News for hackers. Stuff that matters.
Re:Servo’s new home by nvrspyx on Tuesday November 17, 2020 @12:37PM (#25126705)
A third-option that I would like to see is extending the latter to an Electron alternative (aka using Servo as a cross-platform GUI). There's definitely positives to Electron, but it would be nice to see a more performant, less memory hungry, and more battery friendly alternative.
Re:Servo’s new home by devwastaken on Tuesday November 17, 2020 @01:10PM (#25127280)
Servo last time I checked used the same if not more memory (100MB) for basic pages. There doesn't appear to be major differences for that application yet.
Re:Servo’s new home by jhoechtl on Tuesday November 17, 2020 @04:08PM (#25129861)
Why should there be any great difference? Firefox is already written in a language where compilers have been optimized since decades.

Rust will not bring that magic.

Re:Servo’s new home by trevyn on Wednesday November 18, 2020 @12:28AM (#25133526)
Rust makes it much easier to do complex performance optimizations safely, particularly those involving concurrency. The canonical example is the multi-threading of Servo’s CSS processor.

It’s not an automatic magic bullet, but in a world where humans are still writing the code, Rust’s feature set makes many performance optimizations much more practical.

Re:Servo’s new home by jhoechtl on Wednesday November 18, 2020 @07:37AM (#25136080)
When did Firefox last crash on you? Aren't we optimzing on a problem which is practically non-existing?
Re:Servo’s new home by ben-schaaf on Tuesday November 17, 2020 @06:40PM (#25131382)
Rust isn't what can bring that "magic", but a rewrite sure can.
Re:Servo’s new home by nine_k on Tuesday November 17, 2020 @06:43PM (#25131409)
Why, better lifetimes' control could reduce the RSS, and reduce the GC time / energy.

(Provided that the C++ version is not strictly optimal WRT resource allocation; I think it's a rather safe bet.)

Re:Servo’s new home by devwastaken on Thursday November 19, 2020 @10:44AM (#25150621)
Hypothetically there should be possibilities for significant memory reduction by being able to deploy only what you need from the runtime instead of having everything loaded at once.

The memory usage part isn't really apart of rust or C, but rather that projects like chromium require significant hoops to jump through if you wanted to compile electron with only what you need.

I expect that servo will be tackling this issue, the browser works standalone and that's a hefty achievement in itself.

Re:Servo’s new home by asveikau on Tuesday November 17, 2020 @11:38PM (#25133268)
I'm not personally a rust zealot, but I know that rust builds on LLVM and therefore benefits from some of that very same engineering effort that was built to optimize C and C++.
Re:Servo’s new home by Subsentient on Tuesday November 17, 2020 @11:53PM (#25133348)
Clang usually has worse performance than GCC.
Re:Servo’s new home by trevyn on Wednesday November 18, 2020 @12:24AM (#25133500)
Is there anything specific GCC does for performance that Clang/LLVM could adopt? How is this race expected to evolve over time?
Re:Servo’s new home by PeCaN on Wednesday November 18, 2020 @02:42AM (#25134127)
it's not so much that gcc does anything specific so much as LLVM is just really really inefficient—they don't track compilation time at all so it's easy for releases to regress, half the stuff in there is academics implementing their PhD thesis aiming for algorithmic accuracy with little regard for efficiency, and LLVM's design itself is somewhat inefficient (multiple IRs, lots and lots of pointers in IR representation, etc)

that said this makes it an excellent testbed but compilation time will keep getting slower every release until they start caring about it

Re:Servo’s new home by roblabla on Wednesday November 18, 2020 @05:08AM (#25135061)
The rust compiler team now tracks LLVM performance, see [0]. They are able to spot regressions as soon as they get merged, and petition to get the change reverted or performance fixed when the impact is large. This has been fairly successful so far, but obviously needs sustained efforts if we are to ever dream of ever getting halfway decent compilation times.

[0]: https://nikic.github.io/2020/05/10/Make-LLVM-fast-again.html

Re:Servo’s new home by pmarin on Wednesday November 18, 2020 @01:45AM (#25133854)
Not answering your question but in my experience reading assembly output of the two compilers the unoptimized output of Clang is atrocious while GCC is closer of what a human could have wrote. Clang seems to have to do better in optimization passes to archive similar results. Usually Clang was a faster compiler in O1 but I don’t think it’s true anymore.
Re:Servo’s new home by PeCaN on Wednesday November 18, 2020 @02:27AM (#25134066)
that's just because gcc has certain optimization passes that can't be disabled

(that said gcc -O0 is still absolutely nothing like what a human would write)

Re:Servo’s new home by chubot on Tuesday November 17, 2020 @01:25PM (#25127512)
Has anyone tried Cobalt? It looks like something based on it could be an alternative to Electron.

https://cobalt.foo/

It's meant for embedded but that probably just means it's easy to build and uses low memory.

A high-performance, small-footprint platform that implements a subset of HTML5/CSS/JS to run applications, including the YouTube TV app.

