I mean, even if they need to cut costs, isn't Servo part of the team they should keep until the end?
[1] https://blog.mozilla.org/wp-content/uploads/2020/08/Message-...
To put it bluntly: Mozilla is no longer a browser company. It is a patsy for google to point to and say "See we're not a monopoly.". Secondary goals are increasing the supply of devs for big tech and lowering wages.
Except for the leadership.
You see this at companies when engineering teams get it into their heads that they need to rewrite in a new framework. There's a trade-off between that work and building new features for customers. The benefits might be worth it in the long term! They also might not be.
That seems to be the call Mozilla made. I don't know if it's the right one, but it seems to be what they did.
Yes, because all those years the Mozilla executive team has proved that they have a good grasp of where the cost/effort optimum lies... /s
I've seen it several times: Often after years of promises from "the new hot" team, with them cannibalizing resources and top talent but missing milestones and still having failed to deliver, the new beast is cancelled and the value of the "old" system is suddenly recognized.
There's the classic Spolsky essay about this: https://www.joelonsoftware.com/2000/04/06/things-you-should-...
What's better? Refactor the old thing as you go. Include the "legacy" team in the development team of the new thing. Even more so, make the "new" team work on the old thing as well. Rotate developers between the two, build a culture of respect.
Not saying that is what happened at Moz, but I've seen it play out so many times. And without a concrete commitment to be forced to deliver to actual customers projects like these can just go forever noodling around in perfection-land.
I wonder if this is the end of that model, or more an effort to get external people to contribute more to Servo.
People have been claiming Firefox components were written in Rust for ages, usually meaning Servo, but it seems now that those claims were just misleading.
Yes, lots (and increasing). E.g.:
https://searchfox.org/mozilla-central/source/servo/component...
https://searchfox.org/mozilla-central/source/gfx/wr
https://searchfox.org/mozilla-central/source/third_party/rus...
In general, a lot of text handling, and parsing in general, is moving to Rust, although some code is moving far more slowly than others. Additionally, APIs that had been implemented in JS that are too slow (or memory-heavy) are being moved to a Rust implementation instead, such as the l10n implementation.
By niche I mean the myriad of keyboard driven browsers. kiosks, QML, electron, etc.
All I can think is that they took a look at Servo and decided it was cool tech but was really nowhere near delivering things they needed for Firefox aside from Stylo which it had already delivered.