Stubsack: weekly thread for sneers not worth an entire post, week ending Sunday 12 May 2024
Stubsack: weekly thread for sneers not worth an entire post, week ending Sunday 12 May 2024
Have a sneer percolating in your system but not enough time/energy to make a whole post about it? Go forth and be mid!
Any awful.systems sub may be subsneered in this subthread, techtakes or no.
If your sneer seems higher quality than you thought, feel free to cut’n’paste it into its own post, there’s no quota for posting and the bar really isn’t that high
The post Xitter web has spawned soo many “esoteric” right wing freaks, but there’s no appropriate sneer-space for them. I’m talking redscare-ish, reality challenged “culture critics” who write about everything but understand nothing. I’m talking about reply-guys who make the same 6 tweets about the same 3 subjects. They’re inescapable at this point, yet I don’t see them mocked (as much as they should be)
Like, there was one dude a while back who insisted that women couldn’t be surgeons because they didn’t believe in the moon or in stars? I think each and every one of these guys is uniquely fucked up and if I can’t escape them, I would love to sneer at them.
You're viewing a single thread.
I haven't read it yet (and I might not even bother) but Google's probably lying about their ML models again
"I can predict the structure and interactions of all of life’s molecules"
it does seem mildly useful https://www.science.org/content/blog-post/screening-against-alphafold-models-maybe-better-it-looks
note that google's thing is pattern matching, and if there's few similar protein structures in literature, then alphafold is working on poor quality data. comments do notice a red flag or two
Can’t see the comments on there unfortunately, but interesting start to that piece. Will read it in parts today as I get time
if you're using firefox, disable enhanced tracking protection to see disqus comments
Ah my solution to this is different: if the things that I use to protect my shit are blocking some $x, ah well that can just get fucked then
Too exhausted by all the bullshit to try debug someone’s fuckery and specialcase them. Only exception is if it’s something I need (and then I’ll do so very grumpily, and quite possibly loudly)
got you
i mean specifically this one
I don't believe this says anything useful about Alphafold or Cryo structures because I just don't believe these hit-rates. Anyone who reports 50% success from virtual screening has something very wrong with their assay. Before doing anything they should stick a pin in the catalogue 50 times, test those compounds, and when 25 of them come back active they should stop and work out why.
The rest of criticisms is in article
hahaha that's an amazing comment, and yeah agreed/understood
also note that the big failures come from faulty biology, not from faulty chemistry. You can set innumerable piles of cash on fire if you choose to plug a wrong enzyme
elaborating on that. when you can hit a protein cleanly (without some side effects (called off-target, there are also on-target side effects)) but you hit wrong protein, this doesn't have to be a complete loss, but it's likely. maybe you learn something, either about biology of that all or about something else that it interacts with, or maybe you have found a cure to some other disease instead. but this is the actual hard problem of drug development, especially when the biology of it all is little known. there's a few of these drugs that started this way, fentanyl, viagra, thalidomide (as anticancer), minoxidil come to mind, and at worst you have a new tool for biologists. alphafold won't help with that
also note it's from january so not about the new thing