Enjoy the Disruption

Just spotted this as Slack has recently updated their API Terms of Service, though it may have been in there before this update?

Of course, faced with either jail time courtesy of the government or breaking Slack’s TOS, which do you think a developer would choose? Slack legal time just covering their arse with this clause.

https://slack.com/terms-of-service/api

9.1 Government Access: You will not knowingly allow or assist any government entities, law enforcement, or other organizations to conduct surveillance or obtain data using your access to the Slack API in order to avoid serving legal process directly on Slack. Any such use by you for law enforcement purposes is a breach of this API TOS.

Courtesy of JSTOR. Well, hopefully not JSTOR as an organization, but at least one of their employees thinks you should get knocked out. That was just learned from a Freedom of Information Act request obtained from the Aaron Swartz saga. You can view all of the files obtained from the FBI, Secret Service, MIT, JSTOR, US Attorney’s Office and others at swartzfiles.com

The above is from page 4 of more than 3K pages of internal emails from ITHAKA.org, which is the non-profit responsible for the JSTOR digital library (whose mission is “to foster widespread access to the world’s body of scholarly knowledge”). It’s a couple of systems administrators communicating in the context of high download activity going on at MIT (which we later learn was Aaron Swartz downloading academic papers).

Sadly, despite its mission, JSTOR  believes it is no longer capable of sustaining itself in the digital era without resorting to restricting access to knowledge. The Internet and World Wide Web, designed to spread information, have changed everything, sometimes ironically.

For its part, JSTOR settled a civil suit with Aaron Swartz out of court, and later told the US Attorney’s office in Massachusetts that it no longer had an interest in further proceedings. The US Gov’t didn’t stop, however.

Getting punched in the face was the least of Aaron’s worries. The FBI had much worse in mind for him (and succeeded in doing) for the act of downloading PDFs. It culminated in the unnecessary loss of his life.

There’s too much in the FOIA for any one person to really go through entirely, but have a perusal of the documents and let’s remember what Aaron stood for, and that we can do better when it comes to being stewards of the world’s academic knowledge. As I’ve stated before on this blog, without Aaron there may have never been the motivation to start the journal PeerJ. News of Aaron in 2011 was the catalyst to finally say, “Nothing about the outrageous costs in publishing is changing. What can I do?” Thankfully Pete Binfield agreed with me and we set out to make public access to research faster, cheaper to produce, and most importantly free to download. No one deserves a punch in the face for pursuing academic knowledge.

I just don’t understand WTF the Mozilla Foundation was thinking on this one. It’s akin to making someone a CEO who donated to segregation campaigns in the 1960s. You just wouldn’t do that.

Gay rights have not yet achieved the same acceptance as other civil rights have, but they will do so undoubtedly one day. And when that day arrives, and it already has in most of Mozilla’s fan base markets, it is going to haunt the Mozilla Foundation even more than the decision today is doing. Mozilla looks to the future in all it does, except when it comes to its leadership apparently. The future profile of CEOs will not have anti-LGBT beliefs, and Mozilla needs to be skating to where the puck is going, not where it’s been.

What is really troubling is that the origins of the tech industry, the West Coast tech, were about tolerance and civil rights. Personal computers and related technologies were about freedom from oppression. For a tech non-profit to install an anti-LGBT CEO is completely 180 from why and how tech evolved from its early days. Technology isn’t just about what you do, it’s also about who does it. They go hand in hand.

If this was simply an oversight that was missed in due diligence of the CEO’s past then OK, but make the change happen. 

With Google announcing massive price drops (https://developers.google.com/storage/) it has a lot of developers and tech managers re-thinking the use of AWS, Rackspace, etc. Certainly the $.026/GB of monthly storage and lowered compute engine prices make stiff competition. I am not convinced yet, however. 

First, depending on what one is trying to accomplish, the new bandwidth prices announced by Google are still more expensive than AWS once you reach a certain volume. For heavy bandwidth out users then, it may not make sense to use Google over AWS if pricing is your only concern. 

A larger issue is one of trust, and opinion shows many developers and decision makers are in agreement with me on this one. Google has failed users and developers countless times as it has pulled its APIs and services. Google has a one year notification term for its cloud computing services, but that can change, and one year is possibly not enough time if your entire business is structured around the service. The fine details of the cloud computing terms of use with Google could also give one pause compared to AWS.

Further, AWS revenue accounts for ~7% of Amazon’s overall business revenue. In comparison, the Google cloud computing business brought in roughly 1.5% of Google’s overall 2013 revenue. That alone is enough to give me a second thought when considering trusting my business or computing needs with Google due to its lengthy history of pulling services. 

There are of course other reasons to distrust Google, which I won’t go into now.

AWS is likely to follow Google and continue dropping its prices as well. So, any prudent decision-makers should take a wait-and-see approach before jumping ship. And even then make a careful analysis (including future growth scenarios) of how your business or process actually utilizes the various cloud services to determine the real and future costs involved. 

Thanks to this tweet by @CameronNeylon we see a very loaded question about Open Access licensing consequences from NPG. I should say also that there are a few other misleading questions from this NPG survey - which look to be as much as propaganda as (poorly designed) survey material. 

