Ad verba per numeros

Thursday, February 24, 2011, 08:43 PM
Research results are supposed to be disseminated. The reason is simple: knowledge progressively increases starting from prior knowledge; thus, to get new knowledge in the future, current knowledge must be carefully preserved. Nanos gigantium humeris insidentes, as latinists would say.

Interestingly, that idea is not very old (less than 400 years) but that's another story; the fact is that current researchers are supposed to publish the results of their research. Indeed, publishing involves a huge part of the work of any researcher and, unfortunately, publishing for the sake of publishing is a major concern among researchers.

To be better researchers maybe we need to go back to the basics and adapt them to our new context.

First of all, it's clear that a paper is a publication but not every publication should be a paper, or much better, not every research result has to be disseminated by means of a paper, or even better, different channels (including papers) can be used to communicate our research results to different audiences.

Second, to avoid publishing for the sake of publishing we must remember that publishing supposes communication, and communication requires us to be sure that the audience has understood our message.

So in short, scientific dissemination is much more than publishing papers and researchers need to master different communication channels such as visual presentations (e.g. slide decks, screen casts, or video), verbal communication, and even networking and social media. Besides mastering those different channels the message must be crystal clear for the audience.

To achieve those goals I propose the following "10 principles of (good) scientific dissemination":

#1 Good scientific dissemination describes current research and contributes to build the reputation of its authors.

The immediate goal of publishing is describing some recent, even ongoing, work but remember that every single work is a piece building your future reputation.

#2 Good scientific dissemination captures the attention of its audience immediately.

There is an overwhelming number of papers, journals, proceedings, conferences, workshops, symposia, powerpoints, and online videos. If you want to have any chance of your research having an impact you better get their attention from the very title. If you are delivering a paper the abstract has also to be carefully assembled. When using visuals avoid "Death by PowerPoint", and "chicken-chicken" when speaking.

Needless to say, this is a just a fortune cookie, you need professional advice in these regards. Daniel Lemire has got a superb post on writing papers plus several additional links. Regarding the use of visuals I highly recommend you "Presentation Zen" by Garr Reynolds, and "slide:ology" by Nancy Duarte. The best way to improve your speech skills is, in my opinion, enrolling in some course on the matter, and training a lot on your own.

#3 Good scientific dissemination starts from a sound core idea and describes interesting and reproducible results.

A superficial reading of the second principle could imply that fireworks, smoke, and mirrors are the main ingredients of successful deliverables (especially visuals). No way! The meaning of the second principle is simple: Boredom kills. That said, publishing is pointless if there is nothing to communicate and that message has to be a sound idea. Moreover, your audience must obtain something of interest and, on top of that, they should be able to reproduce your results on their own.

#4 Good scientific dissemination is clear.


#5 Good scientific dissemination acknowledges prior work but stands out because of its novelty.

As the song says, no matter who you are, you're going to need somebody to stand by you. Actually, you need somebody to stand on. Certainly, there were, are, and will be a few giants but, frankly, I doubt they'll need this list of principles. Therefore, you must be extremely careful when reviewing the literature, you have to properly contextualize your work within the pertinent prior works and, above all, you must clearly expose the novelty of your work (because it is novel, isn't it?)

#6 Good scientific dissemination is memorable.

It's difficult to have an impact if your audience doesn't remember your work. Hence, your papers, slides, and speeches must be memorable. Fortunately, if you manage to capture their attention with a clear exposition of a sound, novel, and interesting idea you are halfway there!

#7 Good scientific dissemination is relevant to its audience.

Universal appeal is almost unattainable. That said, you are responsible of reaching the proper audience for your research. After all, you choose the venues where you submit your papers, and the title, abstract and contents of the paper, or the metadata associated to your slides or videos implicitly define their audience (remember they'll find your work using a search engine). Therefore, you must think of your target audience before crafting the deliverable, and you must be sure of providing them with something not only relevant but "actionable".

#8 Good scientific dissemination makes its authors approachable to their audience.

Shocking revelation: Most of your audience are other researchers.

This means that if you are doing your job correctly, some of them can be interested in working with you. Certainly, papers provide e-mail, office address, and even phone number. Unfortunately, they are a lot like a "cold approach" and, thus, not very appealing. Of course, you are gentle and would gladly reply their e-mails but they, just, don't, know. Thus, in addition to all that "contact" information you must provide some other ways to approach you: for instance a blog (if you update it frequently, and accept comments), or your Twitter, Facebook, or Linkedin aliases; anything that may help your audience to reach you to say "Hi! You rock!".

I've left conferences and speeches apart on purpose. Take advantage of coffee breaks and informal meetings, and promote the use of some backchannel; you can suggest the audience to send you questions by Twitter, or the hashtag to use when discussing your speech. Of course, you have then to devote the time required to answer any question providing extra information.

#9 Good scientific dissemination is not a single good paper but a continued work in progress towards excellence.

Avoid short-term goals, chose long-term goals to drive your research (and publishing) agenda, aspire to excellence but remember it's an asymptote. Embrace (or get tattooed) this motto: "Today's paper has to be better than yesterday's and worse than tomorrow's".

#10 Good scientific dissemination adapts to different media exploiting their strengths while accepting their limitations.

You produce papers, slide decks, videos, speeches, books, lecture notes, software, press releases, tweets, blog posts, etc. You are comfortable with papers, they are familiar ground, and thus you are tempted to use the same verbosity and style all of the time. That brought us Death by PowerPoint. When using a different channel you must adopt its manners, take advantage of its strengths, and be aware of its limitations. The only way of doing that is getting used to the new channel before using it for dissemination, try and fail and try again, and, in case of doubt, KISS (Keep It Short and Simple).

Disclaimer and credit: This post does not mean I consider myself a good example of scientific dissemination; it's rather a list of good intentions to improve my work day by day. In addition to that, I must recognize that the list is very strongly inspired by "Los diez principios generales de la publicidad" (The ten general principles of advertising) from the great book by Luis Bassat "El libro rojo de la publicidad" (The Red Book of Advertising); unfortunately, that book is not available in English. Last, e-mail discussions with @lemire, this link and the "chicken-chicken" video pointed out by @xamat moved me to write this post.