3 ms·

> Chrome is something users install because it's their preferred choice. We can argue as to why it's their preferred choice but the point remains that users act

> Chrome is something users install because it's their preferred choice. We can argue as to why it's their preferred choice but the point remains that users actively download and install Chrome - usually as one of the first things they do when they purchase a new computer. Chrome is getting bundled with a buttload of software. So many software that comes with an ‘installer’ has a hidden checkbox somewhere that if you don’t uncheck it, it will install Chrome on your windows machine. Also Edge is now the default on Windows which is also Chromium. So a Chromium-only future doesn’t sound so distant anymore.

jpgvm 4y ago
I think there is a big difference between engine monopoly vs browser monopoly, however that does depend on how well governance works for the Chromium project (I really don't know enough to assert anything concrete). Namely an engine isn't a product and most value-add is in the porcelain of the browser, implementation of web standards -shouldn't- be a competitive advantage because doing leads the the sort of fragmentation that we don't want. If all the major browsers shared and worked on a single engine collaboratively (similar to PostgreSQL and Linux) I think the web would be a -much- better place as a result for a number of reasons. 1. Aforementioned standards could be relied on by developers. 2. Browser vendors could spend less time duplicating efforts on low level standards implementation that aren't product differentiation. 3. Platform specific performance optimisation could be shared across browsers instead, i.e Safari like energy performance for everyone. 4. Plugin architecture could potentially be shared across all available browsers which would help plugin authors out significantly. 5. The engine could be linked as a dynamic library so that multiple browsers could be installed without duplicating the engine binaries if they are using the same version (probably only viable on Linux distributions but would be nice) 6. Fuzzing and security efforts wouldn't need to be duplicated across engines. This is important because of just how gnarly the sorts of code you see in browsers. Namely XML/SGML parsers, JS runtimes/isolation, etc. Also with every vendor chipping in this infrastructure could likely be greatly expanded. This is a utopian view and most of this probably wouldn't come to pass because the world sucks at doing the "right" thing but the potential is there. On the other hand the downsides mostly amount to Chromium governance could be bad and Google could abuse their position to make it hard to create new browsers based on Chromium/Blink. However they would be incentivised not to in order to prevent a return to the current fragmented situation.
- dmitriid 4y ago
  Current theoretical way of web standards is: a standard can only become a standard if there are two independent implementations of it. This somewhat prevents browsers from implementing fully proprietary tech that canät be replicated by anyone else. It does lead to problems like WebSQL being scrapped (the idea was to basically run sqlite in the browser, Mozilla said they donät want to reverse-engineer sqlite behaviour for an independent implementation, the idea was scrapped). However, with Google dominance we see more and more things appearing in Chrome only and based on Chrome's own goals, and then advertised as standards. There's an ideal state somewhere, but I don't know how we could reach it.
  - jpgvm 4y ago
    Yeah I think it's a difference in philosophy somewhat. PostgreSQL and Linux aren't standards, they are just shared infrastructure. If Chromium was to become the same thing we could have just had WebSQL assuming maintainers agreed it was worth landing and maintaining.

jpgvm 4y ago

I think there is a big difference between engine monopoly vs browser monopoly, however that does depend on how well governance works for the Chromium project (I really don't know enough to assert anything concrete). Namely an engine isn't a product and most value-add is in the porcelain of the browser, implementation of web standards -shouldn't- be a competitive advantage because doing leads the the sort of fragmentation that we don't want. If all the major browsers shared and worked on a single engine collaboratively (similar to PostgreSQL and Linux) I think the web would be a -much- better place as a result for a number of reasons. 1. Aforementioned standards could be relied on by developers. 2. Browser vendors could spend less time duplicating efforts on low level standards implementation that aren't product differentiation. 3. Platform specific performance optimisation could be shared across browsers instead, i.e Safari like energy performance for everyone. 4. Plugin architecture could potentially be shared across all available browsers which would help plugin authors out significantly. 5. The engine could be linked as a dynamic library so that multiple browsers could be installed without duplicating the engine binaries if they are using the same version (probably only viable on Linux distributions but would be nice) 6. Fuzzing and security efforts wouldn't need to be duplicated across engines. This is important because of just how gnarly the sorts of code you see in browsers. Namely XML/SGML parsers, JS runtimes/isolation, etc. Also with every vendor chipping in this infrastructure could likely be greatly expanded. This is a utopian view and most of this probably wouldn't come to pass because the world sucks at doing the "right" thing but the potential is there. On the other hand the downsides mostly amount to Chromium governance could be bad and Google could abuse their position to make it hard to create new browsers based on Chromium/Blink. However they would be incentivised not to in order to prevent a return to the current fragmented situation.

dmitriid 4y ago
Current theoretical way of web standards is: a standard can only become a standard if there are two independent implementations of it. This somewhat prevents browsers from implementing fully proprietary tech that canät be replicated by anyone else. It does lead to problems like WebSQL being scrapped (the idea was to basically run sqlite in the browser, Mozilla said they donät want to reverse-engineer sqlite behaviour for an independent implementation, the idea was scrapped). However, with Google dominance we see more and more things appearing in Chrome only and based on Chrome's own goals, and then advertised as standards. There's an ideal state somewhere, but I don't know how we could reach it.
- jpgvm 4y ago
  Yeah I think it's a difference in philosophy somewhat. PostgreSQL and Linux aren't standards, they are just shared infrastructure. If Chromium was to become the same thing we could have just had WebSQL assuming maintainers agreed it was worth landing and maintaining.