domain:archive.ph
A short prompt of good news for starting the week- the likelihood of the current Gaza conflict ending just got significantly higher today, as Hamas has released at least the first 7 of 20 surviving hostages to Israel, with more expected later today (or maybe already completed), as part of a Trump-mediated peace deal that is excepted to culminate in a regional summit this week.
Big if carried through, and while there was leadup to it last week, there was a fair bit of (and fair grounds for) skepticism on if the deal would actually be followed through. There were questions on if Hamas even could deliver all the living hostages given how the hostages were often not under Hamas's direct organizational control (but sometimes under other groups), and this deal does not address the bodies of the dead hostages, among other things.
There is also some irony, or possibly some future culture war conspiracy theories, about how this will not get Trump a noble peace prize, since they announced that late last week.
That said- and I think this is good news in general- it's also worth noting this doesn't mean stability or even a lasting peace. While the Yemen-based Houthis have indicated they will stop their Red Sea attacks so long as Israel upholds the ceasefire, this runs into complications like how Hamas has already engaged in gun battles with gazan clans as it tries to re-assert control, which goes significant premise of Hamas being removed as the military and civil authority of Gaza. Which remains a huge, unanswered question which could restart this problem all over again, if Hamas remains in power for lack of anyone actively displacing. The NYT is running a piece on how mediators are already signaling this isn't a comprehensive deal for either side.
One thing that isn't in question, however, is that the return of the still-living hostages is going to reshape the underpinnings of Israeli politics, as the post-October 7 war cabinet coalition that kept Netanyahu in power will lose much of the reason for being. This means political instability, for worse or for better, as Israel rebalances. The next election would be no later than late next year regardless, and could come earlier.
Absent some new (and detrimental to all) nonsense, this means that a lot of the people who only supported Nnetanyahu because of the war will likely be more willing to withdraw their support and trigger early elections, which would be no later than about a year from now anyways. This does not, however, mean a general discrediting of the Israeli right, and a decades-belated return of the Israeli left (whose original decline was after the failure of the gaza withdrawal almost two decades ago). The war was a significant polarizing effect on Israeli politics and society, and while I'd not bet on Netanyahu I'd also not bet on any part of the political left seen as opposing the war for pro-Palestinian reasonings.
I'll end it there. While there is plenty of reasons things could yet again get worse, and while I am sure eventually they will, for the moment I'll encourage people to view this new news as good news, which can well make many people's lives better.
This is my third or fourth attempt to read this book. In the past, I've gotten a few hundred pages deep only to fizzle out as it didn't go anywhere in particular. It's incredibly difficult to read, not so much because of the footnotes or the pure length as because of the structure of the story.
I'd previously enjoyed DFW's shorter work, and to some extent I think Infinite Jest is just a really large short story collection that links together in intricate ways to produce a bigger work. A lot of the chapters, or sub-chapter units of the book, really constitute stories or vignettes or essays of their own, and their significance to any broader plot (indeed the existence of any broader plot) only becomes significant later. DFW's brilliance in writing essays and short stories gives you these really gripping moments throughout that seem to have nothing to do with the entire rest of the story. DFW also uses very non-linear storytelling, with a confusing in world neologism for years, to obfuscate what you are seeing and when.
Then you have the overall setting, which is sort of semi-sci-fi magical realist near future-past in a way that's incredibly difficult to find your bearings in. When I read Tolstoy, I know what the rules of the universe are because they are the rules of my universe. When I read Asimov or Tolkien, I can quickly grok the rules of the universe because they're very different from the rules of our universe in specific fairly well explained ways. Infinite Jest's universe is different in confusing and non-specific ways, and it's not clear when narrators are unreliable or taking the piss, or when we're supposed to take things seriously. At times DFW chooses to be brutally realistic, at times absurdly fantastical, nearly always pessimistic in outlook.
It's also disgusting, viscerally disgusting in a way that only a writer as talented mechanically as DFW can be. Everyone is asymmetrical, everyone is disgusting, bodily acts are described in extensive detail, rape abuse illness and addiction are commonplace, deformity is everywhere. It's just how the book is, but there are significant sections that are just viscerally unpleasant.