This seems to be the new scare tactic for anti-OA activists. Explain one possible commercial use case, one likely to offend or upset academics, while neglecting to state the many other reasons one would want to allow commercial re-use: teaching in academic situations (if the academic is paid that’s commercial use), text/data mining for new cures, in certain cases physician’s may hesitate to use or cite the research after developing new tools based on that information, etc.

Maybe you have a moral reason to not want a biopharma giant to profit off of your Open Access article. Fine, fair enough, but that actually doesn’t prevent them from using the information - facts can’t be copyrighted. More often than not, the use of a Non-Commercial OA license (e.g. CC-BY-NC) has the opposite effect from what the author hoped to achieve. Peter Murray Rust explains this in an excellent writeup here. An NC license doesn’t prevent the publisher you use from profiting off of the material, it won’t stop pharmaceutical companies, but it does deter others with many legitimate use cases.

Had all software development in the early days of the 60s, 70s, and 80s restricted commercial use then we wouldn’t be here today discussing this. Open licensing with explicit reuse for commercial interests has been the foundation of software that powers a majority of the world’s websites, and software that powers research activity in academic institutions. The parallels with Open Access articles and the early days of open sourced software are massive.

For sure, all academics should be aware of the possible uses of their research, but the point is to make them fully aware of all use cases, not just a select few intended to scare. And we also need to understand that choosing a restrictive NC license may have unintended consequences as well. 

Updated to add: Many, including the Budapest Open Access initiative, do not consider OA licenses with an NC clause to actually be Open Access. I agree with this position.

Whoa. Some serious debate flying around after the newly minted Nobel Laureate and Editor-in-Chief of eLife wrote that journals like Nature, Cell, and Science are damaging science. 

On one side you find the supporters, such as co-founder of PLOS and UC Berkeley Professor Mike Eisen, who hopes Randy’s actions can inspire others. In the other corner are the haters shouting hypocrisy.

The way I see it, Randy had two options:

1. Say/do nothing at all, and thus inspire no one to take action.

2. Do what he did.

I’m on Randy’s side here. If we’re going to start making the changes that are needed within academia then someone must speak up, even it comes laden with ad hominem attacks of hypocrisy and conflicts of interest. And note my own COI as a co-founder of the Open Access journal PeerJ.

Let’s examine the fallacies of the naysayers’ arguments:

1. Sheckman’s words ring hollow because eLife, like CNS (Cell, Nature, Science), has a high rejection rate, even though it is Open Access. As editor-in-chief of eLife he has a conflict of interest and should not make such statements.

This argument is ignoring the actual message and its possible impact. Whether eLife is a luxury journal or not doesn’t change the message being told. Same with the conflict of interest. Those are all separate issues from the message and how people can act on it.

Additionally, it’s naive to think that everyone boycotting CNS would all of a sudden 1) start publishing with eLife and 2) that other publishing options (PeerJ, PLOS, small society journals, F1000Research, preprints, etc) wouldn’t grow.

2. Even if everyone boycotts CNS, it won’t change things because the next three highest impact factor journals will replace them.

This is a non sequitur argument and the silliest one of all. It ignores the fact that if CNS actually did go out of business, then Sheckman’s words will have achieved an f'ignly astounding result. Do these naysayers actually believe if everyone boycotted CNS that it wouldn’t have other knock-on effects within the overall academic debate on impact factor? 

What would really happen if everyone were to boycott CNS is that our funding bodies, governments, academic departments, etc would take notice. It will have meant that academics’ habits have actually changed. That will lead to other changes. It won’t just lead to the next three journals replacing CNS; that conclusion is unsupported as can be.

3. Sheckman can only say boycott CNS now that he has secured his Nobel prize after publishing more than 40 times in those journals. Younger scientists don’t have that option.

This is an ad hominem argument. Again, whether Sheckman is being a hypocrite or not has no bearing on the message that things must change in order to improve scientific research. Whether younger scientists have the luxuries that Sheckman has now or not has no bearing on the message. The message is “things must change.”

That we’re now debating the merits of Sheckman’s call means what he said is already having an impact. And let’s remember that most hearing his message are not academics, but the public who are unaware of the issues at hand, but still have the power to change things through their elected officials.

If a Nobel Laureate isn’t allowed to state these things, then who is allowed? Reality is that everyone’s allowed, but not everyone has the voice that Randy now has. He can choose to remain silent, or he can try to have an impact that perhaps may help eLife, but will undoubtedly help advance science and other publishing experiments that are sorely needed. A rising tide raises all ships.

Finally, whether his words will have any real results at the end of the day or not isn’t a reason to stay silent. When we’re trying to push the boundaries we go into action knowing full well that failure is a possibility. If success were guaranteed then we’d have no need for inspiration.

Kudos to Randy Sheckman for having the courage to do what he did, despite knowing the heat he’d take. That makes him more worthy of the Nobel than ever.

I returned yesterday from Birmingham, UK and the 2013 ALPSP international conference. It was great to listen, to present, and of course nice that PeerJ won an award for its publishing innovation (we’ll do a proper post about that on the PeerJ blog shortly). 

