As you may know, Mozilla laid off a bunch of people, including the entire team making the experimental Servo browser engine (written in Rust). Can Redox be part of an effort to save Servo? Servo deserves a future, and Redox could use a browser engine entirely written in Rust.
I suppose that the same question is being asked over WASM generally. I tended to view Servo as an insider project with internal direction and leadership within Mozilla. It could become something taken under the Rust Foundation if such an organization ever were to materialize. I’d certainly like to learn more about the project as this shift within Mozilla evolves and will continue seeking relevant information.
1 Like
Does Redox have a WASM runtime?
Also worth mentioning is that it appears that Servo is modular and you can use pieces of it without having to include the whole thing.
https://github.com/twilco/kosmonaut is a hobbyist web browser written in Rust using Servo’s html5ever and cssparser for HTML and CSS parsing.
Perhaps partial victory is possible if the more finished parts of Servo live on and are improved and maintained.
I find an existing redox/cookbook/recipes/servo.
Depends what you mean by save. Even under stewardship of Mozilla, making a new web browser engine from scratch was too much of sisyphean task to be realistic. There’s no way Servo will by itself ever compete with the mainstream browser engines. Its components could still see a future in projects of smaller scope though.
You could argue the reverse: that the browser is so central to modern casual computing, many people have no standalone productivity suite or email client. Even the standalone apps may be increasingly underpinned by the browser engines through web technology frameworks such as Electron or Tauri. Support for such frameworks has been argued as especially beneficial to Redox because it lacks native apps.
If Haiku is aiming to be a clean sheet operating system reimagined in rust, there are strong arguments for making the browser engine part of the masterplan.
I suspect it would be easier to get Firefox running on Redox than creating a competitive browser from scratch, and it’s also the sort of groundwork that would be a natural goal for Redox to get some form of compatibility with Linux programs in general.
I would welcome a new browser that could compete with the incumbents, but I don’t see it happening and I don’t think it makes any sense for the Redox project to pursue that goal.
Could somebody summarize the current use of Servo in the Redox project related to their conception of a roadmap?
I suspect there isn’t one at present, but second your suggestion there should be.
I see this debate as being between two different philosophies. Servo represents the moonshot - ambitious, and will take years to pay off, whereas porting Mozilla and other linux apps represent what might be considered the more “pragmatic” approach. To this I say that writing an operating system from scratch is already a moonshot, and it would be a shame to start making compromises now.
I have misgivings about making it easy to port Linux apps across - as somebody with less technical knowledge I wonder if that might lead to dilution of the purity of the Redox concept, and erode the reason to use Redox over Linux. I suggest, if we accept the practical need to port tools over from existent systems, Plan 9 might be a better source.
Take text editors (as an example, I know there is already one in Redox). Do we really want Vim and Emacs? Maybe we could do better by porting “acme” from Plan 9. Or, if we had a kick-ass toolkit of web technologies (like Servo), we could simply load Atom running in the Electron framework, or maybe something better still in the Tauri framework.
What purpose does the current Servo library serve in Redox?