fmac
Ask me about bike lanes
No bio...
User ID: 3415
fmac
Ask me about bike lanes
1yr ago
Why is that an improvement? This seems like a terrible way to allocate capital.
fmac
Ask me about bike lanes
1yr ago
different incentives than making the best product or most money
I think they absolutely have these incentives and are just out of touch idiots like most of the """elite""" class in the West at this point in history.
fmac
Ask me about bike lanes
1yr ago
The idea seems to be to make money
Why not buy units in private equity funds then? Much better returns than Intel lmao
Furthermore, if that is the goal, why do we think government bureaucrats and elected officials are better capital allocators than Wall St et. al? Are we pretending government employees are hyper competent now?
fmac
Ask me about bike lanes
1yr ago
"the Government taking a direct share in owning companies isn't a state-owned enterprise, it's simply market-flavored convergent evolution (??????)"
The government owning common shares in a company makes it (partially) state-owned, the state has a share of ownership. No amount of word games will change this underlying fact.
If it walks like a duck, and quacks like a duck, the duck owning shares in something has an amount of ownership proportional to the % of shares owned (moderated by shareholders agreements).
fmac
Ask me about bike lanes
1yr ago
why not get something in return?
The question is how much return. Do we trust the government to be a great asset manager? Why not just have governments dump money into PE funds if you're trying to IRR-maxx?
Do we think government agencies/employees are better capital allocators than the current cohort of capitalists and Wall St et. al?
fmac
Ask me about bike lanes
1yr ago
Stupid government scope creep is fine when my favorite team does it!
I think if you're pro Trump doing this you also need to consider you're implicitly pro Kamala doing this, do you think that sounds good?
Because in 4-8 years the blue tribe might start doing it too, and this is a silly road to go down.
Does this represent a leftist turn in the Republican Party's view on the state's role in the economy, leaning more towards a nationalist democratic socialism?
It's definitely a move away from New England Finance's control over the party
Are there risks of corruption arising in the Trump administration related to government acquisition of major shares in large companies?
Profoundly, this is a risk for every government in perpetuity
Does this represent an expansion of executive authority? What do we expect USG to do with its stakes in these companies?
Yes, I'm not sure but I have serious doubts as to their ability to make better choices than businesses now. If they could, planned economies would have a way better track record than they do.
Does this raise potential conflicts of interest, directly aligning the interests of the Federal government with large firms (rather than their merely influential status today)?
Yes.
I can kind of think of a Steelman here where government shareholdings in major national champions could enhance corporate governance and specifically help align business and government interests over the long term, but I give this roughly a 0% chance of happening with the current status of western governance institutional skill.
Feels like everyone in the political class finally woke up to China's industrial results and going "oh fuck maybe there was something to be said for this" and are now trying to cargo-cult-government to replicate.
fmac
Ask me about bike lanes
1yr ago
I mean I agree with you, and I'm not involved in the field of robotics.
I have seen interesting videos of robot berry pickers (Dyson has a cool experimental farm), I hope commercially viable ones are invented and scaled
I'm just commenting on what I've picked up as an interested observer of the space
The lack of massive robotic automation indicates to me it's quite hard, if it wasn't, we'd see more of it
I hope LLMs help smooth over the "fuzzy" parts that make this so hard
fmac
Ask me about bike lanes
1yr ago
I'm just vibing based on what I read and observe
There's a snappy "X's name law" about this. Basically getting robots to do rehearsed and consistent activities (a backflip) is now "easy", somewhat independent of the complexity of the action itself.
But having them able to handle the profound randomness of life at a 99.999999% accuracy level is really really really hard (see: self driving cars).
fmac
Ask me about bike lanes
1yr ago
I'm in much more agreement with "80% of work is done by 20% of employees" (although I think it's a larger share than 20%)
I disagreed strongly with "most office work is fake email jobs"
"Many employees don't work at full capacity" =/= "many employees have fake jobs"
The shorts would be against whatever companies you think are wasting large sums of money paying people to do nothing, as presumably they're very liable to be disrupted by companies with more competitive cost structures
fmac
Ask me about bike lanes
1yr ago
After reading more of your thoughts I'm actually much more in agreement with them than how I interpreted them initially
fmac
Ask me about bike lanes
1yr ago
we had an order from the higher ups that we must use LLMs, and that they will be tracking how often we use them
And absolute dipshittery like this is why 95% of LLM projects (whatever that actually means) fail, not because LLMs are stupid, but because the people using them are
fmac
Ask me about bike lanes
1yr ago
Training a model to operate robotics for conventionally "hard" problems
This is really hard and if current models could easily be trained to do it they already would have
fmac
Ask me about bike lanes
1yr ago
We really haven't seen a matching boom in productivity.
