Chrome 100 | Hacker News Reader

Chrome 100

chromereleases.googleblog.com

171 pointsNiekvdMaas120 comments

Early chrome was a breath of fresh air. Design was light, it was fast, and it felt like the promise of the web will soon pay off.

Unfortunately, it helped kill the hated IE6/7 dragon only to take it's place. If you value the free internet and open standards, use Firefox. It is not important how's good chrome is or how beautiful it is (it actually isn't, early design was way better), because, after the demise of Opera, Firefox is the last alternative browser engine we have. We don't want one company decide the standards, it'll end badly.

I think the larger tech community should be made more aware of this. Maybe if the likes of LTT or MKBHD (popular YouTube tech channels) could be convinced to do a video on it, it could help get more people into the Firefox camp

Doesn't Chrome already effectively decide standards? What do they do that isn't standards compliant anyway?

AFAIK they push new web standards all the time, and even if they didn't, they have the vast majority of the market share so the boat's sailed on Firefox imo.

Agreed. I also always feel like linux communities or hackernews is gaslighting me into thinking that firefox is just as fast as chrome. It's not. There is a tangible speed difference when using it.

I use it on a M1 and in my (totally subjective) experience it goes in order of fastest to slowest: safari, firefox, then chrome for speed on heavy websites...

I just ran a little test to see what the difference was on my end.

I opened a private tab in Firefox and Microsoft Edge (which is for all intents and purposes the same as chrome) on a 2020 iMac. (I uninstalled Chrome and installed Microsoft Edge when I found that their css styling of the chromium dev tools was easier to read)

I wanted to test a commonly visited heavy website. I settled on reddit.com due to it's famous weight.

Before loading the website, I opened both browser's dev-tools, switched to their network tabs and disabled both of their caches.

These were my results:

-----------------------------

Firefox (with uBlock Origin):

- DOMContentLoaded: 2.84s

- load: 4.19s

- Network requests: 291

-----------------------------

Microsoft Edge:

- DomContentLoaded: 3.99s

- Load: 5.79s

- Network requests: 318

-----------------------------

I double checked the network tabs and noticed that Firefox received 291 network requests while Microsoft Edge received 318 requests. Thinking that uBlock Origin may have blocked some ads and trackers making the site load faster, I disabled uBlock Origin on firefox and reran the test (again with the caches disabled).

-----------------------------

Firefox (without uBlock Origin):

- DomContentLoaded: 3.02s

- load: 5.14s

- Network requests: 305

-----------------------------

By my (admittedly small) testing, it appears that Firefox is faster than Microsoft Edge (which is just a restyled chrome).

Also, this really highlights just how great uBlock Origin is on Firefox and how painful it is for Google to push Manifestv3 on their browser (and by extension, other browsers) since it neuters uBlock Origin.

There's also webkit, though there is no webkit browser on Windows, as far as I know.

I do use Firefox, but Mozilla really goes out of their way to betray their core audience more and more with addons like Pocket being integrated into the browser along with advertisements in the new tab window.

In either case it always feels like no matter what browser I'm using there are huge tradeoffs in one sense or another, which I never felt when I originally started using Firefox early on or early Chrome.

> I do use Firefox, but Mozilla really goes out of their way to betray their core audience more and more with addons like Pocket being integrated into the browser along with advertisements in the new tab window.

They also removed many useful features and hide useful settings... (one example being the RSS feed finding feature .... luckily there's an addon that you can install that does a pretty good job, it is called Awesome RSS)

> there is no webkit browser on Windows

But there is for every other OS.

Macs can use Safari, obviously.

On Linux, KDE users (or other Qt based desktops) can use Falkon, and GNOME users (or other GTK based desktops) can use GNOME Web.

Yeah, I didn't mention Linux because there are actual builds that are baked into many systems or easily installed. On Windows there seems to be a way to run webkit, but not viably as a daily driver web browser.

Webkit browsers in linux feel a bit half baked as well, though, when you compare it against Firefox and Chromium.

> Webkit browsers in linux feel a bit half baked as well, though, when you compare it against Firefox and Chromium.

