evouga

11 ms·

61.
by evouga 4y ago
C sits at a sweet spot of performance vs. expressiveness vs. portability whose Pareto-optimality has withstood a half-century test of time. It’s not going anywhere. The case for replacing C++ is a lot stronger but I don’t see game engines,
62.
by evouga 4y ago
I mentioned DALL-E and Copilot in my post, so I'm not sure why you're linking me to a article summarizing recent high-profile research in large language models...
63.
by evouga 4y ago
At this point, I'm numb from all of the AI overhype. I was extremely excited about DALL-E and convinced myself that concrete fruits of the AI revolution were finally here... until a few seconds after I got the chance to try some querie
64.
by evouga 4y ago
What do you mean? Joel puts easter eggs in his posts?
65.
by evouga 4y ago
I don't see the connection between RMS and Bill Gates? In any case, the backlash against RMS was due to his defense of Minsky, characterization of Epstein's crimes using extremely tone-deaf language (referring to his victims as &q
66.
by evouga 4y ago
Currently the method only works for elliptic, linear PDEs like the heat equation. Keenan and his students are trying to extend the method to more general PDEs (they recently had a paper showing how to do elliptic PDEs with non-constant coef
67.
by evouga 4y ago
The paper is in LaTeX and uses the required ACMTOG template (also used for SIGGRAPH papers); you can get it at https://www.siggraph.org/preparing-your-content/author-instr... The inset figures are probably placed using
68.
by evouga 4y ago
It’s interesting to see the plots of true change in opinions. There has been massive change in favor of gay rights, gender equality, marijuana legalization, and interracial marriage. On pretty much every other topic, change is negligible.
69.
by evouga 4y ago
Agreed. Symbolic AI dominated the field for 50 years: thanks partially to an accident of history (the highly influential work of Minsky, Newell, etc) and partially due to lack of the data needed to try anything else. Now we’ve seen essentia
70.
by evouga 4y ago
The difference is that chess rules are extrinsic to the players at the board, whereas art is intended to communicate (emotions, narrative, ideas) to humans. A neural net that can communicate with intent to human minds as well as a biologica
71.
by evouga 4y ago
Please no. Good science is often boring; this system prioritizes hype and marketing over everything else.
72.
by evouga 4y ago
Yep. I’m slowly canceling all of my streaming services. If I can buy an entire show from Amazon, I’ll do that. Otherwise, back to piracy.
73.
by evouga 4y ago
Huh? Ron Jeffries is a champion of TDD (see for instance https://ronjeffries.com/articles/019-01ff/tdd-one-word/ ). He most certainly wasn't deliberately implementing Sudoku in a stupid way to make TDD
74.
by evouga 4y ago
This seems to be the case with a lot of "methodologies" like TDD, Agile, XP, etc. as well as "XXX considered harmful"-style proscriptions. A simple idea ("hey, I was facing a tricky problem and this new way of appro
75.
by evouga 4y ago
I completely agree. I'll use TDD when implementing a function "where the code is mostly self-contained with few external dependencies and the expected inputs and outputs are well defined and known ahead of time" and where t
76.
by evouga 4y ago
This is a strange article because the strongest counterpoint to its own argument: > The AI can make art. I’ve seen people dispute this, since what the AI makes by definition may not be art, under at least certain definitions. I think tho
77.
by evouga 4y ago
I'm pessimistic. I see DALL-E and other generative algorithms as akin to the camera (or video camera): they are powerful new tools for creating art, and their full impact on the art world remains hard to foresee. But what is clear is t
78.
by evouga 4y ago
"Medium" and "AI" are spot-on IMO, but the "UI" track seems suspect to me. "Click" and "Tap" are essentially the same thing (on a desktop vs. on a mobile device): the user actively selecting
79.
by evouga 4y ago
Under. Way under. Google Glass's spectacular failure had a chilling effect on the entire industry, but it failed due to (1) lack of killer apps, (2) poor aesthetics, and (3) consumer concern about privacy. But ML has revolutionized ima
80.
by evouga 4y ago
Sure, but plenty of people still browse Reddit or Instagram (or HackerNews), listen to podcasts, etc. There is a middle ground between "AR will never take off as an entertainment medium" and "AR will kill video/images&#x
81.
by evouga 4y ago
I mean, adding to a doubly linked list is just four lines of code and deleting is just two, right? A bit more if it’s possible for the list to become completely empty.
82.
by evouga 4y ago
The crystallographically-determined structures are the ground truth ! [1] Saying that AlphaFold makes x-ray crystallography useless is like saying DALL-E makes photography useless or Copilot makes GitHub useless. You've got the depend
83.
by evouga 4y ago
My observation is that breathless hype pieces proclaiming that a new technology will imminently revolutionize area X outnumber the articles expressing common-sense skepticism about the technology, by two to three orders of magnitude.
84.
by evouga 4y ago
Performance degrades when you get many hash collisions, which will happen if you have a lot of points clumped together in space. Sparsity is fine. For points you expect to be unevenly distributed the more advanced spatial partitioning data
85.
by evouga 4y ago
The Delaunay tree link in footnote 14 doesn’t work. But in any case you can build and incrementally update an AABB tree of the triangles to efficiently find the triangle containing query points, right? Though you need to rebuild the AABB tr
86.
by evouga 4y ago
The focus on best paper awards is odd as the major conferences of some CS subfields dole out awards as if they were party favors, and those of other subfields don't have best paper awards at all. For instance SIGCHI 2021 had 28 best pa
87.
by evouga 4y ago
I mean. Stratego is a great game; I had a lot of fun playing it at summer camps when I was a young boy. It's cool there's a good AI for it. But this result feels a bit anticlimactic in a world where AIs can already beat expert hum
88.
by evouga 4y ago
Auto merge using a “semantic diff” sounds line a complete nightmare to me. How many people actually check that git has merged code correctly? When there are no conflicts and the merge passes CI? Exactly. Now imagine the git automatically fi
89.
by evouga 4y ago
But almost nobody actually uses git the way it was originally intended, eg. as decentralized version control? Instead there’s a canonical master repository (on GitHub) everyone pulls from/pushes to.
90.
by evouga 4y ago
> But I suppose convenience always trumps openness and freedom. It's especially sad because the whole point behind FOSS was this. Is the whole FOSS idea getting old? It seems so---I read the post above saying essentially "I don

More ›