nsfmc

you listen to the audio and try to determine if you can either read or understand the audio. it's repeated twice, once formally and once casually, so good for listening practice, but also good for adding new words/phrases into your vocabulary since some of them might be familiar but maybe reconfigured in a way that you don't normally see in your practice.

this is probably only useful if you've started learning a very small amount of grammar, know hiragana well enough to make furigana useful, and have started memorizing enough kanji/vocab to make 'overheard train chatter' useful. probably, generously, something maybe 3-6 months into your japanese language journey, so not good for bootstrapping.

For every HTML 2.0 you might have tried, you were just as likely to have got stuck in the dead-end of Flash.

i'll just say, and i understand this is not the point of the article at all, but for all its faults, if you got in on flash as earl as html 2.0 and you were staring at an upcoming dead-end of flash in say, 2009, you also knew or had been exposed at that time to plenty of javascript, e4x and what were essentially entirely clientside SPAs, providing you a sort of bizarro view of the future of react in a couple of years. honestly, not a bad offramp even if flash itself didn't make it.

unless... maybe... https://www.ox.ac.uk/news/2017-03-28-tetris-used-prevent-pos...

my read as a US person is that math academy is optimized towards students who would otherwise be well served by an in-person supplemental math program. at the earlier grades for math academy (grades 4-5 etc) the main competition i've encountered are in person programs like AoPS, Russian Math, or Kumon. The prices for those range between $450-$100/mo and for a student or student and parent combo that may be looking to supplement their math classes or for somebody who needs to home school for a period of time, mathacademy at $50/mo is a steal.

i suspect that a lot of people applying to a lot of jobs and not hearing back are applying via an inbound webform or even, unknowingly, through an agency or intermediary like Indeed. my experience (on both sides of the hiring process) is that inbound applications are generally considered fairly low-signal unless they are paired with a highly-specific job req or come with a referral attached. i'd be curious to know how you applied to those companies to get such good response rates.

as an anecdote, for some jobs i've applied to, i received an internal referral to a specific recruiter and it just never went anywhere (although i could often observe the recruiter looking at my linkedin profile from linkedin's own upgrade promo emails). in many cases, i would have preferred a 5 minute call to see if i was even a good fit so as to cross a job off a list.

i'm not anybody special, also degree and over a decade in, but it seems like a market where i could be connected 1-1 to an internal recruiter that is actively filling a role but where on several occasions i was ghosted is just a different sort of environment than what you're describing.

this has happened to me once, appropriately enough with some hauntology tracks. the songs are weirdly enough still available in the ios app to play but not to download via the web. presumably they're still somewhere in bandcamp behind a boolean, but i never got around to downloading them (to be fair, i have them on a _cassette_ that originally included bandcamp codes, so i mean, i really can't complain, i knew what i was getting into)

the unicode angle is great for moving sparklines back into the terminal especially making it into a postgres function, since it makes so much ad-hoc analysis easier at the point of query.

it also solves the problem where sparklines feel useful to somebody reading a summary and not so useful to somebody that's just exploring data "in the moment." having it be available in postgres is brilliant.

a bit of shameless self promotion, i had done something similar ages ago but with a custom webfont and some small js to handle scaling of the input dataset, solving the problem of unicode graph characters not being baseline aligned well. http://nsfmc.github.io/chartjunk/ (turns into dust while looking at the last commit date)

i think the library's approach to DI is pretty neat (and meets a team where they are which is worth a lot), but i think you're running into an issue where people are saying that instead of working around the realities of your codebase, team and testing needs, you should have done something like this.

  const useCountdownValue = initialTime => {
    const [time, setTime] = React.useState(initialTime);
    React.useEffect(() => {
      const interval = setInterval(() => {
        setTime(t => t - 1);
        if (t === 1) clearInterval(interval);
      }, 1000);
  
      return () => clearInterval(interval)
    }, []);
    return time;
  }

  const Countdown = ({time}: {time: number}) => {
    return time > 0 ? <>Time left {time}</> : <>Done</>
  }
  
  const ActualCountdownInContextSomewhere = () => {
    const remainingSeconds = useCountdownValue(60)
    return <>
      <Countdown time={remainingSeconds} />
    </>
  }

  <Countdown time={-1} />
  <Countdown time={0} />
  <Countdown time={1} />

but like, whatever works for your team! i think it's fair to argue for either direction, but neither is zero-cost unless you have buy-in for one direction vs another from your coworkers, which honestly is all that matters especially if you have to eat lunch with them occasionally.

i'm pretty sure this is allowed as Transformative Use (in the usa) (unless the science fiction book was full of patented inventions still covered by patent law then i think maybe you're no longer in the clear.)

