, keyboards, iPhones, google’s next phone.
- but it’s faulty at worst, and incomplete at best. if we can’t access it, we can’t tell how and where it’s incomplete or wrong.

Use cases and potential applications for consumer access to data

- send to app provider to interpret data and help me make better choices about my care.
- send to app provider to interpret data and help me make better lifestyle choices (ideally not a panopticon of health provider, ATT, and government - but a decentralized, competitive market)

that’s what this event is mainly concerned with. but there’s more.

- extract the “real” clinical data (compared to the “dry” data from phones etc) from the record. needs a lot of normalization etc but it starts to paint a longitudinal phenotype of me and my life. mapped to my genome, in large in sample sizes, we can start to correlate lifestyle and medical treatment outcomes to individual genomic variation.

Current state of consumer access to data

- raw in every sense of the word. most folks aren’t exactly aware of the things we here are aware of.
- most of the data collection is happening in zones like mobile and social where we have no positive rights to privacy, like health, and where the entire system depends on designing awareness of the data (and its ownership) out of the hands of us as citizens.
- in health, access to data is hamstrung by a combination of rapidly advancing technology and slowly (that’s a nice way of putting it) adapting law.
- we just moved here from Oakland. when i wanted my son’s immunization records to get him into his new child care center here in DC, the provider couldn’t fax them to me because they were afraid it violated hipaa. but it would have been legal to hire a TaskRabbit - total stranger - to go pick them up, take them to Kinkos, and fax them to me.
- that’s insane. we’re not being protected. we’re not getting access at the rate or in the form that we need yet. it’s getting better, but it’s still slow. and we’re less willing to tolerate it, because we’ve been trained to expect more from our institutions by good technology.

Where we have to go next / role of Blue Button+

- the hard part is that most of us, when handed a file of data, don’t know what to do with it. i downloaded my genotype, my fitbit data. No idea what to do with it. compared to my “medical record” in PDF, which is computationally useless but human readable.
- the investment in BB will pay off best when I have the right to direct my file to someone who can do something with it, for whatever reasons i choose. whether to run an app that makes sense of the data or to donate the data to research.
- BB+ is a great example of this. It allows for both market solutions (banks!) to emerge and for pre-competitive or public private solutions to emerge where we can donate data, or share it conditionally.
- i’m looking forward to pushing Sage Bionetworks, the non profit where I work, to be one of the first certified recipients of BB+ precisely to enable the non-market reuse of health records data.

Officially Entering "And Then They Fight" Phase of Open Access

Mon, 20 May 2013 10:41:00 -0700

And, my friends, in this story you have a history of this entire movement. First they ignore you. Then they ridicule you. And then they attack you and want to burn you. And then they build monuments to you.

Although the substance of the above quote is usually attributed to Gandhi, there’s no record that he actually said it. The quote above is by Nicholas Klein, a labor activist, from 1918.

I don’t include it as an example of attribution decay. I use it as a frame for where we are in the open access world right now.

We’ve had a good run. We got the NIH public access mandate. We got the petition to 25,000 signatures. We got the presidential directive extending the NIH policy across the entire federal government. We got multiple examples of open access publishers into sustainable revenue models.

But changing the default from closed to open was always going to involve a phase where those whose revenue models depend on closed really brought the guns out against us. And we’re there now.

There’s Wiley, wallowing in the mud and smearing Public Library of Science’s peer review credentials under the charade of a survey of authors.

There’s Elsevier, proposing a novel license for STM publishers and somehow magically being part of a Netaction “bad legislation” coalition that attacks all open access bills, while denying any knowledge of it (no story coverage, but some conversations on Alicia Wise’s twitter feed). (UPDATE May 24 2013: Times Higher Education UK has a story in which Elsevier sort of distances themselves from Netaction)

There’s crocodile tears covering the emergence of scammy open access journals, none of which mentions the long-time existence of scammy closed access journals. This is not surprising, as so many of the large, “authoritative and important” publishers make money by publishing scammy journals - anyone remember the Merck-Elsevier scammy bone journal? Still waiting to see someone mention that in the same breath.

Then there’s the systemic disadvantage we have as advocates once policies move into implementation phase. The meetings last week at the National Academies are a great example of why it’s so hard to change the system. I had to travel 3 of the 4 days for work, and the fourth day I was in meetings all day that made it impossible for me to attend, or to speak.

We have day jobs, us advocates. But the publishing industry we’re fighting against has no other job. They can hire people who have only the responsibility of making sure the open policies are implemented in the least open way. They can saturate every meeting in DC with hired guns, and claim it as evidence that the public supports them.

But it’s not about being depressed, or complaining. It’s a sign that we’re finally getting close to the bone. We’re enough of a threat not to be ignored, or ridiculed. We’re gonna get hit, and we’re gonna get hit hard.

We have to keep reminding the world that this isn’t about protecting a dinosaur business model, this fight. It’s not about scammy journals, which exist no matter how they get paid for. It’s not about who has the most lobbying money in DC. It’s not about new licenses, or sleazy survey language.

It’s about letting entrepreneurs build businesses on top of open content. It’s about kids building cancer tests on open content. It’s about you and me being able to read what our tax dollars paid for. Don’t let the FUD and mudslinging get in the way of that message, ever.

We have to keep getting up. We have to keep fighting back. Because in the end, we’re on the right side of history. And once we get through this phase we get to the good part, where they build a monument to Heather Joseph and Peter Suber and Mike Eisen and all the heroes of open access.

Business Strategy and Openness

Sat, 13 Apr 2013 05:55:49 -0700

I got a lot of good responses to my post yesterday about Mendeley’s acquisition by Elsevier.

But a theme emerged that I didn’t intend to emerge, which was the idea that because I pegged Mendeley’s investment in open access as a customer acquisition strategy that it made it insincere.

That was most definitely not the point.

Remember that half the time I’m a senior fellow in entrepreneurship at the Kauffman Foundation. We study startups and support entrepreneurs. I can tell you that using openness as a customer acquisition strategy is something I think is a smart move, especially the way that Mendeley did it.

