Some Of What You Can Find On Mozilla's Servo Roadmap

[zamadatix]

Also, of all the browsers I know, Chrome seems to be the only one without an option to reject 3rd party cookies by default.

Menu -> Settings -> (privacy section) Content Settings -> Block 3rd Party cookies. Been there since I can remember Chrome.

[bug77]

Menu -> Settings -> (privacy section) Content Settings -> Block 3rd Party cookies. Been there since I can remember Chrome.

Not on the mobile version.

[Hamsterkill]

I am not following Servo closely, but its development speed seems relatively slow to me. Has there already been a release of Servo as a proper package in some Linux distribution? I don't have a Servo package in Gentoo Linux, neither it is listed at http://gpo.zugaina.org.

It's not slow when you consider that it's an entirely new browser engine -- I think it's been over a decade since anyone tried to do that. It's still pretty far from being able to be put in a general release browser. Last I saw, Servo is able to render some relatively simple sites pretty well. Their current efforts are still focusing on specific site compatibility (I think github, reddit and duckduckgo are their current goals). As mentioned in the article, they intend to put out an initial alpha package (with browser.html frontend) this summer.

[FLHerne]

How will Browser.html integrate with the system and give a native desktop application experience on GTK and Qt based systems?

Will it blend in and look native or will it look alien?

In principle, it could use GTK 3.20's CSS theming, although the GTK element names would need translating into classes on normal HTML elements.

[zamadatix]

Not on the mobile version.

Settings -> Site Settings -> Cookies -> Allow third party cookies. Been there since I can remember mobile Chrome. At least look through the menus or do a search before you spout things as fact.

[Luke]

Well, you may read all that on the forums, but that doesn't mean it's not bull. Chrome may be a lot of things, but it's not lightweight. Also, of all the browsers I know, Chrome seems to be the only one without an option to reject 3rd party cookies by default.

A browser that cannot reject 3ed party cookies is one I could never consider using. OK, so that's only mobile, yet another reason I don't want a smartphone. I would still not use Chrome even if Firefox became unmaintained, and won't use any website that only works on Chrome such as anything requiring Pepper Flash. I normally block flash anyway and look forward to uninstalling it entirely as soon as a few remaining sites dump it. It is an untrustable closed blob, semi-tolerable only because so many keep it under a microscope and because it is never enabled while doing anything sensitive like logging into something.

I ditched Chromium years ago when they started gettting too tightly integrated with Google Services (and thus trackers and ads). In Panopticlick I could never stop it from coming up unique even with JS disabled, so that was that and out it went. A privacy-respecting browser built on Chromium would require an audit of the code, followed by finding the things that make it so easy to track and fixing them. Firefox too now requires actively disabling cloud services, I so wish something like GNOME Web could use the Firefox extensions.

For me, the ideal browser would be a fork of gnome-web that uses CSD only in GNOME and not when over other DE's, that uses a traditional menubar, and for which NoScript's functionality is added by default, which always returns a random value when asked to send canvas data back to the server,and always uses Disconnect's blocklist like firefox does in a private window in "strict" mode. Alternately the whole Disconnect blocklist could go into /etc/hosts so the browser would not need to deal with it. Lastly it would need to pretend to be Chrome on Windows so far as the useragent is concerned. Good luck tracking it..

[bug77]

Settings -> Site Settings -> Cookies -> Allow third party cookies. Been there since I can remember mobile Chrome. At least look through the menus or do a search before you spout things as fact.

I did look, but after finding the setting on the desktop, I guess I tried to find it in the same place. Asked a few friends about it too and nobody pointed me in the right direction.
Still, Firefox is good enough for me, I only use Chrome on occasion.

[Rubble Monkey]

- Other people have repeatedly pointed out that C++11 and C++14 have a lot of the features in Rust, and a disciplined use of those versions of C++ in a new browser layout engine would probably be as good as Servo. I agree, but Rust and Servo were started in 2008. We can argue about the benefits of Rust over C++14 - I maintain there are some because Rust has stronger immutability by default - but it's unquestionable that Rust is a better choice for memory-safe code than C++03.

True, they love to say "also in C++11/14", but can you name a big project that uses them? After all this time QT is adding SOME C++11 code, and have you seen the number of QT programmers? You have to wonder how long it will take anyone to be fully C++14 compliant, especially considering how long it took the compilers to do so.

P.S. will VC++ ever get full C++11/14/17 support?

[Krysto]

More "bye Gecko". Given the amount of branding that's gone into Firefox, they're more likely to wind up with a Servo+WebExtensions-based Firefox than to ditch it altogether.

(Similar to how WebKit was forked from KHTML and now there's a WebKit KPart you can use in Konqueror in place of the KHTML KPart.)

And maybe a stronger sandboxing model as well. I don't like where they're doing with the current patched-up sandboxing architecture meant to go around the XUL add-ons and only isolating all content from the UI. Each tab should be fully isolated, as it is in Chrome and Edge. No wonder Firefox isn't even invited to Pwn2Own anymore. The security of the browser is a joke.

[ssokolow]

Unless this new Firefox can fake the results in a canvas tracking (so it can render a canvas image but can't be tracked by it) like Canvasblocker, block javascript on a per-origin site basis like NoScript, and block connections to trackers like Disconnect or supports existing or new extensions for this purpose I will be unable to use it.

Last I heard, their intent is that extensions which can't be implemented in WebExtensions can write and depend on a "native.js" component to implement a manually-audited private extension to the WebExtensions API which will then be used as a starting point for standardizing a public equivalent in their planned superset of Chrome's extension API.

They're actually in discussion with people like Giorgio Maone to ensure that their WebExtensions efforts won't shut out or cripple extensions like NoScript.