https://en.wikipedia.org/wiki/Transformative_use

oooooph

i'm not sure if you're serious, but the california public utilities commision's public advocates office (what a mouthful) describes california's rates as generally higher than most of the nation[0], with southern california's rates being highest (with both increasing).

you can see, for instance san diego's rates [1] which are $0.38/kWh in the winter and $0.48/kWh in the summer. for context, this means if i pay 11 dollars in electricity generation (because i'm part of a municipal electric generation coop), i'm still paying $36 for distribution/transmission/etc, which is $47 for 106kWh used or ~$.44/kWh which is roughly what electrify america charges ($.48/kWh) when i go to 'fill up my car.' as far as i can tell from talking to people, this is is more than most people anywhere in the country (including hawaii) pay for their electricity.

[0] https://www.publicadvocates.cpuc.ca.gov/-/media/cal-advocate... [1]: https://www.sdge.com/sites/default/files/regulatory/1-1-24%2...

i submitted a few shows to this and included a handful of stills to go along with it, it was an a+ contribution process, but there was still a lot of work i didn't do, so i can appreciate why there's a backlog. great project!

here's another piece in the issue that addresses your concern https://www.newyorker.com/magazine/2023/11/20/a-coder-consid...

i won't speak for gabe, but i (hispanic) had two distinct (and memorable) instances at mit where a classmate told me, to my face, that the only reason i was at mit was because i was hispanic. one passed it off as common knowledge in a group conversation, the other said it to me, unprompted, while we were talking to each other. they were assholes, to be sure, but it's not some hypothetical scenario.

that site is not really suitable for mass consumption, iirc it used to even have a caveat that said it really only existed to make certain leaf nodes sharable. it's basically a personal zettelkasten, but this update was also sent to his patreons as well, where the writing just ended up being whatever your email client made of it.

that's great context, i started having a flashback to hypepitch around thegrid.io when i started reading the copy.

an earlier newsletter shows this distressing graph of teen suicides and, shocker, boys are doing terribly, a suicide rate at least three times more than girls, but the author seems to find the 34% delta less distressing. the data is unsettling, but the author doesn't remotely confront the question about why 2017 is a post-2010 maxima for both boys and girls. (which would probably change the boys delta to at least 60 percent and possibly undermine the premise that social media exclusively harms young women)

https://jonathanhaidt.substack.com/i/101249738/increases-in-...

i believe the slideshow issue is actually a limitation of the apple tv's apis having changed around the time that the aerial screensaver came out and not getting updated to support the old screensaver functionality

great, now you have a native client but now your customers will expect that it works offline, so now you're building a sync system for your app.

additionally, if your app has any design/layout issues, now you need to hire for people that can do visual layout using tcl/tk.

If your app consumes rich text, you're in for a real treat as you are now on the hook for figuring out how to manage that as well (in addition to serialization, you also need it to likely be emitted into some web friendly format anyway. if your app consumes some feed (say new-user facing features or whatever), you need to find some meta-format that your app can use to layout that content (or i mean, i guess you can use a webview, but then...)

if your customers are enterprise-ey, you are now dealing with some overzealous IT dept that is skeptical of your application running with user permissions.

if you're trying to push a fix/update to your users, you now need to build infra around deploying new apps as well as customer support determining if users are somehow running old versions when they report a bug.

the web is a total "mo money, mo problems" situation, but i think people dismiss how many problems the web solves for your developers on a day to day basis: easy to push updates, simple to whitelist your app's domain on some restricted network, easy add dynamic content/layout to portions of your app with stored content, (less sync resolution issues because your app probably requires network to operate) etc. Native apps have their own issues, they're great, don't get me wrong, but people demand a lot from basic apps of any stripe these days, and as those requirements increase, so does the amount of complexity that developers need to manage at all stages of the pipeline.

What a Misunderstanding! ?

I can't tell if the author has done a real analysis course before, but if they haven't that's the one they should choose next... I don't see the utility of re-doing calculus or linear algebra if the author is already strong in both.

having (somehow) completed many of these requirements for my 18c degree, i would say that analysis is not necessary if your interest is actually applied math. There's a great line in rudin's preface that says that his approach is ~"pedagogically sound at the expense of being logically incorrect," and recommending analysis for somebody that's not looking to mainline a pure math degree to me feels "pedagogically unsound (but logically correct)"[0].

I took analysis and i appreciated it, but i really loved the applied classes in my degree: 18.310, 18.311, 18.781 (theory of numbers) along with algebra 18.701/702. If you haven't taken a higher-level algebra class, it will let you know if analysis is right for you because you'll brush up against the edges of it without (what i consider to be, at least) its hallmark punishing density.

