unlikelymordant

6 ms·

1.
by unlikelymordant 8mo ago
In your experience, can you take the tech debt riddled code, and ask claude to come up with an entirely new version that fixes the tech debt/design issues you've identified? Presumably there's a set of tests that you'd k
2.
by unlikelymordant 10mo ago
Would you mind expanding on what you do for anticorruption? It has been something ive been thinking about and wanting to get into lately. It seems like complete poison to democracy, and more should be done to bring it to light wherever it o
3.
by unlikelymordant 2y ago
generally there is a "temperature" parameter that can be used to add some randomness or variety to the LLMs outputs by changing the likelihood of the next word being selected. This means you could just keep regenerating the same r
4.
by unlikelymordant 2y ago
I have been using search for engines for 30 years, my queries are not vague, i put as many keywords and "inurl"s and whatnot in as i can manage. I dont use kagi blocklists. Google results for my specific queries are garbage. I am
5.
by unlikelymordant 2y ago
I haven't found quoting helps much. I also feel like i shouldnt have to craft search queries with a lot of inurl or other tags or quoting. Kagi just seems to work better. Its worth 10$ a month to me to not have to worry about it, I use
6.
by unlikelymordant 2y ago
I have used kagi for quite a while now, and i use it pretty much exclusively. I was unhappy with google ignoring many terms in my search queries and giving me results that I generally considered to be 'intro' pages and generic con
7.
by unlikelymordant 2y ago
are you at all confident that this isn't hallucinated? I'd never trust an answer like this from an LLM
8.
by unlikelymordant 2y ago
You need to break down the lignin without damaging the cellulose. Boiling and mashing doesnt do this very well, you still get bundles of fibres sticking together, you really need them all to separate. The bonds holding lignins together are
9.
by unlikelymordant 2y ago
Of course there has been 'climate change' and 'global warming' before, caused by things like large volcanic eruptions (deccan traps) or meteorites. But they have pretty universally coincided with mass extinctions. And wh
10.
by unlikelymordant 2y ago
Australian houses aren't bought with american dollars though. If this were the case wouldnt we see a big correlation between house prices and USD exchange rates? Why wouldn't the correlation be with other currencies such as china.
11.
by unlikelymordant 3y ago
What is the coconut coir for? How did you stick everything together? Does it all go in a bag of some sort? How does price compare to retail mattresses? Hoe comfortable is it?
12.
by unlikelymordant 3y ago
I haven't tried on ios, but brave browser has an inbuilt ad blocker. Might be worth a try. I am quite happy with brave and kagi on android.
13.
by unlikelymordant 3y ago
Disorganised theft, damage, mislaid, lost
14.
by unlikelymordant 3y ago
If a 'superintelligence' achieves the same outcomes as humans without engaging in the same class of intellectual tasks that humans do, wouldnt it still be a superitelligence? Deep blue was beating everybody at chess without engagi
15.
by unlikelymordant 3y ago
The title "you wont survive as human capital" sounds interesting, but beyond the first paragraph, i cant quite connect the content to the title, it seems to mostly talk about the size of political groups. I feel like it needs a co
16.
by unlikelymordant 3y ago
Humans are currently on track for around 4 degrees of warming by 2100 assuming we hit all our current emission reduction targets (i.e. reaching net zero when governments have said they would). So removals are quite necessary.
17.
by unlikelymordant 3y ago
This whole 'undesireable side reactions' thing is completely overblown for basalts. The effects to soil of volcanic ash from eruptions is quite well studied, not to mention crushed basalt is a recognised fertiliser in organic and
18.
by unlikelymordant 3y ago
The fourier transform converts between time domain and frequency domain. The fourier transform is linear, which means it can be implemented as multiplying your time domain signal by a matrix of complex fourier coefficients. Usually we dont
19.
by unlikelymordant 3y ago
The fourier domain is just a linear transform of the time domain, i.e. just a matrix multiply away!
20.
by unlikelymordant 3y ago
Youd expect much bigger increases in atmospheric methane if that were the case i think: https://gml.noaa.gov/ccgg/trends_ch4/
21.
by unlikelymordant 3y ago
Etymologically there are 3 and 4 way couplings... latin copula just means 'joined', nothing about 2.
22.
by unlikelymordant 3y ago
Im australian, and a "couple of things" is some small number, like 2-4. Using it for exactly 2 seems like people are confusing it with 2 things that are coupled. Why not just say 2 if thats what you mean.
23.
by unlikelymordant 3y ago
That is a completely different usage of the word as far as I'm concerned, like confusing nut ( and bolt) with nut (from a tree), in this case 'a couple' like two people in a relationship implies a link, like they are coupled.
24.
by unlikelymordant 3y ago
you mention climate being a waste, but one of the only extinction events bigger than KT was the permian triassic, and that was caused by high atmospheric CO2. We're already on track for a pretty bad time (if all current government clim
25.
by unlikelymordant 3y ago
This is in the voluntary carbon markets, quite separate to regulated carbon markets. It is unfortunate that so many may be junk, but in the end 380 billion T of carbon needs to be removed from the atmosphere between 2050 and 2100 to keep ea
26.
by unlikelymordant 3y ago
If the sea ice melts and the chicks drown before they develop waterproof feathers, then the melting ice could easily be the culprit, and still captive breeding would work.
27.
by unlikelymordant 3y ago
Sticking aerials in the soil seems so unlikely to work. Even theoretically, an aerial would mostly be picking up human transmissions, resulting in vanishingly small voltages/currents in the soil. If there is indeed an effect, why not j
28.
by unlikelymordant 3y ago
The next trend line to look at is that without pulling co2 out of atmosphere, just quitting fossil fuels still gets us to 3.2 degrees of warming (this is IPCC modelling). This is climate armageddon. So the growth in renewable energy means n
29.
by unlikelymordant 3y ago
A lot of carbon needs to be removed from the atmosphere, something needs to do it at scale. The IPCC says we’ll need global cumulative net-negative emissions of 380 GtCO2 from 2050 to 2100 to return to 1.5°C after a likely overshoot. Focusi
30.
by unlikelymordant 3y ago
The increase in demand drives up the price of offsets, which increases supply and investment in better carbon capture techniques and scaling. Im lumping in things like direct air capture, bioenergy carbon capture and storage, and enhanced w

More ›