You weren't exaggerating about that being an interesting substack. Just spent a couple hours reading around it, thanks.
For me, the main appeal is texture. I've always been someone who prioritizes texture over flavor, though both are important, and sashimi is by far my favorite food. If it weren't cost-prohibitive I would eat it every day fro 2/3 meals, mercury poisoning be damned.
Sashimi has a huge quality range with a bottom-heavy distribution. If your bowl is mostly days old tuna and tilapia it's easy to walk away unimpressed. I'd recommend giving it another chance at a quality location. High quality tuna (maguro) is great, doubly so if its a fatty belly (otoro/chutoro) cut. Yellowtail (hamachi) or amberjack (kanpachi) are amazing as well. Salmon (sake) has a more narrow quality distribution but a high floor so it's pretty reliable wherever you go. In contrast, I've never been a fan of the squid (ika) or urchin (uni).
Maguro usually refers to tuna cuts as a class. Fatty tuna is either otoro or chutoro depending on the specific cut, in contrast to the standard red akami. Akami in particular has a bottom heavy quality distribution, with average and below quality akami having a grainy, stringy texture with a bland flavor.
Are people using "milktoast" [sic] in an ironic way or do people genuinely not know that it's wrong? I've been seeing it more and more lately on places like reddit.
Also, this might just be Baader-Meinhof, but I have a subjective feeling that mistakes of this kind have become more prevalent recently. Use of "should of", or "i.e." instead of "e.g", etc. Anyone able to sanity check me on if this is real (i.e., do people just care less about proper language use nowadays) or imagined?
No disagreements here. To me, he's always been on a lower tier than the other two.
Simple explanation: clickbait listicle is clickbait listicle. They're optimizing for attention, not accuracy. This is an American publication with a primarily American audience. They're incentivized to prioritize names the median American would recognize. I haven't looked through the whole list in detail, but I imagine sports that are bigger worldwide than in the US (e.g., soccer, cricket, rugby) will be underrepresented while something like the NFL might be overrepresented.
Also, side note hot-take since I don't get to talk tennis much: I'd take Federer over Djokovic. Djokovic has the bigger resume, but he's a lot luckier on his age timing. I don't think 2011-2016 Djokovic would have been as dominant as Federer was in 2004-2009 (e.g., I highly doubt he goes 8-0 vs Roddick at hard/grass slams). I also think 2004-2009 Federer would have done better than Djokovic did in 2011-2016 (e.g., he'd go way better than 4-4 vs Wawrinka at hard/clay slams). Late career (post 30s) they'd probably have similar results as each other if age adjusted.
Does anyone have an explanation for why Microsoft products in particular feel so bad?
I've been using a Macbook Pro since 2015 as my personal computer. It does everything I want with nearly no issues. The phrase "it just works" is a perfect encapsulation of my feelings. (This isn't just Apple fanboy-ism speaking; everything else I own like TV, phone, watch, etc. is non-Apple).
I've recently had to use a Windows computer for work purposes and it has been a god awful experience. It frequently won't boot up from sleep, it often overheats while I charge it overnight, sometimes it won't connect to Wi-Fi and the Wi-Fi settings button is completely missing/inaccessible, getting it to change the scroll direction for the mouse wheel takes 5 more steps than it should, etc. I've also had to use Microsoft Teams instead of Zoom and Microsoft Outlook instead of Gmail, both of which have been massive downgrades along every dimension (lag, dropped calls for Teams; awful UI, lag, and terrible search/organizational functions in Outlook).
These are solid points, but I don't think they apply to my case.
I'm generally comparing popular candidates for "best pizza/burger/etc in the city" to non-Western food options and they still come up far short. On top of that, the burgers/etc. are often times recommended by Americans, whereas most of the time I have foreign cuisines with people from the relevant country they'll tell me it's only bad to average compared to what they'd find back home. Based on my experiences traveling to places like Italy/Japan they were absolutely right and "average" was being generous.