As a quick reminder for those who’ve never started a company, they need to acquire customers or they go out of business (as mine did, because mine didn’t). If you don’t have a strategy to do that, you’ll fail.

By selecting open, and fully committing to it, as a strategy, Mendeley helped the Open Access world enormously over the past four years. What they’ve done creates a solid track record in OA, a serious and provable one. They didn’t do this for evil reasons, or in anything close to an insincere way. And by doing it they advanced the movement. We can, and should, thank them. That’s why the anger stunned me, as I mentioned yesterday.

They hired a community manager (William Gunn) personally and professionally dedicated to OA, and let him run. I know, respect, and believe deeply in William and his commitment. That’s not an insincere company move.

Their data’s under CC-BY. That’s not insincere. I made a crack about allowing the database to be downloaded and reposted, and immediately heard back from William that I could go for it. That’s not insincere.

They ran the binary battle with PLoS (I was a judge!). That’s not insincere.

They backed the Access2Research petition, immediately. That’s not insincere.

My point wasn’t to gleefully give Mendeley the finger and say, you were closed all along, or to say, you were just using us to get customers. Every startup needs customers, and if you don’t realize that you’re kind of being a dick to the people in the startup. Mendeley’s making a meaningful commitment to open as a customer acquisition strategy was innovative, and important, and advanced the cause of open access to the scholarly literature.

My point was deeper. It was that because the openness was not tied to revenue, it could be removed now that the product didn’t need an innovative customer acquisition strategy. It’s tied to a massive customer base now. Open can be discarded. When it’s part of the revenue model, it can’t be discarded nearly as easily.

And in my experience, if open *can* be discarded, it usually *is* discarded when monetization becomes the priority. I hope that I’m wrong.

Lessons from Mendeley: Where's The Open In The Model?

Fri, 12 Apr 2013 07:31:00 -0700

So Mendeley got bought by Elsevier. And there was much teeth-gnashing. I won’t link to it but it spawned two solid hashtags: #mendelsevier and #mendelete.

I have a Mendeley account, but never used it other than to test the system against Zotero, which is what I use to track my own work. So I am not affected by this but I’ve been a bit stunned by the depth of the anger against Mendeley. I’ve waited to write this to try and understand it.

Part of it, I assume, is just Elsevier rage. Danah Boyd has summarized why Elsevier is rage-worthy nicely in her post on the acquisition.

But the greater part feels like the anger over what many seem to think is a broken promise made by Mendeley to be an “open” company.

I don’t feel that way. I never thought Mendeley was an open company. I thought they were deploying a strategic approach to openness by exposing their data under CC-BY, but I always thought that openness wasn’t the point of the company. It’s why I didn’t use the product and why I wasn’t surprised, shocked, or saddened by the acquisition.

It’s got me thinking though about companies and “open” - and what matters in deciding whether or not to use a product from a company claiming that mantle. For me it boils down to where the openness lives in a company.

There’s a lot of ways to slice this, but a simple one would be: is the “open” part of the revenue model or is it part of the market acquisition strategy? If the former, like BioMed Central, I have a lot more faith that an acquisition will not destroy the openness, because “open” is part of the way that the company makes money.

But the open access part of Mendeley to me always appeared to be a customer acquisition strategy. It appealed to the OA folks, it appealed to developers, and it never affected any monetization or revenues. There were always visible choices by management to hedge their open bets, as Jason Hoyt has laid out. And that makes it a risky bet to think they’ll stick with it now that they have access to a massively larger customer base while inside a company with traditional antipathy towards openness.

Again, I’m not mad. I either avoid, from a professional basis, companies built on closure, or I mitigate my expectations of them and do a lot of backing up. Because at some point unless the revenues come from open, the customer acquisition strategy of openness will be deprecated. If it isn’t, then the management of the company will be replaced with managers who are willing to shut things down to make money.

Always, always, always examine claims made by companies about openness. I’m not the world’s biggest Evgeny Morozov supporter, but he’s right to examine the way that the words “open” and “sharing” and “free” get co-opted. Facebook lets you share! It’s free, and always will be!

Look the gift horse in the mouth. And if the revenue model of a startup isn’t built on open, then feel free to use the tool. But don’t get emotionally invested, or dependent, no matter how seductive the rhetoric may be. Because at some point your use and attention and content will be monetized, probably in a way that bothers you.

A Fool's Errand, Annotated

Wed, 27 Mar 2013 13:50:00 -0700

So, I have a commentary published in Nature this month about the importance of using a CC-BY license to achieve full open access. I requested the article be made freely available as part of my agreement with Nature, but they paywalled it anyway. It’s freely available now but not until after some embarrassing email and twitter hassling.

I am not particularly mad at any of the parties involved. It just points out the power of the default switch being closed, and how hard it is, even when you’ve negotiated an agreement, to flip it to open. It points out the weakness of the author in negotiation with the journal. Maybe I’m the fool in the fool’s errand.

Also, in the search for brevity that print journals enforce, I didn’t get to be as granular as I wanted. My quarrel is with the publishing industry’s attempt to write a new license and I have no wish to lump those with whom I have a philosophical disagreement with those OA advocates who sincerely dislike CC BY, like Heather Morrison or many in humanities, into the same pool.

Anyway. Below are references for key points I make in the commentary.

