Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We Need Chrome No More

redalemeden.com

Post reply on HN
Print view View original

361–370 of 536 posts

Re: We Need Chrome No More

#361

Post by PrototypeNM1 »

throwaway66666 wrote: post #329

If anyone's to blame it's mozilla's criminal laziness. There I said it. We were stuck with firefox 3 for 3 years (early 2008-2011), on a time where chrome was getting a +1 version every month or 2 months. I remember Mozilla "evangelist" employees hard at work, tweeting "version doesn't matter, it's the changelog that matters, chrome team could have changed two background colors and did +1 to the version counter". Gre…

Addressing your entire post would require further research on my part so I hope you'll excuse that I only address what I can recall off the top of my head.

> Fennec Fox for mobile, what happened to that?

That's just the code name for Firefox for Android. Ime it's fairly pleasant to use since flick scrolling was improved, and browser plugins on mobile are the selling feature.

> FLASH TO JS CONVERTER

Definitely sad to see Shumway die, but I think in retrospect it will be clear it was a forward thinking decision. WASM will obviate a lot of the effort for what has clearly become a historical feature.

> Mozilla did realize their mistakes got back from their 9 year vacation (2009-2018)

The origin of modern Firefox could be dated back to the origin of Servo in 2012 since it layed all the groundwork. I think this is the clear turning point in Mozilla's refocus, maybe push it back to 2015 when Firefox OS was dropped.

Re: We Need Chrome No More

#362

Post by iamaelephant »

pcwalton wrote: post #264
jorvi wrote: post #92

Earlier quoted context omitted.

Yup. I'm using Firefox on macOS as well and suffering from the same bug. If you use your Macbook with the resolution set to 'scaled' Firefox absolutely murders your battery. On top of that there's a bug where H264 video doesn't get properly accelerated. Both these bugs have been in Firefox at least since FF62. Both have alternatingly been marked 'priority 2' or 'fix-optional' which is absolutely baffling to me consid…

It isn't a bug; it's the fact that Firefox uses a transparent window and doesn't yet use Core Animation, which offloads scrolling to the window server. Switching to CA involves heavy lifting inside the compositor and is not an easy task. My planeshift crate does some of the work needed to make the WebRender path use Core Animation. In the meantime, setting gfx.compositor.glcontext.opaque to true in about:config helps…

Unless the intended behaviour is to cut battery life in half then yes it's a bug. Just because it's hard to fix doesn't mean it's not a bug.

Re: We Need Chrome No More

#363

Post by Gene_Parmesan »

clairity wrote: post #342
robocat wrote: post #315

Earlier quoted context omitted.

> online surveillance I think the drive behind Chrome was to compete with Microsoft, by making a browser that was good enough to deliver your applications, and so make your OS irrelevant. They have succeeded to the point where the only thing I would use Windows for now is Excel (and sometimes some obsolete development tools). Chrome (when not on Android) seems far more benign than most applications as far as "surveil…

yes, chrome was part of google's hedge against microsoft, but microsoft poorly understand the web and badly bungled their internet strategy, to the point that it seemed that they were desperately trying to cede the leadership position to anyone but themselves. google really hadn't needed to worry about microsoft there. make no mistake, chrome's primary strategic purpose was to control the web in ways that extended an…

Microsoft has always been a non-innovator. Their whole MO has been to wait to see what innovations catch on, and then dominate that space. The software/products they create in these spaces tend to be competent but not brilliant. They are almost a market-follower which happens to gain significant market share due to shrewd business tactics and, eventually, their sheer size. They have always mostly taken a 'wait and see' approach -- just look at the way Azure has come around as a response to AWS.

I agree with your views on privacy and Chrome. I've stopped using Chrome except at work, where (sadly) the Chrome dev tools are too valuable to ignore.

Re: We Need Chrome No More

#364

Post by furicane »

Aser wrote: post #259
VBprogrammer wrote: post #188

Earlier quoted context omitted.

I'm surprised there aren't more people on here who moved to chome for the dev tools. That is what moved me over there. Firebug was great in it's day but at some point Chrome's dev tools just became too much better to make staying on Firefox possible.

I'm not a professional web developer, but in the experience I've had building a couple of JS heavy sites, I've actually found the Edge developer tools to be the best. If there's one thing microsoft knows how to do well, it's build developer tools.

I've been in frontend for a long time, since 1999. The dev console is my workplace. I have to use and be familiar with all popular browsers. Chrome's dev tools are far better in my opinion.