Or the current march of productivity increases throughout the 2000s is as a result of shit like cloud software tools, and in a counter factual world where tech stagnated, productivity would be much lower than our actual world?
fmac
Ask me about bike lanes
1yr ago
Most office work is fake email jobs.
This is a ridiculous statement. If you think so, you should post your short positions.
Or, you should start up competing white collar companies and undercut all the companies you think have massive amounts of fake email job deadweight. As you'd have a much leaner cost structure.
fmac
Ask me about bike lanes
1yr ago
It's not quite A Connecticut Yankee in King Arthur's Court but along those lines.
What was wrong with this book? In this context
I read it as an early teen (now far too long ago) and I remember loving the concept (also a massive Twain fan) but then finding the ending so bad it retroactively ruined the book for me.
fmac
Ask me about bike lanes
1yr ago
I have like 40 steam mods (happy to share list if anyone cares) so vic3 was an easy $50. I've also never paid for a Paradox game and put an ungodly amount of hours into Eu4/Vic2/Stellaris (less for this one) so it felt fair lol. Still never paying for DLC tho.
CreamAPI account risk is negligible. Huh.
I do worry about this, it seems like it would be easy for Steam to book and aligned with their incentives to do it (they profit from DLC being sold) but they don't seem to care and I love farming free DLC so it's a risk I'll take.
Plus my stream library is pretty small, I'd have made a burner if I paid for more games.
fmac
Ask me about bike lanes
1yr ago
Also, for anyone considering Vic 3
The game has improved significantly with 1.9, but still has some glaring issues. I think by 1.11 it'll be an un-ironically good game.
I recommend buying vanilla on sale, as access to Steam workshop is MANDATORY (seriously, mods make this game so much better). It's incredibly easy to CreamAPI the DLC for free once you own the base game.
fmac
Ask me about bike lanes
1yr ago
Legit, you see people on the forums suffering through issues that they could solve in 10 seconds with a debug_mode mod and "~"
Victoria 3 might actually be the worst for this too. I find myself constantly needing to tag over to other countries to fix whatever insane and dumb shit they (since 1.9, 90% of the time its France) get themselves into.
Also the WORST border gore, oh god the border gore.
fmac
Ask me about bike lanes
1yr ago
This is a really well thought out comment, thank you for writing it.
I think I agree with most of it. I still think the "mechanism of action" for a cat-call vs skimpy shorts (or whatever) is far enough apart that they don't compare well, but I'll concede they're on the same spectrum of human behaviour/motivations.
I'm gonna read this again later when I'm not in motion, thanks again.
I stood @ToaKraka up the other week on Victoria 3 mods.
While I don't have as many as him, this is what I've been playing with:
- Regional HQs are a shitty game mechanic. They absolutely hijack your investment pool, and they're a pain in the ass to remove (made better in 1.9.8) and impossible to remove from Vassals or if they have investments in a third party country. They're also poor game design as they are a "click this button every time if it's available" option, which isn't good game design (good game design is all about making interesting choices).
Thus, I changed AI weights so they never pick them, and I never pick them.
- late game performance is awful, because of pop fragmentation. I increased the defines for both assimilation and pop consolidation by a factor of 10x. I think a lot of the background stuff is hard-coded because it really didn't seem to do much. I have a late game save right now that's basically unplayable now (I also fucked up and conquered too much of the world, so multithread performance dropped), so I'm going to try 1000x and see if that helps, I don't think it will.
I did notice that my assimilation did increase dramatically for a while when I had a +25% assimilation buff, so I will test that out eventually (more on this later).
-
infamy calculation is fucking stupid as it's based off population and not GDP (african states are so expensive it's so stupid). It also scales incredibly poorly in late game. My attempts to rebalance thus far have gone poorly, I can get it to a good place in early game or a good place in late game, but there isn't a good unified set of defines for both I've found yet. I think I could probably add more infamy reduction to later game tech, but my willingness to spend my free time balancing a paradox game for free was low.
-
clicking "build power plant" a million times makes me want to kill myself. I modded power plants to have 4x output and 4x input. This makes them somewhat overpowered, I don't care, I now click 4x less.
-
I increased the amount devastation recovers per state. The war goal system is AWFUL which causes AIs to get locked into wars no one can end for a while, which means if one is occupied that whole time it gets inadvertently genocided (population loss has 0 effect on war exhaustion, lmao, and it wouldn't matter if it did, LMAO). Making devastation recover faster mildly helps with this. I think they just fixed this somewhat in 1.9.8, but frankly the AI needs all the buffs it can get in this game.
-
I changed the law enactment time from 180 days to 50 days. I think this may make the game easier? Although now movements/revolutions are still mad after the law they hate is passed, so I think it's okay. The law passing mechanism is fucking atrocious, and could easily waste YEARS of times bouncing around with "+10%"..."-10%"... etc. This makes it snappy. The challenge of passing a law is maneuvering your political structure into place so it's passable at all. Once you can pass it, it's just a matter of how many de-buffs you take along the way. I'd prefer to pay the cost and move on, not watch a random number generator slowly tick.
