State of web browsers in 2018 • Hacker News Board

Re: State of web browsers in 2018

#161

Post by Wowfunhappy » Thu, Dec 06, 2018, 7:50 PM UTC

Izkata wrote: post #156

untog wrote: post #155

Earlier quoted context omitted.

Broadly speaking, yes they do. It is faster in the vast majority of cases, and non-techie friends of mine speak enthusiastically about the lightning bolt icon. Of course, they don't understand the implications of it.

> and non-techie friends of mine speak enthusiastically about the lightning bolt icon. Which makes me think it's just confirmation bias. They expect it to be faster, and so perceive it as faster. Just like when Windows Vista was presented to users as the new Windows 7 (only the branding was changed for the experiment), and they openly stated it was noticeably better/faster than Windows Vista.

I'm not quite so quick to dismiss this out of hand, despite my own (negative) experiences with AMP.

Google is certainly capable of making mistakes, but I find it hard to believe they would invest so much into AMP without strong metrics to indicate that AMP pages load faster overall. And on a theoretical level, the explanations of why AMP should be faster also make sense.

Re: State of web browsers in 2018

#162

Post by fchristant » Thu, Dec 06, 2018, 7:56 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…

hi, author of the article checking in.

Thanks for commenting in the open from Mozilla's perspective, even if its just one voice, it's much appreciated.

I agree that Firefox will remain an attraction to the group of people caring about openness, standards, and their privacy. My deepest concern is on how small that group is going to be. Will it settle at the current ~10%, or will it be as low as 5%, which I consider very dangerous grounds to be in.

As for Google's abuse of market power, I strongly believe it has already happened, in plain view. There's no doubt in my mind that dominance in one market has been used to capture another market. We don't have to wait for that "mistake", it has already happened.

These abuses are already under investigation, but I'm afraid it won't change the outcome. They'll pay some fine in a few years, but none of the browsers will be uninstalled.

It won't matter much, but added an update to the top of the article suggesting people to install Firefox.

Re: State of web browsers in 2018

#163

Post by fchristant » Thu, Dec 06, 2018, 8:04 PM UTC

austincheney wrote: post #7

> At least those browsers are compatible with the web. Websites will work in them because websites are built to work in Chromium. That is a poorly drawn conclusion. Edge is compatible with the web. Developers are solely developing for Chrome and not caring for cross browser testing. As a Firefox user I sometimes encounter this myself.

hi, I'm the author. Agree with your point when you read it like that, so I'll clarify what I meant.

"Compatible with the web" in the context of this article means compatible with the web that is primarily built to work on Chromium/Webkit. Because at this point of time, the web is Chromium.

You're right that Edge is compatible with web standards, albeit that they are lagging a bit behind in implementing newer standards. By that logic, IE11 is also compatible with web standards.

Re: State of web browsers in 2018

#164

Post by fchristant » Thu, Dec 06, 2018, 8:10 PM UTC

Wowfunhappy wrote: post #141

The author of this article hand-waves away the speed advantages of one browser engine over another. I'd be inclined to agree, except, am I the only one who has seen a marked speed advantage in Firefox Quantum over other browsers? I'm reminded of how I felt about Chrome in the early days. Everything just seems to load in ever-so-slightly more quickly. I still use Safari on macOS due to Apple's nice touchpad gestures,…

hi, author checking in.

Just doubling down on my statement. I've seen hundreds of discussions and flame wars on browser performance, RAM usage, the ability to keep open 200 tabs.

I mean it when I say that I cannot reproduce any of these differences. Perhaps I could if I would actively measure things, but in terms of just experiencing the speed of all major browsers, I really do not see a meaningful difference.

It could be device related, but I'm on a 6 year old PC (a pretty sizable one though), not having any issues on my mediocre work laptop either.

Doesn't mean it doesn't matter in absolute terms, I'm just saying I don't see it. The more important point though is non-technologist do not tend to pick browsers in such rational ways.

Re: State of web browsers in 2018

#165

Post by fchristant » Thu, Dec 06, 2018, 8:15 PM UTC

miles_matthias wrote: post #88

Great article. Discussion question - what incentives need to change to make Apple invest more in Safari? They've always been invested in native apps first and foremost, because that's where they make their money - the app stores. The other place they make their money is hardware. And it's hard to say you buy a MacBook or iPhone because of safari. You don't. So I just don't know what it will take for Apple to make sig…

hi, author here, and thanks.

I have a straightforward answer. Build incredible PWAs that users will miss out on on iOS, because it has partial support at best. You'd still serve 80% of the market.

I'll admit its a theoretical answer, as most commercial organization would not risk losing out on iOS users. They're not the biggest group of users, yet some of the most valuable groups of users, commercially speaking.

