Yesterday I wrote up an outline for this presentation and now I’m going to write up a basic script so I know what I want to design my slides around. I ended that post by starting this one:
I think I want to start the talk by projecting a hypothetical future. In this future, peer review publications are either non-existent as we currently know it, or they are reserved for organizing information scattered across notebooks. I will talk about a future where scientists get information directly from other labs, instead of peer review articles. Research is updated in real-time and scientists have full access to step-by-step protocols, raw data, software and code, thoughts, notes, ideas, and anything that may come up during the scientific process.
That future is beginning now and it’s starting with open notebook science.
I think I still want to begin there, but I also wrote that I wanted to include tools of the trade, and I’m thinking that doesn’t fit so well into the talk. Instead maybe I can present tools of the trade through the examples I wish to demonstrate (my notebook would highlight successes, sharing, wordpress, figshare, benchfly, etc). In any case I’ll get into a writing zone and then see if it makes sense to include that information. Let’s get to it…
The internet has brought about radical changes to the way that we all interact with the world around us. In academia, knowledge is spread at an exponential rate thanks to social media and the people willing to share their ideas with the world. Recently, the open science movement has brought about tools that allow researchers the opportunity to host and publish their data and methods to bring similar changes to the scientific community.
Traditionally scientific information is shared through peer-reviewed journal articles. Researchers spend a great deal of time formatting their methods and data to conform to the editorial process and then spend even more time trying to get their data through the peer-review process. While the intent is good, peer-review which is intended to weed out erroneous information ends up as a major time consuming portion of the publication process. In the end, even faulty or completely erroneous data will eventually get published. And thanks to Google, that data is just as likely to be discovered as much more credible information.
Open notebook science is the next generation of scientific publishing. By maintaining an open notebook scientists keep a detailed account of ideas, methods, data, and results providing a full transcript of an experiment through the entire scientific process. Ideally this record will be kept in real-time allowing collaborators the opportunity to access scientific data directly from a lab as the experiments take place instead of waiting 6 months or longer for the information to be published traditionally.
I treat my open notebook as an ideal system and maintain every aspect of a given experiment. My notebook is hosted on my own website (IheartAnthony) and uses the WordPress platform because it is open sourced and there is already quite a large and developed infrastructure and community.
From my notebook, you can access experiments from initial ideas all the way through to the final results. The information is search engine indexed so others can easily access information that may relate to projects they are working on. Also, by going open, I’ve bridged the gap between taxpayers and funding agencies and the labs that they fund.
(Show examples of the notebook)
Also by using wordpress I’ve built a system that allows me to incorporate other web tools to host scientific information. I can use videos through BenchFly and Youtube to host video protocols, I can host images to detail projects, I can embed cloud word processing (Google Docs) to keep real-time data and share it, and I can use commenting systems like Disqus to engage the community both public and scientific to give my notebook life. This also allows for my notebook to be peer-reviewed in real-time.
The major issue right now is that the open science movement is relatively new and open notebook science doesn’t have many practitioners (show wikipedia article). As such the community of engagement is quite small but extremely diverse. In order to show the effectiveness of ons in a real-world environment I built an open notebook science community in the Physics 308 Electronics Junior Lab that I’m teaching.
Students in the lab were given some initial conditions and were allowed complete freedom otherwise to demonstrate an open scientific system.
Initially they were allowed to pick any online system that could be used to keep notes, but they were offered the suggestions of using Google Docs, the Github wiki, and WordPress. The students picked the platform of their choice and the amount of students representing each platform is pretty uniform.
They were taught the principles of ons: keep a detailed record of your data, clearly write methods, and try to maintain some semblance of organization. Posting more is much better than posting less or worse, nothing at all. In that sense, students were encouraged to post whatever they wanted in their notebooks, because even wrong information could be benefitted from.
The students were also required to collaborate and communicate with one another. Each platform has some form of commenting system built in, and students were encouraged (and required) to read each others notebooks and provide some weekly feedback.
The final rule was that students were allowed to “cheat.” By this I mean, they were allowed to use each other’s software code, they were allowed to find the answers to problems in each others’ notebooks, and they were allowed to Google anything they didn’t understand or couldn’t figure out. By allowing them to do this, we’ve allowed them to perform exactly as they would in the real world.
So in effect, the class became an experiment in open notebook science, which turned out to be pretty successful. (Illustrate successes)
End with details of benefits of ONS.