Chromium-based Edge preview builds
blogs.windows.comI wish they've would have gone with Firefox's Quantum, in order to try and at least balance out web market shares.
MSFT no longer has any leverage in the web, so trying to keep it fair and accessible (no browser monopolies) should be a priority for them (especially since they have quite a few web platforms like office 365)
Agreed. Browser diversity might be a pain for developers, but concentrating 2 of the 4 dominant end-user computing platforms on the same browser engine doesn't sound like a positive step for users [0]. Particularly when the engine in question is developed and maintained by one of two primary surveillance tech firms on the planet.
Perhaps I'm too jaded, but there's an unmistakable smell of the fishy about this. Dominant mobile + dominant desktop OS vendors decide to cosy up together...
I never thought I'd see the day that I mourned IE's decline having lived through the zenith of its hegemony. Like parent, I similarly would much rather have seen MS link up with Firefox for the sake of diversity. Perhaps that's too fanciful.
Perhaps a more interesting pairing would be Apple and Firefox, given that both are increasingly playing the pro-privacy card...
[0] I'm not counting Safari since WebKit and Blink diverged some time ago afaik.
Apple will likely stick to Safari based on their Webkit fork. Look at their stewardship of coreutils and the graphics stack on MacOS, its plain to see that they just don't care if MacOS is 5+ years behind the curve on features and performance if its not hurting battety life for most users.
Gaming on MacOS has become a much more limited experience than on Linux (blame Valve for this) or Windows. Apple wields a big stick with each of these choices, kneecaping HTML5 PWAs, pushing forward their own graphics stack (hence the crap OpenGL support and lack of Mantle support)
I think the blame for the insignificance of macOS as a gaming platform chiefly lies on Apple's shoulders and their staunch refusal in creating adequately powered desktops at reasonable price points. As long as the majority of Macs are laptops with Intel iGPUs porting anything more graphically complex than Dota 2 is pointless. I think there are more people with custom <$900 builds with decent mid-tier AMD or NVIDIA GPUs than people with extremely expensive Macs with Vega or RX580 cards.
I do not agree Valve should be to blame if Apple decided that gaming wasn't important for them. I am the first one to say when they do amazing things but the reality is that gaming never was/is a priority for Apple. Take a look at the history of OpenGL on macOS and it will become obvious that the only reason Linux today has a better support for games than Apple is Apple's fault only.
I truly wish it was different.
I was blaming Valve for making gaming on Linux have such a broad library of titles. They've made Wine very easy to use, to the point that users barely know that it isn't a Linux native game.
So you meant 'Thank valve for this'? :)
Yep, WebKit and Blink are very different not least because of their different JS engines. Lots of developers have trouble targeting safari, whereas chrome is a breeze.
From a health-of-the-web point of view, it absolutely makes sense to go with Gecko.
But from a business point-of-view, there's two opposing sides: going with Gecko devalues your browser on mobile (where WebKit & Blink effectively form a duopoly, and there's still enough content that relies on non-standard (pre-fork) behaviour that it causes problems for Gecko), though going with Gecko would increase their leverage against Google.
There were certainly a fair few of us at Opera when the decision there was made to move to WebKit who argued for Gecko, but that decision was very much made on the basis of growing web compat problems on mobile (given WebKit had >90% marketshare there) and going with Gecko wouldn't solve that, and that situation hasn't really changed (well, the WebKit/Blink fork has improved things somewhat for new features, but that's about it).
From a technical perspective it makes zero sence to go with Gecko, given that it's 1. much harder to embed (there's a reason everyone's going with webkit/blink) and 2. with servo around the corner, a gecko EOL is expected at some point.
It being harder to embed is definitely true, but there's increasingly will within Mozilla to change that, and if someone (e.g. MS) was willing to step up and help provide engineering resources to make it possible it's definitely doable. GeckoView is slowly changing this on mobile, and resultantly something for desktop is easier than it was five years ago.
Servo isn't replacing Gecko, though. There's nowhere near the level of resources being put into Servo for it ever to become a viable browser engine, and there's no sign that Mozilla are moving in that direction. They're still very much using Servo to experiment with new things (WebRender, parallel layout, etc.), but the approach is clearly moving modules from Servo into Gecko if/when they become production-ready. As such, Gecko isn't going anywhere, even if parts of it are getting rewritten/replaced (and that's no different to Blink!).
Servo is a research project to build a fast browser, ignoring older web standards. Project Quantum is Mozilla cleaning up those components from Servo (adding support for older standards) and migrating them into Firefox.
For refactoring a massive codebase, Project Quantum seems totally reasonable rather reasonable versus a total rewrite.
Point #1 is getting less true with GeckoView (at least on mobile), and point #2 is unlikely. Like most tech products that have a large number of users, a transition plan for Firefox would necessarily involve gradual changes (which are arguably already taking place with Quantum), which an embedder could ride.
There is no plan to EOL Gecko.
