Comment by azangru

Comment by azangru 2 days ago

58 replies

> instead of learning Web standards, rather ships Chrome alongside their application

I am confused.

- The "shipping Chrome alongside their application" part seems to refer to Electron; but Electron is hardly guilty of what is described in the article.

- The "learning web standards" bit seems to impune web developers; but how are they guilty of the Chrome monopoly? If anything, they are guilty of shipping react apps instead of learning web standards; but react apps work equally well (or poorly) in all major browsers.

- Finally, how is Chrome incompatible with web standards? It is one of the best implementer of them.

quacksilver a day ago

Devs, particularly those with pressure to ship or who don't know better, unfortunately see 'it works in Chrome' as 'it works', even if it is a quirk of Chrome that causes it to work, or if they use Chrome related hacks that break compatibility with other browsers to get it to work in Chrome.

- Sometimes the standards don't define some exact behavior and it is left for the browser implementer to come up with. Chrome implements it one way and other browsers implement it the other way. Both are compatible with the standards.

- Sometimes the app contains errors, but certain permissive behaviors of Chrome mean it works ok and the app is shipped. The developers work around the guesses that Chrome makes and cobble the app together. (there may be a load of warnings in the console). Other browsers don't make the same guesses so the app is shipped in a state that it will only work on Chrome.

- Sometimes Chrome (or mobile Safari) specific APIs or functions are used as people don't know any better.

- Some security / WAF / anti-bot software relies on Chrome specific JavaScript quirks (that there may be no standards for) and thinks that the user using Firefox or another browser that isn't Chrome or iOS safari is a bot and blocks them.

In many ways, Chrome is the new IE, through no fault of Google or the authors of other browsers.

  • lowwave a day ago

    Before shipping any web site/app, make sure it works in Apple Safari Mobile is usually the one that is dragging it is foot in Web Standards.

    • pjmlp a day ago

      On the contrary, they are the last one standing fighting Google takeover of the Web as ChromeOS development platform.

      Without Safari we are done, just close shop on the Web standards group.

      • judge2020 a day ago

        This is a lesson in capitalism. It’s so much more profitable to ignore small users bases when you can just tell them to “try switching to Chrome”.

        I think you’re wrong about Safari itself being the reason chrome isn’t a 90%+ market owner; rather, it’s apple’s requirement that no other browser engine can exist on iOS.

    • meindnoch a day ago

      Web Standards™ [1]

      __________________

      [1] some feature a Chrome engineer decided to implement, to boost their yearly performance review

    • gus_tpm a day ago

      Even in portugal/spain se have to worry about this. Safari mobile users are a minority here but they usually spend or have more money to spend

      • meindnoch a day ago

        Those stupid rich people don't know what's good for them and keep buying iPhones. I wonder why?

        • flkenosad 20 hours ago

          They have no friends who like them enough to help them troubleshoot their androids.

  • js4ever 21 hours ago

    No, Safari is the new IE, nothing works on it, it's full of bugs and Apple is actively preventing web standards to move on. Do you remember how much Apple prevented web apps to be a thing by blocking web push, and breaking most things if run in PWA mode?

    Apple are by far the worst offender and I can't wait for Safari to die

    • srcreigh 21 hours ago

      It’s death by a million papercuts with safari.

      I made a reader app for learning languages. Wiktionary has audio for a word. Playing the file over web URL works fine, but when I add caching to play from cached audio blob, safari sometimes delays the audio by 0.5-15 seconds. Works fine on every other browser.

      It’s infuriating and it can’t be unintentional.

paulryanrogers a day ago

> how is Chrome incompatible with web standards? It is one of the best implementer of them.

They have so much market share that they control the standards bodies. The tail wags the dog.

  • JimDabell a day ago

    This is not true yet, but it’s getting close.

    The pattern is this:

    - Google publishes a specification.

    - They raise request for feedback from the Mozilla and WebKit teams.

    - Mozilla and WebKit find security and privacy problems.

    - Google deploys their implementation anyway.

    - This functionality gets listed on sites like whatpwacando.today

    - Web developers complain about Safari being behind and accuse Apple of holding back the web.

    - Nobody gives a shit about Firefox.

    So we have two key problems, but neither of them are “Google controls the standards bodies”. The problem is that they don’t need to.

    Firstly, a lot of web developers have stopped caring about the standards process. Whatever functionality Google adds is their definition of “the web”. This happened at the height of Internet Explorer dominance too. A huge number of web developers would happily write Internet Explorer-only sites and this monoculture damaged the web immensely. Chrome is the new Internet Explorer.

    The second problem is that nobody cares about Firefox any more. The standards process doesn’t really work when there are only two main players. At the moment, you can honestly say “Look, the standards process is that any standard needs two interoperable implementations. If Google can’t convince anybody outside of Google to implement something, it can’t be a standard.” This makes the unsuitability of those proposals a lot plainer to see.