On the second point, my family almost never ate out as a child, so I was mainly introduced to these things simultaneously. The only "burgers" I had were McDonalds or Burger King and I only really think of them as "burgers" in the sense that Taco Bell makes "tacos". They were super health-conscious any wouldn't let me eat many hot dogs and the like, it was mostly lightly seasoned fish/chicken/vegetables/grains. There might be some slight effect in the sense that my average meal may have been closer to the new gourmet burger than to a ramen, but I don't think that effect is particularly large. (Funny enough, someone down-thread suggests the opposite - that early exposure creates a nostalgia effect as opposed to a familiarity-breeds-contempt effect).
Both are definitely in my top tier. My first time in Italy completely changed my perception of so many foods I only thought I knew. I can barely eat pasta or pizza in the US anymore.
It seems we have similar tastes. Main differences I'd have are fusing 1/1.5 then bumping up Italy and MENA into Tier 1.
3 countries out of 30+ constitutes "half"? If we go by population, that's still <10% of "the West" (by which I mean broadly Europe + USA + Canada + Australia + New Zealand). Additionally, those three one could argue are non-central examples given their geographic location on the Mediterranean, resulting in heavy influence from MENA regions.
Does anyone here actually consider Western food amongst their favorite cuisines?
Personally, I find American cuisine is downright trash-tier. My city is lauded by many as a "top-tier food city" but the examples people give of great food are pizza, hot dogs, burgers, Italian beef, and cheesesteak. Most of the ones I've tried I would call oil-drenched slop. None were actually delicious enough to justify the health detriments, especially when similarly unhealthy but better tasting options exist like Mexican tacos, Indian curry, Iranian kababs, Japanese ramen, Chinese hot pot, etc.
In my experience, this has applied to Western countries in general. Except for the Mediterranean-adjacent Italian, Spanish, and Greek, I don't think I've ever particularly enjoyed any other Western food. Do Canada, Australia, and New Zealand even have an identifiable cuisine? I don't know of any British, Nordic, or Slavic restaurants in my area. France is stereotyped as the culinary capital, but most of what I've had was overpriced and looked better than it actually tasted.
It may be that most of the hype around Western food is concentrated in fine-dining, in which I'm largely uninterested. When it comes to a more typical meal, I have a hard time putting any country (aside from Italy/Spain/Greece) above bottom tier when comparing to other regional cuisines from East Asia, South Asia, Middle East, Southeast Asia, or Latin America.
So am I eating the wrong things, is my taste atypical for someone raised in the West, or is it relatively common for most Western cuisines to be clustered in the bottom-tier?
Singapore does this by having housing so unaffordable that ~80% of people live in government subsidized housing, then setting approximately-proportional-to-general-population racial quotas in every housing block. For example, if you're looking to sell your apartment, but your building ownership is already >80% Chinese, then you're limited to non-Chinese buyers.
Interestingly, they also have quotas on permanent residents, presumably to encourage integration with citizens.
It's called the "Ethnic Integration Policy" and it's an interesting approach that obviously achieves its desired metric. I don't know enough to say if it manages to achieve anything beyond that.
Probably politically infeasible in most other countries.
The security vulnerability is a great point. I wonder to what extend this affects political leanings. I recall reading that one of the more consistent differences in conservative vs liberal psychology is that zero-sum thinking is stronger in the former while positive-sum thinking is stronger in the latter.
I would imagine that if most of your interactions with the outside world come in the form of positive sum trade, then you'd be more inclined towards liberalism/xenophilia. In contrast, if most of your interactions with the outside world come in the form of security/border disputes, you'd be much less so. (Then again, island/EEZ disputes still happen plenty so maybe not.)
It's certainly true that west Europe and coastal USA are more liberal-minded than their inland neighbors, but it's difficult to tease out if this is the main driver or more a direct economic effect. Also not really sure how this well this trend holds for north vs south India or for coastal vs inland China.