Re: State of web browsers in 2018

#166

Post by fchristant » Thu, Dec 06, 2018, 8:21 PM UTC

Domark wrote: post #12

This blog post was tough to read! Seems to have capped out at a 6th grade level. Re: the topic, the author shows clear bias toward Firefox. It might be better to just acknowledge this: the best browser wins on each platform. Period. Safari is up there because it runs the best/fastest/with least trouble on iPhones and iPads. 20/20 means PERFECT vision. Above that is ‘abnormally good’. In other words, the author has a…

hi, author here.

"This blog post was tough to read! Seems to have capped out at a 6th grade level."

I'm not a native speaker, and this is the best I can do. I was expecting less than 6th grade, so thanks!

"Re: the topic, the author shows clear bias toward Firefox"

Yes, the article literally says I am biased towards Firefox success. You didn't discover a secret plot or anything.

"It might be better to just acknowledge this: the best browser wins on each platform."

No, the one shipped to user's home screens wins. Even crappy ones.

"20/20 means PERFECT vision. Above that is ‘abnormally good’."

If it would mean perfect, there would not be a level above it. I do agree the statement isn't as clever as I planned it to be.

Re: State of web browsers in 2018

#167

Post by Izkata » Thu, Dec 06, 2018, 8:38 PM UTC

Wowfunhappy wrote: post #161

Izkata wrote: post #156

Earlier quoted context omitted.

> and non-techie friends of mine speak enthusiastically about the lightning bolt icon. Which makes me think it's just confirmation bias. They expect it to be faster, and so perceive it as faster. Just like when Windows Vista was presented to users as the new Windows 7 (only the branding was changed for the experiment), and they openly stated it was noticeably better/faster than Windows Vista.

I'm not quite so quick to dismiss this out of hand, despite my own (negative) experiences with AMP. Google is certainly capable of making mistakes, but I find it hard to believe they would invest so much into AMP without strong metrics to indicate that AMP pages load faster overall. And on a theoretical level, the explanations of why AMP should be faster also make sense.

Note the phrasing, that's an important part of what I'm referring to: they're talking about the icon instead of the speed directly.

Re: State of web browsers in 2018

#168

Post by skykooler » Thu, Dec 06, 2018, 8:42 PM UTC

Izkata wrote: post #159

organsnyder wrote: post #48

Earlier quoted context omitted.

During the IE-6 dark ages of the early 2000s, Mozilla's browser (the Mozilla suite, then standalone Firebird, finally renamed to Firefox) was positioned in a similar way: it was the niche browser for people who cared about an open web. It stayed that way for quite a few years, slowly gaining traction by implementing features—tabbed browsing, popup blocking—not found in the dominant IE browser. I could see a similar a…

And they already have (at least) one they could use as a starting point: Containers

Which is an excellent feature. Another one is WebVR.

Re: State of web browsers in 2018

#169

Post by skykooler » Thu, Dec 06, 2018, 8:50 PM UTC

Semaphor wrote: post #130

SamWhited wrote: post #55

Earlier quoted context omitted.

That's actually really interesting; I don't tend to have a ton of tabs, but coworkers always told me Firefox did a terrible job with 100+ tabs and that Chrome worked better. This always made sense to me because of the multi-process architecture of Chrome, but I've never seen it myself. So you're suggesting that Firefox actually performs better now? Maybe all my coworkers complaining about it are just repeating lore t…

I don't understand how anyone can claim FF is faster. I love FF and I switched completely from chrome. But Chrome is far faster. I don't know maybe it is because I have an abundance of RAM (32 GB) and others don't and that's why they think FF is faster. But for me, it's easy to test. Go to http://addic7ed.com/ and open the (very long, nearly 5k entries) [Select a TV Show] drop-down. On FF (and IE/Edge FWIW) there is…

Interesting. I have pretty much the opposite experience on MacOS; with Chrome, if I open a bunch of tabs (~50), my memory usage quickly goes to 15.9 of my 16 GB, and then my whole system slows to a crawl. In Firefox, the same number of tabs are using less than half of that, and everything is still running smoothly.

Re: State of web browsers in 2018

#170

Post by jolmg » Thu, Dec 06, 2018, 8:54 PM UTC

LamaOfRuin wrote: post #154

xte wrote: post #85

Earlier quoted context omitted.

That's true, but at least Firefox born as FOSS so there is a community that potentially see the code growing by the time so while they may have lost something as individuals as a community there is a certain knowledge of the code. Webkit born open (KHtml, Kde's Konqueror html engine) but subsequent evolution happen inside few big companies and sources are release "en mass" so there is essentially "no community" that…

I'm pretty sure that as far as lineage goes, Chrome was actually born more FOSS (KHtml) than Firefox was (Netscape). Firefox is certainly more FOSS in spirit now than Chrome has ever been, but Mozilla is certainly not immune from the kinds of shenanigans possible for any open source project that is entirely controlled by a single large organization.

Chrome's closed source though. You probably mean Chromium. Still, I'm not sure how much in the spirit of FOSS it is to have an open source version to drive development while marketing the closed source version which might be filled with spyware for all we know.