Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)P
Posts
56
Comments
1421
Joined
8 mo. ago

Freedom is the right to tell people what they do not want to hear.

  • George Orwell

  • Assuming you're on browser and have uBlock Origin installed you can filter out posts based on keywords with a custom filter

    lemmy.world##div.post-listing:has(span:has-text(/trump/i))lemmy.world##div.post-listing:has(span:has-text(/maga/i))lemmy.world##div.post-listing:has(span:has-text(/republican/i))

    And so on..

    Just note that filters like "ICE" block every post with a word in it that contains those letters. You can prevent false positives by adding spaces on both sides of the word but then it doesn't trigger if the post title starts or ends with that word. Either way it helps a lot. Half of the threads on my personal front page regularly get filtered out.

  • DeepMind's AlphaFold AI figured out protein folding.

  • Never really paid any attention to it. I've had unlimited data plan since like 2012

  • The Unknown Soldier is quite well made and high budget war movie by Finnish standards and it does nothing to glorify war.

  • Doesn't matter. Most levels aren't "two sided" like that anyway but have a top and bottom.

  • Not mixing it up with the WD40 one?

  • If you have a miscalibrated bubble level on a truly level surface, the bubble will sit more toward one side - say, left. Flip it 180 degrees, and it'll shift to the right.

    On an inclined surface, the level might read level one way and show incline the other way.

    Here you could just adjust the screw until the bubble stays in the exact same spot no matter which way you flip the level.

  • You can use the same method you already used to check accuracy in the first place. It doesn't even need a level surface - just make sure the bubble sits exactly the same distance from the lines when you flip the level the other way around.

  • Tradition is the experiments that worked.

  • Regularly inhaling combustion fumes is bad for your health no matter how you try to rationalize it. Yeah, more smoke is worse than less, and not all smoke is equally toxic - but it's all bad and should be avoided.

  • First you moved the goalposts by pivoting from "there's no want for LLMs" to "okay but how many are paying." You quietly shifted the entire criteria of "want" from voluntary demand to monetization the second evidence of massive adoption showed up.

    When I pointed out that VLC has hundreds of millions of users who also don't pay, you tossed in the irrelevant "it's open source by one person" line - which is a complete non sequitur. Development model or monetization status has zero logical bearing on whether 800 million weekly ChatGPT users demonstrate real desire for LLMs.

    This is classic bad-faith argumentation: throw in red herrings, change the standard whenever your position weakens, and misrepresent what was actually said to avoid engaging with the actual evidence.

  • I need to see some credentials before I take advice from you.

  • That's irrelevant. Nobody pays for VLC player either.

  • Psychedelics probably. Mushrooms specifically.

  • I just tell people not to read into it. I don't even claim to forget - because I don't. I simply don't casually check in when I have nothing important to say.

    You might not hear from me for six months, but when you finally do, I act like we just talked yesterday.

  • This is quite toxic attitude. You should focus on what's good for you - not what's bad for someone else.

  • When pictures get compressed, the data is lost. It's not hidden - it's gone forever.

    When you "increase the quality" with a tool like this, you're not recovering lost data - you're making up new data to fill in the gaps. It takes an intelligent system to figure out what should probably be there and then add it convincingly.

    I don't think you can pull that off without AI.

  • we have to pretend that this is a product we want

    ChatGPT alone has 800 million weekly users.

  • I'm still going to pop an aspirin the next time I got a headache even though I do have an ebike too.