path: root/transcripts/startup-science-2012/john-wilbanks.mdwn

Several years ago, the NIH said that tax-payer funded research - about $30B a
year- must be made freely available within 12 months from their publication.
That's about along enough time for publishers to recoup their investment from
peer review. Then we publish it into pubmedcentral.

This was put into effect. It's quite easy for authors to comply. There is
essentially no adverse data that has presented by publishers ever since this
happened, in the years of this policy. There has been no data publically or
privately presented to prove that this negatively impacts the publishers'
profitability or not. In fact, publishers have increased in profit over the
past 10 years. If they had this data that would prove it's bad for business, I
can't believe they wouldn't show it.

Well why not expand this to the other federal government? NIH is just one of 16
federal agencies that funds research. There is an additional $30B of research
that is uncovered by the mandate: NSF, Department of Defense, National Science
Foundation.. the White House has thought about this in the America Competes
Act. There has been two public requests for information, many of us have
written responses, publishers wrote responses to both. There has been
legislation in both the House and the Senate, introduced that would expand the
coverage to the other agencies.

They thought we were going to take their babies. No publisher has been able to
show adverse data, or adverse impact data. That is enough to move your profit
negatively if it is going to hurt your profit-- it'si mplemented on 50% of it.
They say things like, "It's not just going to kill babies, it will subsidize
foreign governments and players to know what we know." Xenophobia. Wonderful.

And there is legislation introduced that would make public access policies
illegal, like the Research Works Act, sponsored by Reid Elsevier. It came in
right after SOPA and PIPA. It basically said that the signatories would ot be
allowed to serve as editors or reviews as Elsevier, I had to back out of a
special issue of a magazine. I signed this pledge.

The politicians reacted by saying they would not take the Research Works Act
forward. They were stunned by the emergence of a vocal open science community.
So the question remained, after RWA or the Cost of Knowledge petition, why not
extend? I write blog posts and responses to information drafts, I got invited
to attend this meeting on May 8th with the President's Science Advisor. I had
just come back from vacation. So on the 7th I took a redeye flight back from
the east, back from SF to DC, I showed up at 7am, there's no showers at the
airport. We went from that.

And he's a really good guy. He's a scientist. He buys into a lot of the things
we're trying to do. This is what OSTP does. The debate for the last 10 years
for open access has been open access advocates, open science advocates on one
end, and publishers on the other end. And congress in the middle. Publishers
have the louder vvoice, at least in the past. So what is going on here? We were
going to get the charlie brown noise from this. We need other people involved.

This is me in a suit at OSTP. That's us sitting under the sign when we decided
to do something different. There is a platform that the administration created
called WeThePeople.. if you get 25,000 signatures within 30 days, you can get a
formal response from the president. Some of these petitions are like, rename
the sea of japan, or legalize weed. And socially, that's not going to happen in
an election year.

So we saw an opportunity to see an existing topic that they have considered,
and something that- where if we could get that number not just in 30 days, that
we could attach their attention for people who have previously not been part of
the conversation. Like the Obama Social Media Team. If you go back to this
slide, heather and joseph on the left from Spark, put together the
organizational backend and networking to put together the petition. I did the
petition filing, and did the work on the etchnology side. Mike Carol wrote the
text. The press coordinator coordinated with the open access community. We
divided up the work.

We put together the petition and the petition text within 10 days. We wanted to
ddo machine readable access to all of the literature. You only have 120
characters for the headline and 800 characters for the text. We did not go for
the open access definition. These definitions require liberal copyright
license. We didn't think we could get that. We also wanted to be vague to let
the White House respond positively if possible, if they could interpret
language in a way that they could get a policy through. We're trying to win
here, so we have to be pragmatic.

We chose the hashtag called #openaccess. We knew we would get criticized since
it's not OA explicitly, but whatever we wanted to tap that. We posted it on
Sunday, and then on Monday morning the idea would be that there would be a few
thousand signatures by Monday morning. OA Monday was this idea for the first
day push. That's the overall name for the project. Open Access Monday.

On the first day, we have these organizations already. The first 15,000 we hit
within a week. We thought that we would get 5000 total, but we got 5000 on the
first day. Amanda Palmer.. had 500k followers, but we got no bumps. Tim
O'Reilly we got a bump from, and that came during the first day. We got bumps
from committed organizations. To register and to vote, you have to solve a
captcha, get a confirmation email, there was a drupal authentication error, you
have to empty your browser cache, there's no oauth, there's nothing modern
about it. We hit the 25,000 number within 13 days and 24 hours. And we stopped
pushing.

William Gunn put together the google doc where we tracked this. All 50 states.
It was patient advocates and open science advocates that pushed for this.
There's too many people I could thank. Heather was tireless in blogging and
tweeting and if it weren't for 100s like her, we wouldn't have done this. I
just wanted to pick at least one person to mention.

I pledged that I would recoup the .. peanuts video.. so ... Joseph was asking
about the economics of open access. We know that publishers are meeting daily
with OSTP. We are not, though. It's a big day when we get a meeting. We fly in
from California. They have people in Washington that do nothing but take names.
We know there's a battle for the policy. They are saying that jobs will be lost
if you open up access to research.

This is a response to the petition on digitizing federal records. The petition
did not hit the number. The administration replied that you are right, but
we're not going to do anything about it. Many responses look like this. The
form that looks like this is going to be someone from OSTP on the platform.
Hopefully it will be a response that says yes, we will implement a thing across
all 16 agencies. So then we will respond to the response.

