I’ve come a long way.

When I started graduate school, the only thing I knew about publishing was how to write a blog post, and the only thing I knew about my library was that I hated their website. I didn’t know what open access was, and if it wasn’t in Pubmed, it pretty much didn’t exist for me. All I wanted to do was do my research and work in the lab. Back in 2004, I started work on my first paper and was exposed to the academic publishing process for the first time. For someone who was already familiar with blogging, the whole process made no sense to me. If I wanted to cite a fellow blogger, I could just link to their post with a short little a href=”http://theirblog.com/link-to-post”. I could anchor my link to a bit of text in my post, and they’d even get notified that I had linked to them. Likewise, I could subscribe to the RSS feed of their blog and get updates whenever they published. It was easy to see who was reading your stuff because Google Analytics was free (and even before that, there were plenty of log parsers). Why, then, would a group of people, among the smartest in the world, communicating potentially life-saving or economy altering information, use a system that was so inferior to that which people used to post pictures of what they had for dinner? Well, I was the only person in my lab, possibly the whole med school, who blogged, so no one understood what I was complaining about. I eventually found some colleagues online who felt similarly and we’d talk about why academic paper search sucked so bad, why reference management sucked so bad, and occasionally someone would build a new tool, which no one would use. The failure was usually chalked up to not having access to enough data by the developers and, if it required a critical mass of users, it was considered dead because academics wouldn’t take time away from research or writing papers to use it, because they had no incentive. So to me, the reason I had to use wonky, clunky, ugly tools and endure a long, tedious process to get published was down to these two things: lack of open data and impact factor chasing. As I dug into this, which helped me to procrastinate writing my qualifying exam, I learned that the lack of open data was primarily because academic publishing was mostly a for-profit endeavor and the entrenched interests had no desire to loosen their grip on their data. This was in the thick of the music industry’s “sue ’em all, let God sort ’em out” business strategy and newspapers were just starting to get worried, so the idea that you could do better providing a service instead of selling a product wasn’t really on people’s minds that much. Likewise, publishers didn’t do much to discourage impact factor chasing by scientists and there was literally nothing being done on measures of impact beyond citations. It seemed clear to me, then, that if I wanted to be freed from my drudgery, what I needed to do was to get more open metadata and try to establish something that could free research from the tyranny of the impact factor. I plodded along for a few more years, publishing a few more papers and supporting every new tool that arose which I thought had a reasonable chance of success, provided it could result in more open metadata, open access, and impact metrics. By the time I was done with my PhD, I knew that an academic career wasn’t in the cards for me.

I joined a biotech startup in San Diego later that year, early 2008, but later that year the company fell on hard times, along with the rest of the country. By early 2009 my time with the company was nearing an end, but I had still been following what was happening online and had begun to advocate for a new startup that seemed like it had a better idea than the boring old “social-network-for-scientists” clones that were popping up everywhere. As my involvement with the biotech tapered off, I was able to increase my involvement with Mendeley, eventually becoming part of the full-time US staff.

When I began to work for Mendeley, I was quite definitely aware of the possibility that they would get bought at some point. Nonetheless, I was excited to be able to play a role in helping them become a success. At the time my thought process was pretty simple: they were a non-ugly version of Endnote which also happens to be building a collection of research metadata that they can make available under an open license, and they can provide a measure of impact that is distinct from citations. Freedom at last!
Now 4 years later, there’s a lot I have to be pleased about when I look back. I presented one of the few for-profit business use cases for open access(PDF) to the US Office of Science and Technology Policy. We have ~90M documents available via an API with a permissive CC-BY license. We’re one of the leading contributors of data to the growing #altmetrics movement.

Now my career is entering another phase. I’m going to leave all the “we’re so excited” stuff for the official announcement, but I think Mendeley has gotten to a size where it’s no longer a startup, and smart people are predicting open access will be a reality soon. As Victor notes, we could have carried on, but it would have taken longer for us to get to where we needed to be and there’s no guarantee we would have made it. Springer + Papers or Nature + Readcube could put more marketing muscle behind their apps and neither of them have as open a philosophy as we do. What about Zotero? I think if Zotero was going to change things, they would already have done so, but maybe they could team up with the Digital Public Library of America or Center for Open Science.

I do think there’s a possibility that we could do some good as part of Elsevier. Having talked with tons of people, from the CEO of Elsevier on down, I am now convinced that they want to be a part of the changes, instead of trying to fight them off like the recording companies did. There are and will be a couple competing narratives: They bought us to bury us, we got paid tons of money so we said, “Fuck Open Access”, etc. This is going to be put in the context of Google Reader shutting down, Delicious “sunsetting”, etc. However, I’m not personally getting a pile of money from all this, and I never would have stayed unless I was convinced that they legitimately want to be part of the change to an open access publishing system.

So I’ll be staying with Mendeley. I have been told that my day to day job will remain the same, and that my voice is valued. I trust my friends to keep me honest and to call out bullshit when they see it. I’m grateful to have had this opportunity, over the past 9 years, to not only be a voice for a better way of doing and communicating research, but to be a pair of hands. I’ll learn everything I can about working within Elsevier and, after a couple years, if we don’t finally end up with freely available academic paper metadata and more Google Analytics-like research impact information, it won’t be because I didn’t try my best. That’s my promise and I expect – need – anyone who’s reading this to hold me to it.