Finally, I think the book has gotten a lot harder since its publication, in that it represented a fork from the past around the publication date. For my partner in my book club, who was a teenager when it was published and read it for the first time when she was in art school in the early 2000s, there are a lot of references or just moods that make more sense to her than they do to me, ten years her junior. It's very like Stranger in a Strange Land in that way, a retro-futuristic work that projects the current mores and world forward. In your mind you have to back up to 1995, then fast forward to a world where some technologies never develop and others hyper-develop.
That said, my foolproof method for reading difficult books is to just keep swimming. This developed when reading the great Russians, in that way that Americans typically get confused by the use of first name or surname or patronymic or title or affectionate diminutive to refer to the same character, I used to get stuck trying to figure out who exactly was who in Anna Karennina then I decided one day that I should just keep reading and I'd figure that out later, and that worked. I approach everything confusing in DFW the same way, I just keep reading and I figure it out later. I think this is what @Rov_Scam is getting at, knowing that it's an important book he keeps trying to read it while understanding everything said but it's impossible to understand everything so he can't read it; his brother just read it without worrying about understanding everything and was fine.
The other aid getting me through this reading is my book club, in which I meet up with a pretty lady every few hundred pages and we discuss the book and its themes and broader philosophical topics over drinks. And this creates accountability in that as a man I can't let a pretty woman mog me at something, and also makes the book easier in that discussion helps explain things. Marx's famously dense Kapital was originally published in France serially in socialist newspapers for workers to read in clubs, they wouldn't (probably couldn't) understand such a book on their own, but in a group it becomes comprehensible. The lack of reading groups is one of the unfortunate consequences of our world today.
It's truly a work of rarely-reached genius, a fitting heir to the western literary canon (though in my mind the canon truly ends with Joyce). It's highly prescient philosophically, it has a lot to say about modern concerns on AI. Though I also kind of think the whole book is just about internet porn. Everyone on this forum should put in the work to read it, it's worth it, but I can also see how its cultural impact is mostly negative rather than directly influential.
I do wonder if Parker and Stone were influenced by Infinite Jest when they wrote South Park, though.
The government shouldn't have any idea how you spend your money, but wants to incentivize certain spending. The only way that can be reported is with a citizen provided form. That's the original reason.
Do the filing prep firms lobby to keep it that way, long after the government collects a lot of this information elsewhere. Absolutely.
From a revenue maximizing perspective: It's essentially the same as using coupons and discounts for price discrimination at a grocery store. You capture additional transactions from customers willing to put in the work/inconvenience of shopping from the discount rack or using coupons, and broadly speaking these are mostly transactions that otherwise wouldn't have occurred because those customers would not have been getting enough consumer surplus from the transaction at the original price. You offer different prices to people with different willingness to pay by placing slight inconveniences in the way, people who really want to pay less will go through the inconveniences while people who don't care about paying more won't. With taxes, people who really don't want to pay those taxes (or people sufficiently sophisticated that their objections to having to pay those taxes would actually matter to the system) avoid paying those taxes in various ways, through complex deductions and schemes to funnel money one way or the other. People who don't really care about their tax bill (or people who are low-class enough that their complaints won't matter to anyone but their bartender) just pay the taxes because they don't care enough to figure out all the ways to avoid them.
From a freedom maximizing perspective: At the time it was put in place, this method minimized the degree to which the IRS surveilled individual Americans. This is mostly negligible today, when privacy has been so thoroughly compromised under law and custom that it feels irrelevant. But at the time this was an important consideration.
