This is the final part of my take on the question – “Are Scientists Anti-Social”. The first part is here.
Part 2: Scientists sharing their work with the non-scientific public
“The School of Athens” at the Vatican Museum
This is where we have a truly important problem!
In my last post, I argued that any change in the way how scientists communicate with other scientists must preserve the time-tested peer-review model of scientific publication. However, when it comes to scientists communicating with non-scientists, absolutely any change in the current model is welcome and desperately needed.
Scientists need to understand that doing science is not their right. It is a privilege given to them by the taxpayer (or if you are in a private company – by the shareholders). Scientists are ultimately accountable to those who fund them and should make every opportunity possible to make the public understand their work.
The public needs to understand what the scientific method entails. For example, the active ingredient in those anti-allergic pills you took recently – the one that clears your symptoms in 30 minutes and doesn’t make you drowsy (also known as second generation H1-receptor antagonists) – was first discovered in 1933 by scientists while working with guinea pigs to prevent anaphylactic shock in humans, then widely debated in scientific literature, scored a Nobel Prize in 1957, engendered more projects, taken into development by multi-billion dollar drug companies, declared safe for patients by the FDA in 1995, now a lot of them are off-patent, and thus, for a few dollars, you can enjoy allergy season by controlling your immune system, without which you would not be alive. The public cannot take the arduous process behind these discoveries trivially, and then demand more effective treatments for their children (essentially a finer control of their physiology at lower prices), while tacitly allowing their government to cut research spending to an extent that now only one in five grants gets funded.
Science and medicine are under attack in recent times like never before in history. In cases that are much more serious than rapid allergy relief. Be it climate change, vaccinations, or particle physics; any fool with an internet connection can now pass judgement and give advice in areas so far beyond their field of expertise, with so little accountability, that they become extremely dangerous (the irony of posting this on an obscure blog is just hitting me!). Scientists need to step up and communicate to the public effectively regarding what they do, and how they do it. I believe that if such a communication strategy doesn’t feature prominently in your scientific training and mentorship plan, you have failed as a scientist.
Like most important things, effective communication is not easy. And it takes a lot of time. Scientists are so often caught up in their jargon (to communicate with other scientists, you see!) that any ‘normal’ person would lose interest immediately. This is where science writers and communicators come in. Their expertise lies in understanding the scientific process, jargon et al., and present it in a way that is interesting to the non-scientific public without misrepresenting facts, or obfuscating facts and opinions (very important). Most of what passes for science reporting in the media nowadays comes from journalists who are either jumping the gun on the implications of the science they report (to get breaking news), or intentionally twisting the story to be as sensationalistic as possible to get more page hits (yes, Gizmodo, I’m thinking of you).
Well-informed, articulate science journalists are very valuable to society. Fortunately, they happen to be quite prolific on social networks like Twitter and the general blogosphere. Combinations of scientists, communicators and informed citizens (non-immigrant aliens too!) form groups like SoNYC aimed at tackling these issues. At the SoNYC meeting #2, we discussed the role of social networking tools like Twitter in science communication. A prominent tweep said that she used Twitter to troubleshoot a DNA gel with a colleague. I would consider this to be an extreme version of scientific communication though Twitter. I follow scientists who tweet their thoughts on freshly published scientific paper in less than 140 characters, exactly like most people tweet their thoughts on a linked article. I find this tremendously helpful in keeping track of what is published. Another great friend/ scientist and I routinely engage (often combatively) proponents of pseudoscience like homeopathy, astrology, rabid pro-lifers and religious fanatics. These people need to know that their propaganda against science will not go unchallenged. Social networks are unique in their ability to connect people in interesting ways. For example, I started using Twitter first for following technology related news. Then I started connecting with amateur photographers (still my primary use of Twitter). I found a prolific scientist that a famous technology writer follows. She tweeted about the SoNYC meeting, and that is how I came to know of their second meeting. And it was a block away from where I work. There are many such examples of how social networks indirectly, yet more effectively, influence science thinking and communication.