I spent some time talking with different society publishers and staff. This was new for me. My co-founder at PeerJ is much more seasoned in the publishing world than me - I’m the outsider coming from more of a quasi tech/academia/academic software background. Thus, my perspective on the current situation facing publishing is probably refreshing, naive, flat out wrong in some areas, but dead right in other areas. Yes, I’m qualifying what I’m about to say next :) …

If I had to choose one analogy to describe the state of publishing it would be a deer paralyzed in a beam of on-coming headlights. From numerous discussions at the annual ALPSP meeting, it became apparent that society publishers in particular are standing still in fear, unsure of which way to turn, or to make that risky move. From a high-level bit of questioning, it seemed many publishers didn’t have the right mix of people in their organizations for the digital world.

There was an interesting plenary session with Ziyad Marar (SAGE), Timo Hannay (Digital Science), Victor Henning (Mendeley/Elsevier), and Louise Russell (a publishing consultant). Ziyad and Timo seemed to have opposing perspectives on what a publisher today should be composed of or targeting. Ziyad was on the side of focusing on content, while TImo more on the side of focusing on the tech. That’s a simplification, and both of them probably value and implement both in their orgs, but the extreme views are the two sides of what I see in publishers today. Those who do not have people in place, either through empowerment or directly though titled positions, to make technology a center piece of their organization risk being stuck in the headlights.  

I’ll be even more specific than technology, it’s user experience. We can all blame Apple for this one too. It may not be dominant over content just yet, but it’s coming, and those who do not have the tools and people in place will be left behind. This was missing in the organizations of many who I spoke with at ALPSP. And to do user experience right, you need to be focusing on the right technologies and the right product strategies,  with the right people. I gave a high-level talk on cloud computing and many commented how they just didn’t have the people within the society to make it possible. That’s a mistake, not because cloud computing is the answer, but because you can’t then focus on building the tools needed to please the future reader, author, reviewer, etc.

What’s also interesting is that user experience isn’t something new to publishing, it’s been going on for 300 years. We think of publishers as just delivering content, but they’ve been tweaking the layout and typography of that content for centuries to make it more legible, more comprehensible, etc. That’s user experience. To make that happen today though requires people with different skill-sets than even a decade ago, and those people are either avoiding careers in publishing, not given priority, not empowered enough, or not even considered.

Before Pete (PeerJ co-founder) and I announced PeerJ in 2012 I related to him a little research that I had done on PLOS and lack of technology focus. This came about because we wanted people to know how PeerJ would be different than what had come before. I went through the WayBackMachine on the Internet Archive to look at PLOS’s website history. One thing stood out to me - it took several years before any tech-related people started to appear in the staff list and even today (like other publishers) tech empowered employees are not in positions of business strategy. I wanted PeerJ to make engineers equal to the editorial positions, and that’s how we’re different. That’s what’s needed if society publishers are going to continue. 

Really, it isn’t tech versus content. They support each other, the only problem is that there are a lack of people in the position to make it happen today. Yesterday’s typographers are today’s user experience engineers, today’s human-computer interaction experts, today’s software engineers. That’s what scared me the most in all of my conversations at ALPSP, the missing people. 

That’s the question I’ve been trying to answer for myself over the last few months. One would think that without the UK’s Guardian newspaper slowly publishing new information on a weekly basis that we’d have already forgotten about the domestic spying, encryption disabling, etc from the NSA and GCHQ. It would have been news for a week and then turned over in a new cycle with more important headlines such as Mylie Cyrus and such. 

For sure, the revelations from Snowden have caused more debate and action in the U.S. congress (both House and Senate) than Manning’s ever has (and in the UK’s parliament). And that’s something. But the amount of apathy from the general public is baffling. People are outraged, I know, but at the same time, we don’t seem to be doing much about it either - hence apathy. A lot of shouting, but little action. It’s really odd. Why aren’t we doing more? That’s the question I’ve been struggling to understand.

Lately I’ve been thinking about Maslow’s hierarchy of needs in relation to this question (diagram above).  There are a few caveats with the Maslow hierarchy (usually represented as a pyramid). The main caveat is that the needs can be fluid, i.e. some at the top may be at the bottom and vice versa depending on the location, culture, time, age, etc of the person. And to make this easier, I am categorizing all of the revelations that have come out from Snowden as privacy.

As best as I can tell, Maslow would have placed privacy into the highest need or “self-actualization.” The highest need, shown at the top of the pyramid in blue, represents only 2% of the general population according to Maslow (remember too the caveats above). Interestingly, Maslow also considered “self-actualization” to be the future of humanity, i.e. the best that we could become. Those at the top have a need for privacy, not because you’re hiding any thing in particular, but because you value it as an equal attribute to your creativity, your pursuit of intellect, and personal morality. In the strictest interpretation, those who don’t believe in privacy don’t believe in creativity, intellect, morality, ethics, etc either. 