Intertia: But mostly, I think the best steelman is that changing the system would have unpredictable effects on the economy. Between two thirds and three quarters of Americans get a tax refund. The average refund (I couldn't find the median where I looked) is around $3,000. This is essentially a forced savings program by the IRS, in which Americans are forced to save a small amount from each paycheck and then given the money back in a lump sum later. This might have systemically important functions at this point which lead to significant switching costs nationally. For example, it's pretty well known that the best time to sell a cheap used car is around tax refund season, as lots of poor people who otherwise spend as fast as they get it suddenly have a pile of cash and need a car. People also often spend that money on home repairs or security deposits to move houses. Though they also often, of course, blow it on vacations or poor decisions. While the system that leads to a tax refund might be inefficient in and of itself, at this point if we got rid of this system we don't know what impact it might have. Poor people might stop being able to buy used cars, as they go back to saving nothing. Cheap flights to Florida from northern cities might dry up as middle class folks stop getting a tax return in the colder months and eyeing up tickets. Plus, regardless of the total taxes paid, once you get rid of the refund, people won't notice the extra few bucks every week, but they will notice the lack of a big lump sum every year, and will feel worse about it. The tax filing system is a way to trick children into feeling like they're paying less.
It is poorly worded rule. You are totally allowed to build consensus by convincing and persuading people. You are not allowed to assume already existing consensus on a topic. I am not sure that it is philosophical, but mostly practical. Posts that starts with - "We all agree that" usually are not very productive, not consistent and show that the poster has rarely faced any intellectual adversity and mostly has been in circlejerk places.
The eternal problem of “are you sure?” almost universally lowering its previously declared confidence in any subjective answer also remains.
Works on people too though.
Thanks. As expected, it misses several configurations that are critical, like hadoop.security.token.service.use_ip
.
I'm always surprised he survived the Night of the Long Knives. Von Schleicher didn't, so Hitler wasn't afraid of killing off Hindenberg's buddies.
First of all, sorry to hear that and all the best with the tests.
Secondly, about:
Especially exams that require months on end of grinding and memorization, when is rather be doing anything else.
I’m curious about these medical exams and studying. Are there some candidates you’ve met that can just ace them without studying, based solely on general medical knowledge and above average recollection from both medical school and hands-on training in the years before their specialist qualification? Or is it like some legal qualifications, where even a towering intellect needs to rote memorize that the answer is a section 37 part 3 form and not a part 4 and that a certain period is 13 working days and not 12?
hallucination rates are close to negligible
This has not been the case for me, unless you count “yes, you are correct, it seems that x is actually y” follow-ups when specifically prompted as negligible, which I would not. The eternal problem of “are you sure?” almost universally lowering its previously declared confidence in any subjective answer also remains. No specific examples, just my general experience over the past few weeks.
The appropriate response to hallucination handwringing from luddites is “it doesn’t matter”, not “it’s not happening”, by the way.
The real disaster is that the ones who are self aware enough to know they are bad writers went from 2 line emails to paragraphs of AI slop, no doubt promoted by the same 2 lines they would have previously just sent.
Like 1934 maybe, not the later years when he decided to willingly collaborate with the Nazis anyways despite his reservations.
which raises a lot of questions on how that mechanically actually happens.
We still have people who are pronounced dead, but weren't, despite modern medical knowledge and tools. So it probably happened a lot more often in the past.
The bible explains that the Roman soldiers didn't follow the correct crucifixion procedure, where the legs of the person were broken. So it makes sense that Jesus could have barely survived, recovered a bit after being placed in a cool tomb, then wandered around a little in a stupor, and then died.
Then add a bit of embellishment and you have a resurrection narrative, with him transcending to heaven (aka actually dying) shortly after a faux death.
For the same reason why Venice is no longer Italian. Too many tourists driving out the natives.
Imagine that your hobby spot gets disrupted constantly by tourists who gawk at you like you are a zoo animal. I bet that you'd find a more obscure spot that they can't find.
I was riffing off the blood boys thing primarily!
In the early 70s you would have paid the equivalent of about $1000-$1500 in today's money to fly from coast to coast in the US.
Which confirms @fmac's point - a quick search on Kayak found that Alaska Airlines and Hawaiian Airlines* have 1st class transcon fares in that price range. And "Economy with extra legroom" plus the checked bag and other such upcharges is a lot cheaper, and closer to a 1970's economy experience than 1st class is.
* The big 3 carriers use long-haul configured aircraft on the premium transcon routes, and long-haul business class is a lot nicer and a lot more expensive than traditional US domestic first class. So you need to look at smaller airlines to get a fair comparison.