1. Re: definitions of OA, see “Budapest Open Access Initiative” at http://www.opensocietyfoundations.org/openaccess, accessed 03/13/13

2. Re: restrictions in licensing, see Elsevier’s published contract with California Digital Libraries: ““Schedule 1.2(a) General Terms and Conditions “RESTRICTIONS ON USAGE OF THE LICENSED PRODUCTS/ INTELLECTUAL PROPERTY RIGHTS” GTC1] “Subscriber shall not use spider or web-crawling or other software programs, routines, robots or other mechanized devices to continuously and automatically search and index any content accessed online under this Agreement. “” online at http://www.google.com/url?q=http://orpheus-1.ucsd.edu/acq/license/cdlelsevier2004.pdf&usd=2&usg=ALhdy2_FmzOtI3JkKs-fJwirgig4WLA5fA, accessed 03/13/13

3. Re: “CC Plus,” see various comments in lectures at “FACT Seminar No. 1: Licensing in an Open Access Environment: legal niceties, funder mandates and publishing challenges“ at http://www.stm-assoc.org/events/fact-seminar-no-1/?presentations, accessed 3/13/13

4. Re: CC BY, See “Creative Commons Attribution 3.0 Unported” (the “commons deed” with links to complete underlying license) at http://creativecommons.org/licenses/by/3.0/, accessed 3/13/13

5. Re: 70+ requirements, see “Public Policy Requirements, Objectives and Appropriation Mandates” at http://grants.nih.gov/grants/policy/nihgps_2010/nihgps_ch4.htm, accessed 3/13/13

6. Re: community defintions of open things, see “Open Knowledge Definition” at http://opendefinition.org, accessed 03/13/13, and “Open Source Definition (Annotated) at http://opensource.org/osd-annotated, accessed 3/13/13

7. Re: license incompatibility, see “GPL-Incompatible Free Software Licenses” at http://www.gnu.org/licenses/license-list.html#GPLIncompatibleLicenses

8. Re: decomposition of licensed elements and the CC licenseed Time Photo of the Year, see “Trapped Underground,” a CC-BY photograph of the London Bombing aftermath, available at http://en.wikipedia.org/wiki/File:Trapped_underground.jpg accessed 3/13/13

9. Re: technical solutions to provenance, see “Source Attribution in RDF,” http://www.w3.org/2001/12/attributions/ accessed 3/13/13

A Natural Study of Openly Licensed Books

Thu, 14 Mar 2013 15:01:58 -0700

One of my favorite users of CC-BY is Pratham Books in India.

Pratham Books is a non-profit trust that publishes high quality books for children at affordable prices and in multiple Indian languages. They’ve shipped more than 7,000,000 books.

They’ve got a beautiful post about their move from the Attribution-NonCommercial-ShareAlike (BY-NC-SA) CC license to the Attribution (BY) only license and the internal back story of the decision. What’s really fascinating isn’t just that they relicensed 400 books under BY, but that they only managed to post 173 of those books online at Scribd. The other 227 books were not posted. So we have a nice, natural study to analyze of the differences between openly licensed content that is in a stable, well used platform and content that isn’t.

Some brief takeaways. Read their whole post to see the graphs.

The sales of individual books available on Scribd don’t differ greatly than those that aren’t - but they’re definitely not significantly lower.
Cumulatively, the sales of Pratham books on Scribd appear to outsell those that don’t, though not totally outside the error margins. But again, they’re not lower.
For cumulative sales data for CC books that were available on Scribd vs. CC books that were not available on Scribd, the former outsold the latter in almost a 3:1 ratio.

Let’s unpack point #3, because it’s fascinating. These are books that are available online, under the most liberal license offered by Creative Commons. You can one-click download and print them, or even send them off to be reprinted and sell them yourself. Yet the ones that are there radically outsell those (also liberally licensed books) that are in a more controlled technical environment.

There’s a lot of money spent looking for ways to sell books and content on the web, to protect authors, to protect old revenue sources. Sometimes though, the best advertising for the content (and implicitly the author) is the content itself. And there is mounting evidence that at least some people will pay for the authentic version, whether it’s to get the physical artifact, to help the author, or simply because they want to.

This is a tiny data point. What’d be great is if as these experiments happen, more publishers started to release the data as to the outcomes. We live in a world where we can’t even get a publisher to tell us the breakdowns in their revenues, how much they make off subscriptions and new content versus access to the back catalog. Data about where the money and the sales actually happen will drive far better policy than we have. If only the publishers would get that and start opening up some channels.

Solutionism and Sensorism

Thu, 07 Mar 2013 09:59:00 -0800

Evgeny Morozov’s been on a tear lately, with articles about “solutionism” in Slate and the New York Times, and an absolute destruction of Gavin Newsom’s new book on networked politics.

I’m torn by Evgeny’s work. I tend to agree with what I perceive is one of his basic ideas: our culture of deifying the application has a nasty side effect of reducing the perceived importance of political change: don’t bother actually risking anything to protest, just like the Facebook page about the protest. I really liked the concept about the rise of the choosing algorithm and its impact on creativity. I have a soft spot for a truly well written negative book review. And I think it’s vital that those of us who self-identify as “open” advocates engage with his work, because I think he’s onto something in more cases than I’d like to admit.

On a related note, I’ve been reading The Theory That Would Not Die recently. It’s a nice look at the waxing and waning of probability and inference over hundreds of years. When we’re in periods where we don’t know what we’re looking for - and it’s important to be as right as possible as fast as possible - then Bayes’ table becomes a vital tool in the kit. In World War II, that was cracking Enigma.

Unfortunately now, most of its usage is social, mobile, commercial. It offers me rehab in Napa Valley when I buy wine online.

But I worry that in burying the algorithm and the application and the game, to kill its overestimation in culture and in politics, he fails to appreciate the places where those tools are yet to be applied, but might bring real change. By real change, I mean epistemic change, a change in the way that we know we know something.

Science is one of those places. Science has a problem related to, but different than, solutionism. I’d call it “sensorism” perhaps - the belief that because we can make a machine that senses things more finely, more completely, and in massive parallel, we’ll somehow come to a greater understanding of the underlying thing being studied.

Sensorism is rife in the sciences. Pick a data generation task that used to be human centric and odds are someone is trying to automate and parallelize it (often via solutionism, oddly - there’s an app to generate that data). What’s missing is the epistemic transformation that makes the data emerging from sensors actually useful to make a scientific conclusion - or a policy decision supposedly based on a scientific consensus.

One of the reasons I do “open” work is that I think, in the sciences, it’s a philosophical approach that is more likely to lead to that epistemic transformation. If we have more data available about a scientific problem like climate change, or cancer, then the odds of the algorithms figuring something out that is “true” but incomprehensible to us humans go up. Sam Arbesman has written about this nicely both in his book the Half Life of Facts and in another recent Slate article.

