Look, I am as annoyed as you are with the constant barrage of "rewritten in Rust" projects, but if Mozilla did not try various other projects that are not browser, there would be no Rust.
Rust wasn't a Mozilla project per se, it was something a person who happened to be working for Mozilla was messing around with and it got internal traction.
But I'm actually ok with a lot of the non-firefox projects that they have like the VPN.
What I do have an issue with is the foundation, throwing money away at various projects that have very little to do with making firefox better. From "trusworthy AI" research grants to giving 387k to the Mckensie Mack group or 375k to the New Venture Fund (I get Mozilla are lefties but what does this have to do with Firefox?) plus some other organizations that I can't even tell if they aren't just money laundering fronts as they don't appear to actually do anything.
That and the C-Suite being complete parasites. The CEO of Mozilla corp makes almost as much in a year as the Mozilla foundation makes from donations.
Remove the parasites and the senseless spending of the foundation and you could develop Firefox with the ~20% of revenue that doesn't come from Google.
Also the Mozilla originated Fluent project for localization is another example of a stand out approach. It would be interesting to see how localization fits with the Ladybird browser project as a whole. Making use of a custom implementation of Fluent might actually be a good way of moving forward.
Do people actually use Fluent? When I showed it to some professional translators, the reaction was along the lines of: “Hmm, interesting, but does it fit into my existing [roughly speaking XLIFF] tooling? No? Then no.” More generally, a technical translator’s flow is turning a table of strings into a table of strings with minimal distractions and the occasonal look at the reference; I’m not sure Fluent—however nice it looks—facilitates that.
Mostly not, but the formats and limited available tooling is designed to dovetail with existing offerings. Adoption is extremely low despite fairly easy implementation of most features.
Really? So if the software I want localized uses Fluent, do I have ways to work with translators who use Trados or Transit or Déjà Vu or memoQ or whatnot? My initial impression was that Fluent’s data model is way, way too fancy for any of these (or for interoperability via XLIFF or TMX, imperfect as it is), but I’d be happy to learn I was wrong.
If I recall correctly, Rust was born with building a browser engine in mind, or at least it was one of its earliest motivations. So Rust would have been a thing even if Mozilla had focused on their core product.
>So Rust would have been a thing even if Mozilla had focused on their core product.
Plus, while Firefox is their main product, it's been decades since Mozilla has been solely a browser company. It's like saying Microsoft should stop making Office because it detracts from their OS business. Companies can make more than one product. Some of those products are going to have shorter lifespans or smaller userbases than others and that's OK.
