agoose77

9 ms·

31.
by agoose77 3y ago
I suspect that their LFS bandwidth has been exhausted, so now one only sees the LFS metadata.
32.
by agoose77 3y ago
I think this is somewhat restrictive. A significant amount of scientific computing work is done in high-level languages like Python that dispatch to lower-level kernels (like C++, jitted Numba, etc). I wouldn't discount them as "n
33.
by agoose77 3y ago
It shows an overlap, but it doesn't indicate how one should parse it.
34.
by agoose77 3y ago
Yes, before CLING there was CINT!
35.
by agoose77 3y ago
We can argue about what "nearly" and "some" mean in this context ;)
36.
by agoose77 3y ago
Judith Curry is interesting case-in-point; most science should not be conducted in such a manner as to associate it with a particular science; the cult of personality should by virtue of a preference for rationale treatment of evidence be a
37.
by agoose77 3y ago
Huh? These experts absolutely exist - look e.g. at iSAGE during the pandemic in the UK. They combined epidemiology with sociology to provide a more comprehensive view of the situation. It's just that you ultimately have to listen to pe
38.
by agoose77 3y ago
My pet peeve is people interpreting "that's a stupid argument, here's why" as a conspiracy theory.
39.
by agoose77 3y ago
This is a very hyperbolic statement, and makes the common mistake of framing science as built around a central corpus of knowledge, defended by personalities. Whilst science is a human endeavour, and individual scientists suffer all the sam
40.
by agoose77 3y ago
You should check out https://github.com/deathbeds/jupyterlab-deck
41.
by agoose77 3y ago
Also, https://jupyterbook.org/en/stable/ (nb) I collaborate on Jupyter Book
42.
by agoose77 3y ago
Hehe, by you, you mean? ;) This is a really fun way to combine something that is being discussed to death with something that is visually beautiful. I'm fairly disinterested in most ChatGPT usages at the moment, but this is a really go
43.
by agoose77 3y ago
Check out MyST or Quarto for new efforts in this direction!
44.
by agoose77 4y ago
Depends what "cheap" is. You can get MHZ19 < £20-30 last I checked.
45.
by agoose77 4y ago
Has anyone looked at the sketchy figures? Talk about jumping to conclusions ...
46.
by agoose77 4y ago
I can only picture how one of those three terms would work
47.
by agoose77 4y ago
The important question is how much of a sample was lost. Ultimately, AIUI, the health risk depends upon the exposure in a ~ linear fashion. At least, in radiological protection that's used. The LNT model errs on the side of caution; as
48.
by agoose77 4y ago
This is certainly a motivation, but there are other reasons to avoid loops. In many domains, array-at-a-time abstractions offer a higher-level view of the problem. That's why something like `xtensor` exists for a NumPy-like API in C++.
49.
by agoose77 4y ago
Python actually has support for this with `pip-run`. There's a PR to get similar behaviour into `pipx`, which is effectively the `npx` of Python world. I think the single-script-with-dependencies workflow has been somewhat ignored unti
50.
by agoose77 4y ago
This is really cool! Forth is a language that I hadn't really spent much time thinking of until I stumbled across its use in Awkward Array (disclaimer, I am an Awkward contributor now!) We use a C++ Forth VM to run high performance (de
51.
by agoose77 4y ago
IIRC there was a presentation from Karpathy talking about the challenges with sensor fusion, particular in resolving divergence between e.g. the vision and the radar stack: https://www.youtube.com/watch?v=NSDTZQdo6H8&t=1
52.
by agoose77 4y ago
Hah, answering a question with a question ;) A few pain points for me: - No Markdown support. Yes, there's rich input, but I find it more natural to write prose in a markup language rather than with a rich editor. - Single "kernel
53.
by agoose77 4y ago
Could you elaborate on > The front-end is just lightyears in front of Jupyter notebooks in just about every conceivable way. I had the opposite experience, but it's a very broad statement and I suspect we took different interpretati
54.
by agoose77 4y ago
aldanor also mentions another use case which I only really allude to despite it being an important part of the process: exploratory work. Having a live kernel that maintains kernel state with the benefits of rich outputs is a mainstay of re
55.
by agoose77 4y ago
Eh, I think this misses the point of why Jupyter Notebooks are useful, and who is using them. I agree that in terms of literate programming as Knuth defined it, Notebooks are not great. There are tools to improve that story; I wrote http
56.
by agoose77 4y ago
Eh, but Python isn't designed to be fast. In fact, you might say Python is designed to be slow (it's not, but that's a fun way to look at it) because it does so much . If you don't want to have the kind of runtime flexi
57.
by agoose77 4y ago
https://cps.org.uk/media/post/2022/online-safety-bill-still-...
58.
by agoose77 4y ago
I hear this kind of discussion mentioned, particularly in the lead-up to the Brexit referendum. I don't think it accurately reflects the status of contemporary discourse. The UK media is more right-leaning than right: - https:/&#
59.
by agoose77 4y ago
Not entirely. Democracy isn't a fixed concept, and there are different interpretations of it. Direct democracy (via referenda) is fairly rare, and has clear flaws (look at the EU referendum!) Equally, the Good Friday Agreement signed i
60.
by agoose77 4y ago
There have been studies on opinion to immigration: The BSA31 performed in 2013 shows a strong age correlation with sentiments towards immigration: https://www.bsa.natcen.ac.uk/media/38190/bsa31_immigration.p... Th

More ›