Hacker News Reader

Comment by madhadron | Hacker News Reader

This makes me feel sad. Biology textbooks are full of these kinds of diagrams, and my impression from dealing with people who came up through normal biology training is that they are a hindrance to learning to think about biological systems in two ways:

1. They give the impression of a mechanical system like you might make out of an erector set as opposed to the chaotic, roiling mass that is the interior of a cell.

2. They prevent students from learning to think in terms of relations among observed quantities without overspecifying them.

View on HN

Agreed, but I think the real problem is that the medium for communicating science is still basically a static PDF. Even if you use fancy microscopy to capture some of this roiling biology, the best you can do is link to a video file in the supplementary section of your paper, which few people will visit. Or to summarize the information in a bar graph.

This is so disappointing in this day and age where everyone at least has a computer, if not an iPad, capable of displaying video and animations in line with text.

And it would be great if BioRender would apply the principles laid out in "Stop Drawing Dead Fish" by Bret Victor: https://vimeo.com/64895205

On the other hand, biology is so complicated that you have to simplify things down to abstractions that let you communicate and reason about causal stories, and that's what these sorts of figures are good for.

Completely agree, but the clip-art approach doesn't help - it's often easier to understand casual relationships in terms of colored dots for partitions or very basic parameterization like node size in a graph, than to with clip art of tiny livers or brains.

Many of the more successful abstractions seem like they would make for extremely fun games. I've seen a few Fantastic Voyage! bio-themed videogames but they are usually either stupid memory games designed to teach technical terms, recycled space invader games with bacteria clip art, or simulations like FoldIt that are fun puzzles but don't really capture the dynamism and system behaviors that make biology so interesting.

If you'll forgive me venting my spleen, I really really want a game that lets me run the immune system. I don't need it to look like the inside of a bio textbook, I need it to model the information-production-distribution-processes and their strong resemblance to dynamic wargaming.

+1 would like to see this game too.

As a medical student, however, I can vouch for the utility of clip art in visualisation. You get delivered so much visual information that if there is coherence to it, you have more chance of processing it correctly.

I'm soooo happy to read this. I am very interested int eh dynamics of social networks and spend a lot of time abusing software from the bioinformatics fields to try and model behaviors in online ecosystems. In the process I've gotten quite interested in cell biology and the whole field seems like such a weird mix of hard data and hieroglyphic stories.

Levels of abstraction. I agree that there should be more exposure to the roiling dynamics of it all. From the high-fidelity simulations I've seen, you should think of a large macromolecule more as a tumbleweed or raft of seaweed that is being bombarded by water than as something made out of rigid pieces.

Yup. My school-acquired view of cells was always of somewhat squishy, mostly empty structures. It wasn't until recently that I learned they're all densely packed bags of molecules, shaking like crazy, and in which a lot of functionality depends on random walk making everything eventually bump into everything else.

One of my favourite videos: https://www.youtube.com/watch?v=4Z4KwuUfh0A

Reply on news.ycombinator.com