They're just browsers. They don't have built-in web services, they don't dial home, they don't sync your history and bookmarks with some unknown service. Frankly, this is a positive.

Yes, but they’re also not extensible as far as I know, which is the much bigger issue with them. If you don’t find yourself needing extensions that’s just fine but I rely on several to make my web browsing experience better or more useful.

I’m writing this comment in luakit. It doesn’t support extensions from any other browser’s extension repos AFAIK, but the extensions it comes with out of the box (though they’re not all enabled by default) cover ~80% of the things I use extensions for in Firefox, and the things they do cover kind of work better.

I miss cosmetic filters from µBO and should implement something like Privacy Redirect, but vertical tabs work more easily (they don’t require hacking around user.css to disable horizontal ones, though there are no tree-style ones), there’s a simple Stylus alternative, userscript support, JS whitelisting/toggling with a modifier-free binding, and of course native vi/Pentadactyl-like bindings which make extension-provided ones feel viscerally painful (they’re not eviscerated whenever a page is in the process of loading or unloadable, and searching stays incremental).

> there is no webkit browser on Windows

WebKit does build on Windows, and it's possible to use bare WebKit: https://schepp.dev/posts/running-webkit-on-windows/

Just as IE6 took the place of the hated Netscape Navigator 4.7!

Die a hero, live long enough, yadda yadda

> We don't want one company decide the standards

It's already the case, Firefox lost a long time ago

Then do your part regardless. Use Firefox everywhere, and eventually, it'll have enough users to prevent that one company from fully owning the web and the standards.

I was a Firefox user, but Firefox these days pushes all sorts of Firefox advertising. I want a blank page when I load my browser; don't do the Microsoft thing and change it to your ads. I don't care about your color schemes. I don't want your proprietary extensions added after I remove them.

> I want a blank page when I load my browser

Settings->Home->"Homepage and new windows"->Blank Page

I know. I don't want to keep doing that. I don't care how good your new features are.

I set it once per installation, then it's blank forever. If it's getting reset for you, maybe it's a Windows-vs-Linux thing or something?

Part of the reason that IE6/7 was so annoying is that those browsers had a large market share but also lacked the features of the other leading browsers so you always had to bend over backwards to make your code support both.

Chrome does not have that problem. I fully appreciate the monopoly/privacy concerns, but Chrome being the market leading browser is 100x better from a development perspective then when you had to worry about IE6.

Well, I recon safari counts for something? I guess they forked a while back from whatever chrome is using.

But yes, support Firefox. I've been using it for a decade, there have been ups and downs, but at the moment there's nothing I would change with it.

I have to have chrome installed, like everyone else, for that one form that only works in chrome (or, I use it as the browser without any ad-blocking for when I need that..), but besides that I use it for everything work wise. The argument about the devtools isn't true anymore.

> I recon safari counts for something? I guess they forked a while back from whatever chrome is using.

Safari used WebKit, and Chrome initially did as well. In 2013 Chrome forked WebKit to make Blink: https://blog.chromium.org/2013/04/blink-rendering-engine-for...

I use and love Firefox, but I don't think it's really fair to say Chrome is the new IE. The biggest reason why IE was a nightmare was because they were slow to implement new things, and had weird unpredictable behavior and bugs that only showed up in IE. You also had to have Windows in order to test with it. On top of that, crappy or nonexistent dev tools. Also (not their fault, but) many company IT policies required use of IE. Installing Firefox or Chrome made you a rebel ;-)

Chrome is far from perfect but they implement things fast and have great dev tools. If anything, I feel like Safari is the new IE. You have to have a Mac to test it, it's slow to adopt new features (sometimes aggravatingly slow), and it breaks in weird ways that Chrome and Firefox don't. I can't speak to the dev tools in Safari but I'd love to hear from someone how those compare to Chrome.

The issue in this case, I believe, is that Google is using its dominance as the primary rendering engine to strongarm its own standards. If they want something, they just implement it in chrome and everyone must follow, least the smaller browsers fails to load more and more websites.