It’s a curious thing when someone says “innocent people don’t need to hide anything.” Or similarly, “the innocent have nothing to fear [about the privacy invasions].” Such statements come from people who actually haven’t achieved self-actualization for themselves yet. They’re further down the hierarchy is one interpretation. Another interpretation is that they believe all people should be held to a lower level of that pyramid; i.e. you should only be as high as the weakest link. What’s curious is that this is nothing new. We’ve been subjected to this for thousands of years from leadership in republics, monarchies, totalitarian regimes, all of them. No large populous government has truly sought to bring about the highest level of that pyramid. 

If we were to measure government in terms of Maslow’s needs, it would fall into the second to lowest category of “safety.” In thousands of years of human civilization we haven’t really moved beyond that, which is another huge array of “whys” waiting to be answered. I think we see a glimpse of progression in the 4th amendment to the US Constitution (emphasis mine). 

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

For many people, metadata, phone calls, Skype chats, email, Facebook messages, etc are an extension of their persons. If you are not free to do these things securely then you are unable to attain “self-actualization.” You are being deprived of achieving more. But still, why are we, as a populace, not more angry? Why are we not doing more to ensure this security of our persons? I think an answer, and there probably isn’t just one answer, is extremely complex.

In part, we can go back to Maslow’s assertion that only 2% of the population have reached self-actualization. If this is true, then only 2% of the population is concerned about privacy. Put another way, that means privacy is not the primary concern of the population. Remember again the caveats from above, that the needs are fluid and not binary. Everyone, to some degree, is probably concerned about privacy, but it’s not the primary need that they have. Looking at the list of needs in the pyramid above, things like food, shelter, jobs, love, friendship and others all come before privacy, intellectual pursuits, lack of prejudice, and creativity. And in fact, if you’re in power and want to change the debate, the easiest thing to do is to tell people (and remind them) that they need to go to war (Syria), that the economy still needs recovery, that marriage straight or gay (love in the hierarchy) is a more important debate than privacy. It keeps privacy out of the debate, and with limited privacy you have more power.

In effect, if 98% of the population doesn’t consider privacy amongst their primary needs then action will be limited. I actually think more than 2% of us are in the “self-actualization” part of the hierarchy. However, I think we’ve been deceiving ourselves about how filled those needs actually are. Once you fill the first two levels, it becomes more difficult to discern what the important things are (and again to each person there is fluidity in the needs). We are probably deceiving ourselves that access to Facebook, Google, television, etc are filling our needs higher up the hierarchy  And if we’re deceived into thinking we’re filled, then a small thing like the removal of privacy becomes less of a concern. Or focusing the debate on any thing but a “self-actualization” need will mitigate that concern.

Imagine the reverse, where all you had were the first two levels filled, and only one thing at the top (the next three levels). How would you then feel if that one thing, for example privacy, were stripped away? You’d certainly notice it more than if you had basically every other need filled. Perhaps this is why people from countries that are worse off than the US/UK/Canada/Australia/etc, seem to take more action in the news. They’re barely at the second level of needs, sometimes at just the first level, and then they have things like democracy, freedom of expression, privacy (Zimbabwe, Syria, East Germany pre-1989, Somalia, etc) taken away. You had one thing available to you that represented the highest that humanity could achieve and it is taken away from you.

Right now we are being deprived of the highest that humanity has to offer. We should not be merely satisfied that in the West we’ve achieved the first two levels (and yes individuals in the West are suffering on levels 1-2 still). We shouldn’t be satisfied that this is the best our governments have to offer. We shouldn’t believe that we’re still needing to go to war after 10,000 years of civilization (a level-2 need). We shouldn’t be confusing comforts such as prime-time television and iPhones with our higher needs. I believe we can expect more and offer more to ourselves.

Previously I had argued in “Science funding is borked” that we should be giving out many more and smaller grants, similar to the 500 startups approach. 

Now in a new paper published in PLOS ONE and reported in Times Higher Education, it seems that this argument is starting to gain some data-driven support. 

Here’s the PLOS ONE paper Big Science vs. Little Science: How Scientific Impact Scales with Funding

I’m a little baffled by the recent $35M funding round, which included Bill Gates, for social network for scientists ResearchGate. And of course the media baiting quote from RG that it wants to win a Nobel Prize for its efforts.

One look at the traffic stats tells you all you need to know as a potential investor. The graph below is from Alexa, which is known to undercount traffic, but it can be used to reliably compare competitors relative to each other. The redline is Academia.edu and the blue is ResearchGate over the past six months. Alexa ranks RG at the 5,433 most visited site, with Academia.edu at 2,686 most visited.

While RG claims to have 2.9M users, only a very tiny fraction of those are actives. Other traffic analytics stats confirm these numbers. My guess is roughly <5% at most are active each month (active meaning visit the site at least once). Then there is roughly the same amount of non-registered users visting the site per month. Are at most 145K registered users (+150K non users) visiting the site per month, with no revenue other than a small jobs board, worth $35M in funding? This probably values RG north of $150-200M. I must be missing something, or else Academia.edu should ask for $70M in its next funding round.

