The Project Heads Speak: Matt Wedel

In the final installment of our three part interview series, Matt Wedel shares his thoughts about the Open Dinosaur Project. See part 1 and part 2 for interviews with Andy and Mike.

Matt with another sauropod vertebra

[It] seems (it may be totally wrong) that the project might be more appealing for those who are not professional paleontologists (as you mentioned on the blog, “high school students, school teachers, plumbers“). Why would a experienced researcher be interested in contributing to it?

For some experienced researchers, the project might offer an opportunity to do something new and different. That’s the case for two of the three project leaders (Mike Taylor and Matt Wedel), both of whom typically work on sauropod dinosaurs. The project might also appeal to junior faculty members who are working on building their CVs, and hopefully it will appeal to workers at every career stage who are interested in making science more open (and making more open science). We also hope that some people who have large collections of otherwise unpublished measurements would be interested in contributing, to make their data more available and to boost the project. Of course we realize that the project will not appeal to everyone, and that does not necessarily mean that anything is wrong with either the project or the people who decide not to participate. It’s a big world and there are only so many things that one can meaningfully pursue. We’re not trying to coerce or cajole people into participating. We just want to make the opportunity to contribute open to everyone, and we will be grateful to everyone who chooses to participate. The response already has been very gratifying, and confirms our suspicion that the pool of people who are interested in doing science is not limited to those who hold advanced degrees or are employed as scientists.

The Project Heads Speak: Mike Taylor

In the second installment of a three-part interview with Mike, Matt, and me, here’s Mike Taylor! Matt will follow tomorrow.

Mike Taylor with dorsals vertebrae A and B of Brachiosauridae indet. NHM R5937 ("The Archbishop") in right lateral view.

Do you believe you might be setting a trend?
Part of what’s so exciting is that we are in such uncharted territory.  We really have no idea whether this particular project [is] setting a trend — it’s nice to think it might be so, but equally this might end up being the one project of its kind.  It might crash and burn, or it might spiral out of our control into something really far reaching.  Three days in, it’s much too early to make concrete predictions, beyond that we’re certain to learn something.  Whatever comes out of this project in the short term, I hope that one of the long-term results is that if other people do start broadly similar large-scale collaborative projects, they’ll be able to learn from our mistakes and do a better job than we’re doing.

And this project is only a tiny part of a much bigger … I wanted to say movement, but it’s nothing as well defined as that.  It’s an ill-defined and approximate trend, but generally what’s happening is that the whole way we do science is changing in ways whose ends are hard to foresee.  Electronic publishing is no longer exotic, and some prestigious journals have no print edition; open access to online articles is increasingly widespread and journals that don’t do it are looking more and more reactionary; scientific blogging is becoming very serious in places (not least at our own Sauropod Vertebra Picture of the Week blog, http://svpow.wordpress.com/), with original research sometimes “published” in this way without recourse to journals at all; even the most staid journals are accepting citations of online resources.  Generally the trends seem to be towards decentralisation of authority, increasing openness and progressive divorcing of medium from message.  The Open Dinosaur Project fits very comfortably into that loose coalition of ideas: we’re trying to democratise science, open up data, blog the process, and make sure that the final publications are freely available to the world.

So whether or not other groups in the future do things that are similar in detail to the ODP, I do think that we are trending inevitably towards the kind of scientific environment in which those KINDS of thing — things in the same SPIRIT as the ODP — are going to proliferate.  If we’re a small part of taking the world in that direction, I’ll be happy with that.

Finally, will the contribution be limited to the creation of data registers? Can people contribute in other ways?
Again, we’re feeling our way here.  What we know for sure is that volunteers — including those with no academic background — can make a real contribution in aggregating data.  For that reason, and also of course because data collection is one of the first phases of a project, that’s what we’re concentrating on now, but we’re certainly open to pushing more work onto our volunteers down the line!  In particular, there are lots of different kinds of analyses that can be run on the data, and it’s unlikely that the three of us have the necessary background to do all of them — maybe some of the project contributors can do, I don’t know, principle component analysis or something. We’ll also need people to help organize and format the project’s massive bibliography, as well as track down abbreviations and other crucial (but hard-to-find) pieces of information. Figure preparation, and especially artwork, is another obvious area where we can hope to find and and use the talents of ODP  volunteers.