I will wrap up with another good question floated at SoNYC of how we can get more scientists on social networks. I think it is only a matter of time before Twitter takes-off in the scientific community (assuming Twitter doesn’t keep doing crazy things out of fear). Facebook emerged from a sea of social networks (MySpace, Orkut, Hi5 – to name a few) by rapidly delivering what the public wanted, and delivering it well. Facebook now boasts over 500 million users (more than the population of the USA). Twitter is still relatively small in comparison, but a much more powerful service in terms of connecting people based on interest, rather than acquaintance. The early adopters need to tweet more and spread the word to their colleagues.
A veteran of many, many (too many!) scientific meetings, I recently got a chance to attend one that brought science writers and scientists together (important note: science writers and writing scientists aren’t the same thing). Science Online NYC (cutely termed SoNYC) is a newly launched discussion series between anyone interested in science communication, in the NYC area and beyond (via the internets). Its mission is to encourage the discussion of how science is carried out and communicated online. It is a great group that fulfills its mission very well. I look forward to being a regular attendee, and will recommend other scientists and science enthusiasts to do the same.
For their second meeting, the discussion was focused on the question- “Are Scientists Anti-Social?”
Alright. That loaded question should be divided into two sub-questions (and we did go back and forth between the two over the course of the discussions):
1. Are scientists anti-social with other scientists?
2. Are scientists anti-social with non-scientists?
The panelists were an impressive mix of people from academia (NYU), a leading social online reference manager (Mendeley), and science journalists from Nature.com and Wired.com. The tweets using the hashtag of the meeting are available on something (really cool) called Storify. If you want a tweet-by-tweet account of what transpired at SoNYC, see this link. Tweets by yours truly appear with the handle @banmans.
So a general overview of the discussions pertaining to the 2 questions above, mixed with my thoughts (of course!) will be the basis of this post and the next.
Part 1: Scientists sharing data with other scientists
Scientists reading menus, not manuscripts, at a wine tasting event at a conference in Italy
This is a laudable goal, and the fundamental tenet of academic research. A ‘two heads are better than one’ philosophy to problem solving.
The first reason why scientists in even slightly different fields cannot share primary data easily is because scientific software is quite specific to each field, often coded as ‘homebrews’ (labbrews?) with little effort going into interoperability with other more widely used software. A personal case in point is electron microscopy — when I build a structural model, made by combining thousands of images taken at various angles, I cannot quickly share any of the images with a collaborator, unless s/he has the same version of the same software. The images are saved with a .hed extension, because that is what the software needs to be able to refine the structure iteratively. Conversion to a tiff /jpeg/ png/ eps format is a 4 step process (not kidding), involving 2 different software packages.
The second reason is that most scientists will not share primary data with the rest of the field has to do with it essentially being proprietary till it gets published (in the public domain) under your name. Government funded academic research needs to be in the public domain (and most of it is), but when and how to release it is at the discretion of the investigator. Academic research is quite competitive. A good scientific idea is very hard to come by, and the data is even harder to generate, almost always requiring years of training. The first team to publish stands a better chance to secure grants from the already shrinking funding pool that is the National Institutes of Health (NIH), the National Science Foundation (NSF), amongst a few others. The paper publishing model is built in a way that the first to publish a breakthrough gets the story in the highest impact (most widely read) journal, and the rest of the groups, inspite of putting in equal or more effort, have to settle for a journal that is often relatively obscure. Thus the fear is that sharing your data prematurely, even at conferences meant for that very purpose, might get you scooped by another lab working on the same problem. Most scientific conferences I have been to expressly forbid any photography at the poster and talk sessions, mainly to encourage the researchers to show and discuss their data with a comparatively small group of conference attendees.
The published scientific paper is social currency for the research scientist. As a paper is read and cited (sort of like retweets, but much slower!) people on that paper gain recognition and other scientists who like their idea and resources want to collaborate on future projects. This involves a pooling of grant money, ability to apply for special interdisciplinary grants and so on. The question that needs to be asked then is – how can we make the science publication process faster and more transparent?