I work for “open” not because “open” solves a specific scientific problem, but because it increases the overall probability of success in sensorism-driven science. Even if the odds of success themselves don’t change, increasing the sample size of attempts will increase the net number of successes. I have philosophical reasons for liking open as well, and those clearly cause me cognitive bias on the topic, but I deeply believe that the greatest value in open science is precisely the increased sample size of those looking.

I also tend to think there’s a truly, deeply political element to enabling access to knowledge and science. I don’t think it’s openwashing (and you should read this paper recommended by Morozov on the topic) to say that letting individuals read science can have a real political impact.

I look forward to reading Morozov’s new book this weekend at SXSW. It’ll be a bracing antidote to what I expect will be rampant solutionism. My panel is political - it’s about who owns the data these solutions create, and what rights and risks those solutions entail. It’ll get a lot less attendees than a product launch.

And I hope Morozov engages with sensorism one of these days. It’d be fun to see what he does with it.

Nature Haz Open Access

Fri, 01 Mar 2013 10:02:00 -0800

Wednesday featured a double-barreled shot by Nature Publishing Group across the bow of the Open Access conversation. First, an editorial unambiguously endorsing “gold” open access as the proper form of scholarly publishing. Second, a majority investment in Frontiers, a for-profit CC-BY scholarly publisher.

I’m personally just as happy to see this as I was to see the OSTP policy expanding public access across all federal research. Let’s just say some of my friends were surprised at that.

Sorry, but I liked it.

I have enjoyed watching concern trolls wading into the debate warning that Nature’s only doing this for their corporate masters. Well, duh. That’s what for-profit companies do.

And for profit companies choosing to back open systems has a history of working. The open source software movement really exploded when IBM made a full-throated choice to switch to open source. They didn’t do it because they suddenly became devotees of Saint IGNUcius.

Image, ironically, copyright 2002 by Julian Cash

They did it because the corporate masters figured out they were losing in the changing marketplace and they could make more money on services by embracing open, and that it could be a weapon against their competitors.

Turns out they were right, by the way. IBM made $1B on its portfolio of patents in 2012, but it made $20B on global business services in 2011. That 20:1 ratio feels about right to me in terms of value.

If Nature did this because they think they can make money, bully for them. What I care about is a second major for profit publisher has made an unambiguous business choice to back open copyrights on scholarly literature (Springer being the first when they acquired BioMed Central).

Publishers realizing that making scholarly research truly open is a good business decision further tilts the field towards making scholarly research truly open. It’s depressing but not surprising that I have to point out this tautology. And if it means we have to fight them in the halls of government on price, fine.

My personal belief is that if these publishers misprice the cost of libre open access in article processing charges, the market will fix that, and startups like PeerJ will win the day. Scientists are price sensitive.

Rather than insist on one open access regime to rule them all, I think we instead need to cultivate our garden, as Candide learned.

We don’t live in the best of all possible worlds. We live in a world of diverse scholarly disciplines with diverse needs, with competing interests in publishing markets and societies and universities. I think we can advance the whole movement fastest by embracing the “yes, and” approach (HT @DCDave).

Open publishing and self archiving will each solve certain problems at certain times. And each will suffer either real or perceived exploitation by companies new and old.

And that’s ok. That’s part of growth. That’s part of the garden growing.

Reflecting on Public Access

Wed, 27 Feb 2013 08:44:00 -0800

Now that I’ve had some time to process the new White House public access policy, here’s a few thoughts.

First and foremost, I think the policy is worth celebrating. This is something on which there is a difference of opinion. Mike Eisen is the most vocal and eloquent critic of the policy, and his post on the topic is essential reading whether you like his position or not.

I disagree with Mike on the conclusion, but it’s worth examining why. And it’s not worth attacking him - indeed, attacking each other is a feature of the OA movement that sickens me.

But I think his position comes more from a worldview where there is indeed real movement towards true OA (i.e., no embargo, open copyright licenses, and a total change of the publication industry via startups both non profit and for profit). Compared to that progress, the policy is like going back in time to when the NIH had no policy.

But here’s the thing. That progress simply doesn’t exist in most of the other spaces where the US Government invests in research. It doesn’t exist in agriculture (100,000 papers per year which will be key to food supply and debunking bullshit claims about GMO food, for example), or defense, or trade, or patents, or space science, or energy.

I wrote a paper with my dad on open access and energy policy (my dad is a climate change adaption scientist with a piece of the IPCC Nobel Prize) that opened my eyes to just how little of the energy literature is OA. If you look at the links in an IPCC report you’ll be shocked at how few you can read. None of the green, or gold, OA conversation has had a significant impact there at all.

Compared to the world that exists there, a world that is not being wrenched open by change and entrepreneurs and repositories, this policy is indeed enormous progress.

I don’t like the embargo entrenchment at 12 months. I don’t like the lack of reuse rights. I don’t like allowing linking into journal archives as opposed to centralized repositories. I don’t like the praise of the dying traditional publishing industry, but that’s pablum.

So let’s be frank. This isn’t a strong open access policy. But that’s ok by me.

Because it’s an enormous expansion of public access into spheres of research where it was essentially absent from the conversation. It creates a policy environment that tilts the field towards change, towards startups, towards publishers whose embrace to real Open Access becomes a competitive advantage over time - and it does so across an enormous swath of science.

It re-creates the conditions built in the life sciences years ago in other sciences. I believe that’s what will advance open access fastest there. But let’s also not attack Mike, or others, who disagree. We need to be constantly reminded to strive for the most, and attacking one’s own critics is not a healthy sign of an open movement.

This policy is not the end of our work in open access, or even the beginning of the end. But I do believe it’s at last the end of the beginning of the enormous change from closed publication of science to open publication of science. From now on it’s about how we implement access, not if.

Hats off to everyone, and now get back to work, because we’re a long way from done. Time to focus our energy on Congress to finish the job.

