5 ms·

I considered other browsers too (Vivaldi, Brave), but they all seem to be Blink/V8. It seems like Firefox is the only other extant, viable platform for running

I considered other browsers too (Vivaldi, Brave), but they all seem to be Blink/V8. It seems like Firefox is the only other extant, viable platform for running the web. Personally, I'm not one to frequently compromise quality/convenience for some ideal, but the idea that we'll have a future with only one way to render the web is terrifying.

dmarcos 7y ago
Can you articulate why do you think a single engine is terrifying? KHTML is the precursor of WebKit that was used by Chrome and Safari for a while. It didn’t prevent Apple and Google to drive their products. Google eventually fork Blink out of WebKit. Other browser vendors follow a similar path (Brave, Opera, Oculus Browser, Samsung Internet, Supermedium...) Start with Blink and modify as they see fit and eventually fork if needed. I consider a single good reusable engine a level playing field for new browsers akin to the Linux Kernel for OSs. Engines are commoditized and differentiation comes from browser features.
- blattimwind 7y ago
  > Can you articulate why do you think a single engine is terrifying? MSHTML.DLL
- dmitryminkovsky 7y ago
  Thanks for asking. I was wondering if I could. I'll try: > Start with Blink and modify as they see fit and eventually fork if needed One of the foundational ideas behind the web is that you have a set of open standards that specify the web, and then people can go ahead and implement those standards and provide some unique set of features on top of those standards. For example: privacy. I believe that Google and other big-tech actors are working to make the scope of these standards so vast and so fast-moving that it becomes impossible for a small or mid-range operation to implement and maintain a web browser. I mean, who was able to successfully fork and maintain a fork of KHTML/Webkit? Google! No small organization is capable of this. These are massive, complicated codebases that must keep up with evolving standards. Therefore, even though you theoretically have "open standards," we're seeing a future with possibly a single implementation of those standards. And if you only have one implementation, then the whole foundation doesn't hold, i.e. you can't make a privacy-oriented browser because you can't make a browser in the first place. What if some day Google decides to stop contributing to Blink, to fork Blink and only update it closed source, against a set of fast-moving standards that they de-facto control? Or what if, at that point, with total domination, they stop following open standards entirely. I think that would be the end of the web.
  - dmarcos 7y ago
    Notice that a fork doesn’t have to maintain the whole code base, only your tweaks on top. Cases I know like Oculus Browser, Samsung Internet, Brave or Supermedium (I’m co-founder) have very small teams. They start with Chromium and have total freedom to modify privacy policies, ship remove any APIs or standards. At the same time they can take all Web compatibility for granted that I agree is not tenable by a small org.
- duhi88 7y ago
  It means that Google can push technologies that benefit their ad tracking business, or choose to use non-standard APIs to make their apps run faster than competitors. Both of these things they're already doing. While they might not go the route of Microsoft, putting their Active-X plugins in IE, using deprecated and non-standard APIs is pretty damn close. Furthermore, all these forks result in bugfixes and new features making it back into Chromium. That allows Chromium to evolve faster than other browsers. That's not to say that Firefox hasn't been keeping up and pushing past Chrome on some important fronts (CSS Sub-Grid, for one), but I'm afraid a point will come where Firefox can't keep up with all the new web APIs that developers will want to use. Another fork of Chromium doesn't lessen Google's influence over the market.
  - dmitryminkovsky 7y ago
    Way better explanation than mine: > It means that Google can push technologies that benefit their ad tracking business, or choose to use non-standard APIs to make their apps run faster than competitors. Both of these things they're already doing. That's at the core of their whole thing.
webmobdev 7y ago
There's webkit too by Apple. And browsers based on it is available for Linux too. And no, webkit and Blink are not the same engine. Blink is a fork of webkit by Google and while they share a heritage they have different feature set now.
- basch 7y ago
  and as soon as Blink forked, Apple purged a ton of Google/Samsung specific code from Webkit.
  - webmobdev 7y ago
    Really? Where can I read more about this?
    - mikewhy 7y ago
      This Slashdot article has some links to the WebKit mailing list discussing the changes: https://slashdot.org/story/184313 https://slashdot.org/story/184313
- dmitryminkovsky 7y ago
  I was thinking of webkit too, but I'm not aware of any browsers expect Safari and the Linux browsers you mention. Neither seem to be cross platform or practical, "viable" alternatives to Blink domination.
  - duhi88 7y ago
    It's the one reason I'm glad that all web browsers on iOS run Webkit. Now if Apple could just fix their damn flexbox bugs and implement web app features...
  - webmobdev 7y ago
    Check out: - Maxthon - https://www.maxthon.com/mx5/features/dual-core/ https://www.maxthon.com/mx5/features/dual-core/ - Midori - https://www.midori-browser.org/ https://www.midori-browser.org/ I'd say the scenario is similar to Firefox/Gecko and Chrome/Blink. Sure there are clones that use the Gecko and Blink engines, but none of these really are as popular as their original. I am waiting for another Opera / Presto browser platform to emerge. Till it did, the browser market dominated by Internet Explorer was literally stagnant. Opera with a Presto engine was a game changer with their light-weight and super fast browser engine with feature sets that all other browsers copied (and still lack even today).
awkward 7y ago
It depends on the use case, but for a lot of modern render heavy sites, I find Gecko to be strictly better, both in terms of overall time to render and giving a user experience with less flashes of unstyled context.

