This post is a bit different from what Bonnie and I usually post in this blog – an interview with Dr. Richard Price, founder and CEO of Academia.edu, a social network for researchers. Academia.edu is a San Francisco-based start-up, which currently has 1.8 million registered users and 4.5 million unique visitors a month, with about 4,000 new users registering every day. In August 2012, the site added an analytical dashboard, which supplies researchers with various statistics such as the number of profile views, number of paper downloads, and so forth.
I did an email interview with Dr. Price about peer review, the future of academic publishing, and how Academia.edu fits into it.
First of all, it was a surprise to find out you have a doctorate in Philosophy, because bibiliometrically speaking, philosophers are “lone wolves” who prefer to publish alone or with very few coauthors. How did you end up building an academic social network?
I was doing a PhD at Oxford in philosophy, and, while I was doing that, I noticed the inefficiency in academic publishing. I remember there was a three-year time-lag between submitting my first paper to a journal and the journal publishing it. The disparity between the dynamism of the web as a whole and the system of academic communication and publishing was very apparent to me. Just imagine it taking three years for Twitter to publish a tweet; it’s unfathomable.
Philosophy papers tend to be more single-authored than papers in fields like biology, but the process that leads to a paper being written is still very collaborative: chatting about the ideas with colleagues, and so on. Furthermore, once one has published a paper, one is keen for it to be distributed as quickly as possible to the community at large, so others can criticize it, or build on it.
A lot of the world’s innovation in medicine and technology depends on progress in science, and so I thought this was a worthwhile problem to work on. When I finished my PhD, I raised $600,000 from some London-based angel investors and venture capitalists and moved to San Francisco to build the company. I wanted to be in San Francisco because the Bay Area is the main hub for technology companies in the world.
Have you noticed discipline-dependent differences in using the site?
Historically, we have focused on building functionality that serves the whole of the research world. Some common themes that we have focused on are the sharing of papers and analytics around the consumption of those papers.
Moving forward, we expect some aspects of the site to be more popular with some scientists than others, depending on the type of media they typically use to share their research. For instance, a biologist typically works with large data-sets, but a pure mathematician typically doesn’t.
I understand you’ve raised about seven million dollars from investors so far, but I suppose that won’t last forever (and that they’re expecting a return for their investment…). What is the company’s business model?
The goal is to provide trending research data to R&D institutions that can improve the quality of their decisions by 10-20%. The kind of algorithm that R&D companies are looking for is a ‘trending papers’ algorithm, analogous to Twitter’s trending topics algorithm. A trending papers algorithm would tell an R&D company which are the most impactful papers in a given research area in the last 24 hours, 7 days, 30 days, or any time period. Historically it’s been very difficult to get this kind of data. Scientists have printed papers out, and read them in their labs in un-trackable ways. As scientific activity is moving online, it’s becoming easier to track which papers are getting more attention from the top scientists.
There is also an opportunity to make a large economic impact. Around $1 trillion a year is spent on R&D globally: about $200 billion in the academic sector, and about $800 billion in the private sector (pharmaceutical companies, and other R&D companies).
Today, publishing in a top journal is considered a “stamp of approval”. In your opinion, what will be the future Web equivalent of publishing in a high-impact journal?
There will be a family of credibility metrics that reflect the impact of a piece of research on the scientific community. Ultimately, a credibility metric is trying to reflect the sentiment of the scientific community toward a particular piece of content. The historical peer review process ends up taking the opinions of two peer reviewers as a proxy for the opinion of the scientific community. As noted above, 2 people is not a large enough sample size.
One feature of the future of credibility metrics in science is that they will be based on much larger sample sizes. It is going to be possible to see what hundreds of scientists think of a paper, and not just what two people think.
Another feature is that there is going to be a family of signals about the quality of any given paper. Historically, there has been one signal of quality for an academic paper, and that has been the title of the journal that the paper is published in. In the last 5-7 years, citation counts have also emerged as a valid credibility metric, mainly because Google Scholar started making them available for any given paper. Academia.edu’s Analytics Dashboard is helping scientists see usage metrics associated with their work: page view counts, download counts, and related metrics.
Resources are scarce in science, and this means that there is significant competition for any given grant or job. When you are up for a job or a grant, there are typically 200 other people applying who have a similar number of peer-reviewed publications as you. You are incentivized to try to make your application stand out. That competitive spirit has driven the adoption of new credibility metrics in science: the citation counts and the page view metrics that Academia.edu offers. Many Academia.edu users take screenshots of their Analytics Dashboards and include them with their applications for tenure track jobs or grants. These credibility metrics demonstrate across a variety of dimensions the impact of the researcher’s work.
There will be a growing number of credibility metrics in science, each of which reflects a different kind of sentiment. This mirrors the diversity of credibility metrics on the web more broadly. There are credibility metrics as part of Twitter (followers, retweets), Github (repos, followers), YouTube (views), StackOverflow (reputation), Facebook posts (likes, comments). The way a person or a community thinks about a particular piece of content is complex and multidimensional, and increasingly credibility metrics will reflect that multi-dimensionality.
What will drive the adoption of these credibility metrics is the competitive spirit in the scientific community. A scientist is incentivized to add as many strings to his/her bow as possible when applying for a grant or a job. The way the credibility metrics will be introduced will be in a grassroots way, with scientists saying ‘This metric presents my work in a good light, so I am going to use it.’
It’s worth mentioning that any credibility metric in any domain is going to be gamed. The journal publishing system is subject to this as much as anything else. For instance, there is the practice of defensive citation: a scientist is incentivized to cite anyone who might conceivably peer review their paper. That is a way of gaming the journal system.
Looking at the broader web, people try to game Google. Google has a certain amount of built-in resistance to spam because its algorithm is recursive: Google looks not just at the number of inbound links to a website, but the quality of the linking site as well. Nonetheless, link farms exist to try to game PageRank. Any site that runs a credibility metric has to stay one step ahead of people trying to game or spam the system. This is a solvable problem, as many sites have shown. But it is an issue that you have to be prepared for.
Tell us a bit about your new analytical dashboard. How can it benefit its academic users?
The Analytics Dashboard fits into the general trend of scientists wanting a direct relationship with their audience and wanting to track analytics around that relationship. One reason they want to see those analytics is personal: in the researcher’s mind, the analytics validate that the research they are doing is having an impact.
But the more important role that the analytics have is professional: being able to establish to the world at large, and especially grant and hiring committees, that your work is having an impact. There is a need to stand out from the crowd when applying for a grant or a job. The Analytics Dashboard on Academia.edu helps an academic do that.
Should every scientist think about her or his self as a Web brand?
In the past, the journal would sit in between the scientist and his/her audience and mediate that relationship. We are moving toward a world where the personal brands of scientists are starting to eclipse those of journals. This is reflective of a broader trend occurring on the web, where sites like Twitter, Facebook, YouTube, Github, and others have enabled content creators to have direct relationships with their audiences.
We are moving toward a world where the key node in the network of scientific communication is the individual rather than the journal. The individual is increasingly going to be the person who drives the distribution of their own work and also the work of other people they admire.
I’d like to thank Dr. Price for the interview and Paige Schoknecht for arranging it. Photos courtesy of Academia.edu.
The future of science, guest post by Richard Price in TechCrunch