I would say the opposite. IE was popular long ago because they were fastest to implement new features. IE is the reason XMLHttpRequest has such a silly name. IE supported CSS grid 5 years before anyone else[1]. The problem is they tried to force features on everyone instead of collaborating. That started decades ago with the <marquee> tag trying to stand out against netscape, and never really stopped until IE11 got feature frozen. Now IE is left with a "grid" that's different from everyone else's "grid", because everyone else waited for consensus.

Google is playing the exact same game now with things like manifest v3 and federated cohorts, etc. They're winning the game so far. They may keep winning forever, or they may not. We can only wait and see.

[1] https://caniuse.com/css-grid

> I use and love Firefox, but I don't think it's really fair to say Chrome is the new IE.

Nobody wanted ActiveX on the web except Microsoft.

Nobody wants Manifest v3 except Google.

> You also had to have Windows in order to test with it. On top of that, crappy or nonexistent dev tools.

Same with Safari and you can use VMs for both of them.

Eh, I don't think I count macOS/Safari as usable inside a VM. Unless you are running a macOS VM on top of macOS, which defeats the purpose.

You can run a hackintosh in a VM but it's not a trivial matter for someone who doesn't know anything about macOS, or hackintosh's, or configuring hardware emulation. I also had it break on me regularly in ways that took lots of time and ducking. I finally decided that if Apple didn't want me to support macOS/Safari, then I wouldn't.

Apple is working with Google, Microsoft, Mozilla, and others to make web standards work more consistently across different browsers.

https://web.dev/interop-2022/

Congrats to the Chrome team! Obviously with market dominance will come a lot of criticism (some very much deserved) but you don't get there unless you made something truly awesome for people. I remember the outrage for the forced / auto-updates and am so thankful for their insanely aggressive installer today.

One small non-dev nitpick recently... I am a fan of keeping things "zen" or slim for productivity. Basically annoyingly hyper-focused on only having essentials. Today though, my Chrome browser is a mess of things I can't get rid of on my window:

- Share Icon

- Star Icon

- Extensions/Jigsaw Icon

- Reading List / Bookmarks Icon (new?)

- Profile Icon

- Tab arrow icon

- More icon

Less is sometimes more! I miss the minimalism. At least let me remove. Disgusting pic below:

https://imgur.com/a/CHVLaXQ

You just inspired me to dig through the flags -- at least the side panel is removable.

chrome://flags/#side-panel

It'd be useful if this side panel could be at top or bottom instead. I usually have two chrome windows side by side (left and right of the screen), so they're more vertical than horizontal

Or if they did not change the viewport size and instead appeared as a popup like how it is on Edge.

I find the small, label-less, monochrome icons undistracting, personally. If they tack on a few more I might start grousing, though.

On the other hand, for some reason I cannot get myself over Firefox's default UI cruft in comparison.

Firefox has a way to remove the icons though. (Right click, customize toolbar, drag off what you don't want)

It’s always baffled me that Chrome lacked toolbar customization entirely, even in the early days back when such a feature was practically a given in any serious desktop software, and standard across several different platforms (Mac/Cocoa, Win32, Qt).

It’s such a cheap feature to add and maintain and has practically zero overhead for less technically inclined users. Really strange that Chrome doesn’t have it.

They maybe knew they were able to take over IE monetization with their toolbar and someone could do it to them. Same reason you can't set a custom home page for new tabs (they only moved to that later though).

I hate to tell someone with a solution they're happy with to change, but consider firefox. It lets you customize your toolbar and remove/add/resize whatever you want, and anything you still want access to but don't want to be immediately visible you can throw into the overflow menu so you can still get a dropdown with all of them but only have one symbol in your toolbar the rest of the time.

I’ve stopped using Chrome a while back, due to it being a too integrated with Google services.

However, Chrome is an engineering marvel and well polished.

Congrats on version 100.

You could definitely argue it's the most important piece of software in the world, possibly besides whatever popular OS is running it. For the vast majority of humans, it IS the internet.

What other software would appear in the top 100? I'd guess curl, the major C compilers, the major OS kernels, Microsoft Office and Adobe Acrobat to name a few.