The Project Heads Speak: Andy Farke

Mike, Matt, and I recently were interviewed about the ODP by the Brazilian science publication Ciência Hoje On-line. Only a few select bits of the interview were included in the article, so the next few posts here will include our full responses (with permission from our interviewer, Ciência Hoje’s Raquel Oliveira). Exercising privilege of being project lead, and because the interview makes the most sense in this order, here are some of my responses. Mike Taylor’s will follow tomorrow, and Matt Wedel the next day.

Andy Farke

Why did the group launch the project? You meant to distribute the job? From what I understood, you want to analyse a huge amount of data.
We launched this project for three reasons. First, we wanted to do some good and interesting science. In the case of collecting all of the data, we knew right away that it would have to be a task spread across many individuals. And the best way to do this seemed to be just opening up the process to anyone who wanted to help out!

Second, we wanted to get the general public excited about and involved in doing “real” science, working in cooperation with paleontologists. There is a great interest out there in paleontology, particularly dinosaurs. It’s amazing how many non-paleontologists read the technical literature! I thought, “Why not harness this enthusiasm?” There have been many people waiting for this sort of opportunity (even if they didn’t know it), and I think the response speaks for itself. In less than a week, we’ve had over 40 non-specialists indicate interest in the project, and half of these have already submitted data. I look at it as a tremendous outreach opportunity! And the information flow goes both ways–we’ve had some very insightful comments, observations and suggestions from “amateurs.” They’re making substantive contributions to the project, beyond “just” entering data.

Finally, we want to promote the concept of “open science” in paleontology. It can be a pretty secretive field, sometimes for good reason–to prevent theft of fossils from dig sites, for instance, or to avoid being “scooped” in naming a new dinosaur. But, I think there are areas where we can change this culture. For instance, lots of scientific publications are based around huge data sets of measurements. . .but often these measurements are never published, and are eventually lost as people retire, die, or switch careers. Not only does it make it difficult to reproduce the analysis (an important cornerstone of science), but it also means that it is much more difficult to build upon previous work. We’re continually re-inventing the wheel. I would never say that every piece of in-progress research should be blogged, or that every piece of data should be immediately available, but I would like more of us scientists to have data and publication availability as a higher priority. It will take time, but I am optimistic. There are many paleontologists who are starting to buy into the idea of “open science”. If some branches of physics can do it, with the arXiv website, and if molecular biology can do it, with tools like GenBank, surely paleontology can change with the times too.

How do you think the academia will react to the Open Dino?
Already, we’ve noticed an interesting mixture of excitement and skepticism. Some people have looked at the project and said, “Wow, why didn’t I think of that?!” They’re excited about the possibilities. And others have said, “Well, there are so many potential problems, scientifically and organizationally, that it will never succeed.” By opening up the project, by blogging about it, we want to be able to address any concerns about the structure and science of the project essentially in real-time. Some legitimate concern does stem from the fact that we’re pulling data from the published literature – how do you know that all of these scientists measured bones consistently, for instance. So, we’re focusing on a set of relatively simple bone lengths that should hopefully be more consistently measured. For the large-scale evolutionary analysis that we’re conducting, I suspect that minor variation in measurement style from paleontologist-to-paleontologist is a rather small problem.

On another note, I think there is some reaction – positive and negative – to the fact that the project is going to be so transparent. By blogging the project, all of our successes and mistakes will be out there for the world to see. I think some folks will like that, and others won’t.

The ODP in Portuguese

The Open Dinosaur Project is featured as one of the inaugural articles for the newly designed website of the Brazilian science news publication, Ciência Hoje On-line.  The article is in Portuguese, but a somewhat-readable translation is available through the wonders of Google Translate.

Look for expanded versions of the interviews behind the article soon!

