• 2 Posts
  • 45 Comments
Joined 1 year ago
cake
Cake day: July 4th, 2023

help-circle



  • It’s not that we don’t use mode, there are definitely times mode is used. It’s just that mean (and median as well) contain a lot more useful information about distributions that we often care about. For a normal distribution mean, median, and mode should all be identical. So why do we use mean? Because mathematically, the mean is what underpins the formula for the normal distribution, not median or mode, and when you’re talking about doing math with normal distributions mean is the thing to talk about (along with standard deviation).

    We use median a lot too, you probably just don’t hear it called median very often. The median is useful in non-normal distributions, and it defines the 50th percentile, so along with the 25%-ile and 75%-ile you’ve got your quartile distributions. We use these all the time to talk about grades in schools, or when we talk about home prices distributions in a given area, or salaries within a given field.

    We use mode too, again just by a different name most of the time. Any time you’ve asked “what’s the most common blank” you’re basically asking for a mode. When we talk about “average” income in a country, we’re usually actually talking about median or mode. Favorite animal? Answered as a mode.

    You have to use the right statistical tool for your question: unfortunately English doesn’t do a good job of conveying this without math jargon.






  • If I was guessing, in general, I think people who advocate for a pure meritocracy in the USA feel the world should be evaluated in more black and white, objective terms. The financial impact and analytic nature of STEM and finance make it much easier to stratify practitioners “objectively” in comparison to finding, for instance, the “best” photographer. I think there is also a subset of US culture that thinks that STEM is the only “real” academic group of fields worth pursuing, and knowledge in liberal arts is pointless -> not contributing to society -> not a meaningful part of the meritocracy. But I’m no expert.


  • As a general rule, yes. People who are able to better perform a task should be preferentially allocated towards those tasks. That being said, I think this should be a guiding rule, not a law upon which a society is built.

    For one, there should be some accounting for personal preference. No one should be forced to do something by society just because they’re adept at something. I think there is also space within the acceptable performance level of a society for initiatives to relax a meritocracy to some degree to help account for/make up for socioeconomic influences and historical/ongoing systemic discrimination. Meritocracy’s also have to make sure they avoid the application of standardized evaluations at a young age completely determining an individual’s future career prospects. Lastly, and I think this is one of common meritocracy retorhic’s biggest flaws, a person’s intrinsic value and overall value to society is not determined by their contributions to STEM fields and finance, which is where I think a lot of people who advocate for a more meritocracy-based society stand.







  • A lot of the really prestigious medical schools/residency programs have a reputation for a toxic culture, and this means by and large they attract a larger share of toxic applicants. Their programs are really focused not on training great bedside clinicians, but on training people who will attempt to change fields through research and public policy. Unfortunately, the toxic nature of these programs, and their immense emphasis on publishable output and reputation, likely attract a larger fraction of narcissists with the skills necessary to mask their inappropriate behaviors when needed in comparison to other programs. That is not to say that all Harvard trained physicians are horrible people, I’m sure the vast majority of them are fantastic, but I would be money that Harvard attracts a greater fraction of the kookes than your average midtier medical school.



  • All remote based typing is awful, T9 included. I can’t speak for everyone, but I can type with swipe gestures on a virtual keyboard via remote faster than I can input T9 text. I’m unaware of any stock remote for a device with a full keyboard. I would argue Apple has text entry perfected at least as well as any other major manufacturer. You have virtual keyboard entry, solid voice-to-text, and it can be configured to push a notification to your iOS device when you enter a search bar which will auto-open to the remote app and pull up the keyboard. Because of this feature passwords can also be autofilled from Keychain to make logins easier.

    You may personally prefer T9, but I’ve never seen anyone in the last decade input anything into a TV via T9. And you’re asking why it doesn’t have voice input, when it does. You admit to having never used an Apple TV yourself. I hate the idea of app-only interfaces features, but this isn’t a case like that. Maybe you should understand the features of a product before you call it “fucking stupid”.



  • You’ll have to strike a balance between security and ease. Your two major options are reverse proxy and VPN (Tailscale is one option for VPN)

    For reverse proxy, you functionally open the app to the internet. Anyone with the correct web address can access the login page. This is inherently less secure than VPN, but not irresponsibly so. Beyond the reverse proxy itself, you’ll also have to learn how to configure an HTTPS certificate to increase security since it will be open to the internet.

    For VPN, every user you want to be able to access the service has to be tied into the VPN and have the VPN running throughout their access. Tailscale is arguably the easiest way to configure a VPN right now, as you won’t have to manually deal with VPN configuration files for every device. VPN use will functionally make it like you’re on your home network. VPN access to your network should not be given to tons of people if at all possible.