Open Science

Open Science: How the web is changing the way science is done, written and published
Dr.Hemai Parthasarathy (former editor at Nature and PLoS)

-Began work when publishing was totally paper based; coworker using fax to send acceptance letters was novel
-PLoS founded on model that value added services should be funded upfront
-Web has transformed all business models of communication
-No one knew how to sell subscriptions for online models
-OA to science lit can be argued by ROI, taxpayer, cross-disciplinary access, smaller institutions have access
-True OA is not just access but ability to reuse, repackage, translate – do everything you want with lit, subject to proper attribution
-Opens thinking for submission, review, acceptance process to change
-There’s much waste in peer-review process; going from jrnl 1 to 2 to 3… without major change to the paper
-All this evaluation put into peer-review that is lost when papers are bounced
-If one can deemphasize where paper is published as well as perceived quality
-Yes, there are risks
-Where is balance of top-down and bottom-up filter of science
-How can web 2.0 change how science is published?
-Let’s separate process of evaluating rigor and evaluating substance
-What is the value to overworked scientists to interact with lit to make more valuable? (concern)
-Is there a critical mass of expertise that can provide review? (concern)
-Can process be incentivized?
-Hope and promise is info that is now lost in peer-review process can be captured
-Can change concept of what sci paper is from static, done doc – not the end all, be all, but just the start of the discussion
-Are you trying to improve sci publishing or improve sci? Not the same thing
-Trying to improve sci – how would you improve publishing without improving science?
-PLoS contacted journals where previous papers had been rejected to ask for previous reviews; some said yes, some said no
-Nature once looked at paying reviewers and it would’ve been bank-breaking, even to offer paltry sum
-Politics of editors does come into play when editors don’t pick good reviewers; lots of trust comes into play, as well as knowing expertise
-Has anyone contemplated model where both paper and reviews are published in wiki form where rest of sci community can comment?
-PLoS One is sort of this model
-Anonymity is another risk – does anyone want to read an anonymous review of paper? Is there value without authentication?
-Is Nature with 200 articles still Nature? Still have cache value?
-“No longer doomed to obscurity” if you can’t get published via peer-review in big journal
-“Put info out there, make it available, I’ll find it if it’s of use to me”
-Same issue whether you’re publishing data direct from lab or publishing paper – is it a democratic process? Should you read Nature because you trust them
-Sophisticated reuse of OA article would blow it open
-In small fields where topic is of interest to small group of people, will there ever be a critical mass of people looking at paper and commenting on it if open?
-Peer review does provide positive criticism that is worthwhile
-Is web democratizing anything or creating new kind of oligarchy?
-In field such as evolution with “crackpots and pseudoscientists” should we risk open papers that risk great implications from policy being built on poor papers?
-If everyone publishes online, what role should Nature take?
-Faculty of 1000 has tried to do this to an extent
-All papers should be in PLoS One, then PLoS Biology “republish” best papers from top-down assessment/review
-Foundation organized to accelerate drug research; cooperative of labs across country; design experiments as a team and share info in real time; they perceive rate of discovery is much faster; publishing more papers, and publishing more jointly; began in 2004, originally thought it would be 2019 before identifying viable targets, but process has accelerated and they have already identified 14(?) targets in 2008; take these and patent with promise to share revenue with universities (Carol Menake, Myelin Repair Foundation)
-“What’s the deal with embargos in this day and age?” [came back to this question below]
-Patent issue is big concern with open science
-Will drug be usable if doesn’t go through patent process?
-Are tools to search sophisticated enough to get high ratio of what you want to see in info that is put out there?
-This is where journal brands are going to come into their own – they will be aggregators of great information; filtering role
-When all info in your field is delivered to you, when do you browse? If you no longer flip through Nature, how to you learn about other science?
-Rational for embargo for PLoS actually makes sense: if USA Today says cabbage cures cancer, if article is in PLoS, then readers can do directly to source
-In practice, use of publicity to block publication is a rare issue; authors not penalized
-Sci journalism uses embargos to be less investigative, don’t go after the story
-“Better grasp of probability of trueness” is essential for sci publishing to change
-Serendipity is changing, people finding own methods of serendipity
-Peer-review process is becoming difficult as science is being popularized for general public
-Never find enough reviewers to read all the science that’s out there, which will doom the system
-Real peer-review is taking data and reusing/building on it
-Peer-review harder on multidisciplinary papers because hard to get people who have time and expertise to give quality review, whereas in small niche field easier to find three reviewers who are willing to review because they want to read the paper anyway
-Peer-review gets oversold; should be a smell test: “this doesn’t stink, let’s let the community have at it”; “if there’s something good in there, let the community sniff it out”
-Always try to get raw data; make it the standard
-Need a new kind of citation analysis – not only give citation, but WHY
-Very difficult for peer-review to catch fraud without raw data
-Science tends to have dominant lang – was German, now English – which helps globalize science publishing; if scientists in other countries want to be heard, they will find way to publish in dominant lang
-Incentives are answer for getting OA to grow; understanding needed too