I'd guess Firefox would make the list too, despite having a much lower install base than Chrome.

If you include hosted services, then I guess the major ones like Google Search, Facebook, Instagram, WhatsApp, YouTube, TikTok, Twitter, Reddit, Wikipedia, etc. I guess email providers too: Gmail and Hotmail at least.

Nginx, Java jdk/jre

If mobile counts; Safari, Facebook, WhatsApp, Snapchat, TikTok?

There was a recent post on HN where it was tangentially mentioned that Chrome has 3.1 billion users! There can't be much software in the world with that kind of reach. I guess stuff like curl, but certainly not end user software.

Not user facing, but there are estimated to be over a trillion SQLite databases in use

https://www.sqlite.org/mostdeployed.html

Whatever Cisco IOS services comprise the networking stack of core backbone routers are effectively used by every single user that connects to the Internet. That's every human user but also all of the devices that connect without needing a user, which is probably a much greater number than the entire human population of the planet.

cough libc.so..

I mean.. For many many humans, the Internet is Facebook running on a mobile device. Also, do they use Chrome in China?

So I don't know about vast majority.

Definitely one of the most reliable pieces of software that I use though, no doubt about that.

> Also, do they use Chrome in China?

yes https://www.scmp.com/abacus/tech/article/3103747/google-chro...

*for certain definitions of importance

If Chrome disappeared I think everyone would still have access to the internet and a lot of people would argue that we'd all be better off for it.

You are right though, it can and does have a ridiculously large impact on people's lives.

Sure, there are competitors so that if Chrome disappeared, we’d be fine.

But imagine if Chrome had never existed. Chrome was so much better when it came out that it saved us from Internet Explorer hell.

I don’t want to picture what Microsoft would have had in store for us if Chrome hadn’t come around.

I feel like Firefox fumbled a bit in the early days of Chrome that many people switched over to Chrome.

I might be remembering the timeline wrong, but I was perfectly happy using Firefox when I first heard of Chrome. When I switched from Firefox to Chrome I was very impressed with the improvements in design, performance, and greater set of useful features. Definitely some steps in the right direction, but I believe we had already been saved from IE. These days there are many browsers to choose from (though most seem to be built on Chromium, which we wouldn't have without Chrome) and the competition is fun to spectate.

The big improvement for me that Firefox had over IE was tabs, though I'm sure there were many other UX improvements that my young mind didn't recognize or hold onto.

Your timeline is correct. But we all accepted that by switching to Chrome we were sending our entire browsing history to the biggest advertising giant to ever exist.

I would argue that Opera were better than Chrome feature-wise.

The only interesting thing Chrome added was process per tab, which isn't particularly interesting. Firefox had at least 25% market share in 2010.

I'd say the most important software is Excel.

No, that's the most abused software in the world. Poor lil' Excel forced to do things it really never should.

The same could be said for the browser

I mean, I don't think students are forced to run an operating system in Excel... (yet?) ;)

The browser is certainly being tasked as an operating system to run “web apps”

No, but you could argue it's among the most used pieces of software in the world. That's a great distinction. Chrome is entirely replaceable.

I only use it when some crappy site is only tested in Chrome and fails to run in Firefox. 10 years ago, I had to use Internet Explorer sometimes for the same reason.

Nah, that was gcc.

Time to find out how many applications have regexes with a user agent parser with version /d{2} instead of /d+.

I haven't followed exactly where they landed on this but both Firefox and Chrome have been testing user-agent strings with version 100+ for a while, and have (or at least had) a fall back mechanism to present the version as 99.xxx where the xxx contains additional information with the true browser version. There's also a longer term plan to phase out User-Agent headers as they exist now, more details at https://blog.chromium.org/2022/03/chrome-100-beta-reduced-us...

Anyone noticed, how Google Meet screen sharing on macOS is broken on a daily basis? I then always check for an update and then after a restart of chrome, it works again. Next day, it has some weird issue again.

But boy is it annoying. I wonder if it is an Apple Silicon only issue.

