> (specifically: it's the vector that rejects pluralism, divides the polity into "the people" and "the elites", and declares that a particular political agenda, in its entirety, represents the true will of the people)
The elites vs the people distinction is usually the key characteristic of populism in most definitions. Populist movements also tend to be very anti-establishment. Many vectors for influencing politics don't involve viewing society as this battle between the people and the elites. For example, while it's true that left-wing populists exist in the US I wouldn't describe the current Democratic party as a populist party. It is more of an institutionalist party and they don't tend to attack elites, because most groups traditionally considered part of the elite support the party. This is in contrast to the current Republican party which very much casts itself as being opposed to the elite. All the rhetoric about battling the deep state and draining the swamp is very much populist.
It's so buggy right now. It feels like it's gone back to the old fail whale days. "It works" is a pretty low bar to clear. On the upside, it gave me back a lot of free time, because the Twitter experience is just too frustrating to deal with anymore.
I don't think it even technically works as well as before. The iOS app seems mostly fine, but on web the feed or tweet fails to load and shows the blue retry button probably 1/3rd of the time for me now. Combined with all the bot issues you mentioned, that has dramatically reduced my Twitter usage. It's become incredibly frustrating to use. I only use it now when I click on a tweet link that's been shared with me.
Buybacks being more tax efficient are a part of the story, but I don't think that's the primary reason companies use them. They're also a lot more flexible. If you pay a dividend you set the expectation that you will continue to pay that dividend and cutting the dividend crushes your share price. Buybacks don't carry the same expectation so in lean times you're not forced to signal to the market that management is not confident in the company's future earnings. They also offer a way of artificially propping up the company's stock price in bad times by using cash reserves to buy back shares. They also artificially increase earnings per share and a lot of executives have bonus targets tied to that.
Really, no one? Flag participation is already higher for 6 to 12-year-olds than tackle. Lots of girls participate in flag. I suspect they'll be interested in watching the sport they grew up playing. It might never be more popular than tackle, but to say no one will prefer watching it is silly.
While American Football won't end anytime soon -- depending on your definition of soon. It's worth noting that youth participation rates are falling dramatically, and there are big regional differences. It will take a while, but football will eventually end up as a niche regional sport. Unless a drug like this or some revolutionary piece of safety equipment came along. Then I could see the sport staying on top long term.
That's wild. I wonder if you just have a couple of jerks in your area. I grew up in a rural area and my parents still live out there, and I cannot ever remember someone driving on the road with their light bars on. I used to have one on my jeep. Anyone who has them knows they're being a jerk if they're driving with it on the road.
Meta's choice of license doesn't indicate that Yann is clueless about open-source principles. I don't know about Meta specifically, but in most companies choosing the license for open source projects involves working with a lot of different stakeholders. He very easily could have pushed for Apache or MIT and some other interest group within Meta vetoed it.
I don't think that's entirely true. I know a few A+ people who enjoy being in the office and hate working from home. I don't know how many of those folks there are, but I don't think company A would be totally stuck with leftover workers.
Getting the extra reading time is nice, and if I didn't have a family I'd be all about that. But for me I get to spend mornings and evenings with my son, but if I had to commute half an hour then the morning would become just a rush to get him to daycare and the evening I might get to see him for a few minutes before he goes to bed. Sure the hour of reading would be nice but it would greatly diminish the quality time I'm able to spend with the little man.
I have often thought about creating an everything app for myself. Do you have it as a desktop app, or is it a web app you host somewhere? Would love to read a blog post about it if you have one or would like to write one.
It seems to me that it would be in YouTube's best interest to provide me with accurate recommendations. The more content they can get me to watch, the more ads I view, which is how they get paid. It would seem that more accurate recommendations would encourage more view time and more add engagement.
I also find that I often experience the modern version of "no one gets me" quite frequently, which is that none of the algorithms really get me. Maybe I am interested in too many disparate things. Probably the worst offender here is the YouTube algorithm. YouTube really really believes I'm someone who wants a never-ending stream of Joe Rogan, Ben Shapiro, Jordan Peterson, etc. I have no idea how YouTube became convinced of that, and why it won't give up on the idea. I don't click on the videos, When they pop up in my shorts I immediately swipe past them, I can't even figure out how the kinds of videos I do watch would be connected.