Is the explanation for why coastal areas with ocean access are consistently richer than more inland areas really as simple as "ocean access = international trade hub = money"? E.g.,
-south India vs north India
-west Europe vs east Europe
-coastal USA vs middle USA
-Japan/Korea/Taiwan vs coastal China vs inland China
Is there more to it, like say a particular path dependence in each of these regions? Are there significant exceptions (either now or historically)?
The misuse of "exponentially" in particular infuriates me. I can't count the number of times I see "'exponentially' large" as if it just means the same as "really" instead of referring to specific functional relation. It's as nonsensical as saying "linearly large" and immediately indicates to me that I should probably disregard anything else said by the writer.
The restaurant I have in mind had reviews that literally used the word "authentic" and yet the Banh Mi I ordered was basically an Italian Beef with some pickled vegetables.
(I suppose I should have avoided using the term "best" in reference to restaurants, given the inherent subjectivity.)
My only reservation with a binary system would be that it fails to capture things that resonate strongly with a small subset of the population (like cult classic movies).
How do you go about evaluating the competence of someone in a field that you yourself don't know?
One thing I've noticed about my city is that all of the best Asian restaurants have reviews more in the 4.2-4.6 range on Google. Every Asian restaurant that I've been to in the 4.7-4.9 range has been consistently mediocre. This still holds true even accounting for higher variance with lower total number of reviews. This is likely because the majority of reviewers have preferences near the median American palate, which means that Americanized Asian food is often rated higher. Most Americans would find actual Japanese food a bit bland, actual Sichuan food too spicy, etc. I've seen people give negative reviews to hotpot restaurants because they thought the natural numbing sensation of mala flavor was giving them an allergic reaction. I've seen a Vietnamese restaurant with multiple non-Vietnamese confidently exhorting the "authenticity" while several reviewer with a Vietnamese name commented how the food was "barely Vietnamese". (I'd be interested to know if this trend holds in the reverse, but I've never gone to European/American restaurants in Asia because I'm mainly US based, so why would I waste money and satiety like that)
Anyways, returning to the main point: Most reviewers don't know anything and don't realize they don't know anything. But with restaurants, I can just go there and try it for myself. At worst I'm out 30 dollars and I know to go elsewhere next time. How do I know my mechanic/doctor/accountant/dentist/etc. aren't making basic mistakes or misses? If I can't trust public reviews, is there even anything I can do?
I think you thought I was asking about a different comment of yours? I was referring to this comment:
The Fed learned how to operate at the 0 lower bound. The only thing that ended the 2008 expansion was Covid and that was a choice by policymakers to cause a recession. Without getting super long winded because the subject matter is a PhD thesis the evidence seems strong as we have gone thru a 14 year expansion which also included planning a short recession and rapid recovery.
Don't know much about economics (had to google "0 lower bound"). Got any reading suggestions for understanding what you mean by operating under that condition?
One angle I'm somewhat surprised hasn't been brought up much is that this set-up will almost certainly lead to the "problematic" optics of a non-Japanese person running around slaughtering a bunch of Japanese people. Video games are largely a power fantasy so the player character is going to be generally depicted as by far the strongest being. In the best case scenario, the Japanese will be depicted as weak and passive, needing a "foreign savior". I can imagine the outcry on the left would be swift if this were a game starring, say, a European warrior in 15th century Africa.
Even as someone who feels like "problematic" gets overused, I'd say those optics are more than sub-optimal. I remember having a conversation with someone about a similar potential issue arising with the God of War series, which sees former Greek solider Kratos slaughtering a series of mythical gods (first the Greek pantheon, then the Norse). We were discussing whether it would be difficult to continue the series beyond European pantheons while keeping Kratos as the main character because it just seems like the optics of a white guy traveling to Japan, India, and Mexico and killing a bunch of their local gods would be harshly criticized as a narrative faux pas in this day and age.
I'd be curious to know how positions on the two series correlate. If people were largely basing their opinion on "principle" I'd expect the largest groups to be:
1.) "It's a video game. I don't care if it's about an African guy killing a bunch of Japanese people, or a Greek guy killing a bunch of Japanese gods."
