‹ BackHN Continuity

Thread

One year of sponsored Servo development

358 points · 144 comments · AshleysBrain

  1. p-e-w · · focus · HN ↗
    What can one actually do with Servo today?

    It’s apparently still not ready to be used as a browser engine (and may never be), so what exactly is it for?

    1. foresterre · · focus · HN ↗
      Regardless of actual use as an (embedded) browser, it is useful in the sense that it can be used to provide (another) source of validation about whether specs and WPT [1] tests are properly and clearly defined. That helps for long term browser compatibility.

      [1] <a href="https:&#x2F;&#x2F;wpt.fyi&#x2F;results&#x2F;?label=experimental&amp;label=master&amp;product=servo&amp;aligned" rel="nofollow">https:&#x2F;&#x2F;wpt.fyi&#x2F;results&#x2F;?label=experimental&amp;label=master&amp;pro...

      1. charcircuit · · focus · HN ↗
        &gt;long term browser compatibility

        Making a new browser engine hurts browser compatibility more than it helps it. It now becomes one more target that needs to be tested against, limitations need to be kept tracked of, etc. The best thing for long term browser compatibility is focusing resources into Blink. In regards to the standard we can use LLMs to cross reference the spec, with tests, and with the implementation to make sure all match.

        1. teekert · · focus · HN ↗
          Ah, you must be too young to have experienced early Internet Explorer, which attempted just this.
          1. charcircuit · · focus · HN ↗
            Internet Explorer was not open source acting as engine everyone&#x27;s browser could use and it got abandoned by Microsoft. It is not the same situation.
            1. teekert · · focus · HN ↗
              One could argue that effectively Blink is also not open source. At any rate users don&#x27;t have power over it, as witnessed by the manifest v3 stuff jammed down our throats.
              1. pygy_ · · focus · HN ↗
                Churn makes it impossible to keep up a fork, Google weaponizes its manpower, making the &quot;open source&quot; argument moot.
                1. charcircuit · · focus · HN ↗
                  There are many active forks that are an existence proof that it&#x27;s possible. It&#x27;s also a ton of man power to maintain your own browser engine.
            2. le-mark · · focus · HN ↗
              Monocultures are bad full stop. Blink is corporate controlled open source, which has been clearly less than.
            3. mohamedkoubaa · · focus · HN ↗
              At this point Microsoft should just open source it for science.
        2. Oxodao · · focus · HN ↗
          yeah in a perfect world google dictates how the standard will evolve, we kill all competition so we&#x27;re sure that you have only one target to build your software against, I&#x27;m sure a monopoly will greatly improves innovation and not be used to stiffen competition. &#x2F;s
          1. charcircuit · · focus · HN ↗
            Using Blink doesn&#x27;t mean you have to defer all decisions to Google. You just benefit from not having to duplicate work.
            1. samus · · focus · HN ↗
              Except one does since Google is the primary maintainer. Anyway, work on other browser engines is not duplicated or wasted.
              1. charcircuit · · focus · HN ↗
                It&#x27;s easier maintaining a fork of Blink than an entirely different engine. Google being the primary maintainer doesn&#x27;t mean that browsers can&#x27;t make large forks.

                &gt;work on other browser engines is not duplicated or wasted.

                How is having N engine developers implement some new addition to the EMCAScript or CSS standard for each of their engines not duplicated work?

                1. samus · · focus · HN ↗
                  &gt; It&#x27;s easier maintaining a fork of Blink than an entirely different engine. Google being the primary maintainer doesn&#x27;t mean that browsers can&#x27;t make large forks.

                  Maintaining a fork is not fun as taking patches from the origin repo becomes more complicated the longer the fork exists and the bigger the divergence becomes.

                  &gt; How is having N engine developers implement some new addition to the EMCAScript or CSS standard for each of their engines not duplicated work?

                  It would be duplicated only if they all maintained, say, a Blink fork.

            2. kryptiskt · · focus · HN ↗
              And you can&#x27;t do what Google doesn&#x27;t want you to do. That&#x27;s a nice ad blocker you have there, shame if we removed support for it.
              1. charcircuit · · focus · HN ↗
                Brave literally ads an ad blocker into the browser itself. When you fork software you can do whatever you want to it.
              2. armadyl · · focus · HN ↗
                “The benefit of open source is that anyone can fork and modify the code!”

                Yet somehow this argument seems to evaporate since people’s brains short circuit the moment they see the name Google.

                The community can always fork and maintain a separate version of Chromium if necessary. Or does this concept only not apply to Chromium?

        3. latexr · · focus · HN ↗
          &gt; The best thing for long term browser compatibility is focusing resources into Blink.

          That’s what makes Google more dominant, and if you think that’s a good idea you should really put down your LLM and go read up on a bit of history. Do you want to live in a world where a single company controls your access to the web? Are you eager to navigate through ad pages with a little bit of web? To have even more of your data mined?

          Browser diversity may be a bit annoying for developers, but it is good for users, and developers are users too. When there’s a lack of diversity, everyone but the dominant player is squeezed.

          1. charcircuit · · focus · HN ↗
            Google&#x27;s dominance comes from market share not from owning upstream. If a different fork took 90% market share that fork would have the dominance and power.
            1. samus · · focus · HN ↗
              And that market share is maintained by integration with Android and low-key sabotage of others browsers, especially Firefox. Similar to how Microsoft maintained its monopoly decades ago.
              1. charcircuit · · focus · HN ↗
                Whether a new browser uses a brand new engine or not that doesn&#x27;t stop Chrome being integrated with Android.
                1. samus · · focus · HN ↗
                  The argument was not about stopping Chrome from becoming integrated into Android.
            2. latexr · · focus · HN ↗
              &gt; If a different fork took 90%

              That can only happen if we don’t all focus resources on Blink, which is the point. By definition, we can only have other engines if diversity is a thing.

              The problem isn’t specifically Google (though they are a problem), but any one individual company having all the power.

              1. charcircuit · · focus · HN ↗
                &gt;That can only happen if we don’t all focus resources on Blink

                Why? What&#x27;s fundamentally stopping Edge market share from going to 90% that would go away if they switched back to their own browser engine?

                &gt;but any one individual company having all the power.

                Open source means that anyone can make any change at any time. Google doesn&#x27;t have all the power over what a fork does. A fork is still in full control.

        4. embedding-shape · · focus · HN ↗
          &gt; The best thing for long term browser compatibility is focusing resources into Blink.

          Lol, for Blink perhaps, what about the rest of the ecosystem, the web and the users, like the ones who don&#x27;t even use Blink?

          Very idealistic point-of-view with no basis in reality, sadly, because wouldn&#x27;t it be nice if we could care about just one engine?

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.