Then there is the value proposition. I fail to see any with RG. Academia.edu doesn’t have a lot either, but certainly more than RG. A lot of scientists have also compared Mendeley to RG and Academia.edu, which definitely has value proposition to users, and never raised such a round. One could also throw in FigShare, which has some overlapping functionality and more value add, yet it has never achieved anywhere near such funding to date. 

Probably the biggest strike against such a large round however is researcher sentiment. I’ve yet to meet anyone exclaiming RG as beneficial to their work (quite the opposite in fact with many reports of spamming and site scraping). On the other hand, I hear plenty of scientists talking about the value of FigShare, Mendeley, Papers, and even Academia.edu to their work. One has to wonder what kind of due diligence was done by the investors here.

CHORUS is another attempt by subscription publishers to defeat Open Access. Probably no better writeup than Michael Eisen’s of how deceptive the intent and logic of this plan is.

CHORUS claims that it will save the US govt money if implemented, as part of the plan calls for the shuttling of PubMedCentral. The fallacy of course, is that costs to the govt (i.e. taxpayers) will actually INCREASE as publishers now have control of the “Open Access” content via a CrossRef like dispatching service. To maintain this dispatch service requires passing on the costs to their journal subscriptions — that ultimately means the libraries and agencies foot the bill.

If this is really going to save taxpayers money, then why have the publishers that are part of CHORUS not provided a cost break down? Let’s see the expected operating costs, charges to publishers to join this new organization, and the details of the API restrictions and practicality of retrieving the full-text for data mining. Then let’s compare that spreadsheet to the cost of running PubMedCentral. But that’s just the financial cost; more concerning is the cost of giving control of Open Access content to organizations whose business model is counter to the principles of OA.

Are these APIs truly open? What happens if I decide to build an aggregator with this content that is supposed to be Open Access? Will I be restricted or charged for high volume access, because publishers are now losing eyeballs as researchers go to my aggregator search engine? Do we really want publishers in charge of the key to the only source of all embargoed Open Access content? How gullible do they think the Obama Administration is? 

CHORUS is a patronizing plan to researchers, libraries, and the American taxpayer. It’s a coordinated effort to sustain subscription-based publisher revenue streams and falsely paint PubMedCentral as a waste of taxpayer money. It is not about innovating on Open Access content and expanding its accessibility.

I joined Mendeley as head of R&D at the start in early 2009 and left end of 2011. Here are my thoughts about my time there and why I eventually left to start PeerJ. 

First, in terms of business success, regardless of your stance on Elsevier acquiring Mendeley, this was a win for the Mendeley team, so congratulations. I hear many saying that Elsevier overpaid (FT states £45M), but Elsevier is smart and they are getting a huge team in Mendeley that knows how to operate like a startup. That’s very near to impossible to grow organically with an established organization the size of Elsevier. This also gives Elsevier a major operational unit in London, which they did not have before.  That and other factors say to me that the amount paid is about right give or take a few percentage points. 

In terms of mission success, however, I am uncertain if this was a win. Mendeley had become known as the darling of openness, which in my view was already closing off when I left. Selling to Elsevier sets up a new challenge to maintain that open ethos, and unfortunately we can’t immediately gauge what the outcome will look like. 

I joined Mendeley after abandoning my own fledgling online article manager and recommendation tool that I had started while still finishing up my doctorate. I started talking with the Mendeley founders in the Summer of 2008 just before they opened for a beta program. I was really impressed with their plans to “disrupt” the academic market then held by EndNote and also to allow wider discovery of the literature. At Mendeley I was responsible for building out new discovery tools with the data that we were collecting. 

The R&D group did a lot in a few years (deduplicating 200M documents, real-time stats, search, Mendeley Suggest, etc), so I’ll mention just three of my projects below and what type of indication it creates about how Elsevier approaches things that are open.

One of my early projects was the “Open API.” It took some convincing at first to the founders that by releasing the extracted article data we would actually grow, but to their credit it didn’t take long for them to agree. I wrote the spec in late 2009 and worked with Ben Dowling (now an engineer at Facebook) to develop the first beta version of the API that was released in April 2010. After Ben left, Rosario jumped in to improve and expand the API and still heads it up in engineering. The API is now used internally at Mendeley for its own tools. From the start, the API was very controversial as we were now giving away metadata for free (titles, abstracts, etc) that publishers had always held onto tightly unless you were willing to pay for it. 

Next up were “PDF previews” that display anywhere from 1-3 pages of a PDF on each of the article landing pages on Mendeley web. The inspiration here was the 30 second previews that iTunes gives (and now every other digital music provider). The thinking was that this would benefit publishers by sending them traffic that was ready to engage with their articles, something that only seeing the abstract can limit. And we were right, the data we saw showed that articles with previews sent more traffic to publishers than without previews. To many publishers however, this was even more controversial than the Open API, as we were now showing even more content, although you couldn’t get access to it via the API. 

The third project was a JISC funded grant, DURA, that we co-wrote with Symplectic (now owned by Digital Science) and the University of Cambridge. The idea was to utilize the Mendeley Desktop client to connect to institutional repositories to allow drag & drop depositing. Put your authored papers into Mendeley and they are automatically deposited with your institution’s Green Open Access archive. 

