|
EVERY MORNING I WAKE UP AND OPEN PALM SLAM 36 HYDROPONIC HEADS OF LETTUCE
|
#
?
Aug 2, 2022 05:34
|
|
|
|
| # ? Apr 25, 2024 07:44 |
|
|
the secret is out nmeofthestate 2 hours ago | root | parent | next [–] When the article is about something you're familiar with, you do start to wonder if all HN comment threads are similarly topped by blowhards missing the point. reply
|
|
#
?
Aug 2, 2022 11:14
|
|
|
AinderS 1 hour ago | parent | context | favorite | on: What the Hell Is Up with Dilbert? Simple. Workplace complaints and management priorities have changed, and Dilbert followed. When Dilbert started, the government wasn't yet training employees how to "interrupt whiteness" [1], or Coca-Cola how to "be less white" [2], or Cigna simply forbidding hiring whites [3], nor did academia require mandatory diversity pledges for new hires and promotions [4]. Dilbert is doing what we are told art is supposed to do - hold a mirror to society. Do you like what you see? [1] https://www.city-journal.org/seattle-interrupting-whiteness-training [2] https://www.newsweek.com/coca-cola-facing-backlash-says-less-white-learning-plan-was-about-workplace-inclusion-1570875 (Note that in all the "debunkings" of this story, Coca-Cola never claims the presentation wasn't shown by their hired diversity experts as part of its diversity training. Merely that Coca-Cola the company didn't require those specific slides. But the slides are completely in-line with rhetoric championed by diversity experts routinely hired to train employees.) [3] https://www.washingtonexaminer.com/news/cigna-critical-race-theory-training-dont-say-brown-bag-lunch-mindful-religious-privilege [4] https://www.wsj.com/articles/the-universitys-new-loyalty-oath-11576799749
|
|
#
?
Aug 2, 2022 15:44
|
|
|
The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > hn thread: blowhards missing the point
|
|
#
?
Aug 2, 2022 16:58
|
|
|
dilberts holding a mirror to scott adams nose
|
|
#
?
Aug 2, 2022 17:29
|
|
|
oceanghost 2 hours ago | parent | context | favorite | on: What the Hell Is Up with Dilbert? Yep. I was once advised by HR to "try not to offend anyone." I asked the HR woman, "How do I know what offends any particular person?" She looked genuinely stumped.
|
|
#
?
Aug 2, 2022 17:46
|
|
|
toyg 2 hours ago | parent | context | prev | next [–] | on: Coding as a greybeard > I see no prospect of it happening in the next 10 years It's happening right now: the field is being flooded by data-kids who reason in different terms. They are statisticians first and programmers later. Their models can do stuff that bit-pushers like me will always struggle with. 10-15 years from now they'll run everything, and you won't be able to code a helloworld without specifying a ML model. reply
|
|
#
?
Aug 2, 2022 18:18
|
|
|
NihilCredo posted:there has been only one so far, hasn't there? nancy reagan
|
|
#
?
Aug 2, 2022 18:54
|
|
|
Standish posted:oceanghost 2 hours ago | parent | context | favorite | on: What the Hell Is Up with Dilbert? in fairness if someone asked me this I’d probably give them a confused look but not for the reason this guy thinks
|
|
#
?
Aug 3, 2022 00:44
|
|
|
kitten emergency posted:in fairness if someone asked me this I’d probably give them a confused look but not for the reason this guy thinks I can't imagine even talking with HR beyond the on-boarding during my first day.
|
|
#
?
Aug 3, 2022 00:46
|
|
|
i assume he was talking to hr because someone complained to hr about him and that was the moment where the hr person realized he was an idiot
|
|
#
?
Aug 3, 2022 01:22
|
|
|
I need someone's own personal EULA before I engage with them
|
|
#
?
Aug 3, 2022 01:35
|
|
|
MononcQc posted:I need someone's own personal EULA before I engage with them ah, a googler
|
|
#
?
Aug 3, 2022 02:19
|
|
|
LeonTheremin 6 hours ago [flagged] [dead] | parent | prev | next [–] >schizophrenic product management Probably because good employees are driven away by Electronic Harassment (schizophrenia is fake and the real issue is beaming AM signals to people's heads and bodies to hurt them until they submit to Mind Control). I can introduce you to one such ex-employee if you want. Happens a lot more than people think and Microsoft itself is clueless still.
|
|
#
?