    But now that Firefox market share has vanished, that argument is turning into “Google and Apple disagree about whether to add functionality to the web”. This hides the unsuitability of those proposals. This too has happened before – this is how the web worked when Internet Explorer was battling Netscape Navigator for dominance in the 90s, where browsers were adding all kinds of stupid things unilaterally. Again, Chrome is the new Internet Explorer.

    The web standards process desperately needs either Firefox to regain standing or for a new independent rendering engine (maybe Ladybird?) to arise. And web developers need to stop treating everything that Google craps out as if it’s a done deal. Google don’t and shouldn’t control the definition of the web. We’ve seen that before, and a monoculture like that paralyses the industry.

    • pjmlp a day ago

      They are at the edge of transforming the Web into ChromeOS Platform, with the complacency of everyone that helped it become a reality.

      • flkenosad 20 hours ago

        At least chromeos is open source. We can fork it anytime. You'd rather everyone run ios or windows?

        • carlhjerpe 20 hours ago

          Open source in code but not in spirit, you "can't" contribute to ChromeOS without being a Google employee or some special person

    • GoblinSlayer 20 hours ago

      webdev in 2005: webapp spa just werk everywhere, and werk fast and efficiently, only add these 20 lines of code for compatibility :3

      webdev in 2025: OMGWTF NOTHING WORKS WITHOUT THIS NEW SHINY FEATURE RELEASED YESTERDAY AAAAAAAAA!!!!!111

    • bergfest 21 hours ago

      Why not forbid them to ship any non-standard feature in their pre-installed default build of Chrome? Experimental features could be made available in a developer build, that would have to be manually installed in a non-obvious way, so that they cannot gain traction before standardization.

    • carlosjobim a day ago

      > Firstly, a lot of web developers have stopped caring about the standards process. Whatever functionality Google adds is their definition of “the web”.

      Businesses who hire such web developers will lose huge amounts of sales, since 90% of visitors are on mobile and half of those are on Safari.

      • JimDabell a day ago

        How do you think that’s going to play out once Apple are legally barred from mandating WebKit on iOS?

    • whywhywhywhy a day ago

      > Mozilla and WebKit find security and privacy problems

      This is a little disingenuous because Apple often falsely claims security when it’s to hold back tech that could loosen the App Store grasp.

      • JimDabell a day ago

        Can you give an example?

        Generally speaking, when Apple rejects a proposal, Mozilla do too. What’s Mozilla’s motivation for doing this and lying about it?

      • immibis 21 hours ago

        PWA is an antifeature anyway; it's an operating system inside a browser. This benefits companies that have market-dominant browsers and do not have operating systems; on a technical level it's just stupid.

        • carlhjerpe 20 hours ago

          I love PWAs when the alternative is Electron, I'd rather let one browser instance run my crapps since it improves memory sharing and other resource utilization.

          I really like being able to install websites as apps too so my WM can manage them independently.

pjmlp a day ago

Web features being pushed by Google via Chrome, aren't standards, unless everyone actually agrees they are worthy of becoming one.

Shipping Electron junk, strengthens Google and Chrome market presence, and the reference to Web standards, why bother when it is whatever Chrome is capable of.

Web devs with worthy skills of forgotten times, would rather use regular processes alongside the default system browser.

  • duped a day ago

    There are no realistic alternatives to Electron. So calling it "junk" when its the baseline for "cross platform GUI application" is nonsense.

    I get that you don't like it, so go build an alternative.

    • johnnyanmac a day ago

      Are we really trying to argue about cross platform GUI in 2025? This was solved decades ago. Just not in ways that are trying to directly appeal to modern webdevs by jamming a browser into every desktop application.

      I don't even hate Electron that much. I'm working on a toy project using Electron right now for various reasons. This was just a bizarre angle to approach from.

    • pjmlp a day ago

      The alternative already exists, processes using the system browser, for several decades now.

      Or actually learn how we use to ship software on the glory days of 8, 16 and 32 bit home platforms.

      Now I do agree there are no alternatives for people that only care about shipping ChromeOS all over the place.

      • JimDabell a day ago

        > The alternative already exists, processes using the system browser, for several decades now.

        Yes, Windows supported Electron-like applications back in the 90s with HTAs. If you want something modern and cross-platform, Tauri does this:

        https://v2.tauri.app

      • duped 13 hours ago

        The system browser doesn't work on iOS, MacOS, or Linux (under certain, but common, conditions).

      • charcircuit a day ago

        You can't trust the system browser to be up to date and secure or for it to render things how you want. You can not guarantee a good user experience unless you ship the browser engine with your app.

        • carlhjerpe 20 hours ago

          Yeah sure but I use most web apps through the browser either way so I'm already in "possibly incompatible land" and you can reasonably expect any user facing device to have an updated browser OR one specific browser in case of embedded. We're not in Windows XP software distribution times anymore.

badgersnake a day ago

> how is Chrome incompatible with web standards? It is one of the best implementer of them.

Easy when they make Chrome do whatever they want and call it a living standard (whatever that is). There is no such thing as web standards now.