dmarcos 7y ago

Can you articulate why do you think a single engine is terrifying? KHTML is the precursor of WebKit that was used by Chrome and Safari for a while. It didn’t prevent Apple and Google to drive their products. Google eventually fork Blink out of WebKit. Other browser vendors follow a similar path (Brave, Opera, Oculus Browser, Samsung Internet, Supermedium...) Start with Blink and modify as they see fit and eventually fork if needed. I consider a single good reusable engine a level playing field for new browsers akin to the Linux Kernel for OSs. Engines are commoditized and differentiation comes from browser features.

blattimwind 7y ago
> Can you articulate why do you think a single engine is terrifying? MSHTML.DLL
dmitryminkovsky 7y ago
Thanks for asking. I was wondering if I could. I'll try: > Start with Blink and modify as they see fit and eventually fork if needed One of the foundational ideas behind the web is that you have a set of open standards that specify the web, and then people can go ahead and implement those standards and provide some unique set of features on top of those standards. For example: privacy. I believe that Google and other big-tech actors are working to make the scope of these standards so vast and so fast-moving that it becomes impossible for a small or mid-range operation to implement and maintain a web browser. I mean, who was able to successfully fork and maintain a fork of KHTML/Webkit? Google! No small organization is capable of this. These are massive, complicated codebases that must keep up with evolving standards. Therefore, even though you theoretically have "open standards," we're seeing a future with possibly a single implementation of those standards. And if you only have one implementation, then the whole foundation doesn't hold, i.e. you can't make a privacy-oriented browser because you can't make a browser in the first place. What if some day Google decides to stop contributing to Blink, to fork Blink and only update it closed source, against a set of fast-moving standards that they de-facto control? Or what if, at that point, with total domination, they stop following open standards entirely. I think that would be the end of the web.
- dmarcos 7y ago
  Notice that a fork doesn’t have to maintain the whole code base, only your tweaks on top. Cases I know like Oculus Browser, Samsung Internet, Brave or Supermedium (I’m co-founder) have very small teams. They start with Chromium and have total freedom to modify privacy policies, ship remove any APIs or standards. At the same time they can take all Web compatibility for granted that I agree is not tenable by a small org.
duhi88 7y ago
It means that Google can push technologies that benefit their ad tracking business, or choose to use non-standard APIs to make their apps run faster than competitors. Both of these things they're already doing. While they might not go the route of Microsoft, putting their Active-X plugins in IE, using deprecated and non-standard APIs is pretty damn close. Furthermore, all these forks result in bugfixes and new features making it back into Chromium. That allows Chromium to evolve faster than other browsers. That's not to say that Firefox hasn't been keeping up and pushing past Chrome on some important fronts (CSS Sub-Grid, for one), but I'm afraid a point will come where Firefox can't keep up with all the new web APIs that developers will want to use. Another fork of Chromium doesn't lessen Google's influence over the market.
- dmitryminkovsky 7y ago
  Way better explanation than mine: > It means that Google can push technologies that benefit their ad tracking business, or choose to use non-standard APIs to make their apps run faster than competitors. Both of these things they're already doing. That's at the core of their whole thing.

webmobdev 7y ago

There's webkit too by Apple. And browsers based on it is available for Linux too. And no, webkit and Blink are not the same engine. Blink is a fork of webkit by Google and while they share a heritage they have different feature set now.

basch 7y ago
and as soon as Blink forked, Apple purged a ton of Google/Samsung specific code from Webkit.
- webmobdev 7y ago
  Really? Where can I read more about this?
  - mikewhy 7y ago
    This Slashdot article has some links to the WebKit mailing list discussing the changes: https://slashdot.org/story/184313 https://slashdot.org/story/184313
dmitryminkovsky 7y ago
I was thinking of webkit too, but I'm not aware of any browsers expect Safari and the Linux browsers you mention. Neither seem to be cross platform or practical, "viable" alternatives to Blink domination.
- duhi88 7y ago
  It's the one reason I'm glad that all web browsers on iOS run Webkit. Now if Apple could just fix their damn flexbox bugs and implement web app features...
- webmobdev 7y ago
  Check out: - Maxthon - https://www.maxthon.com/mx5/features/dual-core/ https://www.maxthon.com/mx5/features/dual-core/ - Midori - https://www.midori-browser.org/ https://www.midori-browser.org/ I'd say the scenario is similar to Firefox/Gecko and Chrome/Blink. Sure there are clones that use the Gecko and Blink engines, but none of these really are as popular as their original. I am waiting for another Opera / Presto browser platform to emerge. Till it did, the browser market dominated by Internet Explorer was literally stagnant. Opera with a Presto engine was a game changer with their light-weight and super fast browser engine with feature sets that all other browsers copied (and still lack even today).

awkward 7y ago

It depends on the use case, but for a lot of modern render heavy sites, I find Gecko to be strictly better, both in terms of overall time to render and giving a user experience with less flashes of unstyled context.