4 ms·

I love it when people do take the time to do good information visualisation, especially when it helps everyone understand a complex idea. However, I do think t

I love it when people do take the time to do good information visualisation, especially when it helps everyone understand a complex idea. However, I do think that those who believe that all academic papers should take this form fundamentally misunderstand the purpose of academic papers. Sadly, most of these papers are not about improving universal knowledge - instead they are simply attempting to convey to their peers some insight into a commonly understood problem or question. Because they are aiming at such a small target audience who have a deep understanding of the problem domain, shorthand and skimping on explainatory visualisations is not only an acceptable choice - it can make the difference between a paper being published or not. Few academics have the time, experience or funding to make beautiful visualisations - and those that can afford to spend them on that aspect often do so in exchange for depth of insight - and thus they are often seen as less valuable.

twtw 9y ago
This is consistent with my experience. Many papers seem to rely on unstated conventions regarding the problem definition and use of variables. Say I'm not in the "small target audience who have a deep understanding of the problem domain," but would like to read and understand a paper. What's the usual approach to overcome this barrier?
- tincholio 9y ago
  You'll most likely need to get acquainted with a significant chunk of the literature in that domain. For the most part, papers are self-contained, but they build on top of past work. The best way is to work your way through the references cited. Often you'll find a few that come up again and again. Read them, and you'll have gained a bit more understanding on the topic. Rinse and repeat until you do understand the problem and how it is commonly treated in the literature.
toxik 9y ago
Another important distinction is that absolutely not everything is even close to visualizable in short sweet diagrams.

twtw 9y ago

This is consistent with my experience. Many papers seem to rely on unstated conventions regarding the problem definition and use of variables. Say I'm not in the "small target audience who have a deep understanding of the problem domain," but would like to read and understand a paper. What's the usual approach to overcome this barrier?

tincholio 9y ago
You'll most likely need to get acquainted with a significant chunk of the literature in that domain. For the most part, papers are self-contained, but they build on top of past work. The best way is to work your way through the references cited. Often you'll find a few that come up again and again. Read them, and you'll have gained a bit more understanding on the topic. Rinse and repeat until you do understand the problem and how it is commonly treated in the literature.

toxik 9y ago

Another important distinction is that absolutely not everything is even close to visualizable in short sweet diagrams.