Chrome 57 Arrives For Linux Users, WebAssembly By Default

[phoronix]

Phoronix: Chrome 57 Arrives For Linux Users, WebAssembly By Default

Days after the Firefox 52 release, Google has officially released Chrome 57.0...

http://www.phoronix.com/scan.php?page=news_item&px=Chrome-57-Released

[uid313]

When is it getting built with GTK3?

[wswartzendruber]

I wonder if one can port OpenJDK's JRE to WebAssembly with any reasonable amount of final success (performance).

[s_j_newbury]

When is it getting built with GTK3?

I've been using gtk3 builds of Chromium for a long time. I have, on occasion, had to fix up the build but it has always worked well. The only significant (and I think, still unfixed) bug that I've noticed is that running a gtk3 build under Wayland crashes with key input if GDK_BACKEND=x11 isn't specified.

Chromium 57 has good gtk3 theme integration with menus etc, which unfortunately is gone with Chromium 58.

[mmstick]

Now someone needs to write a FLIF decoder in WebAssembly so the web can save a ton of bandwidth.

[starshipeleven]

Now someone needs to write a FLIF decoder in WebAssembly so the web can save a ton of bandwidth.

There are 2 javascript-based FLIF decoders that can probably be converted easily

https://github.com/saschanaz/libflif.js/

https://github.com/UprootLabs/poly-flif

[Pajn]

There are 2 javascript-based FLIF decoders that can probably be converted easily

https://github.com/saschanaz/libflif.js/

https://github.com/UprootLabs/poly-flif

WebAssembly is a compile target, you doesn't convert JS to it. However C++ or Rust decoders should probably be compilable to it (maybe with some tinkering) with modifications for the API differences.

EDIT: Ugh, the poly-filf is already written in C++ and compiles to ASM.js, it should probably just be a compiler option to output to WASM then...

[markg85]

CSS Grid! Finally!

I've been waiting for that for years. It allows one to very easily make a "golden layout". That is a layout where your header is on top, your footer is at the bottom and the content within automatically adjusts. So even if you have no content, that gives you a content section that stretches to the position where your footer would be (at the bottom of your view area). That is something that *is* possible in pre grid times, but much more trickier!

There is a bit of irony on the microsoft side of this. Microsoft initially proposed the CSS Grid and implemented it in IE10 and IE11. By now, the spec they had implemented back than wildly differs from the final version that the other vendors implemented. And Edge doesn't have support for the final version yet. So, while microsoft was first, it still is last in implementing the final version (firefox already has it, opera gets it as soon as there core updates to chrome 57).

This also means the value of the css layout engines (like for instance bootstrap) becomes... well, a lot less Bootstrap obviously offers more then just flexible layout arrangement, but it's based on outdated tech. Even the to be released version 4 (it uses flexbox).

[abral]

I wonder if one can port OpenJDK's JRE to WebAssembly with any reasonable amount of final success (performance).

https://github.com/mozilla/pluotsorbet was written in JS (with some parts in asm.js) and the performance was reasonable.

[kneekoo]

CSS Grid! Finally!
[...] the value of the css layout engines (like for instance bootstrap) becomes... well, a lot less Bootstrap obviously offers more then just flexible layout arrangement, but it's based on outdated tech. Even the to be released version 4 (it uses flexbox).

It will take quite a while until the majority of people will update to css grid-compatible browsers. So the frameworks will be necessary for a long time now, to maintain compatibility. As for Microsoft, I hope they will step up soon, unless they want Edge to become the new IE6 regarding the web standards.