pupperino

As a Latin American, I find it funnny it's basically impossible to tell if an American with this vocabulary and framing is like ultra MAGA or "communist".

which our Western governments are invariably complicit in

You are not the center of the universe, my friend. People in the Global South have agency too. This sort of performative anti-imperialism goes around itself and becomes an Empire-centered account of world politics.

Yes, they do. Just ctrl+F "identification" if you don't feel like reading the whole thing.

In fact, this is such a common practice that it's nigh impossible to publich an empirical paper without discussing identification and employing an identification strategy.

You can always tell people who are not familiar with modern Economics because of it too, they will bring "but correlation is no causation" up and feel pretty good with themselves. Then you mention the identification strategy, and they just stare at you blankly without getting what you mean by that. I've seen this happen so many times.

Well, let's try to take a step back and approach this sort of stuff with a more scientific, less politically hysterical way. Models are simply mathematical fiction, useful in as much as they constrain our thinking and remove wordplay from it. They're ways of imposing some discipline on our thinking. When a paper pours over the data, makes the case for an identification strategy and its associated causality graph, and estimates some parameters, these are not Objective Truth. You get this, obviously, amazing. Ok, so, are they? Samples from a high-dimensionality probability distribution encoding the "real" parameters for the effects of immigration on any desired outcome.

So maybe try to consider this yet another data point. A paper estimating a certain effect size in a certain context shouldn't flip your entire mental model of a certain phenomenon, but it's also totally irrational to handwave away empirical results that don't match your intuition.

This is usually what's known in Econometrics as Identification. Any applied econ paper written in the last two or three decades has at least a short section discussing causality and making the case for its identification strategy.

{renv} is pretty solid, I've been using it in production for years now and have no complaints.

This is every country, buddy.

I'm confident the people pursuing this mad agenda don't know the first thing about US budget laws and think they can just cut spending DOGE-style forever.

Baked in this plan is the idea that the revenue from tariffs will be much smaller than income tax revenues, and therefore that the federal government will also get smaller.

As a Latin American, I can't help but buy the American Exceptionalism thesis, and seeing you guys in this situation humanizes you so much. The US has such a strong bureaucracy and high bar for competence that you generally lack the immune system required for detecting con artists and social climbers - at least in politics. Ask a random Argentinian or Brazilian what politicians Trump reminds them of, and you'll get a seemingly endless lists of genuinely stupid, borderline sociopath populists. Follow up with a question on what are the consequences of having arbitrary, ever-changing tariffs on most goods for the purpose of industrialization, and you'll get a laugh, then a sad face.

Import-substitution is bad for econ 101 reasons that most people who have an axe to grind against Dems would've been absolutely capable of grasping only a few years ago. Now, it seems so many are willing to turn off a part of their brains for this short-sighted wishful thinking. "Well, the official narrative is that we're doing this to get manufacturing back, so let's wait and see" is something most people would immediately perceive as bs if a Dem was sitting in the Oval Office. Seeing a politician campaign on a stupid platform, and then getting surprised he actually shoots himself in the foot spending political capital pursuing is also very Latin American.

The good part is that the soon-to-be-coming recession the federal government just fabricated out of thin air is fully self-inflicted, and therefore somewhat easy to fix. The bad part is you have (at least) 4 more years of this lunacy, so it might take a while.

Is this a joke that I'm not getting? "Planned economy" and "command economy" are synonyms.

Well, those Indians living in the US will have families of their own, and over time become part of the community you claim to be a part of. Very much like your ancestors did, except they likely didn't face the arbitrary constraints on immigration that Indians (and any other nationality) face today.

Well, living near and airport or railroad is priced in, and not really a problem. Chemical plants... A bit of a stretch, isn't it?

That being said, in the US you can and should absolutely should build more, and basically get rid of most zoning regulations. You'd have a hard time finding anything as touristic and dense as Barcelona in the US.

I can only see my own video :(

Michael Levin [1] does brillant work on this "cellular language" space and the key point is that genetics is more like biological IaC, it only encodes proteins. Most of the actual control, anatomic and physiological homestasis happens at the higher level of bioelectricity - which works as a computational medium.

[1] https://www.youtube.com/watch?v=7FGM33sz25k

This links to a pre-print, but it has recently been published in the Review of Economics and Statistics [1]

[1] https://direct.mit.edu/rest/article-abstract/doi/10.1162/res...

"Massive" is a bit of a strong word. One year of 8-16% inflation with historic averages close to 2-3% is actually quite ok and demonstrates the European Central Bank's credibility.

Sure, but is this what we want, as engineers? Can computers be held accountable?

And an unchained LLM trained on reality is far more capable of finding solutions to that problem than a bunch of squabbling politicians.