2.) "Optics are bad. I'd prefer a story set in Japan to star a Japanese samurai and I'd prefer a story set in mythical Japan to star a Japanese character."
Does anyone here fall outside of these two and if so why?
I suppose, but in the context of the original post, polyglot feats were given as a supporting example for there not being much of a "crowding out" effect on other skills/knowledge.
I imagine these senior clerics tend to specialize in specific subregions of the world with similar languages (though do correct me if I'm wrong), which is why I emphasized "highly distinct languages". Furthermore, their conversations in those secondary languages are likely to be limited to a well-trodden Catholicism-focused subset of all possible topics. Sure, speaking about Catholicism in English plus 5 Romance languages is impressive, but it doesn't compare to the cognitive load of having depth and breadth of proficiency in English + Mandarin + Arabic, where I'm positing that you will start to see crowding out effects at least on the level of "quickly, effortlessly accessible" if not on the level of having the actual knowledge somewhere buried in the brain.
I have a baseline high level of suspicion towards Youtube polyglot videos. These people are like magicians in that they give an illusion of an ability when their real talent is for something somewhat orthogonal (that is, less about being actually proficient, but rather about looking proficient in select settings that they tune). A few tricks include:
-
Controlling the conversation: The main skill many of these "polyglots" have is in pushing conversations towards topics about which they have the appropriate stock phrases well rehearsed. Sentences about how much they "love the culture", "I always thought [insert country] was so beautiful", "the [insert cuisine] is delicious" etc. If someone says something the "polyglot" doesn't understand they'll smile, nod, say "that's great" or "hmm, I'm not sure" and quickly try to change the subject. The better ones can do this more subtly, but even the clumsier ones can get away with it since most viewers aren't watching critically.
-
Selective editing: The format lets them use staged videos or simply to selectively include footage of their best performances. Given the incentives on Youtube I don't really trust most to not do these things. For every free-flowing Mandarin conversation there may have been 10 where the polyglot just totally misunderstood the native speaker.
-
Optimizing study: Words and phrases are Pareto distributed, so you can get to a basic conversational level in most languages with about 3000. If you're good at point 1 above you can probably get away with much fewer. For comparison, a native speaker is estimated to have a vocabulary of 20k to 35k. If you're loose with the definition of "fluent" you could just study these most optimal 3k for several languages instead of becoming highly proficient in one.
-
Piling up on highly similar languages: The distinction between a "language" and a "dialect of a language" is more political than anything objective to the forms of speech themselves. An English-only speaker could likely become fluent in Spanish, Portuguese, Catalan, Italian, and French with less effort and time than it would take to get equivalently fluent in just Japanese (or Mandarin or Arabic or Korean). By largely focusing on clusters of highly similar languages, they more easily inflate their "language count".
Most genuine "polyglots" (high proficiency in 3 or more) I've spoken with say that they can only have thorough, complex conversations in 2-3 highly distinct languages at any given moment. Even a few months of disuse is enough for them to feel significantly sluggish retrieving words and forming sentences in their native languages, albeit still highly proficient. If they know they'll need a particular one soon (like they'll be traveling to [insert country] next month), then they can review for a week or two and revive the quick access (like putting it in RAM), but trying to keep all of them active simultaneously makes organizing thoughts a bit chaotic.
Is there a good analysis somewhere of the modal ideology/thought process of people by their voting patterns in the last three presidential elections? As in people who voted:
Obama-Clinton-Trump
Obama-Trump-Trump
Obama-Trump-Biden
Romney-Clinton-Trump
Romney-Clinton-Biden
Romney-Trump-Biden
I've found some articles about Obama-Trump voters or Romney-Clinton voters between 2012 to 2016, but couldn't get much that extends the analysis to 2020. Are these groups just too small or internally ideologically heterogeneous to say anything substantive? Who and where are these people?
More options
Context Copy link