• 5 Posts
  • 14 Comments
Joined 1 year ago
cake
Cake day: June 13th, 2023

help-circle
  • fiasco@possumpat.iotoSelfhosted@lemmy.worldHow much swap?
    link
    fedilink
    English
    arrow-up
    9
    arrow-down
    4
    ·
    edit-2
    1 year ago

    I think it’s better to think about what swap is, and the right answer might well be zero. If you try to allocate memory and there isn’t any available, then existing stuff in memory is transferred to the swap file/partition. This is incredibly slow. If there isn’t enough memory or swap available, then at least one process (one hopes the one that made the unfulfillable request for memory) is killed.

    If you ever do start swapping memory to disk, your computer will grind to a halt.

    Maybe someone will disagree with me, and if someone does I’m curious why, but unless you’re in some sort of very high memory utilization situation, processes being killed is probably easier to deal with than the huge delays caused by swapping.

    Edit: Didn’t notice what community this was. Since it’s a webserver, the answer requires some understanding of utilization. You might want to look into swap files rather than swap partitions, since I’m pretty sure they’re easier to resize as conditions change.







  • It’s older, but The Longest Journey is good. Unfortunately, the final game in the series kinda sucks.

    While it’s an ensemble, most people would agree that the main character of Final Fantasy VI is a woman—they just might disagree about which woman is the lead.

    I also liked the first Xenosaga game, but again, it’s a series that goes pretty badly downhill.




  • I guess the important thing to understand about spurious output (what gets called “hallucinations”) is that it’s neither a bug nor a feature, it’s just the nature of the program. Deep learning language models are just probabilities of co-occurrence of words; there’s no meaning in that. Deep learning can’t be said to generate “true” or “false” information, or rather, it can’t be meaningfully said to generate information at all.

    So then people say that deep learning is helping out in this or that industry. I can tell you that it’s pretty useless in my industry, though people are trying. Knowing a lot about the algorithms behind deep learning, and also knowing how fucking gullible people are, I assume that—if someone tells me deep learning has ended up being useful in some field, they’re either buying the hype or witnessing an odd series of coincidences.




  • I hope this will help somewhat…

    When an NT interacts with another person, I have to try and make sense of what they say and do. For the most part, this’ll instinctively be done by imagining that they’re basically like me. This leads to some hilariously wrong judgments, because the line of thinking goes something like—if my imaginary version of myself were doing or saying this thing, what would have to be going on in my/their head for their words/actions to make sense?

    People tend to have pretty poor imaginations, so it’s pretty easy for someone to get to, what they’re saying/doing must be malicious. I think this is also what people mean when they say “projection,” that projection isn’t always intended as an attack, but it is almost always a failure of imagination.

    It’s said, “when someone shows you who they are, believe them.” What’s fascinating is that this needs to be said; it should be obvious. But this is basically my point: in my experience, people don’t have a great conception of the reality of others, and laziness makes people fall back on just imagining (their concept of) themselves but for being a woman, being black, being ND, being trans…

    But remember, for the NT, this is all done automatically, this is just the baseline NT social perception. Or at least my understanding of it. So for example, when you ask someone, are you competing in a beauty contest? who cares what people think? many an NT will take that as an accusation of frivolity; after all, why is that the thing you focused on? What’s going on in your head, or rather, what would have to be going on in my head for me to ask someone that? I must not like them very much, so they must not like me very much. And so on and so forth.

    I’m (probably) NT and I find this incredibly tedious, I can’t imagine what it’s like for NDs.



  • Okay, so I’m reasonably neurotypical I guess, but I’m browsing All. And a neurotypical perspective may help, or maybe it won’t. And I’ll warn you in advance, this may be a bit harsh.

    The first basic fact is that the agenda is set by the people who actually do the work (and to a lesser extent, the people who fund the work). The quoted post says the poster is not a developer. So what we’re talking about here is “backseat driving,” someone wanting to impose direction without providing either work or money. I don’t use the term “impose” lightly; the quoted post accuses everyone else of not being open to discussion, of being narcissistic.

    The other basic fact is, unless you’re in the thick of it, you don’t know what’s really going on. There are usually reasons things are the way they are. Sometimes those reasons are bad, sometimes they’re good. But particularly when we’re talking about complex engineered systems, and doubly so when we’re talking about computer software, even modest changes usually ripple out and have systemic effects, or require systemic reengineering.

    But this is why advice usually isn’t welcome, because an advice giver doesn’t know the details of what they’re advising on. Unless they begin by learning the problem inside and out, obviously, but that takes a ton of time and effort.

    Finally, speaking as someone who knows programming very well, the gulf between “why don’t you just do X” and the actual work required to do X, if X is even feasible and possible, is enormous. Furthermore, everything comes with tradeoffs, and someone suggesting X is unlikely to understand the tradeoffs, or the tradeoffs that have already been made, and how X might affect those.

    All this said, yes sometimes suggestions are ignored or rejected because of ego. This is doubly true when someone is part of an institution, government for example, and wants to defend their turf or they don’t wanna spend “political capital” on something outside their personal agenda. This is also true of open source software; if you really wanna see some gnarly shit, try and figure out why LibAV split off from ffmpeg.

    If you want the real answer to the question, it is possible to be in charge. The danger with being in charge is that you become accountable for the things you’ve overlooked. You have to be able to survive your mistakes, then figure out how to avoid them in the future. Being in charge is incredibly taxing, but this is a choice we’re all condemned to make: accept things more or less as they are, or put yourself on the line.