“The paper publishing industry is dead”, claimed one panelist, “so don’t worry about being scooped and move on”.
And it has been replaced by what, exactly?
The current scientific paper publishing model does one vital task, and does it well. That task is peer review. Before a paper is published, it is reviewed by scientists working in the same field, but at different institutions, chosen by the journal editor. This step is critical in checking the fallibility of the hypotheses, experimental design, and if the data actually support the conclusions. This process is not perfect – for instance, it is hard to publish a radical theory that goes against conventional wisdom until you dilute it by consensus to a state that the reviewer will accept. Further, getting the ‘right’ reviewer is critical for eventual publication in a reasonable timeframe – some journals can keep a manuscript stuck in the review process for several months! But getting your data interpretation vetted by someone with credentials, experience (often surpassing yours), and hopefully no conflict of interests, prior to it being released to the public remains the most important safety check before it is used by the field as a basis of future research. It also collectively increases the responsibility for the data in the scientific community. In contrast, a ‘crowd-sourced’ model (like Digg or Reddit) will simply not work in science because you cannot verify whether the person up or down-voting your research has a sufficient level of understanding that comes from years spent in a lab.
The solution to the first question, therefore, may be to simply streamline the peer reviewed publication process. Communication between the authors, the journal editor and the reviewers can be made faster with current generation tools. Most biology journals now have a completely online submission, review and notification system. I have reviewed manuscripts as a pdf synced between my iPad and iPhone which gives me (the reviewer) more flexibility of when and how I think about it, going back to the text whenever needed. The space constraint of how much data that can be fit into a weekly journal of a 100 pages is disappearing too, as more journals move completely online. There already exist thousands of journals where you can publish various parts of your research story, if you so choose. What is needed is a way of rapidly disseminating these articles as they are published (after surviving peer-review) by leveraging metadata tools like RSS feeds across platforms and devices. These steps will increase the visibility of research coming out of lesser known labs, eliminate country-specific biases (real or perceived), and truly accelerate good science.
To be continued to Part 2. This scientist needs to do some science first!
The Brightest – A convocation for PhDs at one of the leading scientific institutions in the world
The esteemed scientific publication– Nature – has a feature in its latest issue that discusses fixing the PhD system. In its zeal in doing so, it has unfortunately chosen to include the crazy with the more pragmatic. Here is my take on the situation.
Different PhD programs are structured in different ways, and to club all of them together might make for sensationalistic newsbytes, but is good for little else.
Historically, PhD programs have been designed to train the next generation of academics, using a mentorship model of close, hands-on apprenticeship of the student with the PhD mentor. The number of PhDs produced by a country is considered a benchmark of how good the higher education system of that country is. A typical PhD program has several key strengths which are not yet replicated in any other kind of training – it cuts you from funding worries, shuttling between meetings, and constant traveling (to name a few), by ensconcing you in an academic ivory tower so that you can, presumably, be creative in tackling difficult problems plaguing society at a fundamental level. Such problem solving (for a PhD) often starts from first principles, without the din and distractions of ‘day-to-day’ life. First and foremost, the investment of both the institution and the student for a PhD is meant to be geared towards a creative pursuit to produce a novel body of work.
This is, of course, in the ideal system.
The very advantages of a PhD program become a major detriment when the traditional scope of a graduate is expanded to now compete with others in the job market. They have no idea how to generate revenue for projects, how to negotiate effectively in a boardroom, or how to function while living out of a suitcase with only four hours of sleep. This is where tweaking of some PhD programs can be beneficial without the proverbial ‘throwing the baby out with the bathwater’ strategy (of shuttering PhD programs altogether). A PhD student, more often than not, will have a higher IQ and greater passion about his/her field than the general populace. In my experience, if anyone can be trained to pick up the critical essence of any topic quickly, while disregarding the fluff, it is a PhD student that has read and discussed numerous scientific papers at journal clubs. If a re-tooling is required in some fields to make graduates available to work industry, why not include classes related to economics, communication and media in our graduate curriculum? Something along the lines of what Germany is doing (German academia, however, still has an archaic system of induction and promotion, which I will save for a future post). Won’t most of what is taught in “Advanced Biology 709: Cell Signaling” be learnt by the student while working in the lab, or attending talks during the course of their PhD anyway?
