"standardizing on a single rendering engine would go a long way towards unifying the web."
I would agree, but I don't think it would be a good thing to do. Sure, if there were only one rendering engine, all of the web would be rendered in the way. That isn't a good thing though. It would commit the web to being Webkit, warts and all, for a very long time. IE is a good example of this, except it is we who brought it upon ourselves as web developers for using IE specific behavior.
"One of the core evils behind Internet Explorer is that it is a closed platform."
Again, agreed. That said, being closed isn't why we dislike it; its because it has a lot of non standard behavior that people have come to rely on. If the IE source code were to be entirely open sourced tomorrow, how much easier would it be for web developers to work around its idiosyncrasies? I posit that it would still be hard.
"Now Gecko is old, buggy and slow - at least in comparison to WebKit."
Even if this is true, its only true today. It will eventually get better, leaving this argument irrelevant.
This article seems to say that the author disagrees with the "program to interfaces, not implementations" advice. The interface (HTML & CSS) should be how we describe web documents, rather than Webkit's particular implementation of the interface. The author's article seems shortsighted - both forward looking and retrospectively.
> "Now Gecko is old, buggy and slow - at least in comparison to WebKit."
> Even if this is true, its only true today. It will eventually get better, leaving this >argument irrelevant.
Even if it is true today, Mozilla is working on Servo (which will hopefully prove to be a success). There isn't a single reason to not develop competing engines.
Competition will drive all sides to improve their engines. And standardizing on WebKit is like standardizing Linux (it's just a kernel not overall browser). Slow and very hard, if not down-right impossible.
I wish Mozilla all the success in the future but I don't want a monoculture of any engine (Servo/WebKit/Presto/Gecko/<insert engine name> included).
I would agree, but I don't think it would be a good thing to do. Sure, if there were only one rendering engine, all of the web would be rendered in the way. That isn't a good thing though. It would commit the web to being Webkit, warts and all, for a very long time. IE is a good example of this, except it is we who brought it upon ourselves as web developers for using IE specific behavior.
"One of the core evils behind Internet Explorer is that it is a closed platform."
Again, agreed. That said, being closed isn't why we dislike it; its because it has a lot of non standard behavior that people have come to rely on. If the IE source code were to be entirely open sourced tomorrow, how much easier would it be for web developers to work around its idiosyncrasies? I posit that it would still be hard.
"Now Gecko is old, buggy and slow - at least in comparison to WebKit."
Even if this is true, its only true today. It will eventually get better, leaving this argument irrelevant.
This article seems to say that the author disagrees with the "program to interfaces, not implementations" advice. The interface (HTML & CSS) should be how we describe web documents, rather than Webkit's particular implementation of the interface. The author's article seems shortsighted - both forward looking and retrospectively.