There are other great electives in math at mit, shop around the 18.4* classes and dial in by interest, most of them only require a prereq of 18.02/18.03/18.06 and you can sort of figure the rest out along the way.

Something to be aware of is that for a while 18.310 didn't have a dedicated instructor, so it really was all over the place. 18.311 was also somewhat hastily structured the semester i took it, but it is actually pretty good material.

You may find that after you've done all these classes that you are actually interested in pure math and at that point i would suggest looking at 18.100b (analysis), 18.700 (linear algebra), 18.100c(real analysis), 18.901(topology) and the rest of the "hard math classes," but i really do think that you'll find that the rationale for those classes doesn't click if you haven't taken a few classes like 310 or 701 first.

just my two cents! good luck, have fun!

[0]: this is the actual quote, it's in the preface rudin's principles of mathematical analysis 3rd ed. which is the 'textbook' for 100b.

i perhaps have diminished standards, but i found steam link over ethernet to be totally serviceable for playing through most of hyper light drifter on a controller. the input lag was fairly minimal and the computer was already quite long in the tooth, but to your point, i wasn't getting 60fps, that's for sure (probably realistically it was a firm 30fps).

i don't doubt that consoles will remain useful, but i think that services like this will satisfy a pretty legit niche for a vast swath of games that aren't really dependent on low latency input (i.e. puzzle/turn based/rpg/simulators/board game conversions) and that are often 'discovered' by people finding letsplays on youtube.

for some variety of mmorpgs, i can imagine devs being excited about the reduced surface area for cheating/exploits. for somebody like me that uses a mac, i'm looking forward to playing a version of Civ that doesn't cause my laptop to sound like it's about ready to take flight. i don't think the idea is meant to replace consoles, though more, it seems like a way to grease the wheels of commerce and get people playing (and buying) games that they've been traditionally priced out of because of the not-insignificant startup cost of building and maintaining a gaming pc/console.

i tried this and found it to be pretty satisfying. i played entirely on a macbook while traveling during my winter vacation. It worked surprisingly well even with mixed wifi, i would compare it to the experience you get from using PS4 Remote Play or Steam Link over ethernet.

The only time i found it lagging was similar to the times when both of the above lag: during particularly complex visual scenes (i.e. you're circle strafing around a target and the entire screen is constantly redrawing). I thought it was great for playing a game casually: i.e. story mode. Lots of people use that phrase as a put-down, but the system is well suited for a game like ACO where you are mostly being tactical, planning, exploring, and moving the story forward.

I think of a game like 2016 hitman: i hesitated to install 30G of it to my ps4, but if you told me i could drop into the demo/prelude in less than a minute, even at 720, it's a very appealing concept for somebody like me that plays video games the way other people watch netflix while they're eating dinner: basically whenever i have some downtime and want to dip into a story or mechanic i like for ~30 mins.

in california, you can apply to get your first year of community college free ( https://home.cccapply.org/money/california-college-promise-g... ), there's an assembly bill ( http://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?... ) which would add a second year to that.

ok, i love that series, but i am not sure i would necessarily recommend it as a starting point (plus i felt like i had maxed out on recs) but i strongly endorse it as a great series!

Adam Curtis' short "Oh Dearism" (https://thoughtmaybe.com/oh-dearism/) 12 minute mini-documentary is a good ramp up to his longer movies and is a critique of the failings of television news media in the late 20th century. My favorite of his is "Bitter Lake" (https://thoughtmaybe.com/bitter-lake/), a haunting "how did we get here" view of the middle east. His 3 part miniseries "All Watched Over By Machines of Loving Grace" (https://thoughtmaybe.com/all-watched-over-by-machines-of-lov...) is a pretty good examination of 1970s silicon valley ruh-roh cybernetics culture and how it collided with the rest of the world vis-a-vis objectivism, finance, etc.

Somebody in this thread mentioned Wiseman's titicut follies, a lesser known but equally fascinating documentary of his is called The Store (http://www.zipporah.com/films/19) and just follows and examines the goings on at the Nieman-Marcus flagship in Dallas during holiday season of 1983.

