Computer Science is no more about computers than astronomy is about telescopes. - Edsger W. Dijkstra
4,819 karma · joined May 23, 2010
https://www.aaup.org/sites/default/files/AAUP-2021-SurveyTab...
Baumol's cost disease theory makes sense in general. But empirically, for a specific service, there's always a lot of other stuff going on, are they providing a superior good with rising demand as society gets richer, how easy is it to substitute lower quality adjuncts, etc.
- be richer than God
- don't rescue a venerable, functional institution that is modestly, temporarily impaired, partly due to terrible risk management and partly due to a tsunami of cash flowing in and then starting to flow out. (SVB didn't have to buy Treasurys, could have lent overnight, deposited at Fed and taken interest on excess reserves. Instead they stretched for a relatively puny yield pickup and got badly burned.)
- don't calm the waters. instead, tell people to take money out
- short SVB stock and make a killing
- say it's Biden's fault and he has to do something and bail the startup ecosystem out
- https://tenor.com/view/the-big-lebowski-youre-not-wrong-just...
Bill Ackman didn't do all these things but as a group that's what people like him and Thiel and Musk 100% did. They want to be disruption chaos monkeys and then have the taxpayers bail them out of a mess they created. Remember than when they go bananas about Biden wanting them to pay similar tax rates to middle class folk.
One UX instead of accessing FRED, Polygon, FMP, AlphaVantage etc. with pandas datareader in a Jupyter notebook is pretty good. The OpenBB name is a bit of pushy marketing. It beats GameStonk Terminal which was (no joke) the original name.
But tbh, might be just a matter of time before they have to change it for legal reasons. But also, there will be some publicity in the forced name change.
Geopolitically France is an interesting place with one foot in Mediterranean civilization and one foot in Northern Europe, good natural barriers, and a fairly unified culture.
After the religious wars of the Reformation, there was a period of pretty good kings and increasingly strong state, prosperity importing tech from Muslim world via Italy, and liberalization. During this period, Italy, Germany, UK, low countries had their own problems and France was pretty dominant.
Before the French Revolution and the Industrial Revolution, France had 4x the population of Great Britain. After Napoleon's defeat, was never that dominant again. Mediocre kings and Napoleon III, played second fiddle to UK in industrial revolution, political unrest, and Germany unified under Prussia and Rhine industrialists and defeated France in 1870.
Gradient descent and reinforcement learning work eventually, but they take a lot of iteration and spectacular failures. Different exploration/exploitation tradeoffs are needed in different environments, the market signal isn't really strong enough to differentiate and the incentives don't actually line up perfectly.
The standard economics answer is more like, the market (i.e. regs) evolved to address coordination problems and align the market incentives better with the true long-term welfare function.
Not that different from 'effective altruism', really. Maybe running a Ponzi is moral if you are taking the money from people who are not effective altruists, but you are saving the planet or something?
Or Kant's categorical imperative, act in a way that you would want to make universal law. Then in order to act morally, you have to solve for all downstream impacts of the action over all time, if everyone acts that way.
Seems computationally intractable, possible multiple local equilibria, problems with which discount rates to use and whatnot.
Hence the popularity of simple ethical systems, just do what God says on the tablets and be done with it.
That's the cool thing about shaky theoretical foundations and fuzzy data, you can project anything you want onto them and act with a great deal of moral certainty about everything.
in general probably a good idea, because it's always something with that guy and his fanbois, but this has implications and it's weird that lesser-voted stuff is on the front page.
there are actual Nazis out there who want to put people in ovens, so it's taboo to make jokes about how they may have a point.
similar thing with trans people, you mention them and trolls come out of the woodwork saying they should kill themselves.
there are a lot of horrible people and trollish behaviors online, and sometimes the only way to deal is to avoid the topic that brings it out, which kind of aligns with the objectives the trolls wanted in the first place. social media isn't making these calls as some kind of woke social engineering, they're making them to avoid a death spiral into a cesspool that destroys their business and civil society itself. Yeah, it's silly to read "AI comic canceled for transphobia", but maybe it's not just the community standards that are the issue.
"Evolution of democracy is not possible if we are not prepared to hear the other side."
"Democracy is a great institution and, therefore, it is liable to be greatly abused."
- pandas
- polars
- modin with various backends
- DuckDB and Postgres SQL from Jupyter/python with the usual SQL dataframe magic
- R
- on medium and 'pretty big data' i.e. 10GB dataframes with millions of rows.
- maybe on consumer Intel, M2, and Epyc with lots of cores
What used to be 'big data' fits on an instance, you only need shards/clusters for 'ludicrous data'.
Life doesn't require anyone to confront bad stuff that only grows worse if it is tolerated.
Life does, however, require everyone to live with the consequences of our choices.
Certainly more than suspending journalists on trumped-up pretexts.
I'm less worried about a possibly incorrect 'misleading' tag than I am about weirdo elected officials who make false claims about things I can see with my own eyes, say everyone else is lying, say anyone who disagrees is un-American or a traitor, push divisive conspiracy theories and call for imprisoning our own medical professionals. And given all their other claims and agitprop, it's hard to take seriously a claim that the label violates their First Amendment rights or is a breach of trust.
there is so much absolute garbage bad faith BS on social media that is poison to a civilized society and a democracy, that there is no choice but to do the best we can to try to spread less of it. you don't get to throw up your hands and say, there is no perfect arbiter of truth and decency so let's say, anything goes. of course that's what all the worst burn-it-down extremists would like to see.
