Bridging the Gap between InfoVis Research and Practice

bridging the gapOne of the main goals of this blog, as explicitly stated in my about page, is to expose what happens in the world of research to a larger public and to bridge the gap between practitioners and researchers (in both directions). I don’t know if I am already meeting my goals, and to tell you the truth I am not even sure whether you, my fellow readers, are really that passionate about the idea. But let me experiment with it. I am convinced we really need some initial baby steps because there are benefits for everyone.

My research paper posts

I am sure many of you didn’t even notice it, but I already started to take this business quite seriously. Thanks to Andrew Vande Moere (really, thanks Andrew!), the mind behind Infosthetics, I had the possibility to reach a very large audience with research paper posts. What is a research paper post? Basically, every two weeks or so I select one single paper from top-notch conferences or journals and publish a summary report with few comments and guidelines. The summary report has two main features: (1) must be easy to read (2) must offer simple and handy guidelines. The intent is clear: make science accessible and ready to use. Here is the list of posts published so far:

As you can see they are all quite theoretical and all with the potential to add new tools to your data visualization mental toolbox. This is the intent. Give them a read, I’d be happy to hear if you learned anything.

I need your help

My plan is to continue along these lines but … I need your feedback!

  • Do you like the idea of research papers posts?
  • Do you think they are or may be useful to you?
  • Would you like to see them appearing in FILWD?
  • What about having an archive with research papers posts in FILWD?
  • Any suggestions on how to make things better?

Your answers would be very helpful. I really need to decide what shape to give to this idea and I’d like to hear from you. Please comment below or let me know on twitter. Thanks a lot in advance!

But why is research important?

As the whole field of visualization, visual analytics, and data analysis, more in general, reach a larger and larger public, it is important to offer a solid platform on top of which people can build excellent products and innovations. We might be tempted to think research in this field is only a way to produce new technologies, but this is only one side of the coin. Research is also very much needed as a vehicle to reason about what we do and how.

The temptation is high to limit the practice of data visualization to an endless iteration and rediscovery of pretty and joyful designs. Don’t get me wrong, I love this side of this discipline: the colors we can show, the impression we can make. But this is only one aspect of it, the part that makes visualization easier to market, but no one will ever take this stuff seriously or adopt it  on a longer term just for the fun of it.

We need to build a body of knowledge and make sure practitioners will use it. Please note: not only we need to build this knowledge, we also need to spread it. Yes, it is not only important that people do good research but also that its effect is heard. I noticed it so many times: people re-discovering things over and over. This is a huge waste of resources and we just cannot afford it.

What researchers can do

There are a number of great things researchers could and should do in order to reduce the gap.

  • Write a blog. There are an endless number of benefits related to writing an academic blog. I am not the first one to point this out. The number one excuse for not blogging is the limited amount of time. The second, is the alleged lack of academic reward. Both are false. Read: You Arent Blogging Yet?!? – The Scientist and 6 tips for low-cost academic blogging.
  • Write more readable papers. One of the deadly sins of research is its extreme autoreferentiality, that is, researchers do research and write for their fellow colleagues. This is true in every academic circle but it’s especially bad in engineering/design fields where lots of practitioners could/would learn more if the papers were more readable. I also suspect one thing: if one writes papers with a larger audience in mind the end results would be overall better, also to the eyes of fellow researchers.
  • Learn how to market research products. I am always surprised to see how many people in academic circles don’t take care of this fundamental aspect. Research is not different to any other kind of product (I hope nobody is offended by this sentence), in order to spread and influence people it has to be marketed, you like it or not. We live in the years 2000, there’s no more space for the hard-working lab rats with genius ideas. And ideas spread on the web, with specific marketing techniques.
  • Use web exposure, not citations as a metric for success. It’s funny how attached to the number of citations researchers are. This is again related to how auto-referential people in academy are. But if a broader impact measure is used there are benefits for everyone. If you don’t exist on the web you just don’t exist. Do people talk about you? Do they tweet about you? Do they go to you web page? I always laugh when I think that a single successful post in this blog is cited a number of times more than any of my research papers.

What practitioners can do

  • Refuse to reinvent the wheel. This is a killer mindset for a visualization designer. If you are contented with the first solution coming into your mind without any prior research, chances are you are not doing excellent work. InfoVis researchers have been inventing solutions for any kind of data for ages and there are very high chances that somebody else has dealt with it before. Do your homeworks, search around, be open to be inspired from what others have done before.
  • Learn how to learn from research. My guess is that research looks intimidating to people who are not accustomed to it. Maybe some people don’t even feel like they are good enough to read and understand a research paper, but this is bullshit. Everyone can learn to swim this sea, it just takes the patience to learn the rules. The most important part is to learn where the authoritative sources are. What are the best labs? The best publications? The best conferences?
  • Learn where to find great papers. Some people might object this is not always true and that there are notable exceptions, but listen to my advice: the best conferences and journals are the best. They can guarantee high standards over and over and you have  less chances to waste your time if you focus on them (e.g., VisWeek).
  • Participate to research conferences. Similarly, if you have time and money, participate to research conferences. I am sure you won’t regret your choice. Personally I am always amazed to see how open this community is and how many cool things happen during research conferences. And there are plenty of opportunities to participate actively even if you are not from research: panels, birds of feathers, exhibits, challenges, etc. Take my advice: go to VisWeek next year and you won’t regret it.

