jcattle

6 ms·

1.
by jcattle 14d ago
R really is/was the superior traditional data science language. Python ecosystem is slowly catching up though. ggplot vs matplotlib dplyr vs pandas And I loved that everything in RStudio was so easily inspectable. Have a huge dataframe
2.
by jcattle 22d ago
Just to put Bill Gates thoughts on AI into perspective, this was his comment on an AMA 9 years ago to the question "What kind of technological advancement do you wish to see in your lifetime?": > The big milestone is when compu
3.
by jcattle 1mo ago
You are not stopping people from taking in calories. In your analogy it would be the same as saying: you are not allowed to heat your home with a coal burner. Same as saying you are not allowed to eat beef. You can still eat lots of other s
4.
by jcattle 1mo ago
How do you make beef and milk more sustainable? I'm honestly wondering. To me it seems similar to saying: we can make coal power plants more efficient. Equivalent to more "efficient" (however you want to define that) forms of
5.
by jcattle 1mo ago
Or Buschenschank in Styria (where you traditionally don't get warm food, only "Brettljause")
6.
by jcattle 1mo ago
I thought you'd mention plant poaching. Which is just evil.
7.
by jcattle 2mo ago
Ah, I thought that was the main vegetation present beforehand? What kind of forest was it before it was cut down?
8.
by jcattle 2mo ago
None of what you are writing is applicable to Landes. Outside of scandinavia Landes region is some of the most intensely managed forest in Europe. From planting to cutting is only around 30-50 years. With multiple thinnings in between. Also
9.
by jcattle 2mo ago
With freely available satellite imagery you're looking at a 10x10m resolution per pixel. So basically entire tree stands have to die or crowns have to lose their leafs for it to be detectable form space. Main insect of concern are bark
10.
by jcattle 2mo ago
Yes, I'm well aware that Landes naturally should be an extensive sandy marsh. With how natural forest disturbances are more and more starting to affect the plantation, it proves to be seen what the response will be. The thing is that i
11.
by jcattle 2mo ago
No, at the moment classifying disturbances is only done retrospective. Predicting fire likelihood is its own large area of research. An open case is more in terms of temperate forests which are increasingly starting to burn, for example in
12.
by jcattle 2mo ago
Just an aside: I'm a researcher trying to classify forest disturbance types from satellite imagery. The Landes Forest which is currently burning is really interesting in that you have all 4 of the major European disturbance types there
13.
by jcattle 2mo ago
In this case no, all the water is drained. I don't think meaningful amounts are retained for agriculture.
14.
by jcattle 2mo ago
What I can not wrap my head around: How are these models trained? What training mechanism or model architecture provides the glue to go from human text to images? Don't you need to have millions of really descriptively labelled images?
15.
by jcattle 2mo ago
Ah I thought you would link https://xkcd.com/605/
16.
by jcattle 2mo ago
I was wondering why I don't get that animation. Turns out I turned on reduced motion in Windows 11, to get snappier UI and they actually set some accessibility tag correctly for the animation!
17.
by jcattle 2mo ago
Same as for banks. Downloading some verification app with a confidence inspiring 1.2 rating on the app store, getting on a call with some random gig worker looking like they are taking the call in their living room and wiggling your ID arou
18.
by jcattle 2mo ago
The thing is that LLMs can only ever go forward. It might go into one direction, once that direction is in the context it will realize that it doesn't work, but can't delete. So it has to backtrack (or double down, see the whole s
19.
by jcattle 3mo ago
Something something bit flips, something something stateful
20.
by jcattle 3mo ago
Yep. I don't see 500 million being even close to enough to develop broad-spectrum preventatives that would be easy enough and safe enough to administer, to get over 67% of the population to take them.
21.
by jcattle 3mo ago
Yes all good points showing issues that academia has at the moment. However I often see this going from "there's issues" to discounting academia altogether and positioning private labs as a good or only alternative. After all
22.
by jcattle 3mo ago
There's this crowd on HN which is very vocal against academia. From what I've seen, the main points are that academia isn't efficient, most of the science coming out of academia is useless and that the whole system is just a
23.
by jcattle 3mo ago
It's tech from the 80s. Look up the Nishika N8000 and Nimslo 3D. Basically it's multiple lenses next to each other, each capturing a small slice on the 35mm film. Every lens has it's own shutter, which is triggered at exactly
24.
by jcattle 3mo ago
there's also https://www.myvocab.info/en From what I can tell they actually have a bit more robust science behind their algorithm (and a lot less questions to answer)
25.
by jcattle 3mo ago
Then I have a backup yubikey at home for services which allow to register two keys. For other's there's still good old password+some second factor.
26.
by jcattle 3mo ago
Pretty happy with having a yubikey on my keychain. Log in someplace new? plonk in your yubikey and off you go!
27.
by jcattle 3mo ago
Yes, you could do a lot more with the data. The only other thing I did was a wordle style guessing game: https://guess.landshade.com But the data is open source, so if you want to dig around my badly documented code and raw sate
28.
by jcattle 3mo ago
That's pretty cool. In a similar but very different vein: A few years ago I took twenty years of daily satellite imagery and computed the mean color for countries and the world https://www.landshade.com/ But in doing
29.
by jcattle 3mo ago
A coding agent should make short work of that. However I'm a bit doubtful if the results would actually be meaningful. I'm thinking that some of the LLM-isms are a bit more complex than just repeated phrases. It's often more
30.
by jcattle 3mo ago
This blurb gave me the idea to try and quantize this. Scrape the top HN blogs over the last few years and see how occurences of common phrases change. I'd expect to see a huge increase in "solving real problems" over the last

More ›