Another favorite documentary that's more like watching a really good lecture is Thom Andersen's Los Angeles Plays Itself (https://en.wikipedia.org/wiki/Los_Angeles_Plays_Itself ). it's a survey and exploration of the history of Los Angeles and its relationship and portrayal in hollywood and pop culture in general.

around 2005-6 i considered switching over to dvorak, but trying it out, doing tutorials, etc, never had much of an effect. i would say i was a pretty fast qwerty typer, but i hadn't learned to properly touch type. instead, i was fast enough to not need to see the keys once i had started typing but it relied on a bunch of terrible hand positions.

around 2007, i broke my right thumb by hitting a pothole and that put my typing (and all my bad habits) in a weird spot. At that point i decided to try one-handed dvorak and failed, but i found that two handed dvorak was very doable, so i spent about two weeks of my convalescence doing http://web.archive.org/web/20040803054502/http://www.gigliwo... on and off and watching movies. About a week in, i swapped all my keycaps to make the transition easier, but found that i missed having the home row nubs on f/j (u/h in dvorak) because at this point i had started actually relying on them for positioning as i had finally learned to touch type.

at this point, about two weeks in, i was solidly able to to use my modded keyboard, but i still peeked at the keys occasionally so that i would re-position my hands correctly at rest. it took me about a full month of occasional typing in dvorak while i was in physical therapy to "make the switch" mentally.

this probably marks my eleventh year typing in dvorak, by far the first few months were the most awkward, especially since at the time i was using vi and some of its key commands are optimized for qwerty while others have semantic meanings. the most annoying thing in those early months is not so much the letter positions, but recalling where all the special characters live: .,'-;/=[] are all characters that you possibly associate with certain regions of the keyboard, but in dvorak they are a punctuation-diaspora, scattered across the edges of the keyboard. by around six months in, i think i was back to my regular typing speed, but i believe i exceeded my qwerty speed easily a year into dvorak simply because i had finally learned to properly touch type and had developed muscle memory for all keys.

how hard is switching back? like you, i wondered about being keymap-bilingual, but it turns out to not be all that important day-to-day. i definitely type noticeably slower when i go back to qwerty keyboard (usually because i'm using a friend's computer). in practice this happens remarkably infrequently. one of the strange things of learning dvorak but leaving your keycaps in qwerty is that i look at a `t` but i "see" a `p`. it's strange, it's the closest thing i have experienced to having a sixth sense. because ios doesn't allow you to remap its keyboard to dvorak, i use qwerty there all the time and it's not at all disorienting.

One place, however, where you do need to remain somewhat sharp is pc gaming, where games inconsistently pick up custom keymaps (i.e. in js this is the distinction b/w evt.key and evt.code). in most cases, you just end up using qwerty because it's a pain to remap (this was especially true for games like dota2, but starcraft 2's grid layout remedies much of that). again, many apps like illustrator/photoshop also have keyboard shortcuts located for physical reasons and less for logical "this letter is short for that idea" reasons, which can make the switch harder

that said, it's doable and i believe in you! at the very least, you'll become a better typer and develop better habits from your time in dvorak. good luck!

curious why you are scraping instagram for this purpose and not something like flickr which has a reasonable public api and tagged creative commons licensed images that are suitable for your ML purposes. at the very least, it's worth investigating archive.org's many freely licensed archives for this sort of thing.

as somebody that has fielded numerous emails from friends asking me to remove tagged photos of them from flickr, i sort of wonder about the ethics of harvesting these sorts of images from instagram, a community whose norms sort of revolve around semi-public sharing of photos. I don't doubt that there's some rationale for harvesting the images from ig, but aside from thumbing your nose at their TOS, it feels like it's a greater violation of trust to harvest your friends and strangers photos for an ML project without their informed consent.

at the very least, it's worth considering pointing your app's gaze at a set of images licensed for any purpose whatsoever rather than ones that are explicitly licensed All Rights Reserved by their respective photographers.

i think the point is that all react-likes were (with high probability) infringing on at least one of facebook's patents. People were free to move to vue or whatever, but licensing wouldn't have mattered because they would still have been infringing on a software patent that didn't belong to them and that they hadn't licensed.

the closest comparison i can think of is like the mp3 patent situation. Up until this year you were free to use LAME non-commercially even though its technology was covered by patents held by fraunhoffer & thomson.

even though vue's api is different from react's in lots of significant ways, it's not really clear that vue doesn't infringe on some of facebook's patents anyway (at least i don't know what the patents are) but the impression you get from reading every single post by an ex-fb (or current) employee is that using other react-like library put you in a similar legal situation as commercial users of LAME: while it was highly unlikely that facebook would take any action against your hackathon heroku app, it was never clear that any of those libraries (or their users) were free of actual risk. At least users of react were assured they weren't going to be sued regarding any of those vdom patents held by facebook.

This reads like a strong warning to the already cautious payment processors that legalization hasn’t eased any of the uncertainty around processing payments for dispensaries or made banking any simpler for them despite a massive market having just opened.