• 2 Posts
  • 340 Comments
Joined 2 years ago
cake
Cake day: July 26th, 2023

help-circle



  • Most of these aren’t even news. They’re entertainment companies that happen to include reporting-like segments and opinion pieces, whether they’re accurate or not. And they’re for-profit; their main customers are advertisers, not viewers/readers.

    A non-profit that simply reports without spin would be “real” news - actually trustworthy information from a source that isn’t trying to brainwash you or milk you for ad revenue. There are very, very few organizations that fit this criteria. The biggest is AP (Associated Press). The entertainment companies license AP content (and content from for-profit competitors like Reuters) but filter and distort it however they wish.



  • I feel like an ethics-optional group that wants to really end the use of coal could run a campaign of breaking into or drilling down to the most profitable commercial coal seams and lighting them on fire. More Centralia, Pennsylvanias, especially in populated areas, would probably permanently dampen the domestic industry and drive prices far upwards leading to the faster growth of alternatives. The obvious tradeoff is, those seams your group lit are going to burn for a very very long time, so you’re causing some emissions in exchange for a global reduction.





  • “I don’t live on this property, I just use it for storing my RV.”

    Tall shrubs/trees all around, get a PO box. Don’t show your neighbors around.

    Exactly what is going to compel local police to get a warrant to enter the property and somehow confirm you live in that RV full time?

    These kinds of laws are crafted by elected legislators who are either truly inept, or only interested in performative gestures to boost their image. A law without realistic enforceability is mostly imaginary, and primarily serves just to tack extra charges onto some other sentence a person got for breaking a law local police actually enforce.



  • And who’s going to make these arrests?

    Make arrests by police that don’t identify themselves or their badge number ineligible for jail time, and their statements should be ineligible in court. Police don’t want detainees getting off free and wasting their time? Guess the body cam needs to be on when you declare their arrest and your badge number.

    Tie state and municipal police department bonuses, benefits, and overtime pay to adherence to identification policy. Department has credible reports of violations and fails to properly investigate, discipline, or fire the offending officers? Sorry, your department is ineligible for their usual handouts this quarter. Your department is doing a good job policing themselves in good faith? Looks like someone’s getting all that leftover bonus money.



  • Nobody is forcing athletes competing in this event to seek medical approval and prescription for the use of performance enhancing drugs by physicians with no moral qualms doing so. Witold is not any athlete’s personal physician, and is not qualified to speak about medical concerns. He is a former sprinter turned politician. WADA’s role is to serve sporting events by preventing performance enhancing drug usage that those events consider unfair during competition.

    If someone is choosing to take a performance enhancing medication in this “Enhanced Games” event and is harmed by it, that’s between them (or their estate 😂) and their doctor, and whatever malpractice suit is brought about to a court of law.





  • At the end of the day, most internet forums exist as entertainment. It doesn’t really matter whether the text is from real people or not, what matters to Reddit and other platforms is that real users show up to pump numbers and get served ads. It’s kinda like how pro sports is not latently about athleticism, it’s just another entertainment platform to get butts into seats and eyes onscreen to make a buck.

    If bot content gets the eyeballs, so what? Guess the content being fictional doesn’t really matter. People don’t like being lied to, but then nobody is claiming everything you read on the internet is true in the first place.

    Anyway, to all the people who imagined humanity’s future would end up looking different than WALL-E: sorry to disappoint. You’ll get a hover recliner, at least.


  • How do two people decide they like each other enough to form a long term relationship?

    Part of that has to do with cues that really do require physical interaction to gauge. Apps will never bridge that gap.

    Another element of it is the inundation of choice. Because there are so many options, and new options every day, it’s tempting to hold out waiting for Mr./Ms. Perfect. (Frankly, women tend to fall more into this camp though. Desperate men often try to match as many women as possible, and the exact opposite is true of attractive women on algorithm based apps.)

    Finally, I think what people think they want, and what they actually settle for, are very different when it comes to relationships. Two people might become infatuated with one another because of regular contact (work, volunteering, hobby, etc) and end up in a successful relationship despite maybe not being an instant match. Apps will also never replicate this, and its sort of a distillation of the above two points.


  • LC blows. I’ve been on the side of interviewing and hiring engineers, and the only thing LC is good for is filtering down when you have way too many decent candidates to reasonably interview, like on the order of hundreds for every position. It’s like how ivy league college admissions are very arbitrarily selective - getting in is way harder than actually doing the work once you’re there. They just need a way to toss out most applications without too much risk of dropping the most promising ones because they don’t have the time to critically assess everyone.

    Whether someone can remember the algorithm for depth first search they learned 10 years ago but never needed to use is totally useless. System design knowledge is much more relevant, but not as easy to auto-grade a candidate on. But more than either of those is someone’s ability to…you know, actually do the daily work and take responsibility for the systems they work on. I actually hope the AI interviewer pre-screenings can take the place of unhelpful algo centric automated assessments.

    Like: What does the candidate think about testing? How do they plan a large project? Can they form opinions and drive the outcome of a technical discussion towards a robust solution? Are they capable of organizing complexity into manageable and composable units? Can they quickly debug a problem to identify the root cause? Can they write code to do something practical and similar to the actual work my team does on a regular basis?

    Anyway, I’m rambling. If I were in your shoes, I’d look into a trade certification program. Go out and repair wind turbines or something. Sitting in a chair all day looking at screens leaves people depressed.