If we get a public access policy, our response should be celebratory. You
should be getting a press release ready if something is about to go out. We
should talk with each other. We will get a spreadsheet going with a list of
companies that will do press releases about this, about this celebration. If we
don't get what we want, we have to get more radical. So maybe we will paywall
Wikipedia for 24 hours. Like pay so that you can access Wikipedia.

We could do that with open access publishers or even Wikipedia. They haven't
agreed to this but we should push for this. This is something that Washington
is deafly affraid of. I worked on the hill in 1994 and that's when legislature
was a verb for the last time. To be BTU'd, to lose because of vote.. so now
there's to be SOPA'd, so we have to put the fear of SOPA if we want the White
House to give us a positive response.

We have to ask all the time. There are may claims about open access versus
publishing. We know we have tons of data locked in the publishing companies.
Their profits are not being hurt by the NIH policy. Half of their business is
from STM from NIH grants. In the absence of data, we have to look for data
elsewhere. PeerJ is a for-profit VC-backed model. The non-profit PLOS One model
is too expensive. When is the last time when a VC backed company was started
because the non-profit was priced too high?

We're starting to separate the functions of publishers. We have talked about
this a lot: dissemination, certification, registration, validation. These are
getting torn apart. The ability of the research funder to step in and make a
change about the access principles, makes it easier to tear these pieces apart.
That's what the internet does for content. The institutional funding
environment- where the scientist doesn't see the pain- because it's faded into
her library.. but if you're starting a company built on scholarly content,
you're screwed because you have to engotiate with all the maor players. Only
Google has been able to do that.

If you want to integrate 10k public health data with the papers, you're
screwed. The transaction costs for you on a startup are higher than anywhere
else, they should be the lowest because we need the most innovation. We need
all the materials research so you can start solar companies. We need materials
research available so that battery companies can get started. It's not just
health and biology. It's been hell. This is not my day job. I have a fulfilling
day job, but this isn't it.

The reason we did this is because we thought we had a chance. We will know by
the Democratic Convention in August as to whether or not we have accomplished
this. We've increased the number of people who are aware of this from 10,000 to
at least 100,000 worldwide. Many people in this audience have done far more
than I have, so that's great.

Keep asking. Until we see adverse event data I am not going to believe that
open access hurts anyone. Keep asking.

----------

So, the NIH policy includes funding to pay for the costs to make things open
access. There's OJS from the Public Knowledge Project. The traditional
information in a paper is knowledge. You will see many different pieces in
publishing, you will see figures published on figure-supporting sites, and
texts published on other sites with varying degrees of certification... most
papers now are advertisements for multiple years of research, and scientists
will pay for that. Reagents will go on credit cards.. not on the level of deal
where MIT pays $10 million to a publisher for access to 5000 journals. I think
the funding will come directly from individual scientists who have a vested
interest in promoting their own work. That's why I like to buy into
experimental ideas of PeerJ where the lab credit card will pay for publishing.

New battery production, let's say DARPA and DOD might fund that. This
information is published and somebody in China based on that information,
because of the availability of that information, then develops a new tool. So,
on the surface there is some .. there is some validity to that concern?

If it's been published at all, all the Chinese have to do is subscribe. They
already have the article. The tool that is used in this is that you don't
publish until you file a patent or something. These things are already
published, they are just paywalled. The main people that this benefit is the
publishing companies. It doesn't benefit the researcher. It doesn't benefit the
technology company that licenses that, because it gets harder to build
incrementally.. the law of averages isn't that we have better odds of people
breaking through, 1 in 1000 breaking through is still the same. 1 in a million
is still that risk. So you get better results. It's mathematically one of the
best way to accelerate the rate of innovation is increase the number of people
that participate. But competitive advantage is a red herring; published things
are alerady published. It's not about disadvantaging the US in a competitive
way. It's open access versus vested interest by the publishers.

Economics analysis.. the beginning Internet.. there are two structures:
vertical and horizontal structures. The battle is between the two. The internet
is a horizontal project. The problem with open/close, why .. why is closed
items have that? They don't. They don't have a loss. They cannot demonstrate a
loss for a reason that.. then they illustrate the weakness in how they can be
attacked. When we did the financial analysis back in doing strategic analysis
fo the internet, there's a war that goes on between these two forces. People
lose hidden advantage.. how do you make the horizontal win over the vertical?
The vertical's primary protection is through hiding things as much as possible.
That's why you don't get anything out. The horizontal can show that the
velocity of the transactions.. if you look at the velocity of money going
through the organization. So say... the money going into the publishers, then
what is the total impact of money going into that, you would see that small
amounts of horizontal activity is 100x more powerful than any particular
vertical organization. That's a very powerful argument.. the problem is that
most scientists aren't aware of the impact they have, and the monetary aspect
and being able to trace that. That will let them integrate it into the total.
You can have a real-time display with twitter, but you have dollars attached to
activity. That's one way of being able to press the advantage in this
particular environment.

Quick question.. about that White House meeting. Did they say what their
response would be? It was after we left the meeting that we got the idea. I was
worn out and pissed off. I was in DC for six hours and I was on a plane for 2x
as long as I was there. It was another meeting that was just.. the same thing.
The four of us got a meeting, and we put on our suits.. and those meetings
happen every day for the publishers. We have no idea what their response will
be. We want people to be in that room that aren't in there already. I can tell
you that they are very much aware that we exist in the West. So at least that
has benefited.

We should all give Joseph a hand because doing these sorts of conferences
sucks.