Re:You want to know something about how bullshit insane our ...
by eli_gottlieb on Monday January 07, 2019 @07:49PM
(#18851877)
It's amazing how many of the "bullshit insane" things in neuroscience make perfect damn sense from a Bayesian brain standpoint.
If you're saccading to update an approximate posterior model of the visual scene, you don't need to "pause" or "blank out" the blur of the saccade. Each fixation is a sampling of the environment under generative assumptions, so yeah, your previous posterior becomes your new prior, and as long as the saccades don't take too long, your new prior is accurate enough to make decent predictions! Since what we consciously experience is the predictions themselves, we don't notice any gaps until the bottom-up prediction error signals become huge.
Re:You want to know something about how bullshit insane our ...
by parliament32 on Monday January 07, 2019 @06:03PM
(#18851189)
Why do people think it's a good idea to post a 1700 word writeup on Twitter? I'm not trying to be snarky, I'm legitimately curious why anyone would want do this. Do you think your audience enjoys reading tiny post after tiny post? Do you just love posting short messages in sequence SMS-style?
Re:You want to know something about how bullshit insane our ...
by hombre_fatal on Monday January 07, 2019 @06:53PM
(#18851560)
Well, it has 28k retweets, 72k likes, and a hundreds of 1-on-1 engagements and Q&A. More than I've seen in pretty much any Disqus comment section.
Doesn't seem like the best tweet to die on to make your point. The results surely answer all of your questions.
It's a weird format, but I encourage you to at least try and appreciate it instead of dismissing it. I'll try: It's kind of like what a blog post might look when annotated on genius.com: each point can spawn its own discussion and people can engage the author on any subtopic right there alongside the content.
Another benefit of the anti-narrative format is that it has an immediate quickstart. One second you're reading the opening tweet about how "bullshit our brains are", and the next you've learned something interesting. You might not have even clicked into a longer-form format about how the eyes work, there's just no commitment.
Re:You want to know something about how bullshit insane our ...
by jimfleming on Monday January 07, 2019 @07:15PM
(#18851700)
Yep, people post long-form content to Twitter because there's engagement there. Literally no other reason because it is a terrible medium for reading and writing long-form content like this. But that level of engagement is rare on proper "long-form" sites and personal blogs.
Re:You want to know something about how bullshit insane our ...
by justtopost on Monday January 07, 2019 @08:37PM
(#18852220)
Agreed. When I see a link to a twitter feed instead of an article I pass on by. If your best outlet, professionally, is twitter, you need to reevaluate.
Re:You want to know something about how bullshit insane our ...
by alternaught on Monday January 07, 2019 @10:55PM
(#18853004)
Uh, well, it’s not about whether the platform serves the content well, or whether the platform introduces disruptive artifacts into the delivery.
It’s really about finding the eyeballs where they live, because there might be no other way to get in front of them.
Some platforms really can claim a monopoly on a cross section of eyeballs that only use their site. Some people only use Twitter. Some people only use Facebook.
Much in the same way some people absolutely refuse all name brand social media. How do you barge in and get in front of them? Well, that’s the trick, isn’t it? Most people engage in this sort of isolation, because they don’t want advertisers barging in, and prying into their maxed-out attention span.
If you're saccading to update an approximate posterior model of the visual scene, you don't need to "pause" or "blank out" the blur of the saccade. Each fixation is a sampling of the environment under generative assumptions, so yeah, your previous posterior becomes your new prior, and as long as the saccades don't take too long, your new prior is accurate enough to make decent predictions! Since what we consciously experience is the predictions themselves, we don't notice any gaps until the bottom-up prediction error signals become huge.
Doesn't seem like the best tweet to die on to make your point. The results surely answer all of your questions.
It's a weird format, but I encourage you to at least try and appreciate it instead of dismissing it. I'll try: It's kind of like what a blog post might look when annotated on genius.com: each point can spawn its own discussion and people can engage the author on any subtopic right there alongside the content.
Another benefit of the anti-narrative format is that it has an immediate quickstart. One second you're reading the opening tweet about how "bullshit our brains are", and the next you've learned something interesting. You might not have even clicked into a longer-form format about how the eyes work, there's just no commitment.
It’s really about finding the eyeballs where they live, because there might be no other way to get in front of them.
Some platforms really can claim a monopoly on a cross section of eyeballs that only use their site. Some people only use Twitter. Some people only use Facebook.
Much in the same way some people absolutely refuse all name brand social media. How do you barge in and get in front of them? Well, that’s the trick, isn’t it? Most people engage in this sort of isolation, because they don’t want advertisers barging in, and prying into their maxed-out attention span.