All three of these projects were impacted by Elsevier. With the PDF Previews out, Elsevier came out hard to limit what we could do. The Mendeley founders, citing little choice, cave to removing Elsevier abstracts from the API and taking down PDF previews of Elsevier articles. Meanwhile, another publisher, Springer, surprisingly took the opposite approach and wanted us to do more with the PDF previews.

With the JISC funded DURA project, I got word that behind closed doors Elsevier was allegedly trying to stop the project altogether at JISC. That project has since stalled since I left Mendeley. 

These events were entirely against my personal open ethos and why I had originally agreed to join Mendeley. It was clear that the founders and I no longer agreed on the future (a blog post I had written calling out Elsevier’s actions was censored during the intense Elsevier talks) and tensions led me to reconsider my time there. It eventually came to a head and we agreed to part amicably. 

If one is honest, from a business perspective the Mendeley founders did the right thing to comply with Elsevier’s demands. My personal passions about Open Access hindered that, so no surprise it didn’t work out for more than a few years. What I learned was that my next project had to have open at its core, rather than just tacked onto the side. For that reason then I co-founded PeerJ, an Open Access journal, with one aim of never being in the position to take shit ever again from a closed publisher. 

I think that Mendeley as it stands today will continue to be useful even at Elsevier. That said, I think it will be challenging for Mendeley to become a truly transformative tool in science, which is what had originally convinced me to move from San Francisco to London four years ago. I cannot take all of the credit for the “open street-cred” Mendeley has gained over the years, but the projects I worked on had a big contribution to that and I can’t help to think that the day I left was the last day of further open innovation. It will be interesting to see what happens. I sincerely wish Jan, Paul, Victor and the team at Mendeley well.

Updated (10 April 2013): Let’s keep it classy Internet. Since I’ve witnessed some personal attacks against a few current Mendeley staff I thought I’d step in and state that there are still voices for change at Mendeley. I don’t want to leave anyone out, but since I saw attacks against them; three such people are William Gunn, Steve Dennis, and Ricardo Vidal. I know their character having worked together for years and know that they will do their best to make changes from within and wish them the best.

This is the latest paper, published in PLOS Genetics, from one of my former PhD advisors in graduate school.

Recurrent Rearrangement during Adaptive Evolution in an Interspecific Yeast Hybrid Suggests a Model for Rapid Introgression

This one isn’t even all that bad, from submission to publication took seven months. However, I didn’t have the heart to ask him if it had been submitted elsewhere and been rejected. I didn’t want to shit all over what is a great feeling to have another publication out the door.

Sadly, most papers do go through multiple rejections from other journals before finding a home. And that’s if they are not “scooped” while doing the long journal dance. This means that advances in research are tied up for 1-2 years in many cases. 

Why I am upset and concerned

Delaying research helps no one. The paper above is an important one, that doesn’t immediately translate to saving lives, but is core basic research that one day could be applied as such. And this happens every day.

I cannot wait for PeerJ PrePrints to come out this Spring, which will put a stop to this nonsense.

It isn’t lost on me that what I am about to say is rather cliche and pronounced at least twice a month somewhere in the world. I feel that biological research isn’t progressing like it should be. First some background.

The two paragraph background

In circa 2000, about the time I was looking at grad schools, I read the book “Gene Dreams.” The book rehashes the biotech scene of the 1980’s and 90’s, along with all of the promises that biotech + gene therapy would bring to the world. It notes that all of those ‘dreams’ failed, but that the 2000’s would make it right. I was a believer! I entered graduate school determined to use gene and stem cell therapy to cure the entire F'ing planet. Spoiler alert, I failed.

I wasn’t alone though. Many others have failed in that pursuit as well. Gene and stem cell therapy is largely an 'applied science’ versus being a 'basic science’ (aka learning about the building blocks of what makes things tick). So, for a long time I felt that it was just applied science that had issues. The “issues” being that no solid or common frameworks to make discoveries or breakthroughs exist. And that it was better to do basic science if you wanted to succeed more easily. Years after completing grad school though, I think my bitterness was too narrow. I’m an equal opportunist, and would therefore like to bash basic science as well for lacking a modern toolset to take “it” (whatever “it” is) to the next level.

That’s the quick, slightly incoherent two paragraph background, now on to the (somewhat) present thinking.

I was at FooCamp last June 2012 and wound up in a session about biology (don’t remember the title). I think a discussion was brought up originally over the question of how programmers could get more involved with science in a DIY approach by Jesse Robbins. By this time, I had been several years removed from doing actual science on an actual bench with an actual pipette. Instead, I had taken up what I originally started out doing (at age 5) by programming shit; albeit shit related to science now.

Something that was said during that discussion brought me back to my graduate school days of failed experiment after failed experiment. Jesse asked, “Jason, how can I get more involved in science?”

My first reaction was, “Don’t! You’ll end up depressed.” All of those failed experiments, and all of the failed work by others in gene therapy, and most of the time we had no-f-ing-clue-why-they-were-failing. That was the biggest drain on me in grad school, not knowing why. Equally, if I did some kind of PCR directed-evolution experiment and found ONE good result amongst a thousand failed wells in a plate, my next step would be to toss those 999 failed wells and evolve the good well in another PCR round. How, stupid, was, I.