Re: We Need Chrome No More

#366

Post by paulryanrogers »

tschellenbach wrote: post #257

Chrome is just a better browser. Also, Google already has all my information, so who better to trust with my browsing history? Firefox has >10% market share, it's not like anybody can ignore 10% of users. So this article to me sounds like pure content marketing for Firefox.

Article recommends Vivaldi and Safari too.

Re: We Need Chrome No More

#367

Post by zackmorris »

c-smile wrote: post #290

The post is missing one simple point: the complexity of browser's engine. The complexity of current web client stack is comparable with the average OS. Any modern browser has all components of Android OS for example - internal file systems, threads, VM, virtual native code execution, independent windowing and graphics stack, message routing, etc. That's like situation with Windows and WINE. Yes, you may run some win…

IMHO we got here because we chose the design-by-committee philosophy of organizations like the W3C over more of a waterfall approach built from first principles.

I would like to see a core HTML/CSS/Javascript spec that is a union of features proven to be available and work consistently across all browsers. The spec should incorporate ideas that are obvious in hindsight, such as the DOM diffing optimizations in frameworks like React, or running each tab in its own process.

I think that it should also take a broad perspective approach on things like padding and margin. For example, use simpler abstractions from layout engines like Qt or the layout-as-a-matrix-solver math behind iOS's Auto Layout constraints.

Find the commonalities to give us a high-level abstraction over all the div/span/table concepts of the web, then show a mapping from the abstraction to the quirks of how a div flows, for example. Maybe we could allow one transpiler to go from the solvable/predictable rules to the quirks of say HTML5.

The steps in this process could be relatively tiny and easy to test. It should render the DOM at > 60 fps on a Pentium 100 or equivalent, since games used to do that before they even had OpenGL.

I don't see any step of this process that is intractable. Once we had that, it should be relatively straightforward to build an open source browser as a base spec implementation. This is the sort of thing I fantasized about working on before I got so burned out.

Re: We Need Chrome No More

#368

Post by marcosdumay »

behringer wrote: post #150
leftyted wrote: post #35

Earlier quoted context omitted.

Chrome's main appeal was that it was much faster than firefox and IE. At everything. I had no technical knowledge when chrome came out, was using firefox, tried chrome, never looked back. It's not really possible to separate out the effects of advertising from the effects of the product actually being better. Maybe I installed chrome because I saw an ad, but I stopped using firefox because it was clear that chrome wa…

I've used firefox and never bothered with chrome as a daily driver. The speed issue is entirely marketing driven. Nobody cares about a few tenths of a second of load times until the ads were shoving stats down our throats that chrome was so much faster Firefox is and has always been perfectly adequate, except maybe for a few html5 canvas type apps.

Firefox was incredibly slow by that time. It was having maintainability problems for a few years, with unfixed memory leaks, deadlocks and core instabilities. Mozilla had a large ongoing effort for fixing those things, but it took some time until they got results.

Javascript heavy sites were also very slow on every browser, except for Chrome that invested in a good JS interpreter.

Re: We Need Chrome No More

#369

Post by kevin_thibedeau »

AsyncAwait wrote: post #340
Laforet wrote: post #288

Earlier quoted context omitted.

The decision to revamp their extensions API also cost a lot of goodwill from the community. FF today may be faster and more secure than two years ago, but from my perspective it has permanently lost some of its utility and appeal.

It could never deploy multi-process with XUL cruft. Then the argument would be that Firefox is too far behind. You can never win everybody.

Taking away XUL is not the problem. Leaving users with a crippled replacement is.

Re: We Need Chrome No More

#370

Post by cromwellian »

pcwalton wrote: post #270
cromwellian wrote: post #46

Earlier quoted context omitted.

It still is behind technically isn't it? Does Firefox support Site Isolation yet?

There are features Chrome has that Firefox doesn't. There are also features that Firefox has that Chrome doesn't, such as Stylo and WebRender. You can argue that Site Isolation is more important than large performance improvements (I think that's a losing argument, given that Site Isolation as presently deployed in Chrome doesn't even defend against compromised renderers and as such is restricted to being another Spe…

I'm not being glib, my current work involves security and privacy, and so I'm more sensitive / concerned about those issues than rendering performance. Current JITs like V8 even give me the willies in terms of attack surface area, such that I'm willing to stomach something like a WASM interpreter for my use cases.

I recognize that WebRender is a huge technique achievement and I absolutely love it when people achieve silky smooth 60fps renders, and if I was working on games or graphics or UX, I'd probably rank that over top of Site Isolation.

The original post mentioned process isolation and flash, so my comment was more a long the lines of 'it's still behind chrome (in isolation)'

Post reply on HN