I would actually love to have a good algorithm that would scour all of the content of the web and recommend just the things I would like, but all the algorithms that exist seem bad at doing that.
I went to grad school with a guy who was in that situation. He worked at a bank and got laid off during the financial crisis. The plan was for the bank to port the old system over to Java or something like that, and they were "close" to the end of the project and they thought at that point the rewrite team was comfortable enough with COBOL that they could do the rest of the rewrite without him. Turns out that yes they understood COBOL just fine, but they desperately needed his institutional knowledge. He ended up agreeing to come back on a part-time basis at some obscene consulting rate for however long the transition took and in the meantime, he did grad school part-time to skill up.
I think that's a compensation problem more than anything else. I've known some QA folks who enjoyed QA and would have stayed in that role if they could have justified the massive differential in comp between QA and SWE or product development. If we valued QA and compensated it at the same level we do those other roles then there would be a lot less difficulty retaining good QA folks.
> OK are you suggesting that people who don't have kids of their own don't care about other peoples' kids as much as those who do? It seems you're pretty deep into logical fallacies.
That's a bit of a leap, but let's pretend for a minute that that is what I was suggesting. What specifically would the logical fallacy be?
I used to consistently read a book a week. Then I had kids and now I feel pretty lucky to get through 10 or 12 a year. I'm consistently amazed by folks with young children who are still able to read a ton. My time management skills aren't that good.
This is next-level cynicism. Zuckerberg has 3 kids, most of their execs and a lot of their employees are parents. They've also employed a team to try and keep that stuff off their platform for a long time. The idea that they don't really care about this kind of stuff is asinine.
I actually do see DL research on tabular problems come out, but it seems like they generally aren't better than xgboost or similar boosted tree models on the same dataset, and when they are better it's only marginal and not worth the interpretability tradeoff in the real world. Boosted trees are already really good for most tabular tasks, so it makes sense that most of the focus would be on images and natural language where they are a big improvement over traditional methods.
Additionally, it seems to me that most of the money spent on research in the DL space goes to places like OpenAI whose main goal is to develop AGI. They see large transformer models as the path to AGI, and once you've got AGI solved the AGI can solve all the other problems.
Divide and conquer actually does seem to be the strategy. Each department within the DoD does its own audit, and a few of those departments actually have passed. But most of them have failed with a few still ongoing.
The bigger issue is the no political will part. If congress mandated that senior pentagon officials lose their jobs when they fail an audit they would get the process figured out really quickly. As is there's no urgency because there are no consequences for failure.
Also, audits are not 100% about saving money. They're also there to enforce conformity to standard accounting practices and try to make sure the books aren't getting cooked. Even if the books aren't getting cooked the only way you can be sure of that is to go through the audit process.
All of that sounds like bog standard kinds of stuff that we and every other private company have to go through during our annual audits. I don't totally understand why these things are so burdensome to DoD personnel that they can't pass an audit. I would have been fired a long time ago if our company regularly failed audits over this kind of stuff.
This has become my workflow too. Admittedly though I've spent most of my career writing large amounts of SQL, and was a pretty heavy Tidyverse user for a while, so that all makes a lot more sense to me than Pandas. I generally get my data into whatever shape I need it in and then load it into pandas.
But as we've seen with human societies once you reach a certain level of affluence you tend to reproduce less not more. Why would we assume that civilizations that are presumably far more advanced and affluent than ourselves would desire to expand? Expansion is fast on galactic timescales, but it's still very slow in the scale of a human lifetime. We don't tend to undertake projects knowing that we won't finish them within our lifetimes.
Rent prices in Florida were skyrocketing long before interest rates started rising. Our last rent increase before we moved was going to be around 25% back when interest rates were still 0. If anything rent increases -- in South Florida at least -- are partly a zero interest rate phenomenon. There are a ton of half full midrise luxury condos that priced lots of people out of previously affordable communities built as the result of access to cheap capitol and money laundering. There was also the massive influx of people moving into the area during covid. All that to say rent prices in Florida are the perfect example of monocausal explanations for complex phenomenon almost always being wildly incorrect.
You just nailed the number one reason I'm always skeptical of these claims.
With our current level of technology we have not lost a major airframe in the US since 2001. Yet these beings advanced enough to fly across interstellar space get to earth and suddenly can't keep their ships in the air?