Aug 4, 2022 06:07
|
|
|
DARPA posted:I can't imagine even talking with HR beyond the on-boarding during my first day. if you're a functional human being, you probably started to figure out how to not offend people slightly before you were a teenager. like, even if you had way more to learn, that's where it should have loving started
|
|
#
?
Aug 4, 2022 06:14
|
|
|
Plorkyeran posted:LeonTheremin 6 hours ago [flagged] [dead] | parent | prev | next [–] Redmond Syndrome.
|
|
#
?
Aug 4, 2022 15:04
|
|
|
HN truly has A Guy for everything.quote:gkbrk 4 hours ago | parent | prev | next [–]
|
|
#
?
Aug 4, 2022 15:44
|
|
|
tracecomplete posted:HN truly has A Guy for everything. Is HTTP/2.0 a google thing? I thought the thing Google was trying to push was QUIC?
|
|
#
?
Aug 4, 2022 16:29
|
|
|
HTTP/2 is derived from SPDY, which was a Google thing. QUIC is an underlying UDP-based layer developed by Google that's being used by HTTP/3 and is largely "HTTP/2 over QUIC". HTTP/2 added multiplexing over TCP, which was better but had bad failure conditions; HTTP/3 mostly improves things by doing multiplexing in the application rather than relying on TCP, and by so doing reduces data loss/retransmission for unreliable (read: mobile) clients and nontrivially reduces latency for most clients. It is a good thing for everyone, not just Google. That particular A Guy is...uh...very A Guyish.
|
|
#
?
Aug 4, 2022 17:33
|
|
|
if only the gains in latency wouldn’t be eaten up immediately by javascript and ads
|
|
#
?
Aug 4, 2022 18:27
|
|
|
that’s why all you actually need is HTTP/1.1, if you block ads and turn off JavaScript everything is plenty fast
|
|
#
?
Aug 4, 2022 18:54
|
|
|
eschaton posted:that’s why all you actually need is HTTP/1.1, if you block ads and turn off JavaScript everything is plenty fast Fast to show you a blank screen most of the time
|
|
#
?
Aug 4, 2022 21:36
|
|
|
Even faster if you turn off your monitor
|
|
#
?
Aug 4, 2022 21:41
|
|
|
Armitag3 posted:Fast to show you a blank screen most of the time and nothing of value was lost
|
|
#
?
Aug 5, 2022 00:18
|
|
|
It's just really funny that he's claiming "HTTP/1.1 has worked for ages and it will continue to do so for the foreseeable future" literally in response to someone encouraging switching off support for HTTP/1.1
|
|
#
?
Aug 5, 2022 00:26
|
|
|
some guy on HN complaining about how easy it is to get banned/probed on SA today, with a strong intimation from his comment history that it's because he's a chud face us, coward!! 4lokos basilisk posted:if only the gains in latency wouldn’t be eaten up immediately by javascript and ads block better, friend (my only tryhard home technology is an always-on VPN for mobile/laptop and so everything is piholed, it is wonderful.)
|
|
#
?
Aug 5, 2022 00:27
|
|
|
http 2 was basically not worth anyone’s time. it’s massively more complex and its main feature (multiplexing) makes tail latency much worse due to head-of-line blocking. http/3 and quic on the other hand is a genuine improvement and the web will be better off for it
|
|
#
?
Aug 5, 2022 10:23
|
|
|
tracecomplete posted:block better, friend i too have pihole set up and i browse with ublock origin. websites are still bloaty imo because a lot of the functionality is tied to javascript libraries doing something in an overcomplicated manner while what i really care about is images and text. sure, janitoring every website’s js imports and scripts one by one can get better results but this is not a viable way of life for 99.99% of internet users
|
|
#
?
Aug 5, 2022 14:59
|
|
|
tracecomplete posted:some guy on HN complaining about how easy it is to get banned/probed on SA today, with a strong intimation from his comment history that it's because he's a chud link, or would that be touching the poop?
|
|
#
?
Aug 5, 2022 15:02
|
|
|
4lokos basilisk posted:websites are still You can totally make a monster of a website that's actually slow to use but I just don't see them in the wild very often? The worst offender on my computer is gmail, as it has been pretty much since gmail was invented. I wouldn't be surprised if this experience isn't representative, but sites' JS loads just don't impact my personal experience very much. Neito posted:link, or would that be touching the poop? It's in here, the full "SA should have gone for recurring revenues" subthread is a time: https://news.ycombinator.com/item?id=32339314
|
|
#
?