White House Public Access Policy Is Out

Fri, 22 Feb 2013 10:29:37 -0800

The White House has published a response to our We the People petition, commonly known as the OA Monday petition.

I’ll have a longer post up once I’ve had time to thoroughly review the policy, but the multi-year campaign to turn the NIH public access policy into something that applied to the entire federal research system is over.

I’d like to thank the White House, especially the Office of Science and Technology policy, for listening to us through the various RFIs and then finally to the petition. This is good policy. And it’s nice to see something substantive emerge from the We the People platform, too.

To the squeals of the publishers, I say only this. Jobs will be created by open content. I personally built a company years ago on the open data published by the NIH. I’m looking forward to watching others build companies on the open literature that will emerge under this policy. And the fight’s not over - it won’t be until we have reuse rights, not just free downloads.

But for now, let’s dance. We won.

The Strange State Politics Of Arabic OER

Tue, 29 Jan 2013 09:12:00 -0800

Yesterday, U.S. Secretary of State Hillary Clinton announced the Open Book Project, a major initiative to make Open Educational Resources available in Arabic. There was a big launch program at the State Department, and some remarkable words from Mrs. Clinton:

You can look around the world and see young adults in remote villages and towns huddling around a computer watching videotaped physics lessons by MIT professors. Top universities like Rice University are creating free online textbooks and saving students money in their studies. Science education websites like Khan Academy go viral. There are other examples, and these are all fruits of technological progress, but also of a commitment to make more learning materials open – free, open licensing for anyone to use, adapt, and share.

(from the text of Mrs. Clinton’s remarks - emphasis at the end added by me)

I was one of many people excited to see these kind of words. I’m always glad to see this kind of phrasing about copyright in the halls of government, because it indicates at least some understanding of the potential of open copyright licenses to create innovation.

But then I started thinking more about it, and the uncritical acceptance of the project without a conversation about the politics of it started to bother me. I still think it’s a worthy, and wonderful project.

But let’s unpack what we’re doing here, as a government.

It seems to me that we are applying state power to create digital knowledge resources that (probably, maybe, definitely) subvert the attempts of both state and non-state (social, religious) actors to control the kind of educational materials available in the Arabic speaking world. That we are practicing, as statecraft, what Nils Gilman has called deviant globalization, using an open source approach.

And that’s worth contemplating. I do not think that this is an embrace of open copyright because of philosophical shift in favor of openness. I think it’s a pragmatic piece of realpolitik - getting textbooks into a language and format where they can evade the controls imposed by actors whose interests are felt to be contrary to the US.

The theory that more knowledge in more hands is better for the world is one I happen to subscribe to, and agree with, but let’s not pretend this isn’t an exercise of state power. It is far more like the Voice of America (especially at its founding, when it was not required to provide “unbiased” news) than it is like Free Software.

Note: I edited this post at the end with the bold text in the final sentence at the suggestion of Dave Clifford after it was first posted.

Emperor's New Short Tandem Repeats

Thu, 17 Jan 2013 11:49:21 -0800

Yaniv Ehrlich’s lab at MIT has a new paper out in Science today, with a companion policy piece from the National Human Genome Research Institute at NIH. Apologies for icky paywalls but these are important papers.

The gist is that a savvy computational scientist can find enough breadcrumbs in a genome to figure out the surnames for participants in supposedly de-identified studies. The methods for the paper are reminiscent of the re-identification approaches against the Netflix database, the AOL search database, and the Massachusetts health records database: cross-referencing de-identified information with other public information, and then using that against other records associated with surnames.

And it gets bigger - 135,000 or so records may potentially identify millions more through inheritance of short tandem repeats. I’ve probably mangled the approach. The short version is perhaps better summed up by a picture:

The dude in red is mathematics. The dude in white is your anonymity.

These two papers force us to be honest in talking about genomic information and identifiability: the basics to re-identify significant portions of people from their genomes alone are already in place. And they’re already strongly able to lead to surnames, long before we hit the mythical $100 genome.

So what does honesty in this space mean? It means we shouldn’t promise people that we can both de-identify data and make it useful. It means that we should also celebrate the benefits that de-identification brings, and think of it in a risk-reward context for those joining studies that involve genome publication online. It doesn’t mean stop sharing, stop sequencing. It means stop pretending the methods for de-identification work very well. A lot of people will go away anyway, but a lot will share.

Both papers reject the idea of ceasing data sharing as a result of the research, which is heartening. We are in a world where we are simply less anonymous than we used to be, than we’ve ever been. There are enough unique things about us all, and enough devices capturing them, and powerful enough algorithms, that this stuff is simply doable now.

We need to develop a whole spectrum of ways to manage privacy. My own work on consent is just a piece of a tapestry, for those who really want to donate, to share, to be exposed. Hopefully this opens up more space at the table for the new approaches that are bubbling in health privacy management. We need data markets. Data banks (you know, like old-school community banks). Data conservation trusts (like land trusts - I’m going to publish something soon on this topic). We need entrepreneurs to fill the gaps between all open and all closed, to provide products that make someone’s data alive to her, not just to a gearhead with a taste for naive bayesian inference.

Pandora’s box didn’t open with this paper. It’s been sitting open for quite a while now, just waiting for the right eyes to see it.

Inspired by Aaron?

Mon, 14 Jan 2013 12:35:03 -0800

If you’re new to the idea of open access and public access, brought by the sad energy of Aaron Swartz’s suicide, I urge you to try to be inspired by him and learn about the space. He was a voracious reader, an insatiable scholar.

Follow his example. Do your research, read the overviews, follow the links. Learn is the first part of his four-part method. This is a well developed field, and though we need your energy, we need it very much in a learned and focused manner.

There is already legislation. There are already organizations, advocates, resources. Get involved.

Learn. Then Try. Then Gab. Then Build.

But always, always, start by learning.

It's Time For An Answer to #OAMonday

Mon, 14 Jan 2013 12:33:00 -0800

On May 20, 2012, I joined Mike Carroll, Mike Rossner, and Heather Joseph to start a campaign for a federal policy to require free access to scholarly articles emerging from taxpayer funded research. We decided to launch a petition on the We the People website.

