I am somewhat concerned about the volatility. All three languages have their merits and each has a stable foundation that has been developed and established over many years. The fact that the programming language has been “changed” within a short period of time, or rather that the direction has been altered, does not inspire confidence in the overall continuity of Ladybird's design decisions.
back
3 comments
Ladybird as a project is not that old, and it's still in pre-alpha, if they are going to make important changes then it's better now than later.
> I am somewhat concerned about the volatility.
Not just volatility but also flip-flopping. Rust was explicitly a contender when they decided to go with Swift 18 months ago, and they've already done a 180 on it despite the language being more or less the same as it was.
they tried swift, it didn't work, and they figured rust was the best remaining option. that's not "flip-flopping" (by which I assume you mean random indecisiveness that leads to them changing their mind for no reason)
Yup, this was not flip-flopping, it was willingness to be open to options, even if it means going back on a decision branch made earlier in the process.
For the Ladybird project, now is the best time to be making a big decision like this, and it's commendable that the project lead was honest to recognize when an earlier attempt was not working, to be able to re-think and come to a better decision. I'm no fan of Rust, but for this project I think most of us would agree it's a better language than Swift for their purpose.
They made a very pragmatic and sensible decision after reviewing Swift that it wouldn't be suitable for their purposes, so they shifted to the next best alternative. I think they reasoned it very well and made a great decision.
I guess they bet on Swift being more than Apple's blessed way of writing UI software.
It's not that they are loving Rust, but they realized going all-in on Swift means becoming sharecroppers on massa Tim Apple's plantation.
There's been some fun volatility with the author over the years. I told him once that he might want to consider another language to which he replied slightly insultingly. Then he tried to write another language. Then he tried to switch from C++ to Swift, and now to Rust :P
Upside: he's learning?
Indeed, and as a school those 18 months are well worth it, but it is in many ways also 18 months wasted. There is a strong sense of NIH with the Ladybird dev(s), and I wonder if that isn't their whole reason for doing this.
I've seen another team doing something similar, they went through endless rewrite cycles of a major package but never shipped, and eventually the project was axed when they proposed to do it all over again, but this time even better.
> Indeed, and as a school those 18 months are well worth it, but it is in many ways also 18 months wasted.
* of course i'm just an internet person, i don't deserve anything from anybody ^^
The sense of NIH is from Serenity, and that was probably the reason for Jakt's existence too. Now it's spun off into its own project there is a lot more pragmatism.
Well, here's to hoping because we really need a stand-in for FF. I realize the irony here in terms of that being the ultimate 'NIH' project but that one I can get behind because the browser landscape is much too fragile. Of course they might end up taking users away from FF rather than from Chrome, Edge or Safari.
In case you didn't know they're using a lot of third-party libraries now for pretty major things: libcurl for http, Skia/Harfbuzz for rendering, libxml, OpenSSL, ffmpeg, etc:
https://github.com/LadybirdBrowser/ladybird/tree/8017f8a7ed3...
The core browser engine, JS/CSS/layout etc will always be original.