Aug 5, 2022 15:09
|
|
|
tracecomplete posted:I hear this a lot and tbh don't see it all that often. My new laptop is a future computer from beyond the moon but before that my normal browsing machine was a 2014 MBP and it's capital-F Fine; if I'm on my 2008 MBP (dedicated Logic Pro machine usually) it's definitely slow but that computer is also ancient. maybe i am a grognard but “fine” is not “instantly blazing fast” which it should be at this point. stuff like rendering pure text and images should be blazing fast also on 10y old machines. like one of the data points is http://www.csszengarden.com: does your js interactive website load as fast and react as snappily as some of those pure css examples? probably not maybe my impression is also colored by being in the eu and yeah loading websites based in the us will never be as fast. even more so when i do it on mobile im absolutely a grognard though
|
|
#
?
Aug 5, 2022 16:25
|
|
|
4lokos basilisk posted:maybe i am a grognard but “fine” is not “instantly blazing fast” which it should be at this point. stuff like rendering pure text and images should be blazing fast also on 10y old machines. And for me a page that slowly loads different parts of the screen is a lot more frustrating that waiting a moment for backend loading and then displaying everything. It's so common now that a page loads elements at random and it can be easy to misclick because things jump around or you may see the content but it's unresponsive until it loads a ton of functionality.
|
|
#
?
Aug 5, 2022 16:51
|
|
|
lobsterminator posted:And for me a page that slowly loads different parts of the screen is a lot more frustrating that waiting a moment for backend loading and then displaying everything. It's so common now that a page loads elements at random and it can be easy to misclick because things jump around or you may see the content but it's unresponsive until it loads a ton of functionality. the contrast here on sa is so stark lol. each page loads completely in under a second... except for all the twitter, youtube, and tiktok embeds which when make the page unusable for the next 10 seconds while they all inject and load their loving SPAs and analytics framework garbage.
|
|
#
?
Aug 5, 2022 18:54
|
|
|
HTTPs 2 and 3 are stupid poo poo for idiots the hackernoid had a broken clock moment
|
|
#
?
Aug 5, 2022 23:06
|
|
|
QUIC and HTTP3 are extremely google-quality ideas: let's break a universally accepted standard and gently caress up an untold amount of existing infrastructure so we can squeeze out mild performance gains in the use cases google cares about
|
|
#
?
Aug 5, 2022 23:11
|
|
|
what if we took library-level components in multiple languages, and made them also have to reimplement TCP as well
|
|
#
?
Aug 6, 2022 02:28
|
|
|
they should simply make dedicated javascript accelerator cards that you put into your computer easy peasy
|
|
#
?
Aug 6, 2022 03:13
|
|
|
from working in HTTP proxy support for far too long, the main hallmarks of http/2 and http/3 is that their implementations are both much stricter than http/1.1 and do gently caress all to provide detailed errors. everything's just GENERIC_PROTOCOL_ERROR and you gotta just squint at the request log to try and determine that it's specifically that the content-length is mismatched or the server included a Host header in their response. you'd think that google would be able to track common stack traces and fill out the error details with their whole near-monopoly on desktop web clients but nope, the only way you're figuring that out is digging through the chromium or firefox source
|
|
#
?
Aug 6, 2022 03:14
|
|
|
Best Bi Geek Squid posted:they should simply make dedicated javascript accelerator cards that you put into your computer Don't AMD microprocessors have dedicated instructions for "cast 64-bit floats to 32-bit signed ints, do binary arithmetic, cast the 32-bit signed int result back to 64-bit float", which is how binary operators work on numbers in JavaScript? Or something like that...
|
|
#
?
Aug 6, 2022 10:29
|
|
|
|
| # ? Apr 25, 2024 07:44 |
|
|
Doom Mathematic posted:Don't AMD microprocessors have dedicated instructions for "cast 64-bit floats to 32-bit signed ints, do binary arithmetic, cast the 32-bit signed int result back to 64-bit float", which is how binary operators work on numbers in JavaScript? Or something like that... idk about amd but arm64 has something along those lines, although it's much more limited in scope than what you describe (just a specialized float->sint32 conversion which does everything exactly the way a JS interpreter or jit would prefer it to)
|
|
#
?
Aug 6, 2022 11:15
|
|