I think this may be a somewhat more specific to your environment...I'm running on a new generation macbook pro (so M1+) and haven't had any trouble screen sharing

No, this is definitely a much more widespread issue than that. Every single person that has tried to screenshare in my Google Meet meetings the past week has experienced the same problem.

Same issue here on Intel MacBook Pro 2020.

It seems the OSX permissions for Chrome Screenshare, Mic and Video get reset quite frequently. I can't pinpoint exactly what triggers it. If it's a Chrome update or OSX update.

EDIT: Just updated Chrome to 100 and sure enough it lost screen recording permissions. Camera and Mic were kept though.

If I had to guess it probably has to do with how Chrome updates itself, which is likely causing the OS to see it as a new binary that hasn’t had permissions set after each update.

I've been seeing this problem too at $JOB. Amusingly, the people on my team that use Firefox don't have these issues, but the Chrome users are seemingly losing the ability to screenshare daily.

I believe they hit this milestone a little faster than you might expect (if you've been using Chrome for a long time) because previously Chrome was on a 6-week release cycle and they recently (within the last year or 2) switched to a 4-week cadence.

https://chromium.googlesource.com/chromium/src/+/master/docs...

Is the phenomenon described at http://www.chromeisbad.com still happening? Does the Keystone updater cause slow performance on MacOS when using external displays?

There was never really any good evidence backing it.

To add, the author only profiled within Resource Monitor, which was like the cause: WindowServer shouldn't have a massive increase of CPU, but a bug since Catalina sometimes does cause it to do super inefficient WindowServer stuff: https://news.ycombinator.com/item?id=25401681

The site was made mostly to rile up HN users. So HN crowd can indulge their daily ritual of bashing Google. AFAIR the post reached the top spot of front page. Although it was flagged later.

Really wish they'd bring back being able to edit shortcut keys yourself. I end up having to compile my own version so that I can have the shortcuts I want.

https://github.com/ungoogled-software/contrib/blob/master/tw...

Under macOS at least you can set custom key shortcuts for any function in any app that has a corresponding menu item under System Preferences > Keyboards > Shortcuts.

Really wish other OSes had a similar feature, though it’d be more hit or miss there due to how often standard menubars are eschewed under Windows and Linux.

I think Chrome may be the single most impactful piece of software with regards to end-user security. Click to play plugins, sandboxed flash, sandboxed renderer, process isolation, auto updates, etc.

Chrome completely changed the threat landscape in just a few years.

What about privacy? Does it not phone home with every single browser request?

It doesn't, no.

What a horrible web browser, created by an abhorrent company. Safari, Firefox, and Edge should be recommended to our friends and family at every opportunity to help counter its market dominance.

Tell us how you really feel

last time I checked Edge was running in chromium as well

and if you wanna do yourself service use Vivaldi instead Edge

Honestly one of the first steps that has to happen when Google is broken up is put Chrome in a foundation with open governance.

Pretty neat to see them list out all the bug reports and bug bounties paid. I don't know if that's common but it's the first time I've seen it.

It's also a little odd to see a release announcement where they call out that none of the new features are in this post but will be announced later on.

Congrats to the Chrome team! In my opinion, easily the most important tech project/product in the world. The amount of innovation built on top of it is simply mind boggling. We've all benefitted from their incredibly hard work.

I look forward to seeing the progress made over the next 100 versions.

On a side note, Chrome for iOS is much faster and easier to use than Safari, which always cracks me up.

Chrome for iOS is Safari with a skin on top

The API for loading web pages, yes, so it's not running Blink or V8, but people choose to use (or not to use) Chrome nowadays for its integration with Google services and sync across devices. The only difference is that you don't get the wide compatibility Chrome affords via its engine, but IMO Apple's control over blocking JIT-compiled code allows them to maintain enough global browser marketshare to ensure developers at least test against two browser engines and not just one.

Not until chrome implements extensions.