Die Staufer: Glanz und Elend eines deutschen Kaisergeschlechts, by Johannes Lehmann.
It's a historical work about the medieval house of Hohenstaufen, and especially about the emperors, with Friedrich I., known as Barbarossa, serving as the headliner. It's pop-sci, but very dry, just the way I like it. Tries to draw from as many different sources as possible, compares them dispassionately, usually shrugs and admits that there's no way to know for sure, and then goes on to recount the history of those rulers in a workmanlike fashion. It's unexciting, calm and goes for accuracy instead of shock value. Thus my teutonic heart gets all warm and fuzzy. What a comfy read.
Also, my wife pulled it out of a book stop, so it didn't cost me anything.
I would be surprised if no other highly compensated industries did that.
In finance it is part of the deal at mid-levels and above. As a quant VP I get my bonus in cash, but a VP-level trader or corporate financier would be getting part of their (larger) bonus in RSU's, as do my bosses at director and MD level.
For a different extreme, look at Tesla. 125,000 employees right now. Market cap 1.3 trillion. Elon Musk right now earns about 8$ billion a year under current conditions, but it goes up by 1% of Teslas total market cap for each additional trillion in their market cap. If he hits just the easiest goals, he gets $36 billion a year. If it goes up to 8 trillion, he clears an eye watering $878 billion in 10 years, almost $90 billion a year. (yes that's in stock, not cash, if that makes a difference)
Musk's CEO pay at Tesla is uniquely generous even compared to other overly-generous CEO pay packages - to the point where the Delaware courts ruled it illegal. Tesla has the most liquid options of any single stock so it is quite easy to measure the ex ante value of Tesla share options. The pay package Musk "agreed" with Tesla in 2018 that was rescinded after it turned out to be worth $56 billion ex post was worth $2 billion ex ante back in 2018 - at a time when Tesla was only a $50 billion market cap. The number of public-company CEOs who made $2 billion from CEO compensation ex post is small enough to count them on your fingers. But even as the highest-paid CEO ever (by an order of magnitude) Musk made more as an owner than he did as CEO.
But the typical fat-cat non-owner CEO retires with a net worth in the high double-figure millions. Overpaid non-owner CEOs get a grossly disproportionate amount of public attention relative to how relevant they are to rising inequality, or falling living standards for line workers. I was particularly amused by the press coverage of Andy Jassy's $40 million payday as CEO of Amazon - none of which made the comparison to how much Jeff Bezos made off Jassy's work (about $80 billion, so 2000 times as much).
If there's no aid on the ships, in what sense are they humanitarian? Because they're full of good vibes and well-wishers?
Marina Abramaovic
She's mostly pre-woke, isn't she?
Sure, she's still alive, but her heyday and peak relevancy are long behind her. Now she's just one of those old ladies that gets trotted out whenever there is a slow news week and people need to be reminded of what a Legendary and Influential Artistᵀᴹ she is.
I've heard Infinite Jest is quite the doorstopper. Are you finding it difficult to read?
The Outsiders by S. E. Hinton. It's the book which is widely credited with inventing YA fiction, for better and for worse. An easy read which I know I'll never read again, and probably the best book written by a 17-year-old girl I've ever read. The name on the protagonist's birth cert is "Ponyboy"; now there's another reference I understand, RIP.
it also might depend on what you mean by 'faster' or what you are doing. but if you are multiplexing streams inside of TCP like HTTP2 then this can be slower than separate HTTP/1.1 streams because a single missing packet on the HTTP2 TCP stream will block all the substreams whereas a single missing packet on a HTTP/1.1 TCP stream will only effect that one HTTP/1.1 TCP stream. by 'block' i mean the data can't be delivered to the application until the missing packet arrives. the data can still be buffered in the OS so you can imagine if you were just looking at a very large transfer with a very small amount of missing packets and you were only worried about the overall transfer time then this is not really 'slower'. but if you are very worried about the time it takes for small amounts of data to reach the other side then this can be 'slower'. a good example of this would be some kind of request-response protocol.
More options
Context Copy link