Not that I disagree with this statement, I don't, but this is not a silver bullet. Technology is, ultimately, operated by humans and no amount of frontier research and development can overcome collective action problems. At some point, you do have to sit down with these stupid politicians and get everyone on board. The loom was invented hundreds of years before the industrial revolution, in fact it was nearly forgotten and the designed survived due to a few happy accidents. It was only after the English Civil War and the establishment of checks on royal power that widespread adoption was possible.

A lot of people are quite unhappy with it, yes, but... Consciousness is still very much a philosophical problem. Think of Marr's levels, functionally, LLMs and humans are quite similar, but algorithmically? How about the implementation layer? There are still many, many mysteries to be solved in the realm of philosophy of mind and LLMs should serve as reminders that Consciousness and Intelligence are not the same.

May I suggest this reading? https://qualiacomputing.com/2022/06/19/digital-computers-wil...

I acknowledge they're shipping changes to the frontend - some of which are not bad at all - but it sounds surreal to me that we keep talking about UI changes as if they were major features, and pretend twitter circles aren't broken, that notifications are as in near real time as they were, etc. There were major losses of service quality in basically everything backend related, and no amount of frontend tweaking can compensate that.

There was definitely development and new features - remember tweetpic and "RT @something"? They did the huge work it takes to go from 99% to >99,99% uptime. What did rebranding bookmarks into likes did that you think was so bad? I'm indifferent to it, but I could be wrong. Chronological TL needed only a couple of button clicks - then, after the acquisition, a couple of clicks every time you opened the TL because why not - so "hidden" seems like an exaggeration. Now it's 1, so, kudos to twtr 2.0 I guess?

Let's suppose for a moment none of the objections to these "features" I raise are valid. Why are people not talking about twitter circles being flat out broken? What about the constant changes to the UI which no sane design team would approve - like removing the "RT by @account" text out of nowhere, then putting it back up a few days later. Improvisation and rollbacks to major decisions? Remember when a couple of weeks in they just let people pay to get bluechecks and major advertisers had to deal with verified accounts posing as them? The whole "remove legacy bluecheck" thing?

Twitter was afloat and since 2020Q3, posting profits every quarter, but revenue has crashed as a result of all this. I can't really get why these details like "improved community notes" are more relevant to some people than the blatant mismanagement.

Yeah, I overstated that. It's not relevant to my main point, but I'll give you that.

And yet the process to grant bluecheck was to literally check personal id, and grant it to whoever had media presence. For years, you could request verification, send a couple of links citing you, provide a government-issued ID and that was it.

Twitter verified literally every public figure it could - regardless of politics - even those who didn't seem to need verification. However, being a niche contrarian blogger does not make you a public figure, and that's where I feel the resentment is located.

When having conversations in HN does it bother you that you have no way to know if you're talking to a real human, not hiding behind fake usernames? What do you think underlies this feeling you seem to have?

EDIT: what I do think was a tremendous success is the publicity stunt. I've never seen any platform where people felt they should pay to let people know they were who they claimed to be. Kudos to the people doing the whole "twitter is filled with bots" psyop, it worked.

Chronological feed was always a thing, I've used it for years. In fact, one of the first noticeable changes was that this config stopped being persistent. I had to manually switch to chronological for months and still get "For You" as the default. Community notes precedes Musk too. View count is not only unreliable (because counting things is hard) but of dubious value. What do you get from it? Longform text is literally what twitter is not about, and a major turn-off for so many people.

Twitter as of 2020 was fantastic, and these features added essentially no value to it. On the other hand, the public API is now a mess, twitter circle tweets are being shown to people who were not supposed to see them, I get logged out of all my accounts at least once a week and most notifications have serious delays. I've been using twitter since the whale days back in 2010, and it feels like we lost a decade of SRE progress.

It always makes me laugh how, for years, we mocked journalists because of their inflated blue check egos, and now every "free thinker" out there is eager to pay to be just like them. Verification is meaningless for most individuals since there's no real need to verify if @ronny2938742 is indeed Mr. Ron Whatever.

However, it's crucial to know if an account claiming to be a big company, government agency, journalist, politician, etc. is legitimate. Opening verification to subscribers is absolutely scammy, just a way to make a couple of bucks off the victims of culture war brainwashing.

What really matters is how human institutions and organizations shape each other's incentives and costs [1]. Instead of aiming for abstract, intangible goals like "small" or "big" government, how about developing a sense of institutional design and relating that to a set of moral and political values. That way, it's actually possible to debug disagreements, either reducing them down to moral values and hopefully leaving it out of the public sphere or locating actual policy questions with less room for aesthetics and more for evidence, studies and sensible debate.

[1] https://www.nber.org/system/files/working_papers/w10481/w104...

Nowadays, there's lots of tooling around for nested data, like roomba [1] or the list-column workflow enabled by purrr and dplyr [2].

[1] https://github.com/cstawitz/roomba [2] https://r4ds.had.co.nz/many-models.html#nested-data

People really did love "synergy" in the 90s.