Speaking of the Bipedal / Quadrupedal Transition. . .

Aardonyx, after Yates et al. in press

Dedicated readers of the blog likely remember that one of the core research goals of this project is to examine the bipedal/quadrupedal transition in ornithischian dinosaurs. Of course, ornithischians weren’t the only group to experience such locomotor changes during their evolution! A new paper on the bipedal/quadrupedal transition in sauropodomorphs (the saurischian dinosaur group including animals like Plateosaurus, Apatosaurus, and Brachiosaurus) has just appeared on Proceedings of the Royal Society B’s FirstCite. This paper, headed up by Adam Yates, details the anatomy of Aardonyx, an early sauropodomorph from the Early Jurassic of South Africa. The authors posit that the new critter was a habitual biped, although it had many features that presaged the anatomy of later quadrupedal forms.

The paper has already been covered in some depth elsewhere in the blogosphere – check out posts by senior author Adam Yates, Dave Hone, Sid Perkins, and Bill Parker for more.

If you haven’t contributed data to the Open Dinosaur Project yet, and are looking for something to do, this might be a good one! The supplementary information (freely available) is chock-full of measurements.

Citation
Yates, A. M., Bonnan, M. F., Neveling, J., Chinsamy, A., and M. G. Blackbeard. In press. A new transitional sauropodomorph dinosaur from the Early Jurassic of South Africa and the evolution of sauropod feeding and quadrupedalism. Proceedings of the Royal Society B. doi:10.1098/rspb.2009.1440. Published online 10 November 2009. [subscription required for full access]

Big Announcement – A Whole Pile of Data!

Awhile back, I had alluded to some big announcements for the project. Announcement 1 was, of course, our little contribution in Nature. Now it’s time for Announcement 2.

We at the ODP are very, very happy to announce the contribution of a huge chunk of original measurements from Matt Carrano at the Smithsonian Institution. Matt has graciously offered the core of his dissertation data set (and then some), with measurements from over 600 specimens relevant to the ornithischian component of this project. The database is particularly full of measurements from the hindlimb, including a number of never-before-published data points.

Combining the data submitted by all of us with the Carrano ornithischian dataset, we now have nearly 1,200 individual verified entries in the ODP database! That’s a 21-fold increase over what we had at the launch of the project. Great work everyone, and a huge thank you to Matt for his generosity. As before, the entire dataset is freely available for download and use by anyone.

So what does a plot generated with the gigantic database look like? Now you know!

Log Length of MT III regressed on Log Length of Tibia. Slope=0.852 (significantly different from 1); r^2=0.928

The ODP at ScienceOnline2010

For those who might be interested, I will be giving a short “demo” of the Open Dinosaur Project at ScienceOnline2010, an annual conference on science and the web (Matt and Mike can’t make it, unfortunately). Because the event is so small, registration filled almost immediately (sorry! There is a wait list, though). You may not be completely out of luck, however–previous conferences have been live-blogged, streamed, twittered, or some combination thereof; look for more information on this as the time approaches. In the spirit of open science, I’ll likely be previewing the talk and soliciting input from all of you (you’re all co-authors of a sort, after all) as the time approaches.

The event is January 15-17, 2010, in Research Triangle Park, North Carolina. Find more information at the conference wiki.

Meet the Project Participants: Diane (DeDe) Dawson

Continuing our series of interviews, we’re happy to introduce all of you to Diane (DeDe) Dawson. In a project that relies so heavily upon the literature, it is a good thing to have a science librarian on our side!

