Average Is Boring: How Similarity Kills a Meme's Success | Hacker News Reader

Average Is Boring: How Similarity Kills a Meme's Success

nature.com

25 pointsmikk146 comments

I was expecting this to be an article about why "I Don't Always X, But When I Do I Y" type image macros have a short shelf-life before we tire of them.

The internet and its attendant peculiar uaage of terms has killed my brain.

Killed your brain? Nah. This was about those types of image macros, though it was more about their relative shelf-lives compared to eachother than short life in general. The supplemental materials are informative:

http://www.nature.com/srep/2014/140926/srep06477/extref/srep...

Exactly (I am the paper's author). While it is dangerous to identify the scientific term "meme" with those image macros, to study them has a lot of advantages. For example, it is very easy to quantify how popular they are, where and when they get their popularity spikes and so on. I am working on follow-up research enlarging a bit my narrow set of memes, to get it closer to the actual scientific term.

Ooh! Nice. I've been doing a bit of reading on meme theory lately and it's definitely caught my imagination. Still, I'm pretty skeptical.

I think my general problem with memetics is that it's so hard to quantify a meme. Using image macro jokes is definitely a good way to get something that's easy to measure—but does that simplification go too far? Is it too artificial? What makes an image macro joke "atomic"? Sorry if you've answered this in the paper, but I started skimming when I realized that you were talking about image macro jokes.

As far as I know the criticism of meme theory is that the mutation rate is too high for the sort of stable evolution that we see in biological life. I don't doubt that natural selection is at work, but the more I read about meme theory, the more I think that looking at linguistic units as gene-like is a poor abstraction—precisely because of the difficulty in measurement.

I'm not sure how to phrase the questions I have since I am a complete layman in regards to evolutionary biology, linguistics, and meme theory. But any high level overview of your rationale and pointers to further reading would be awesome.

Using a biological metaphor to study memes is actually an ongoing debate. This strategy has been questioned by some researchers [1], while others (me included) found that it fits the observation quite nicely [2]. I haven't much more to say than "We just need more research". Sorry if it is not very enlightening :-)

My rationale for this paper comes from the fact that the usage image macro templates is pretty stable, as observed with two years worth of data. I have previous work [3] where I use image macro memes and I show evidences of competition, collaboration, clustering tendencies (which could be interpreted as the creation of proto-organisms, more research needed). On top of those observations, I find that these image macros satisfy at least the weakest requirements to be considered a proper subject to be studied.

Bear in mind the difference between what I call a "meme implementation" (the image with superimposed text, like in http://i.imgur.com/j9CTxt8.jpg) with the "meme template" (in this case the collection of all Socially Awkward Penguins). The latter is the one that's relatively stable, the former is very dynamic. In a paper I just submitted I analyze meme usage in Reddit's post titles and I observe just that. So that's definitely going too low.

[1] E.g. Weng, L., Flammini, A., Vespignani, A. & Menczer, F. Competition among memes in a world with limited attention. Scientific reports 2, 335; doi:10.1038/srep00335 (2013)

[2] E.g. Adamic, L., Lento, T., Adar, E. & Ng, P. Information evolution in social networks. http://www.ladamic.com/papers/infoevolution/MemeEvolutionFac...

[3] Michele Coscia: Competition and Success in the Meme Pool: A Case Study on Quickmeme.com. ICWSM 2013

The irony of a scientific paper dissecting the popularity of "Socially Awkward Penguin" is thick.

Reply on news.ycombinator.com