We hit the 25,000 signature number pretty quickly, and I got a few calls from DC people asking what the hell was going on. This was before every maniac who wanted to secede started a petition - it was, and still is, a backwater, and we figured we could make some noise there. It worked.

Since then, petitions on beer recipes, gun control, and the death star have received official responses from the White House.

We have not.

I am told there is a conversation going on. I am asked to be patient. I am tired of being patient. I’m tired of the power of publisher money carrying the day, delaying the policy, blocking the flow of knowledge.

It’s time to make taxpayer funds turn into taxpayer goods. It’s up to the businesses to figure out how to make money in the new world - if you can’t figure out how to adapt your business to the network, you will merely be the latest in a long line of dinosaurs dating back to Digital Equipment Corp and running straight through to Newsweek. I do not cry for you.

It’s time to bring an end to the barriers to new businesses. Time to enable new entrepreneurs, who make money from increasing the free flow of information rather than by restricting it. Time for search engines that make it easy to find and integrate the literature. Time to enable people outside the institutional system who might take scholarly knowledge and turn it into products, into policies, into something of use.

It’s time for an answer. And if we don’t get the answer that says, “you paid for it, it’s yours,” it’s going to be time to escalate this from something on a sleepy backwater petition site to a protest that the whole world will notice.

Gone Fishin

Fri, 07 Dec 2012 00:15:50 -0800

Have a lot of work to do. Then a little bit of vacation to do. Be back here in the new year.

Patient Control - Good for Open Data

Fri, 30 Nov 2012 13:45:00 -0800

Olin Hyde noted a nifty survey on patient control of electronic health record information the other day. But he interpreted it in a dour way, at least for my own worldview of openness. Here’s his tweet:

Bad news for #opendata advocated by @wilbanks “@berniemd31: Study: Patients Want To Control #EHR Information - bit.ly/ToOglB”

— Olin Hyde (@olinhyde) November 29, 2012

The article talks about how patients want control over the privacy of their data, so I can see why this is an easy interpretation to make. You’re a patient, someone tells you how much crap is happening with your data, and your gut reaction is to protect it. Makes sense.

But actually I believe it’s the opposite conclusion. I think this study virtually guarantees an open data commons built of health records.

Because, you see, the very thing that we need to build a health data commons is patient control of health data.

Right now, patients don’t control. Large institutions do. Large companies do. Large hospitals do. So we can’t control our privacy, whether to keep something private forever, or to donate it to science.

But if we get the control that this study says we want, then at least some of us will make that donation. And it doesn’t take very many people making the choice to contribute to create a glorious resource.

Take Wikipedia. It’s remarkable how few people, as a function of total Wikipedia users, actually make Wikipedia. From the Quora page on Wikipedia contributions:

According to Wikimedia’s estimates, the larger Wikipedias (e.g. English, German, French) have 0.02-0.03% of visitors actively contribute.

Wikipedia has roughly 100,000 active monthly contributors, with about 40,000 of those on the English version. It looks like about 10% of users (~4000 for English) are very active (defined as 100+ edits per month).

If you don’t want to log into Quora, you can see the graphs on which these conclusions are based (Core Editors Are Small and WM Articles V Contributions). But .03%? That’s tiny. I bet we can get .03% of all patients to donate their health records to research. Hell, Buzzfeed is built on the bet that we could get .03% of the population to watch a GIF of a cat chasing a laser.

At the current population of the United States (311,591,917), .03% nets us 93,477 people. That’s a massive clinical research cohort - more than 10x bigger than Framingham. My instinct is that we’ll do better than that, but even if we don’t, that’s a big enough number to change the game of mathematical health modeling.

You see, the secret sauce of the commons is asymmetry. A small number of people making an unreasonable choice, a choice to share, to be digitally naked - that’s all it takes. But if we don’t control the privacy over our own health records, that small number will never get the chance to make that choice.

So I view this study as unmitigated Good News. It means that people are going to get more and more pissed that they don’t have control, and that makes it more and more likely that they get control. Patient empowerment is the first step on the road to open data. It doesn’t take all of us. It just takes all of some of us.

Data Versus Pundits (Science Edition)

Mon, 26 Nov 2012 14:10:00 -0800

I’m more than 20 years away from my last math classes. And though I spent my startup experience around some very smart mathematical types, I didn’t learn enough at any point to actually use data in models, make models, or otherwise become one of those “big data” types you read about…well, fucking everywhere.

But I did learn a healthy dose of respect for what data, and math, when put together by the right people, are capable of. And since signing up with Sage Bionetworks, first as a Board member, and more recently as a member of the management team, I’ve learned a lot about what data can do now that processors, storage, bandwidth, and sequencing are all brutally cheap.

The key is, as per @DCDave, to reset your expectations. Cheap, plentiful data changes the epistemology of fields. It did so in baseball, which went from trusting a scout’s “gut” instinct to an intensely data driven science. It did so in weather. This year, it did so in politics.

But it’s very hard to reset your expectations when you’re at the top of a traditional industry. The punditocracy’s dismissive handwaving towards Nate Silver is all I can think of when I go to traditional events on science and health these days.

There is a total unwillingness to reset expectations in the sciences, to go from the certainty of the expert class to the probabilistic worldview of a world overrun by data and new entrants. The idea that non-credentialed experts can generate hypotheses, that roles will be fluid between citizen and researcher, that “normal” people (as if a data scientist is normal - she just might not have a SCOPUS identifier or a PhD) can generate useful science insight, that publishing on the web might be a complement to publishing in peer reviewed journals? Those are all easy to accept with expectations reset by data.

They’re very hard to accept if you attempt to retain the world as it was before sequencing costs were crashing, cloud storage and processing were cheap, and publishing business models were broken to hell by the internet. Those scientists (and many pundits of science) at best ignore and at worst demonstrate bitter anger towards the relentless march of data and networks that “sail blithely on regardless of the carefully worded communiqués that emerge from a parade of meetings and consultations,” as Jonathan Zittrain once wrote of the internet.