DeDe Dawson, collecting fossil graptolites in Nova Scotia for her M.Sc. research

Tell us a little bit about yourself. Where are you from? What do you do (professionally)? Any other interesting facts?
I am originally from the Toronto area, but have just recently moved to Saskatoon (Saskatchewan, Canada) to work at the University of Saskatchewan as the Natural Sciences Liaison Librarian. I am responsible for Geology and Chemistry subject areas, so I maintain the collections in these areas as well as provide reference and “information literacy” instruction to students and faculty. Basically, I help them find the information that they need and teach them effective literature searching skills.
I have quite a varied background. My undergraduate degree is in Biology, my main interest being evolution and zoology. I completed an honours thesis project on several trilobite species found locally (in Ontario) – collecting all the specimens myself and performing some basic morphometric analyses. Certainly I’ve always had an interest in natural history in general, and a particular fascination with fossils. So, during the summers I worked in the Vertebrate Paleontology lab of the Royal Ontario Museum in Toronto. After graduation, I remained at the ROM as a technician for several years and was lucky enough to be involved in an excursion to collect Jurassic marine reptile fossils in northern British Columbia! But it turns out though that my interests are much older… and without backbones! I ended up returning to invertebrates for my MSc thesis in Earth Sciences: graptolite phylogenetics. During the process of my Master’s thesis research I began to realize that I also really enjoyed the literature search, and while I loved paleo, I didn’t see a career in it for me. Academic librarianship presented itself as a profession more suited to me and a way to maintain my connection with the academy.

Why did you decide to participate in the ODP?
Besides my longstanding interest in all things paleo, I also have a very strong interest in open access issues and what we librarians call “scholarly communication.” Being a librarian I am all too aware of how expensive some library resources are, particularly scientific journals. The exorbitant costs often put the research published by scientists out of reach of the general public, and increasingly out of reach of budget-restricted institutions too. I realize that I am in a privileged position in having access to many of the articles that the public does not, so I want to be able to help the ODP gather this kind of data. I also want to support any effort that shares knowledge, increases collaboration across disciplines, and challenges barriers to accessing information. I am very intrigued by the concept of the ODP as a potentially new form of scholarly communication and collaboration in the sciences, and I’m eager to see how the project unfolds!

So far, what has been the best part of the ODP for you, and why?
The best part so far has been to put my library’s collection to use! I am always assisting others to use the collections here, but now I have a purpose for the collection too. I am relatively new to this library, so have not had a lot of time (or reason) to explore the stacks yet. It is very satisfying to be able to pull some of the dusty old journal volumes from the shelves and make use of them! I have also been able to identify and correct some access problems with electronic journals during this process. So I’m actually doing my job at the same time!

What have you learned from your participation in the ODP?
I have learned what a scapulocoracoid is!

What advice would you give those who might be interested in helping out with the ODP?
First spend some time reading through the ODP website, especially some of the tutorial postings about anatomy and what the abbreviations in the forms refer to. This has saved me from hassling the project leaders with lots of questions! Also, I found it easiest to start with verifying some of the articles already entered. This has given me a better feel for the process, and what to look for in the articles. Soon I might look for some new articles myself…

The Open Dinosaur Project in Nature

We at the ODP are excited to announce that the project is getting some publicity this week courtesy of a little journal by the name of Nature. A short letter to the editor [subscription required*], written by me, Mike, and Matt, appears in the current issue.

The letter was crafted (at Mike’s suggestion) in response to a series of articles and editorials (freely available here) in the 9 September 2009 issue of Nature, focused on the issue of data sharing and archiving. Right now, data sharing (either pre- or post-publication) is a huge concern in many fields of science. Without accessibility to original data, it is much tougher to verify results, incorporate new data into previous analyses, or use the data for new, potentially unrelated analyses. So, it should be a no-brainer to release the data underlying publications, right?

Not so fast. Some scientists are worried about being scooped (if they release data prior to publication), or losing a competitive edge to colleagues once the “exclusive” data are released (if they release data after publication). In other cases, there is no institutional support for the release of data (which requires time and money, however minimal). Furthermore, the rewards for actually releasing data can be unclear (a critical factor for scholars at the beginning of their careers, who need citations and recognition for hiring, tenure, and promotion).

All of these problems were covered in some depth in the special issue of Nature. Where does the ODP fit into this, and what does our letter add to the discussion?

We felt that the Nature articles did an excellent job of making the case for why data sharing is important to the scientific community. But, it left out one key ingredient – why data sharing is important for the world outside of professional scientists. So, we focused our correspondence on this problem.