A lot of US institutions were exposed as unfit for purpose by the pandemic and we have to do better. But I'm not sure Twitter would be near the top of the list.
And somehow, the new regime seems able to make everything worse.
Suppose the collision avoidance is perfect. Now put it on the roads of NYC, or New Delhi.
There are a lot of people who will just walk in front of a car going 40mph, if they know for sure it will brake hard and stop.
The problem isn't technology, it's humanity.
The solution is to change the rules of the road, have protected lanes for self-driving buses and taxis and cars, and enforcement.
Let vehicles that can take full advantage of communicating with each other and the road go fast and use the infrastructure to maximum theoretical capacity, without having to worry about dumb human drivers.
normally, if you disturb a system three things can happen, it can be stable, it can explode, or it can oscillate. corresponding to e^x where x is a complex number or a matrix.
but actually in the liminal area at the edge of cyclical behavior there is a mode that is neither cyclical nor stable and it happens quite a bit and is responsible for many interesting and important phenomena and systems.
there isn't much regularity, the only way to predict is a model of the evolution of the whole system which is highly sensitive to small disturbances or measurement errors.
the biggest problem with NYT is they bend over backwards to appear neutral and give all sides the benefit of the doubt. it becomes doublespeak and provides cover for bad faith. for instance today's article on SBF - https://www.nytimes.com/2022/11/14/technology/ftx-sam-bankma... . it's an embarrassment to let him call looting client funds 'a margin position' without hard pushback.
but in the NYT they distinguish fact from opinion, and take reasonable steps to verify the stuff they print as fact. so even if they don't discount what SBF says as they should and draw the appropriate conclusions, a critical reader can. (even if it's while they are yelling at the screen)
so they are not going to print something like the Sequoia garbage that is completely one-sided, with technical absurdities and factual inaccuracies (and they will copy-edit properly and trim 14K-word vanity pieces).
people say the NYT is 'biased' but I think that accusation is misguided. Of course they have a point of view (like, democracy and government are usually good). the word 'bias' presumes the existence of some unbiased reference frame like 19th century ether theory. many people who call things 'biased' are just announcing, my own point of view is the one that must be privileged.
the question is not, do they have a point of view, but what is the point of view, is it transparent, do they check facts and distinguish between fact and opinion and dig regardless of who benefits. and the NYT is, despite its infuriating God-like view from nowhere, hard professional journalism.
the crap media is this stuff that has, not just a point of view, but an agenda, bogus facts, one-sided and self-serving, doesn't distinguish between fact and opinion.
and yes, it's important to know that Sequoia and Musk and other people want to control the narrative with wholly-owned crap media like this, and a Twitter zone they control and flood with BS. they want it so bad they hire 'private historians', try to discredit the real journalists, and spend megabucks for their own ecosytem.
join some real life groups outside your professional world, meetups etc.
cultivate something you like outside of work, movies, sport, cooking, gardening, travel etc.
kind of like Zuck's personal challenges, make a habit to do something annually, monthly
maybe he can turn Twitter into a super-app on the console of every Tesla. a plan is not in evidence other than throw spaghetti at the wall and see what sticks. 'move fast and break things' might really unlock opportunities, or destroy trust.
Zuck was unusually good at fake apologies while continuing to data-rape and strip-mine users, but trust is a problem for Facebook and will be exponentially worse for Musk.
it could indeed turn into MuskChan. if he is really going to turn a blue check into something anyone can buy without providing ID it's just a matter of time before impersonation scams multiply.
it will be an incredible distraction from challenges at his other ventures. Saudis and pedophiles are going to use it to abduct people. what is he going to do when people/nations go nuts about something on Twitter and take it out on Tesla?
I'm also skeptical of the weird turnabout between 'get me out of this deal at any cost' and rushing to pay up in full. either a side deal with e.g. Saudis or there was something existential about to come out in discovery.
layoffs shouldn't be celebrated because they cause pain and they mean someone screwed up.
time will tell whether Musk is putting the business on a sustainable footing or killing a golden goose.
Hey Elon: Let Me Help You Speed Run The Content Moderation Learning Curve
https://www.techdirt.com/2022/11/02/hey-elon-let-me-help-you...
before (I think) the 386, microprocessors didn't support proper memory protection, any OS that attempted to run multiple programs was at risk that one program would crash the system. You might see the 'bomb' dialog on Mac or it might just freeze. Blue screen I think started with Windows 95. Until (I think) Windows NT and Mac OSX the OS didn't implement memory protection. Original Mac had cooperative multitasking, each program was supposed to run in a loop that yielded to the OS so it could e.g. move the mouse pointer, process clicks. Of course at that time DOS didn't have any multitasking. although there was the infamous "TSR" terminate and stay resident feature which was used by e.g. DesqView to allow you to switch programs via a hotkey.
typically software can't fry hardware, even the famous 'halt and catch fire' instruction didn't literally cause a fire https://en.wikipedia.org/wiki/Halt_and_Catch_Fire_(computing...