• 2 Posts
  • 350 Comments
Joined 2 years ago
cake
Cake day: June 20th, 2023

help-circle
  • “I want predictable behavior for all possible inputs” is hardly a requirement that requires a fortune teller to see coming.

    JavaScript has a particularly insane stdlib because this language wasn’t designed, it is a botched chimera with deformities so severe it should have died 15 times over but people just won’t let it.

    Then to rub salt in the wound this horrific mess became the most popular language in the world by virtue of being the only language for the most popular application ecosystem in the world (the web). So the cancer is spreading and now you can find JavaScript in servers and fucking desktop environments and now your windows start menu takes five seconds to load because fucking react.js is loading the 75 polyfills necessary to make up for the fact that JS’s “standard” library looks like it was designed by 3 cocained-up gibbons.



  • It’s one of a plethora of scripting languages from the '90s which were designed to be the antithesis of “fail fast” and kept going no matter what.

    I guess what with C/C++ being the Mainstream Option at the time, not having to deal with a strict compiler must have felt like freedom. As someone who has had to maintain, cleanup and migrate ancient PHP code, I call it folly. That mindset of “let the programmer just do whatever and keep trucking” breeds awful programming practices and renders static analysis varying degrees of useless, which makes large-scale refactoring hard to automate which is just amazing when your major versions aren’t even remotely FUCKING BACKWARDS COMPATIBLE.

    PHP’s original design is just fundamentally atrocious. It became popular in large part because unmaintainable code is usually someone else’s problem.

    A language that I would definitely use for server-side rendering and that was already good from its first stable release is Go. It was thoughtfully designed and lends itself really well to static analysis, while still being easy to write and decently performant.


  • There were definitely a couple literal demented sociopath rapists in the mix. What changed wasn’t the law, but the political context and institutions.

    It took decades for the GOP to systematically destroy faith in institutions.

    It took years of Trump presidency followed by a strong reaffirmation of popular support in the last election.

    It took Obama and Biden abdicating their duty to their electorate by respectively refusing to nominate a new Justice and refusing to prosecute Trump for sedition.

    It took the media failing their duty to inform voters of Trump’s past, intentions, and state of mind.

    It took decades of slow work by the right to reframe the media landscape to be less truthful and more obedient.

    It took social media and their algorithms to galvanize fascism.

    It took an entire cold war and war on terror to normalize an absolutely abnormal and near insurmountable militarization of domestic law enforcement.

    The US constitution is not to blame. That’s a cop-out answer, a lame scapegoat. America wouldn’t be saved by passing amendments alone. The rot goes far deeper than that. Just like the 13th amendment didn’t do much to fix the system of racial injustice the US was built on. If it was just a matter of wording, a silly loophole, it wouldn’t have worked. It worked because the vast majority of Americans abdicated their allegiance to the Bill of Rights, to Human Rights, and to Democracy.



  • The central bank facilitating electronic cash flow makes so much more sense than letting random foreign corporations siphon billions in profit they clearly don’t deserve in the slightest from your economy.

    Good on Brazil for breaking free. There’s finally been some push here in the EU for EPI/Wero, but progress has been frigid and online payment processing remains extremely fragmented to the point that if I buy something online outside the Benelux with a small vendor, chances are very high I will have to fall back to an American payment processor, which is insane.


  • And literally not a single one of them is useful for the purpose of quick, efficient, and secure transactions.

    Blockchains are slow and inefficient by design, since they need to build consensus. On any sufficiently popular blockchain, transactions are either fast or secure, never both.

    The “fix” that the crypto industry has come up with is to re-invent banks, except with even more crime and virtually no regulations. Now you’re just entrusting FTX with your coins to enjoy “immediate” transfers, how could that possibly go wrong?



  • If I am not mistaken the 47.0.0.0/8 ip block is for Alibaba cloud

    That’s an ARIN block according to Wikipedia so North America, under Northen Telecom until 2010. It does look like Alibaba operate many networks under that /8, but I very much doubt it’s the whole /8 which would be worth a lot; a /16 is apparently worth around $3-4M, so a /8 can be extrapolated to be worth upwards of a billion dollars! I doubt they put all their eggs into that particular basket. So you’re probably matching a lot of innocent North American IPs with this.



  • Americans: “Best we can do is one large-ish peaceful protest a couple months ago. Ah well, we tried everything, and we’re all out of ideas.”

    Thousands of kids died for Americans’ right to bear arms and yet when the Gestapo jumps out of an unmarked van to kidnap their neighbor the most the average American can manage is to duck and film a TikTok. Absolute disgrace of a country.

    “American ideals”? All bark, no bite. You’ll spend billions making horny patriotic Hollywood movies about how Great and Free you are, but a couple masked bitch-ass Nazi shows up to round up the brown people and the whole neighborhood is suddenly like pwease don’t huwt me mistaw officaw, wight this way, also pwease fuck my wife.

    Your country disgusts me. Not so much because 30 % of y’all are actual Nazis. But because virtually everyone else is a complete coward about it.