For a field like paleontology, there is a tremendous interest in the latest (and even the not-so-latest) research findings. If you follow the Dinosaur Mailing List, the blogosphere, or any other internet venues, you will notice an active and engaged community of both professionals and amateurs. Requests for PDFs, photographs, measurements, and all sorts of additional information are quite common, reflecting an intense fascination with the field.

Public talks, popular web pages, and blogs are great for public outreach. . .but there is a demand for something more. Folks are interested in high resolution specimen photographs. . .specimen measurements. . .cladistic data matrices. So, we argue that for a field with broad public appeal like paleontology, the release of data should be a part of our public outreach efforts in addition to the immediate scientific role of data availability. People get excited by these data!

This is why we started the Open Dinosaur Project, and why we run it as openly as possible. First and foremost, it is a research project. We are collecting data to investigate some interesting questions – and we want these data to be available to other researchers! In our view, the database doesn’t do anyone any good if it stays locked up on one person’s hard drive.

Charles W. Gilmore, accidental contributor to the Open Dinosaur Project.

We also benefit when other researchers are open with their data – this project would not be possible if original specimen measurements weren’t published in papers. The ODP thrives on data sharing. Principal components analysis and phylogenetically independent contrasts either didn’t exist or were decades away from paleontological application when Charles W. Gilmore published his monograph on Stegosaurus back in 1914. Without putting too many words in his mouth, I think it is fair to say that he would be quite pleasantly surprised that we’re using his data for something he never intended.

Second, and just as important, the ODP is a public outreach effort. There are many, many “amateurs” who want a chance to participate in real, substantive scientific investigations, and many professional paleontologists who want to contribute to collaborative research efforts. Without the combined efforts of nearly 40 contributors from all walks of life, the ODP would only be a blog with a catchy title and cool logo. All of our project participants have made important contributions, and not just in the realm of data entry. The ongoing discussions on this blog are proof positive of the deep involvement of many, many individuals.

In sum, data sharing is something we believe in. It moves science forward, both by allowing new things to be done with old data, as well as offering opportunities to involve new people in the scientific process. If we scientists are serious about our work, and about communicating our work, data sharing is not just an option. It’s the only option.

Citation: Farke, A. A., M. P. Taylor, and M. J. Wedel. 2009. Public databases offer one solution to mistrust and secrecy. Nature 461: 1053. [link*]

*subscription required; the publication agreement with Nature prohibits us from posting the full text of the article for six months; please contact Andy if you would like a copy of the text.

Big News Tomorrow (and a Plot)

Late tomorrow we look forward to bringing you a Big Project Announcement (well, more of a Big Project Promotion). So, tune in then for more details.

In the meantime, enjoy this new plot of our data. The other day Tor Bertin suggested we take a look at MT III and MC III ratios. So, here’s a chart showing the MT III:tibia ratio versus the MC III: ulna ratio. A few things to note:

• Hadrosaurs are still very, very bizarre. Their MC IIIs are relatively longer than in any other ornithischian.
• The vertical spread in ceratopsians is probably because both small, bipedal forms as well as large, quadrupedal forms are included in that category.
• Basal ornithischians have very long MT IIIs. Is this a consequence of their generally small size?
• We need more thyreophorans (stegosaurs, ankylosaurs, and the like) with MC and MT data! This group, for various reasons, is one of the remaining holes in the data set.

In the interests of full transparency, I deleted the point for for the ornithopod Lurdusaurus arenatus. Its metatarsal III length seems far too small for the size of the tibia. This may be an error in the original paper (I even went back to triple-check it!), or it could be that this taxon has tibia:metatarsal proportions completely unlike any other ornithopod.

Want to See Some Other Plots?

We’re always looking for new ways to present the data on the blog. If you have any requests, feel free to note them in the comments section.

Or better yet. . .don’t forget that the data are freely available to everyone! There is absolutely nothing stopping you from playing with the data yourself. In fact, we encourage it. Odds are quite good that, like many project participants, you’ll pick up something nobody else has noticed yet.