Do you have a dumb question that you're kind of embarrassed to ask in the main thread? Is there something you're just not sure about?
This is your opportunity to ask questions. No question too simple or too silly.
Culture war topics are accepted, and proposals for a better intro post are appreciated.
Jump in the discussion.
No email address required.
Notes -
Dunbar himself is still kicking, btw. Moreover, he's producing papers at an insane rate (number of Dunbar's articles about Dunbar's number). Here's the paper from 2016 on Facebook and Twitter, but they used rather old datasets -- 2009 and 2012 resp -- which reflected social media interface at that time.
Social brain hypothesis, which he's been studying, is about existence of several layers of contacts within any social communities. 150 is a size of one of the layers. After clustering social media data he found similar layered structure:
I read about this experiment from another author, who said similar results were obtained in online game communities. But he was rather skeptical, saying that the data is limited and there are many built-in artificial structures, forcing certain clusters. Those clusters might be stretched to fit the hypothesis.
Are big name researchers like him actually doing any of the grunt work or just get handed authorship for having answered one question asked by a PhD student?
Yeah yeah not all researchers, but in the ML space I usually see some very big names appear as 5th or 6th authors in inconsequential papers and am left wondering, why are they even there?
The biggest contribution might even be "asked one question answered by a PhD student" rather than the other way around. My first original math discovery as a grad student turned out to be an idea that was published before I was born, but was just esoteric enough that I hadn't heard of it. My third was something that I got beaten to, because I didn't realize the problem was that interesting (I thought it was just another test case for my second) and I blabbed about it at a conference to someone who turned out to be 10% closer to it than I was but who just hadn't considered using his research on my question before.
I eventually learned to run more by my advisor. Having someone who's been in a field a few decades to tell you what's more or less worth working on is invaluable. Even just having someone who knows what to look for and where to look can be useful. I taught a student once who invented a numerical method I'd never seen before, one with just the right mix of "tricky enough that it might have been missed" and "valuable enough that it ought to be published", but by this point I knew which book to pull off the shelf and which chapter to hunt through to find the prior discovery without having to spend days on the literature search.
More options
Context Copy link
More options
Context Copy link
Wow thanks for the link, had no idea that Dunbar was so prolific. This is interesting, I would definitely need to do some more reading before I could judge whether it was accurate.
I'm quite curious to see how online communities are different from physical communities. I wonder if VR will have a noticeable effect - I could see communities shrinking again due to the somewhat 'in-person' nature of VR.
More options
Context Copy link
More options
Context Copy link