Another criticism commonly leveled against the current PhD model is that they super specialize graduates in a narrow field that is not relevant in the ‘real’ world. I find this argument to be the most ridiculous.
Consider for a moment, the field of nanotechnology, that is supposedly going to revolutionize biology. Which of the following would you consider more valuable to science –
- A thesis (by thesis, I am referring to a substantial body of work, including peer-reviewed papers) that studies how a carbon nanotube attaches to specific proteins in the cell (knowing nothing about cells), or
- A thesis that uses carbon nanotubes to deliver molecules inside a cell to look at how the cells respond to a large number of molecules in a petri dish (knowing nothing about carbon nanotubes)?
The correct answer is that both are equally valuable!
The first one is a vertical problem that looks at how the cell contents respond (in a novel way) to a carbon nanotube at a molecular level. The second applies carbon nanotubes to perform (a novel) assay on a more complex system. Both are interdisciplinary in some sense, but one is more fundamental than the other. Who is more likely to get a job outside academia? Who would DuPont hire if it wanted to test its carbon nanotubes for biocompatibility? Who would Invitrogen hire if it wanted to develop a new cellular assay kit using carbon nanotubes? Both researchers have a set of niche skills that are valuable for a commercial company. However, landing this job, or a job in academia, or any job for that matter, will require both of them to articulate their skill set clearly, and negotiate effectively. This is exactly what graduate students needs to be trained for, that they cannot pick up while performing experiments in the clean/ confocal room.
The important point here is that the key point that makes both our PhD graduates above ‘employable’ in industry (if that is to be the measure of a PhD’s success), is the fact that both worked in a promising new field of nanotechnology. That is a decision of the school offering the program, and the interest of the potential graduate student in identifying current scientific trends. Other factors, like relationship with the thesis mentor, how long the project lasted, number and quality of publications required to finish the PhD, will largely vary across the board.
There are undoubtedly hardships associated with this kind of lifestyle. Money is the first (perhaps the most important) thing. Interestingly, it is worth looking around and noting that sufficient remuneration in the early stages of pretty much any creative field is always lacking – be it writing, painting, photography, or acting. All of these fields, including science, carry the hope that if you are passionate enough to survive the formative obstacles, and confident enough to produce the next big thing, you get to keep ownership of that work – either in the form of a book, canvas, print or role. One could argue then, that an effective PhD program gives one a much better roadmap of nurturing creativity than simply floundering on your own.
The PhD programs at US universities are the best in the world. But they can be improved effectively by including modern knowledge and techniques. Discarding the mentorship model between the thesis advisor and the student entirely, or by spreading PhD theses too thin over multiple disciplines, so much so that the likelihood of discovering anything fundamental about a system becomes harder than it already is, will be a shortsighted, market-driven strategy that is likely to cause more harm to the PhD than good.
“We admittedly made this observation with the benefit of hindsight.
This science is known as retrospectroscopy.”
~ William Lipscomb Jr., Nobel prize-winning chemist. Taken from here.
Welcome to my new blog. This is where I will be writing about science issues, armed with the benefit of hindsight. My other blog is a photography blog called Writelights. I have been thinking of starting a science blog for a while. Last weekend, while reading the New York Times, the name jumped at me.
This blog aims to communicate to non-scientists, aspiring scientists and lost-in-the-way scientists. I will write about events and ideas very closely related to my day job. All photography related posts will continue on Writelights. Eventually, both will co-habitate happily on the web, or one will survive by a process of Darwinian selection. I will have (moderated) comments on this site for the time being.
My identity is fairly Google-able. You can also follow me on Twitter if you like doing such things.
So, here goes …
Retrospectra 