ArjinFerman
Tinfoil Gigachad
No bio...
User ID: 626
ArjinFerman
Tinfoil Gigachad
1d ago
One big thing I see missing from all these "the solution is simple: force women to have children by coercion!" answers is - are men ready to be fathers?
I mean, isn't that a solved problem? You coerce them too.
ArjinFerman
Tinfoil Gigachad
1d ago
Been in a bit of a pre-Christmas crunch, I started working on migrating the data from v1, but didn't get very far.
How have you been doing @Southkraut?
ArjinFerman
Tinfoil Gigachad
1d ago
I used to see the "You like politics? Here have some John Oliver" algo in all it's glory, nowadays I mostly get slop that's neither here nor there. Then again I mostly watch it through FreeTube (effectively watching everything via an anonymous tab). What you're getting sounds looks more like profiling than outright boosting, though you probably were on the edges of the cluster they put you in originally.
Lately my grandpa was directed to this gem.
Man... what do I have to do to get this recommended to me?
ArjinFerman
Tinfoil Gigachad
1d ago
How do you figure? There's been leaks of the Google algorithm where it came out they explicitly derank small independant websites. There was also an analysis of YouTube's recommendations that shows they're trying to pull people to the mainstream. I wouldn't be surprised if Elon was boosting his favorite causes, but it's definitely not what I'd call "boosting diasident voices".
ArjinFerman
Tinfoil Gigachad
2d ago
I'm converting old SQR code into SQL now, and pre-LLM this would have required me to have at least a basic knowledge of SQR, but that's no longer really the case.
Using it in that direction is fine becauae you can check the output, I'm not sure it's going to work so well in a "I'm used to language X and they're making me write in language Y" scenario.
ArjinFerman
Tinfoil Gigachad
2d ago
I don't know if the split is going to be 90/10 when you're messing around with the kernel. Also, when you're rewriting old code from scratch, the risk of introducing new bugs is pretty high. When you want to replace something that's been in production for years, if not decades, you'll need a better argument than "it's perfectly safe 90% of the time".
There lies the rub, though: the way Rust is being introduced defeats the best (and possibly only) argument for it. If they wanted to move fast and break thing, they can just stick to C. Hell in some of these cases the C code is even the thorough option as it's already been in use for many years, so it's well tested. Rewriting it in a completely new language, marketed entirely on memory safety, only to disable the safety features throughout the codebase is supposed to achieve what, exactly?
ArjinFerman
Tinfoil Gigachad
2d ago
I swear I saw an actual study on the subject a few years, but I can't recall any identifying details that would help me find it again.
ArjinFerman
Tinfoil Gigachad
3d ago
As far as I can tell from the literature I have available to me, nuclear has an EROI ranging from 5-15 when you ignore the studies that give a wildly inflated amount by leaving out key steps of the process
Can you link one of these papers that you're reading? I think it would make a much better argument then pivoting to financials.
ArjinFerman
Tinfoil Gigachad
4d ago
You're right, I don't actually go and prove that it is a good proxy. But the reason why I believe that is simple - that's the only real way we have to determine if a given power source can function viably in a modern western society, and the points where that connection breaks are fairly easy to identify and take precautions around. Actually measuring KWh is definitely worthwhile and there are analyses you can make with regards to it, but it gets exceedingly complicated in a way that finances avoid.
Well, while it's true that financial profit is the only way to have a power source function in a modern society, but the reasons are completely different than "Energy Returned on Energy Invested" and are subject to rather dramatic changes. Something as simple as an increase of energy prices can completely change the result, and that's without going into all the way by which costs of nuclear power have been artificially increased. By contrast a negative EROEI means you're up against the laws of thermodynamics and there's literally nothing you can do to make that power source work. Yes, it's complicated to calculate it in KWh, but it's really the only way to properly make the argument. A financial analysis does not avoid any of the pitfalls of actually measuring energy requirements, and it introduces many inaccuracies of it's own.
How reliably was that power generated? Was it limited to specific times (power that can only be generated in off-peak hours and can't be stored isn't as useful)? For the purposes of determining whether or not a given power source can feasibly supply power to a first world economy, finances are one of the best tools we have.
I don't think financies measure any of these things at all, which is why I think you should stop claiming it's about EREOI (or actually show the negative energy returns). Finances measure opportunity costs, and not even relative to other energy sources, but relative to all other things in the economy, so it's not really a good tool for this sort of analysis at all.
Mining, transporting and enriching uranium tends to consume enough energy that the return isn't terribly worthwhile.
Transport costs next to nothing in terms of energy. If you want to make the argument that it's because of mining or refining, be my guest, by provide numbers in terms of actual energy.
ArjinFerman
Tinfoil Gigachad
4d ago
He said October 2025, that's not included on your chart. How is covid supposed to affect that anyway?
ArjinFerman
Tinfoil Gigachad
5d ago
The main focus of my arguments on this topic is simply Energy Returned on Energy Invested. Financial viability is a fairly good proxy for whether or not a given source of energy provides enough of a return to make its exploitation viable.
That's not what I'd call focusing on Energy Returned on Energy Invested, that's focusing on financials, claiming, but never proving, it's a good proxy, and calling it a day. Make the case in KWh, and we might get somewhere.
I too would love to have a Multiverse Viewer, but I don't think the stock market is a particularly good measure of economic performance.
More options
Context Copy link