I understand why a “normal” scientist might not want to see this. Imagine: you’re 40, you went from college in 1994 straight to grad school, graduated in 2000 with your PhD, postdoc til 2005, just got an assistant professorship and a piece of a lab to call your own, and your first NIH R01 grant last year. Now suddenly you have to deal with big data everywhere?

But it’s the pundits that are the worst. Like retail politicians and baseball scouts, regular scientists will hop on board fast with using data, as soon as it’s shown to work better than ignoring data. It’s the elite that will change the slowest. They’re the ones with the most to lose with a change to the status quo because they are already in charge.

The journal editors. The old guard. Those who treasure their hypotheses as precious snowflakes never to be shared, unaware that their ideas often exist in a matrix of data that can be generated as a commodity, findable via algorithm, and often disprovable through algorithm. They have the most to lose in the transition. And they clearly see themselves as the shoulders in the Newtonian equation when we transfer to data.

But that’s not what this is about. In the end, it’s about increasing the probability that we get things right over the way that we try to get things right today. And the philosophies of science that served us so well in our history of expensive, unavailable data will not serve us well in our current history and its inheritors, who have to figure out what data to delete, not what data to save.

Models will help. Open data will help. But a more supple, more finely tuned epistemology of science, one that understands how data can drive the creation of hypothesis, not simply emerge as part of its proof, is the real key. Because the data, and the quantitative research that can emerge from it, is often going to be better at making predictions than the gut instinct that’s driven many of the narrative sciences (and I’m including both biology and health in there, though not chemistry or physics).

But it will take the establishment class either getting with the program, or getting swept away by the tide. Because in science and health, the funding is controlled by the elites. And they’re still deep in the grips of a pre-data way of knowing what they know.

YODA yada yada

Thu, 15 Nov 2012 22:12:00 -0800

Yale University published a draft “Open Data Access policy” on November 15^th to make available rhBMP-2 clinical trial data that they created in conjunction with Medtronic.

Let me get this out of the way before I go negative: the fact that this policy exists, is on the web, and is open to comment is an unalloyed Good Thing. It’s progress. It’s to be commended.

For those who aren’t initiates in the arcane world of open data or pharmaceutical testing, this is data related to a clinical trial of a kind of biological chemical found in our bodies that plays a role in bone development. Most of the time, this kind of data is collected once, then locked away, never to be seen or used again.

One of the reasons drugs cost so much is that we never re-use the data from clinical trials, so we never learn anything from failures, or from secondary uses of data. It’s an incredibly inefficient system. This project at Yale is an attempt to address that inefficiency by making the data available.

But that’s the extent of my nice words. What follows is a point by point review of the policy. The short version: this is not an “open data” policy. It’s a data access policy, and if they’re not going to fix it, they need to rename it. Because those of us who care about the definition of “open data” actually meaning something are going to criticize, persistently and loudly, if there’s any attempt to claim the title for data under this policy.

I apologize in advance for the length of this rant, but it’s a long policy. Also, they’re using the name of Yoda in vain.

I. Decision: There Will Be a Data Registration Process For Data Access

OK, I can live with this. We do this at Sage Bionetworks with Synapse, for example. Whether or not it’s a good thing depends on what terms and conditions researchers are forced to accept as part of registration, and who is allowed to register. Which brings us to the next point.

II. Decision: Registration For Data Access Will Include Investigator Disclosure And Submission Of A Study Proposal

This is not great. You’ll need all of the below to even apply for registration:

·      Principal investigator’s: name, degree(s), SCOPUS ID, primary employer, and contact information, including phone, mailing address, and email address. 
·      Name, degree, and SCOPUS ID of other key investigative team members. 
·      Funding source and conflict of interest statement (using a modified version of the ICMJE disclosure form) for all team members.
·      Certification of IRB approval (or waiver) from academic/university partner [see section III].
·      Project specific aims, main and secondary outcome of interest and analysis plan [an example proposal will be available on the YODA project web site], and timeline.

Huh. I guess this would basically rule out…well, every data scientist in the world that doesn’t have a PhD and a SCOPUS ID. Which is pretty much the vast majority of them. It’s a lockout policy intended to limit liability and contain research to the stuff Yale and Medtronic think is ok.

The only acceptable part of this is that they will publish the registration applications – so we’ll know what they’re turning down, I suppose. But that can also create a disincentive to even apply, as it means that you’ll have told the world the questions you wish to ask before you’ve even had a chance to ask them.

This decision also basically guarantees the data will never be integrated with other data, because then these requirements would have to be syndicated over to all the other data. So trials about similar diseases, genetic networks including rhBMP-2, computational networks? Segregated by this decision, forever.

This decision further creates the possibility of catastrophic success. Should this data actually become essential, the transaction costs of reviewing applications will skyrocket. I’m doubtful that the faculty reviewers involved will enjoy spending their time looking at incrementally different applications to access the data rather than doing, yaknow, novel research that helps them get more funding.

III. Decision: Registration For Data Access Will Require At Least One Key Investigative Team Member From An Academic/University-Based Partner

The justification for this is that “This requirement strengthens the likelihood that the data requester (and eventual user) will have the scientific capability to conduct the proposed analysis” and securely store the data.

Based on my experience, the odds that someone can use data are not strongly correlated with academic credentials. For every whipsmart data scientist in academia there are a dozen more with strong chops, secure Amazon web systems, and killer Bayesian models outside academia.

Imagine if Nate Silver had been required to go through this kind of process to access the polls? Only mathematicians with a political partner need apply! I’m sure all the pundits who called him a wizard would have loved to sponsor his application.

IV. Decision: Requests For Data Access Will Be Reviewed For ‘Completeness’ and to Ensure that Data Use Limitations are Met

This is an unnecessary bit of overhead, but if you’re going to make people file applications, and make faculty members review them, then this makes sense. Might as well make the review process as complete (not to mention time-consuming) for your own faculty as possible.

