LOL. I love how full of his own crap Tom Dale has become.
Ember is a good framework with a lot of potential, but more than anything it needs a good PM. Ember devs just spent nearly 14 months rewriting the layout engine (Glimmer 2.0) that they had spent the prior 8 months writing (Glimmer). And what do they have to show for it? A 100ms or so speed boost on complex views. Instead they should focus on fixing up the crufty parts of their API and adding feature they've been promising users for years. Instead Tom Dale has been off writing FastBoot, an even more niche component of Ember that very few apps will even use.
React and Angular are far from perfect, but one upside is they have parent companies to help keep them focused.
Having worked on Glimmer 2 myself, I don't think it's accurate to say that Ember devs worked on it for 14 months (that is, it took 14 months of work to same 100ms). It had initial work done for a long time and then it sat for a while unchanged because it's an open-source project that people work on when they have time. Month later work started up again to get it out the door.
I think that since LinkedIn is "all in" on Ember it will help to get some of the cruft removed and things moving forward. LinkedIn has probably the largest and most complex Ember app in existence, so there are a lot of learnings to take an use to improve the framework and ecosystem.
Well the Glimmer 2.0 effort started around the same time Ember 2.0 shipped (Aug 2015). That was 16 months ago, Glimmer 2.0 shipped about 2 months ago, ergo 14 months. I agree that it wasn't under development the entire time, but it was a blocker for many other features the entire time.
Basically when 2.0 landed, routable components couldn't be worked on until glimmer components were completed. Angle brackets, Improved Pods, etc, same.
So even if it wasn't under dev that whole time, it held back other features that would have really improved Ember.
Ember is a good framework with a lot of potential, but more than anything it needs a good PM. Ember devs just spent nearly 14 months rewriting the layout engine (Glimmer 2.0) that they had spent the prior 8 months writing (Glimmer). And what do they have to show for it? A 100ms or so speed boost on complex views. Instead they should focus on fixing up the crufty parts of their API and adding feature they've been promising users for years. Instead Tom Dale has been off writing FastBoot, an even more niche component of Ember that very few apps will even use.
React and Angular are far from perfect, but one upside is they have parent companies to help keep them focused.