Having gotten back into programming after grad school, I immediately realized my error. Instead of concentrating on the one good well in a thousand, I should have taken the 999 bad wells to see why they went wrong. In programming, this is debugging by looking at your error logs, or similarly, test-driven development.

Moreover, what was missing was some sort of stack trace that would tell me exactly at which point the antibiotic I was testing started to kill the cell all the way up to the point of death (antibiotics were often used to test PCR evolved enzymes used in gene/cell therapy). It was then that I realized that not just applied science, but basic science was missing these “error logging” tools as well. We usually don’t know why experiments fail because 1) we hardly care about failures in science and 2) no one, AFAIK, has built the tools to examine the “why” stuff fails. Such an approach would be laughed at in programming.

Instead, in programming you check your error logs to figure out how to make things work as expected. And the reason we can do that is because someone in the past spent the time building error reporting tools. What is missing in biology (again, AFAIK) is a comprehensive set of error logging within cells.

This is why we’re in a progression slump; we can’t make the huge mental leaps in science because we’re now at the point where better tools are needed. This happens once a generation or so, where we’ve surpassed the infrastructure that was built by the previous generation. A more methodical approach within a framework is needed. A new respect for failed experiments is needed.

And I wouldn’t be surprised if building such error reporting within cells was more universal than we might expect at first, no matter what the cell type or species whence it came. Error reporting I suspect, much like DNA itself across species, could be generalized enough to require little modification per cell, protein interaction, or whatever. A common set of error messages could bubble up to a logging system to be investigated post-experiment. Even the act of building such error reporting would result in a giant number of new discoveries.

Of course, like any idea, good or bad, this has probably been “thunk” before, possibly at the same time. The thing is, I’ve yet to see anyone do it. Perhaps because they did and it failed, and so the research was tossed per usual.

Probably the most disturbing thing to me about the Aaron Swartz tragedy is this statement in 2011 from US Attorney Carmen Ortiz:

“Stealing is stealing whether you use a computer command or a crowbar, and whether you take documents, data or dollars.”

That is teaching our children that the law is always correct and that discretion should not be used when enforcing the law. It’s teaching our children not to question what they are being told by those in power. Had the American fore-fathers believed that “treason is treason” then the United States would have never had its Revolution and founding.

There is no physical law that governs the Universe that outlines stealing, killing, lying, etc. These are human fabrications to govern us as a society, tribe, and culture. We equally have the capacity to dictate when stealing isn’t stealing, or when an act of treason is the right thing to do as the American fore-founders discovered. That is how we advance as a  civilization.

There is a tremendous difference between stealing for personal gain, and “stealing” [1] to release academic papers paid for with tax-payer money. A true leader would recognize that. It’s been reported that Carmen Ortiz had political ambitions to one day run as Governor of Massachusetts. Is that the kind of leader a state would want? A false leader who doesn’t recognize when an act has morally justified grounds? A real leader would act to make changes, not throw the book at someone.

MIT should be ashamed as well, whether they were actively pressing charges or passively standing by [reports are conflicted over this]. MIT as well is supposed to be leading us. In the past they were one of the first universities to offer free and open classroom lessons online. Here, they failed miserably to lead by example that academic research should be made open.

The Aaron Swartz story is bigger than just a 26 year-old doing some computer hijinks and getting bent-over by those in a position of power. It’s even bigger than the importance of Open Access to academic research. It’s surfacing some major issues that we have in society in both the U.S. and beyond about true leadership [note: I am a US citizen currently residing in London, UK]. Ortiz was put into a position to use her discretion. Instead, she let her ambitions dictate Aaron’s fate.

At the end of the day, if it is against the law to steal whether morally motivated or not, then you’ve broken the law. Laws can be changed though. New countries can be formed. And leaders in power can use their discretion to apply fair judgment, not to further their own ambitions. Where have all the true leaders gone?

footnote

1. Note that Aaron wasn’t even technically stealing in terms of the law, at most it was breach of contract [according to several reports].

If it weren’t for Aaron’s heroic actions to release academic research articles in 2011, I am unsure if PeerJ would have ever been born. 

Today’s news was shocking. Aaron Swartz was found dead from apparent suicide on the 11th of January, 2013 at the age of just 26. For the science community and Open Access advocates, Aaron was the man responsible for the (near) liberation of all pay-walled JSTOR content in mid 2011. He also co-wrote the first RSS 1.0 specification at the age of 14 and led the early development of Reddit. 

JSTOR and MIT eventually dropped the civil case against him (publicly anyway), but the U.S. government continued criminal proceedings against him. JSTOR, it should be noted, was not his first attempt at freeing information. Aaron was facing up to 35 years in prison for the act of setting academic research free. It’s unknown if this was the reason for his suicide, but that’s not why I am writing. 

The events around JSTOR and Aaron’s prosecution were probably the final straw for me. What kind of world do we live in, where such harsh punishment is sought for liberators of publicly funded information? The indictment of Aaron and the severity of the probable punishment angered me.