Side note, but the paradox communities' acceptance of "save scumming" as an acceptable part of the game, but console commands as "cheating" absolutely sends me. It's literally the same thing, except one is a horrible use of your time. I once (only once) spent 30-45 minutes micro-ing and reloading to pass a very low % law. It was probably the worst use of 30-45 minutes of my free time. I don't understand why people do this in a single player game.
Something I would like to do, is get better at event modding. I want to set up some AI only global events that would trigger in ~early mid game to kick the AI all off traditionalism/serfdom as those laws are crippling and AI sucks at government management.
I also am considering setting up a global event for like +50% assimilation as that seemed to be much more effective than my define tweaks.
I also would like to make a mod that gives the AI +10 goods production of every good in their capital. I think it'll help them bootstrap supply/demand cycles for new goods. I'm not sure how they're doing at electricity production, etc
Finally, I don't know if I can, but I really want to 1) make the AI better at upgrading units as it current SUCKS at doing this and 2) make the AI stop making 30 individual armies, as that provides 0 tactical benefits to the AI, and makes the game noticeably laggier.
EDIT: ONE MORE
migration in this game is in an AWFUL state right now. If you get a Chinese or Indian country in your market, they flood your country at absurd rates (although I do plan to do an India run where I take Canada/Australia and fill em up, for historical accuracy purposes lol).
And mass migrations are basically "human only" because you GDP/capita and SoL maxxxxx and then absorb like 80% of the mass migrations regardless of you're Brazil or Vietnam, which is so fucking stupid.
I need to figure out how to give the new world a massive buff to mass migration attraction.
fmac
Ask me about bike lanes
1yr ago
While I agree with a lot of what you say, I do wonder if maybe my active/passive definition didn't work.
Catcalling is a specific act, targeted/focused at a specific person. One that is in the overwhelming super-majority of cases is not desired. I would also posit that many a cat-caller does it not just because they think someone is hot, but because they enjoy the fact they get to flex "power" over someone by making them uncomfortable with no recourse against them (dovetails nicely with everyone's discussion about lower class men, they don't get to flex power often).
Having ones ass out is an unfocused act, it is not targeted at anyone. While it may make some uncomfortable, it does so at a much lower rate (and makes people happy at a much higher rate). There is no intent to cause distress.
Finally, while I agree that society is teaching and reinforcing women to be far more paranoid than is warranted, the Venn diagram between "is willing to break social norms by cat calling" and "is willing to go for a cheeky bottom pinch or other form of personal assault" has overlap, there is a small but credible possibility of violence from that person. The Venn diagram of "has ass out in Lululemon" and "will grab your dick through your shorts" is 0, unfortunately.
I'm unconvinced cat calling should be an indictable offense, but comparing it to skimpy clothing is ridiculous.
fmac
Ask me about bike lanes
1yr ago
Watching Trump's approval rating fall even more will affect my mental state by making me smile and laugh as Americans realize they fell for it again
fmac
Ask me about bike lanes
1yr ago
I do think there is highly significant asymmetry of discomfort between a woman being catcalled and a pious man seeing some legging-clad ass
This seems super culturally mediated, though--I'm not sure I'm in a good position to just tell a pious Muslim or devout Amish that his feelings about bikinis simply don't count the way that a modern woman's feelings about wolf whistles does.
I think it's more an active vs passive thing.
A cat-caller actively intrudes into the life of the random passerby. They do this intentionally by inserting (hah) themselves into the life of another.
The bikini clad ass may upset the Amish or Muslim man, but it doesn't force them to look. It's a passive object in their life they can choose to interact (hah) with or not.
I guess the counter is you have to first notice the bikini to then ignore it, but I again just have a very hard time not finding a someone deliberately taking action (making noise that is in 99% of cases unwanted and coded as threatening) to be anywhere near equivalent as someone getting annoyed as to what someone else is wearing.
Subsidies don't have to lose money if they have a positive multiplier.
I'm not entirely sure what treasury ownership you're referring to? Social security?
Because in that case, that's the government owning a Treasury issued by the government. The interest paid is real it's paid by the way everything else is, taxes or debt.
I fully agree social security is a shitshow nightmare from numerous perspectives. I think they should move to the "Canadian model" a la CPP, where investments are managed by an extremely competent and largely independent team.
I would like to reiterate that the executive branch borderline randomly scooping up equity stakes in flavor of the month companies is not this. Also, it's likely a transfer of welfare from taxpayers to equity holders AND it will definitely fuck heavily with equity price discovery, leading to a less efficient market overall.
More options
Context Copy link