Other posts:
TC: http://techcrunch.com/2013/04/08/confirmed-elsevier-has-bought-mendeley-for-69m-100m-to-expand-open-social-education-data-efforts/
Q&A: http://blog.mendeley.com/press-release/qa-team-mendeley-joins-elsevier/
SciAm: http://blogs.scientificamerican.com/information-culture/2013/04/09/elsevier-giant-for-profit-scholarly-publisher-buys-mendeley-free-citation-manager-and-discovery-tool/
Jason Hoyt: http://enjoythedisruption.com/post/47527556151/my-thoughts-on-mendeley-elsevier-why-i-left-to-start

Real innovation in scientific publishing

Many attempts have been made to re-imagine a scientific article, but just adding semantic markup or visualizing the document in a different way has never quite felt right. Previous efforts have felt like they’re just trying to prop up a print idiom whose usefulness is limited in the new medium of the web. Cameron Neylon has come up with a re-imagining that’s truly useful and truly innovative. The idea is to break down a publication into its component parts, so that the smallest unit of publication is no longer a document. This allows publication to move beyond the limitations of the print era and enables the info overload management practices that work best online to be applied to research output.

For me, a paper is an aggregation of objects. It contains, text, divided up into sections, often with references to other pieces of work. Some of these references are internal, to figures and tables, which are representations of data in some form or another. The paper world of journals has led us to think about these as images but a much better mental model for figures on the web is of an embedded object, perhaps a visualisation from a service like Many Eyes, Swivel, and Tableau Public. Why is this better? It is better because it maps more effectively onto what we want to do with the figure. We want to use it to absorb the data it represents, and to do this we might want to zoom, pan, re-colour, or re-draw the data. But we want to know if we do this that we are using the same underlying data, so the data needs a home, an address somewhere on the web, perhaps with the journal, or perhaps somewhere else entirely, that we can refer to with confidence.cameronneylon.net, Science in the Open » Blog Archive » The future of research communication is aggregation, May 2010

Conference notes from Genostem International Symposium on Mesenchymal Stem Cells

Genostem is a consortium of 160 European scientists from 11 EC member countries comprising 11 academic institutions and 5 private enterprises. The project, founded in 2004, is focused on developing and applying advances in cell based therapies for the regeneration of connective tissue. However, judging from the research reported in the recent meeting, they remain competitive.

We had some interesting discussion within our group, so I thought I would share some brief thoughts.

Is stemness an artifact of tissue culture?

The first speaker was was Laure Coulombel from INSERM France. Her introductory talk was titled: “Stem cells: a single word for multiple entities”. The hypothesis presented here was that potency, as the term is usually used in the context of stem cells, is a fluid state. It’s a little like a ball on a hill, being pushed in many downhill directions at once. It only stays perched on top of the hill if the vector sum of all the forces cancel out. Apoptosis, bone differentiation, and fat differentiation all represent possible lower potential energy states but are mutually exclusive. Later on in the talk, Rocky Tuan from the NIH presented some data that seems to support this. Tuan’s experiments went like this: They transduced MSCs with the osteocalcin promoter fused to GFP. This would cause the cells to produce GFP under conditions where the OC promoter is active, which is in the late stage of osteoblast differentiation. Then they cultured cells under conditions where the cells undergo osteogenesis in vitro and the GFP increased orders of magnitude. Next, they sorted out the green cells cells by FACS, cultured them under non-inducing conditions, then switched them to fat-inducing conditions. GFP levels dropped near zero and the cells differentiated into adipocytes. After doing this, they repeated the whole thing backwards, using a PPARγ-GFP construct and going from fat to bone.

Kinda puts Yamanaka and Thomson’s work in a new light, doesn’t it?

Other items of note:

  • Pierre Charbord, the Director of Genostem, presented some information about CD200, an immunoregulatory protein which is only found on the surface of undifferentiated MSCs.
  • Rocky Tuan also presented his tissue engineering work on chondrocyte constructs. poly-ε-caprolactone is electrospun into a scaffold with fibers about half a cell width wide. MSCs are seeded on these scaffolds and cultured in a rotating wall culture vessel, where they grow into chondrocyte constructs the size of a tomato(~5 cm). They’re still using 8 mm punch biopsies in their cartilage repair model, however, so it could be that a significant portion of the construct is inert or merely structural; However, he did say that the scaffolding material is absorbed and replaced with proteoglycan as the construct develops.
  • Katerina LeBlanc presented her work on MSCs immunomodulation in graft vs. host disease. She cautions that in some cases MSCs express MHC Class I and II and therefore can act as antigen-presenting cells in some cases, depending on the amount of IFNγ present. She also reports that MSCs produce a humoral response when cultured in medium containing FBS, but is able to reduce this by short-term culture of the cells in medium where the FBS is replaced with a small amount of the patient’s serum using a protocol our lab developed1, 2.
    Apparently they have also found (and don’t ask me how, I don’t think I want to know) that MSCs are killed by HIV infection, but are resistant to EBV.

I’m pleased to see such active and interesting work being done by the new organization. One thing that is particularly striking about Genostem is the collaborative nature of the member labs. All 30 labs share the €18 million in research funding, so instead of competing with one another, they’ve joined forces and seem to be making particularly good use of what would be considered a paltry sum for an American consortium of that size.

Is it open science?

There’s no blog, and I can’t find anything on a Connotea, but Mayetic Village, a Parisian “collaborative workspace” company lists them among its customers. Unfortunately, it’s closed to nonmembers.

I’d be happy to add anyone to the MSC Biology group on Connotea, I made, but there doesn’t appear to be a way to add anyone to it once it’s been created.