What else can we do?

This is just a handful of ideas that came into my mind when I started writing the post. I am sure there are many more useful ones. What do you think? Is there anything else researchers should or could do to help you in your job? Is there anything practitioners should or could do to make the infovis practice better? Or maybe, why not, there’s no real need to bridge any gaps and this is just my paranoia? I’d be happy to hear from you. You can write a comment, send me an email or send a message on twitter.

Have fun. Take care.

7 thoughts on “Bridging the Gap between InfoVis Research and Practice

  1. g.

    Hi,

    I think it would be great, so yes to all. As a student, I certainly have access to many ressources, but it’s hard to find interesting articles without knowing specialized journal. If you resume them, it would be great if you don’t, as usually in vulgarisation, do it by removing the process which leads the scientists to their conclusions. That’s what give these articles all there value !

    Thanks.

    Reply
    1. Enrico Post author

      Thanks for your feedback, this is certainly one point in favor of posting this stuff in FILWD!

      Reply
  2. Martin

    Hey,

    again a nice blog post. The part “What researchers can do” is exactly the same what we are actually discussing at our chair – and what our prof already realizes in contrast to other at our faculty. Especially interesting would be “how to do a good web exposure?” A nice hint is the link to the 6 tips for low-cost academic blogging. Maybe there are other which need to discover.

    Thanks,
    Martin

    Reply
    1. Enrico Post author

      Martin, it’s nice to hear the same issues are discussed in the context of a chair. I understand what you mean when you say some people are not ready to listen to these voices. But this is what the future will be. No web exposure = no impact. I see over and over people discussing how to build a public image on the web and gosh … it’s damn hard! I realize how difficult is to let people adopt the right mindset. For instance, people fear their ideas will be stolen if published on the web, but there’s no space for plagiarism on the web! If you do it, people will discover it and you will be dead. The web is a win-win system. there’s no need to keep things for yourself. The more you give the more you get.

      Reply
  3. Andy Kirk

    Hi Enrico

    To your questions raised above,

    Do you like the idea of research papers posts?
    Absolutely

    Do you think they are or may be useful to you?
    Absolutely

    Would you like to see them appearing in filwd?
    Absolutely

    What about having an archive with research papers posts in filwd?
    Absolutely

    Any suggestions on how to make things better?
    Off the top of my head it would be great to have this pursuit, particularly the archive concept, as a highly collaborative process, with opportunity for recommendations – some of the best research papers can be buried and almost lost without largely being cited. Perhaps you could establish a network of data viz blogs to help coordinate a call out? I’d be more than happy to be involved, especially as i am currently wishing to top up my awareness of some of the contemporary papers.

    Keep up the great work!

    Best wishes
    Andy

    Reply
    1. Enrico Post author

      Andy, thanks a lot for your encouraging answers! I really need this kind of support. I like your suggestions. One possibility could be to couple the papers archive with a simple form to suggest papers to add/review? That sounds really really nice. Regarding the network of blogs, this is also a great idea I had in mind for a while, but it looks a bit too early to me. As you may know well, maintaining a blog is a damn s**t amount of work and I am not sure whether I have the time and infrastructure to do that NOW. But I will keep it in mind. One interesting option I’ve learned about in other unrelated blogs is to form a mastermind group. That would be really great.

      Reply
  4. simone

    Hi Enrico, i’m reading your blog since your latest guest posts around the web and first of all YES they are very important. I’m a SEO (search engine optimizator) here in italy and every time i talk to a prospect or to a customer they need to be fully convinced of what i’m saying it’s actually true. The aspects of what’s true may vary from the situation, but if i give data and a context to understand it, i instantly get trust for my words too.

    So this is the second point: trust. The equation for Trust is not that easy to write down, anyway you somehow managed to get a little of mine, so i subscribed to your posts. This is the third point. yes: please post them on your blog, since i started to trust to your efforts in the public sphere (because trust is contagious, like a virus) and i like to find all your published posts in the same place.

    The fourth thing is Open Data. Since we all work with data, we need a full change of the common paradigms of data communication. Infographics are cool to inform people that have few contextual information, what is very good for professionals is the ability to tell their own point of view (it means to put their context) on your data, for example and infographic about schools quality is cool, but having the ability to display data only for the schools i’m interested in it’s also cool. If you put online the whole database i can write a software to take the exact data i need for each school, and i’m able to draw my own infochart, with your data. There is a lot of interest in the world for Linked Open Data since TimBerners Lee’s speech some year ago, so i will not explain further.

    fifth: your need for feedback. If you are going to use your blog to publish, please learn that requesting feedback is one of the worst thing to do, because people like we will come and waste a lot of your time.

    lol

    thanks for your amazing work, and keep posting. Just put a small excerpt to explain WHY you think an article would be cool to read, so that i can learn a little from your context.

    ceers.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *