State of web browsers in 2018 • Hacker News Board

Re: State of web browsers in 2018

#91

Post by kitsunesoba » Thu, Dec 06, 2018, 5:18 PM UTC

jcranmer wrote: post #71

sebazzz wrote: post #24

Earlier quoted context omitted.

How well is Gecko/Servo suitable for embedding in another application as a web browser? Apparently it is very easy with Chrome, why we have 25+ Chrome derivatives.

I can't speak for Servo, but Mozilla pretty firmly gave up on trying to have a good embedding story for Gecko and SpiderMonkey (in the name of being able to rapidly iterate on APIs, which never ended up settling down) somewhere around 2009. I contend that one of Mozilla's big mistakes was to give up on embedding.

RIP Camino, which served as many Mac users’ favored browser for years even after it stopped getting updates.

Re: State of web browsers in 2018

#92

Post by azangru » Thu, Dec 06, 2018, 5:18 PM UTC

vanadium wrote: post #79

azangru wrote: post #70

Earlier quoted context omitted.

> I'm a developer and a Firefox user. By developer, do you mean web developer? I tried to use Chrome dev tools, and they were incredibly easy to pick up; however whenever I try to use Firefox dev tools, they feel embarrassingly awkward. Perhaps it's just the familiarity with the tools (and if so, I am very curious to hear whether you prefer Firefox dev tools over Chrome's), but have a nagging feeling that it's just a…

Both have their pluses and minuses. It was delightful when auditing the experiences from a potential new client in Firefox Dev Tools the other day, and realizing that I could see, inline to the target DOM element, JS events tied to them while looking through the source. Saved me a good amount of time. When I'm looking for pure-play performance auditing, Lighthouse integrated into Chrome's Dev Tools are an obvious win…

> It was delightful when auditing the experiences from a potential new client in Firefox Dev Tools the other day, and realizing that I could see, inline to the target DOM element, JS events tied to them while looking through the source. Saved me a good amount of time.

Just tried that, and wow, that's really cool!

Re: State of web browsers in 2018

#93

Post by lostmsu » Thu, Dec 06, 2018, 5:18 PM UTC

I stopped reading the article after the statement about Edge and FF, that a lot of sites do not work in them. I am yet to see any of those websites outside of google.com. And the author did not name any examples. That made me think the author does not know what he is talking about.

Re: State of web browsers in 2018

#94

Post by wvenable » Thu, Dec 06, 2018, 5:18 PM UTC

Reinmar wrote: post #30

bergie wrote: post #19

Earlier quoted context omitted.

> I'm afraid that it will look like a Cold War's arms race – one of the parties will invest so much that the other will just lose economically (not being able to afford that many developers). Given all the "Safari is the new IE6" posts that have been here over the past few years, that seems to be already happening. Apple is lagging behind quite a bit with implementing newer web standards, especially in areas where th…

I'm not sure how long they can keep doing so. At what point it may affect iOS's reception in general? From my own experience, making the app I'm working on (CKEditor 5 - a web based rich-text editor) compatible with mobile Safari is right now impossible. None of the RTEs work well in mobile Safari because of its quirks and bugs. If Apple will overdo this (just like Microsoft did with IE6), that may affect iOS's usefu…

If you care about iOS, you don't build a web application you build a native one (perhaps using web technologies). This ultimately benefits users and Apple so there is no incentive to improve this.

Re: State of web browsers in 2018

#95

Post by xte » Thu, Dec 06, 2018, 5:18 PM UTC

dtech wrote: post #77

xte wrote: post #63

Earlier quoted context omitted.

Only a sidenote: did you think any human being can have properly reviewed webkit source code before using it? If not, how can be anything based on webkit considered to be safe, privacy oriented etc?

This is a pretty bad argument, as exactly the same can be said for the Servo/Firefox codebase

See above: it's true, but at least FOSS born and evolved projects may be known by "early" devs that see project evolution during it's time, on contrary when a company release million SLOC no one really know anything...

Anyway, in general my line is that all "modern browsers" must die, because browsers should be browsers, not "platforms" and websites should be hypertext, not application... I really dream a modern Plan9 even if I know nobody with enough competence, time and money to develop something like that exists today. I only can hope that a scandal and a disaster at a time we start to be tired, damaged and threatened enough that we support something like GNU/FSF to a level that produce free software and open hardware can be done by a community for the community itself... Well... A bit utopic...

Re: State of web browsers in 2018

#96

Post by jnurmine » Thu, Dec 06, 2018, 5:20 PM UTC

mike22223333 wrote: post #65

As a publisher, most of us do not care about Firefox either. They have done us no good. So we don't even bother to test on it. Their so called "Tracking Protection" is nothing more then a gimmick. It can be bypassed easily using methods other then cookies. All it did was reduce the revenue from ads because of no cookies (the legitimate way to track id's). It did an insignificant harm to Google. But it reduced the rev…

The tracking protection using DNT and the like is indeed a gimmick.

It is an ineffectual and passive way to hinder tracking.

Better ways are to be more active: to reject third-party cookies, disallow sites from storing anything, uninstall Flash and such, run AdBlock Plus, run uBlock Origin and so on.

I mean, if I visit a random site which is full of non-original content and/or simple regurgitation of press releases, why should I care about its business model and ability to monetize through tracking me? Sorry, show me value from the get-go and I might come back and let you "monetize" through whitelisting or something.

Re: State of web browsers in 2018

#97

Post by mrweasel » Thu, Dec 06, 2018, 5:21 PM UTC

asituop wrote: post #69

tapoxi wrote: post #64

Earlier quoted context omitted.

There are enough large, interested parties (Samsung, Microsoft, Mozilla) out there that they could literally fork Blink if organizational issues arise and gain a sizable install base.

If Mozilla can't reach enough user now, why would they do with a fork of Blink ? It's not a web engine problem, it's just that Google is too powerful and Chrome is everywhere, so they decide. And actually it is better for innovation not to use to Blink. WebRender is potentially much more performant than any other engine and it would never had existed if Mozilla were just aligning with Blink

Exactly, people seems to only remember that Mozilla was a little behind Chrome some years back. That by no means the case anymore. Unless you're somehow embedded in the Google ecosystem, there's no reason to prefer Chrome over Firefox.

It's not like the other Blink or Chrome based browsers are doing great, sure Brave and Vivaldi is getting some press, but they have even less users than Firefox.

Sadly Firefox isn't peoples mind anymore when they think browser. It's IE all over again, but with Chrome, Only Chrome is moving forward. I don't know if there's anything Firefox can do, other than marketing. It's currently the best browser out there, but that was the case when IE6 dominated the web as well.

Firefox doesn't need to dominate, but hitting some like 20% market share would be fantastic.

Re: State of web browsers in 2018

#98

Post by larrik » Thu, Dec 06, 2018, 5:21 PM UTC

jcranmer wrote: post #73

Yizahi wrote: post #46

Earlier quoted context omitted.

Arguably Google is already abusing their monopoly power but in the way that people like. I'm talking about AMP. In the future AMP/regular page boundary will blur and Chrome/GSearch/AMPpages will merge in the ubiquitous googlenet hosted on google servers (because it will work better than regular internet pages). Almost everybody will be happy after this and Mozilla will fall into obscurity. Basically you can't create…

> PPS: also to add to less known evil moves by Google - their U2F websites refuse to work in Firefox, despite it supporting U2F for a long time. The one that still sticks in my craw is that they never bothered to add support for Hangouts in Firefox. Hangouts was modified to use WebRTC in Chrome, but relied on a plugin in Firefox--despite Firefox actually implementing the relevant WebRTC features first.

Hmm, I use Hangouts in Firefox...

Re: State of web browsers in 2018

#99

Post by wvenable » Thu, Dec 06, 2018, 5:21 PM UTC

mike22223333 wrote: post #65

As a publisher, most of us do not care about Firefox either. They have done us no good. So we don't even bother to test on it. Their so called "Tracking Protection" is nothing more then a gimmick. It can be bypassed easily using methods other then cookies. All it did was reduce the revenue from ads because of no cookies (the legitimate way to track id's). It did an insignificant harm to Google. But it reduced the rev…

> DNT should have always required manual user intervention to be respected.

What sane person would agree to be tracked?

Re: State of web browsers in 2018

#100

Post by dbolgheroni » Thu, Dec 06, 2018, 5:22 PM UTC

sstangl wrote: post #11

As a Mozilla employee, speaking from my own limited perspective, it seems the company as a whole broadly agrees with this analysis even if it isn't vocalized quite this succinctly. Chrome started out being force-installed alongside Flash, and there is no solution to the bundling problem -- the problem is even larger in scale than the article suggests because of Chromebooks affecting desktop. People who use Firefox in…

As a long Firefox user, I did switch to Chrome/Iridium lately.

When Firefox decided to attack the problem of market share with engineering, 2 things were happening (almost) at the same time[1]: 1) the deprecation of the old add-on engine and 2) the introduction of the new Quantum engine.

For the 1), while I understand there are some reasons in doing so (it was not a stable API as it was always breaking the extensions and security-wise), I think these problems could be managed. On the other hand, there were lots of good add-on depended on this API.

Particularly, even when Firefox was behind technically-speaking, that was the main reason I used it. Firefox replaced the old API without offering all the new interfaces with the new API. So, I lost what made Firefox really unique.

For the 2), I think it was a consequence for the 1). Firefox was finally "better" with Quantum. Of course everyone likes a responsible browser, but I actually didn't see that much difference. And there was another browser which was already really good: Chrome.

Suddenly, there was no real advantage in using Firefox anymore. Someone my say privacy. But in some corners of the Internet, it's already impossible to navigate without some privacy-wise add-on/extension, which usually work on both browsers.

Added to these, if you go deep down, security-wise there are some real technical advantages in running Chrome instead of Firefox in some OSs. For instance:

https://marc.info/?l=openbsd-misc&m=152872551609819&w=2

[1] I'm pretty sure a lot of other things were happening, but I'm in the perspective of the user here.