I wrote the following in July 2011 after learning about Aaron’s fate:

Will the JSTOR/PirateBay news be Academic Publishing’s Napster moment? i.e. end of the paywall era in favor of new biz models?

(Twitter reference)

Something had to be done. I wanted to turn Aaron’s technically illegal, but moral, act into something that could not be so easily thwarted by incumbent publishers, agendas or governments. Over the next few months I let that desire build up inside, until one day the answer came in the Fall of 2011.

It was then that the groundwork for PeerJ was first laid; a new way to cheaply publish primary academic research and let others read it for free. Aaron was significantly responsible for inspiring the birth of PeerJ and what I do now trying to make research freely available to anyone who wants it. 

I hope that when the history books are written in decades time, that Open Access crusaders like Aaron will still be remembered. My thoughts go out to Aaron’s friends and family. Know that Aaron’s light and efforts will live on. Thank you, Aaron, for inspiring us.

Michael Jordan, arguably the greatest basketball player yet. His son? Not so much. Wayne Gretsky, also arguably one of the greatest hockey players ever. His father? Not so much. The list of superstar athletes goes on. Similarly, children of the world’s tallest people are never as tall as their parents. What’s going on here?

It’s a well known statistical ‘rule’ known as regression toward the mean. In any sampling with extreme outliers, a second measurement will not have such extreme outliers. It happens throughout nature and is also visible athletics. Though less proven, it is also visible in business and in culture.

We can take the investment world as an example. For every Google or Facebook, there are at least 10x the number of failed investments or 'small wins’ at best. The venture capital industry is at an all-time low over the past decade in terms of return on investment, ROI. The big VC firms will sink hundreds of millions of capital investment into approximately 10 companies per year in hopes that ONE will be a 'homerun.’ Over the past decade, even that one in ten strategy has often failed. This part is important: These are not just a random 10 companies either. They are companies with past success (e.g. revenue or customers) and often run by experienced founders with past successes. Do you see what is going on here? Regression towards the mean.

That statistical phenomenon is why I am troubled by the strategy of science funding agencies today. An article in last week’s UK Guardian highlights how funding agencies such as the Wellcome Trust, RCUK, European Research Council, Royal Society, etc all have the strategy of funding the 'superstars’ of academia. The same strategy is occurring in the U.S. as well. The Guardian article highlights that this strategy is terrible for up-and-coming academics who find it increasingly difficult to be awarded funding. It is also drying up funding money across a breadth of universities in favor of concentrating that money into just a few. This is certainly a bad strategy in terms of what it means to individual researchers, but perhaps even more importantly, it is a bad strategy for science as a whole. Why? That whole regression towards the mean thing as discussed above.

'Superstar’ academics are similar to the Michael Jordans and the successful startup entrepreneurs. Regression towards the mean indicates that follow-on success is rare. One could argue that since it is the post-docs doing the research in most superstar academics’ labs that this statistical principle no longer applies. I would argue that is still does, as often only 'superstar’ grad students or post-docs are selected to join the 'superstar’ lab. That’s speaking about the people, but we must also remember that the projects themselves become concentrated and are subject to regression towards the mean. With fewer projects funded (in the venture capital example above, fewer companies invested in) it means fewer breakthroughs will happen. This is troubling. 

Regression towards the mean is really just the statistical principle behind such clichés such as “don’t put all your eggs into one basket”, “lightning never strikes twice,” and “one-hit wonder.” Why are funding agencies choosing to ignore a well-known principle? It makes no sense.

Here is what I propose. First, we investigate more fully the consequences of current funding strategies that concentrate research money into just a few individuals or universities. Personally I’d forgo the first point, but this is government or large private foundations who don’t change things without further study, so I’m being realistic here. The second thing that we do, and where I would personally start, is to start copying the “500 startups model." 

500 startups is based on observations about the traditional investment strategies of VCs. The observation of a regression towards the mean principle taking place. 500 startups has done something most other investors have not done, which is to invest small chunks into a TON of companies. And its success rate? Much higher than the old investment model of pouring huge amounts into a tiny number of companies. Yet, 500 startups does more than just give money to new ideas. It also nurtures those ideas within a framework and network of mentors who have 'been there, done that.’ This makes a lot of sense.

Can we apply the 500 startups strategy to science funding then? I think we can. Start by awarding smaller chunks of funding to many more academics*. BUT, we must also create a mentorship framework, which today is virtually non-existant at the faculty level in academia. The software industry has 'best practices’ for programming, etc. Surely by now, we have best practices for the scientific method, writing, thinking, navigating the tenure track, etc within academia. Yet, I hardly see all of this codified and combined with personal case studies from successful academics. Funding agencies should not leave it to individual departments to set that up, because it will rarely happen. If funding agencies, and more importantly the public, want a larger ROI then we need to rethink how we distribute and utilize grant money. 

Let’s unbork the funding crisis.

*[This seemingly would upset many academics who already have large research budgets. There are solutions to this as well, which I won’t go into now.]

jason