Re:Servo’s new home by spankalee on Tuesday November 17, 2020 @01:32PM (#25127627)
Cobalt lags on modern web APIs pretty badly. I wouldn't recommend it.
Re:Servo’s new home by chubot on Tuesday November 17, 2020 @02:53PM (#25128790)
Can you elaborate on that?

To be clear I wasn't suggesting that Cobalt is a replacement for Electron. Electronic has a lot of its own APIs too.

Rather, I wonder if Cobalt could be better than Servo along some dimensions as an engine for something like Electron.

[dead]
Re:Servo’s new home by spankalee on Tuesday November 17, 2020 @02:08PM (#25128173)
I have to make code work in Cobalt as part of my job. You?
Re:Servo’s new home by kortilla on Tuesday November 17, 2020 @02:19PM (#25128340)
Parent was tone deaf but the point is that if you want a bunch of cutting edge APIs then Cobalt isn’t really for you. It’s light weight.
[dead]
Re:Servo’s new home by dang on Tuesday November 17, 2020 @03:10PM (#25129057)
Hey, please don't be a jerk in HN comments. If you know more than others, that's great, but the thing to do is to share some of what you know so the rest of us can learn something. If you don't want to do that, not posting anything is always an option. But please don't post unsubstantive comments and especially not nasty unsubstantive comments.

https://news.ycombinator.com/newsguidelines.html

Re:Servo’s new home by andrewmcwatters on Tuesday November 17, 2020 @03:19PM (#25129212)
I welcome you to go read the rest of my posts in this thread and get back to me on whether or not my posts are substantive, thanks!
Re:Servo’s new home by dang on Tuesday November 17, 2020 @05:46PM (#25130956)
I didn't mean that all of your posts were that way, only some of them.

Can I try my luck at persuading you a little? The reason we moderate like this is not because of morality or taste. It's because of the dynamics of a community of HN's size. The goal is simply to have a community that doesn't suck. That is in all our interests, including yours, as you're participating here.

There are other internet communities, smaller and more cohesive ones, and usually with a specific technical focus, that have a culture of shaming people for being wrong or knowing less. Arguably this can help to keep signal/noise ratio high by punishing low-quality, low-effort posts, demanding that people do their homework first, and so on. Some of these communities have functioned pretty well for many years.

But it's not an option for HN—this community is too large and incohesive. If we were to allow it, such tactics would soon spread, not in service of technical accuracy, signal/noise ratio, or anything like that, but just general douchery. Ignorant users would see knowledgeable users treating others that way and would follow the worst part of their example—not the becoming-knowledgeable part (that would require an effort!) — just the being-a-jerk part.

Then we'd see an exodus of users who don't want to be surrounded by that kind of thing. Then general douchery would comprise a higher percentage of what remained, making the next exodus even bigger. For a community like HN, this is how you get catastrophic decline.

In order to keep HN worth participating in, we have to have a culture that is on the good side of these nonlinear effects—i.e. we need interactions that raise overall quality rather than damaging it. Situations when a user posts something wrong or ignorant are a classic example. Correcting misinformation is good, but some ways of doing it harm the ecosystem while others benefit it. You have to consider not only the local (comment-to-comment) interaction but the systemic effect of such interactions as they compound.

That's why, in situations like this, we ask people who know more to share some of what they know, so others can learn. That way you're not just executing a local destroy operation on some mistake, you're helping to raise the level of knowledge of the community and the quality of the thread. These actions have systemically beneficial effects instead of toxic ones. If you (i.e. we all) do that repeatedly, you end up with a community that you actually want to be part of.

Re:Servo’s new home by Frondo on Wednesday November 18, 2020 @07:53AM (#25136178)
Holy shit, Dan! I just gained so much respect for you as a moderator (and therefore participant) in this community.

Taking the time to write out this very thoughtful, insightful reply to someone who may be knowledgeable but was throwing out hostile one-liners, that is impressive and gives such a great deal of credibility to the work you're doing here -- work that largely goes unseen, of course.

Amazing.

(I actually put down the piece of food I was eating, mid-bite, to write this out. You made my day!)

Re:Servo’s new home by chubot on Wednesday November 18, 2020 @03:16PM (#25141861)
Yeah honestly someone should try to make an AI clone of dang :) I think that would go along way toward solving the problems of the Internet...

I'm sorta joking but also serious because there is a big corpus to train from :)

Re:Servo’s new home by dang on Thursday November 19, 2020 @01:30PM (#25152650)
Re:Servo’s new home by Razengan on Tuesday November 17, 2020 @11:49PM (#25133321)
I wonder if they could also adapt into a game UI framework; design fast, cross platform, scalable game HUDs with HTML/JS.
Re:Servo’s new home by NetOpWibby on Wednesday November 18, 2020 @01:32AM (#25133796)
I’ve posed this question several times online and in the Servo repo and the response was always, “Umm, no, not interested.”
Re:Servo’s new home by pjmlp on Wednesday November 18, 2020 @09:41AM (#25137243)
The alternative exists since 20 years now, run a background daemon with local browser, or ping back in the platform WebWidgets.