  • azertyfun@sh.itjust.workstolinuxmemes@lemmy.worldAny more?
    link
    fedilink
    arrow-up
    15
    arrow-down
    1
    ·
    9 days ago

    Classful IPv4 was obsoleted 32 years ago. Only 8 years left before it’s literally older than a standard career.

    It’s fascinating the sheer inertia that leads formally-trained IT professionals to use and perpetuate such profoundly useless and obsolete nomenclature. You’d think that having an incorrect use of the term “class A” and not having any use for classes B and C would tip off academia that they should cordon off classful networking to the “History of Computing” course next to ARPANET.

    Maybe next time someone refers to 10.0.0.0/8 as a Class A network I’ll refer to it as the ARPANET Network. That’s only very slightly more anachronistic (3 years).



  • Reading the article, where did you get “audience rewards” == “maximal extraction of cash from the audience”?

    IMO having a very profitable game that will comfortably fund your studio for the next 5-10 years AND that has universal critical acclaim and a devoted fanbase is reward enough. You didn’t lose because you didn’t make the most money out of all your competitors.

    Different games have different audiences. Some people want arcade slop and slot machines to play with friends, they were never going to play BG3 or E33 anyway.

    Important to the conversation as well is the fact that plenty of live-service games have recently failed spectacularly. Remember Concord? Within the industry, that is a clear signal that very high budget online slop isn’t as risk-free as previously assumed, which makes ambitious narrative-driven single player games an interesting diversification strategy for studios.

    It’s not either or. Executives could spend 100M€ on “nearly guaranteed” online slop, or 80M€ on online slop and 20M€ on a good narrative game. And the critical and commercial success of games like BG3 and E33 are definitely moving the needle.
    Especially when micro-economically, there are diminish returns when scaling dev teams. It’s kind of obvious but the first million euros does a lot more for a project than the 100th million. That further strengthens the case for a move away for big players from ONLY funding live-service slop.


  • (They’ve already stated they won’t do Portal: VR because of the nausea issue.)

    I completely agree with your analysis, they would need to completely switch up the ambitions from a writing perspective for Portal 3 to make any sense. There are plenty of super interesting stories to be told in Aperture Labs, but I don’t think that Valve is structured to write any of them

    Valve has always been “gameplay/tech first, story second”, and it just happened that Portal 2 delivered unexpectedly well on the writing. But I don’t think they can make a game with gameplay/tech twice as ambitious as Portal 2, and at the same time double down on Portal 2’s amazing writing. They’re just human and most of the people involved have moved on with their lives; in fact Portal 2 was their last truly ambitious narrative-heavy game, and they had to hire the old writers as consultants to make Alyx (which I haven’t played but from what I heard the narrative wasn’t on HL2’s level).

    I’d love to be proved wrong but IMO there won’t be a Portal 3 for as long as Valve exists in its current form.


  • It’s one of my favorite games of all time, but I don’t think Portal 2’s basic formula would be culturally relevant if it was reused today. The quippy writing is very 2010s-coded (à la Guardians of the Galaxy), the gameplay is a bit too simple to be re-used as is in 2025, and the sweet&short linear storyline of Portal 2 would ironically be lacking ambition for a successor to Portal 2.

    Like all truly Great pieces of classic media, Portal 2 is a product of a skilled and truly passionate team getting together at the perfect time with the right idea, and reaching its public at a culturally relevant time.

    The Portal universe still has stories to tell, and there are still test chambers to solve, so I obviously wouldn’t complain if Portal 3 came out, but I understand why Valve wouldn’t want to make a barely decent game in the shadow of Portal 2.


  • A key feature of authoritarianism. Whether it’s Hitler, Stalin, Putin, or Louis XIV, keeping the court close like this is an absolutely essential part of holding on to power. For one they’re too busy with the king to have time to get bored and start scheming against him. For two the courtesans are around each other and competing for attention so they scheme against each other instead. We know that Trump listens to his advisors very haphazardly; it keeps them on their toes, constantly begging for attention (even if the end result is unbelievable political flip-flopping, that’s irrelevant to Trump himself).

    People have this image of the Third Reich as super organized, but in reality the top command was a complete mess as everybody was trying to backstab each other and to please Hitler who didn’t necessarily even have a clue what was going on. The utter incompetence of Nazi leadership was always going to cost them the war, but it did keep Hitler in power until the very end even though the outcome of the war was long considered inevitable by his own generals.

    Putin does the same. Remember the feud between Wagner guy and Shoigu? Putin intentionally encourages internal squabbles because it means in an environment where everyone mostly hates everyone, the only consistent loyalty is to him.

    Anyway, there’s plenty of reason to be concerned about Mr. biggest-nuclear-arsenal-on-the-planet going at a Hitler speedrun, but the only saving grace right now is that the whole thing is an inefficient mess and a large chunk (but not all) of them are too dumb to be truly dangerous. When he starts exclusively listening to his war hawks or the project 2025 guys… We’re fucked.




  • It can do both, lossiness is toggleable.

    If you’ve seen a picture on Lemmy, you’ve almost certainly seen a WebP. A fair bit of software – most egregiously from Microsoft – refuses to decode them still, but every major browser has supported WebP for years and since superior data efficiency compared to JPG/PNG means is already very widely used on the web. Bandwidth is not that cheap.