Right click to reopen tab is missing and I will never forgive them

Yes I know I have to click outside the tabs itself, it is stupid and cumbersome.

what ever happened to the plan for chrome to get its own certificate store, anyway? It was announced a while back but never happened

feels like it would be a good thing especially for old android devices (although at least one can use firefox which does have its own store)

The usual collection of bug fixes:

- Use after free (9)

- Type Confusion (1)

- Heap buffer overflow (2)

I've switched to Edge.

So, yeah.

I'd switch too but that bloated right click menu gives me scares.

I'm with you.

The early builds, after they just released the canary, were so refreshing. From the UI to the UX, it just right. It's a shame they have integrated so much bloatware.

The PDF reader, however, is my favorite feature. It has the right UI and tools that other browsers are missing.

That's a lot of memory safety issues. Makes you wonder how many would exist if Chromium weren't written in C++.

What would you want Chrome all written in? Rust?

Yes please. And everything should be proved using the Coq programming language.

I prefer my web browsers to be written exclusively in provably-safe ADA, for maximum security

On that note, has the Rust compiler/runtime ever been formally verified?

A subset of the type system and standard library has been, but not the whole thing. It’s foundational stuff though so that was really great.

iiuc the Chrome team is actively investigating it[0] and (again) iiuc recent work with Cxx is paving the way for things like that to happen.

[0]: https://www.chromium.org/Home/chromium-security/memory-safet...

With 9 known high severity vulns in release 99 this replaces, looks like they need a few more dozen releases to really get this polished.

If version numbers were still sane, this would probably be closer to 10.0.

They are completely sane. It's just not what you're used to.

Version numbers haven't really ever been meaningful. Calendrical versioning is pretty reasonable, and Semver at least makes an effort to give meaning to the version number. But for nearly all software throughout nearly all of computer history, version numbers have been utterly arbitrary. By that metric, I see no reason why 100.0 is any more ridiculous than 10.0.

I think you're both wrong!

Parent is wrong that Chrome would benefit from semver, and I believe you are wrong that semver is meaningless.

Semver is useful for the kinds of tooling that need minimal marketing, have other software designed against it, and have multiple levels of improvement going on.

Some software which are developed against benefit greatly from a x.y.z to let other developers know what class of change has been made.

Web browsers and operating systems do not benefit in the same way, because they are a different class of software.

You may have misunderstood my comment, as I said that Semver does attempt to make the number meaningful. Though I am not asserting that, on its own, it is very meaningful; it relies on users to define what a "breaking change" actually means, and it leaves it up to third-party tools to enforce any semblance of it. Meanwhile, Semver fails to say anything at all about useful concepts like version ranges. Ultimately, I like Semver! But it's the bare minimum of what it means to have meaningful version numbers, and anyone looking to make them truly meaningful has a lot of work to do. Still better that what we had before, of course.

Version numbers of commercial software, or software competing intentionally with commercial software grow rapidly until they become cumbersome for the users to remember and interact with. 100+ is definitely getting there. I imagine a rebranding will come soon with some "Chrome Plus version 1" and a restart. FF will, of course, follow.

Chrome version numbers aren't something that normal users have to know or care about.

It's not like you can use SemVer on a browser. Too many changes would make it hard to decide whether things are breaking or not. And due to backwards compatibility, it's possible that we'd actually be on 3.100.0 or whatever. It's a lot easier to just bump the major every 6 weeks with releases.

Is it the best idea? Probably not, but it's something simple that works and is consistent.

Chrome does have an API - Selenium uses it. And selenium's chrome driver has strict rules about which Chrome versions it supports as it is versioned. So actually Chrome's rapid versioning does reflect the pace of changes to their API.

But is Selenium’s requirements due to breaking changes actually existing? Or just a lack of effort to test multiple versions?

Arguably with SemVer as html 1.0 pages still work with no compatibility issues, it will still be 1.x if that is the gate for "breaking" changes.

The other way, if you rely on experimental features, it's likely you'll be breaking something every release, ending up with effectively a different major version every release - which is arguably pretty much what we have now.

The issue is that for a sufficiently complex environment, the definition of a "Breaking" api change depends on what you use - it's pretty hard to end up making any changes that don't have some visibility across the API after all, even internal details end up leaking.

Reply on news.ycombinator.com