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.

  • self@awful.systemsEnglish
    0·
    7 months ago

    404 media revisited the worthless DeepMind materials science dataset, featuring some world-class marketing gymnastics:

    Google DeepMind told me in a statement, “We stand by all claims made in Google DeepMind’s GNoME paper.”

    “Our GNoME research represents orders of magnitude more candidate materials than were previously known to science, and hundreds of the materials we’ve predicted have already been independently synthesized by scientists around the world,” it added.

    […]

    Google said that some of the criticisms in the Chemical Materials analysis, like the fact that many of the new materials have already known structures but use different elements, were done by DeepMind by design.

    hundreds of the materials have already been independently synthesized you say?

    ”We spent quite a lot of time on this going through a very small subset of the things that they propose and we realize not only was there no functionality, but most of them might be credible, but they’re not very novel because they’re simple derivatives of things that are already known.”

    this just in, DeepMind’s output is worthless by design. but about that credibility point…

    “In the DeepMind paper there are many examples of predicted materials that are clearly nonsensical. Not only to subject experts, but most high school students could say that compounds like H2O11 (which is a Deepmind prediction) do not look right,” Palgrave told me.

    by far the most depressing part of this article is that all of the scientists involved go to some lengths to defend this bullshit — every criticism is hedged with a “but we don’t hate AI and Google’s technology is still probably revolutionary, we swear!” and I don’t know if that’s due to AI companies attributing the successes of machine learning in research to unrelated LLM and generative AI tech (a form of reputation laundering they do constantly) or because the scientists in question are afraid of getting their lab’s cloud compute credits yanked if they’re too critical. knowing the banality of the technofascist evil in play at Google, it’s probably both.

    • skillissuer@discuss.tchncs.deEnglish
      0·
      7 months ago

      While the above analysis may seem to be critical, we do believe that many of our points could be adopted in the next version of this work. More scrutiny of the “new” materials needs to be performed prior to putting them into a database and claiming “…an order-of-magnitude expansion in stable materials known to humanity”. In fact, we have yet to find any strikingly novel compounds in the GNoME and Stable Structure listings, although we anticipate that there must be some among the 384,870 compositions.

      This brings us to our final point concerning the claim of “an order-of-magnitude expansion in stable materials known to humanity”. We would respectfully suggest that the work by Merchant et al. (1) does not report any new materials but reports a list of proposed compounds. In our view, a compound can be called a material when it exhibits some functionality and, therefore, has potential utility. Since no functionality has been demonstrated for the 384,870 compositions in the Stable Structure database, they cannot yet be regarded as materials.

      then they proceed to explain how badly have they fucked up in the only one example where they tried to find some utility of “new” “material”

      The few examples of functionality mentioned in the article are associated with Li±ion conductors. While the proposed materials are encouraging, their compositions leave much to be desired since they incorporate chemically soft anions. These anions are usually associated with narrow electrochemical stability windows, which renders materials that incorporate them somewhat pointless as Li+ solid electrolytes. (29)

      this is basically closest you can get to “you fucked up, do better” in a published article. saying “you fucked up, actually don’t even try to do better, go home” is not what i’ve seen ever really in published piece, excluding obvious cases of cooked data, even if it’s warranted this time. it’s in conclusions section https://pubs.acs.org/doi/10.1021/acs.chemmater.4c00643

      • self@awful.systemsEnglish
        0·
        7 months ago

        this is a lot more damning than the 404 media article let on, and I’m very happy that’s the case

        • skillissuer@discuss.tchncs.deEnglish
          0·
          7 months ago

          it is really a damning with a slight hint of praise

          We also note that, while many of the new compositions are trivial adaptations of known materials, the computational approach delivers credible overall compositions, which gives us confidence that the underlying approach is sound

          In closing, we hope the comments presented here will usefully serve the large community of materials scientists and engineers in their continued quest to develop the next generation of useful materials. While we are confident that the tools of Artificial Intelligence and Machine Learning have a bright future in the field of materials discovery, more work needs to be done before that promise is fulfilled.

          the other paper cited is that preprint from el reg article from some two months ago