V. Decision: Limitations Will Be Placed On Data Use

“Data requestors will be required to certify that they will meet the following expectations:

·      The data will be used to create generalizable scientific knowledge. 
·      The findings derived from analysis of the data will only be publicly disseminated through the peer-reviewed biomedical literature or a scientific meeting.
·      The data will explicitly not be used for commercial purposes or pursuant of litigation.

What the hell is generalizable scientific knowledge? Where does it stop and start? Who gets to enforce the definition? Are we talking Karl Popper, or Kuhn, or Feyerabend – or Arbesman – here? This is so vague it gives those who would deny a data access request total power to say no under its rubric. An exploratory data scientist simply looking to test a model? Door, locked.

The findings will only be disseminated through the literature and meetings? No social media? No blogging? They won’t be published as new algorithms directly into R clients? Seriously? Utterly myopic view of how knowledge is now communicated.

The litigation thing I’m actually willing to give them. There’s too many law firms that would descend on this with fangs bared to start class action lawsuits, at least now. Until we have social norms and judicial precedent to deal intelligently with clinical data this kind of constraint might be part of the deal.

“Data will not be released to applicants whose intent is clearly based on commercial or legal purposes.”

OK, so if I keep a blog with Google ads, is that commercial? If I work on cancer as my 20% project at Google? Who decides what “clearly” means? What if I’m an academic sponsored by a Medtronic competitor (this appears, ironically, to be okey-dokey)? Again, broad reasons to say no to those wishing to exploratory computational research.

VI. Decision: There Will Not Be A Data Use Fee

Yay! Of course that’s just for a year.

VII. Decision: Medtronic Will Be A Party To The Data Use Agreement, With Authority To Enforce It

Oh, awesome.

Just go read all the terms and conditions and then don’t even bother applying. Anyone who actually goes through this is either going to be someone they already know or a total masochist.

Worse, the data use agreement gives Medtronic the right to snoop into your research. It’s a party to the deal. It has the right to enforce the deal. Sign with care!

“DATA DISTRIBUTION”

Data will be distributed via an “encrypted USB flash drive via FedEx (or similarly secure shipping company)”.

Glad they specified a “secure” shipping company.

Seriously? A USB stick? The thing that gets left everywhere, falls out of pockets, carries viruses into Iranian nuclear facilities, and is used for corporate espionage? After all this detail to secure the data the actual delivery method is very likely to result in the leak of at least an encrypted copy of the data into the wild (when there is a Wikipedia heading on “USB drive data leakage” you may have a problem with it).

If you’re going to go to all this effort to lock it down, just keep it simple and use a secure cloud service and encryption. At least then you can monitor the download and there’s not a physical copy floating around for an exhausted postdoc to drop on the floor and get picked up by a janitor.

CONCLUSIONS.

Well, I doubt seriously that this is going to change significantly. This reads as if it were written by Medtronic and then a well-meaning committee of scientists attempted to ameliorate its worst excesses. And I’m glad for the fact that it was created, placed online, and comment was requested.

We have definitions for openness precisely so the words are not used in ways that mislead us. To make sure that when something claims the mantle of “open” that it is indeed open. To create a certain level of quality control in the open world.

There’s nothing wrong with not being open. What’s wrong is claiming to be open, when one is in fact not open. It is important that we hold all to task against the open definition, and that we call something what it is. Because this is not an open data policy. It’s an Yale Access to Data Agreement – a YADA, not a YODA.

And there’s nothing inherently wrong with that. It’s not bad to make data available under kind of insane terms, because at least it’s better than not making it available at all. But this isn’t open. Not even close.

Thanks, Pete

Wed, 07 Nov 2012 12:01:00 -0800

Pete Stark lost his House seat last night. He’d held it for 40 years.

He is a divisive figure to some. He’s a hero for his legislative work to others. And I’m not going to address any of that, though I believe history will judge him very kindly on the merits of his record.

I just want to thank a man who did more for me than I did for him.

Pete hired me in 1996, when I was so wet behind the ears I could water plants just by sitting next to them. I was a legislative assistant in his Washington office, which is a fancy way to say I read through the legislation that the new Republican majority was ramming down our throats and tried to find some of the nastier bits and get them taken out.

One little anecdote from those days. It was my first few weeks on the job and I was overwhelmed. A previous staffer dropped a massive set of files on my desk and said, “this is the Headwaters file - you’ll want to start reading it soon.”

That night I got a call at home, after midnight, that a clause had been inserted into an appropriations bill that would allow the logging of Headwaters. I had to taxi into the office, start writing Dear Colleague letters, and help whip a campaign to get the clause taken out when the bill came to the floor. I had no idea what I was doing. I was 23.

When the floor debate came, I was watching it on CSPAN, which ran 24 hours in the office. We were just across the street. Pete grabbed his 2-minute comments that I’d drafted off the printer and walked over.

I saw him stride up on the TV, and to my horror, he started reading the press release, not the speech for the chambers. The press release was completely over the top. It called Charles Hurwitz, the then-owner of much of Headwaters, a “blood sucking corporate vampire” (not my words, but the product of a gifted, and awesome, press secretary).

You can’t attack an individual like that on the floor of the House. Pete was immediately called for censure. I nearly wet my pants. I was surely going to be fired for this mistake.

The time it took for him to walk back to the office was epic torture.

He came back, clapped me on the back and said, “Aw hell, I decided to read the release because I thought it would be more fun - relax, kid.”

He was a great boss.

Pete gave me my first big professional break. Pete’s office is where I learned how to run a database. His Mac is where I wrote my first webpage, in plain text in 1996. He gave me my first bottle of great wine. His chief of staff got me backstage at a Jerry Jeff Walker show at the Birchmere, and took me to the National Democratic Club on the night Clinton was re-elected.

Pete, I left before I ever got good at working for you. But you were good to me, and you served your country well. I hope somehow this gets to you. Thanks for your service, and thanks for giving a kid a lot of breaks a long time ago.

ps - we won the debate on Headwaters, and I look forward to taking my son to see those trees, when he’s old enough, and telling him this story.