I was actually gunning for them to continue with edge, and hoping that they'd finally make it cross platform. That they are trying out a drop in compatible nodejs using their js (chakra) was really appealing too. [0]
Ironically, it was one of the few technologies coming from ms that I was interested in.
I hope Apple would turn away from webkit and invested into firefox. They're both privacy focused so it seems like a logical choice.
Why would they do that when they essentially control an entire browser engine?
Because their browser engine sucks, people just use it on mobile (iOS) because they basically have to in order to open any link.
Every full time job I've had I've been issued a MacBook and step 1 of setup was "install Firefox or Chrome".
Safari is one of the few things that truly outright blows about the Apple ecosystem.
Safari is the first thing I launch when I open my Mac and it's the last thing that I quit before shutting it down. If you're looking for battery life and RAM to spare for other applications, use Safari.
Safari is a fantastic browser, actually.
It's as fast as Chrome and Firefox. (Firefox is curiously sluggish on the Mac, even when the latest Quantum improvements.)
Sure, it tends to lag behind when it comes to adopting certain newfangled web technologies, but it supports everything that's important (even web workers now? I think?), and it's super stable.
I use Chrome for web development, though, because of the built-in dev tools. (The tools are pretty gnarly for 2019, but they get the job done, for the most part.)
Because safari will sooner or later become another Edge. Apple doesn't invest enough into safari and now it's a browser which is most behind in terms of technologies.
Apple's Safari investment is focused on features like tracking prevention and anti-fingerprinting. Stuff for users at the expense of developers.
I've seen what happens to web browsers that trail behind the industry in features.
It doesn't matter how nice the privacy features are, when a user encounters web site after web site that doesn't work right in their browser, they eventually each out to a more technical friend that recommends trying browser X. When browser X fixes their problems, they never return.
iPad / iPhone are captive rich markets, where only Safari is available. So companies/devs will keep on testing/supporting for it.
Sadly Apple took the wind out of my all-time favorite browser, Camino, née Chimera, when they started up the WebKit/Safari project by hiring away key people like Dave Hyatt: https://donmelton.com/2013/01/10/safari-is-released-to-the-w...
Stupid question probably... since I haven't had an iPhone for a couple years. Is Safari there still a thing? On my Mac I'm not using Safari at all. Havent opened it a single time since i got my latest MacBook - well I must have to download firefox but I dont remember.
I remember at one point Apple wouldnt allow any other browsers rendering engines on iOS - is that still the case?
Safari is still a thing and still under active development.
All iOS browsers use Safari WebKit under the hood.
iOS apps can use whatever rendering engine they want. The restriction is on JIT -- Apple doesn't allow apps to compile native code on the fly, which modern JavaScript engines like V8 do. The exception is Apple's own JavaScriptCore, which works with WebKit.
In theory you could have Firefox Quantum as a tenderer with JavaScriptCore as the JS engine, but that might be more work than it's worth.
> iOS apps can use whatever rendering engine they want. The restriction is on JIT
Wrong.
> 4.7 HTML5 Games, Bots, etc.
> your app must use WebKit and JavaScript Core to run third party software and should not attempt to extend or expose native platform APIs to third party software
You're not contradicting me. The section you're quoting is not about this, by the way. You want section 2.5.2:
"Apps should be self-contained in their bundles, and may not read or write data outside the designated container area, nor may they download, install, or execute code which introduces or changes features or functionality of the app, including other apps."
Yeah, but if you read carefully, section 4.7 explicitly says that you must use WebKit and JSC to run third party software. Therefore, you can use your own engine to render your app (see games or Google's Flutter), but you cannot use it to render third-party content (thus making impossible to create a browser)
Oh interesting.
There is 2.5.6 > Apps that browse the web must use the appropriate WebKit framework and WebKit Javascript.
Can we finally stop saying safari is the new ie, and recognize that chrome is now literally the new IE?
Seriously, this means that chrome has more or less complete market dominance at this point.
It wasn't dominance that people were referring to, it's the intentional heel dragging to benefit other parts of their corporate family at the expense of the web that makes Safari the new IE.
when people say "Safari is the new IE" they mean that Safari has numerous bugs and missing features that take years to fix or implement. there's probably a ton of CSS stuff that doesn't need -webkit prefixes anymore in any browser (because it has been standardized). except of course in the outdated and widely-used Safari/iOS. For instance, they only now got Intersection Observer (super important to efficiently lazy load on mobile devices) in the latest iOS tech preview. many browsers have had it since 2017: https://caniuse.com/#search=intersectionobserver
Let's not beat around the bush: Intersection Observer can and will be used to detect ad impressions too. Calling it "super important to lazy load images" doesn't show the whole picture.
ummm, did you get the impression that i was beating around the bush? i stated exactly why i've personally needed it and have had to resort to polyfills even for the latest iOS.
> Intersection Observer can and will be used to detect ad impressions too.
yes, and? would you rather it be done using inefficient, non-passive onscroll, onresize and onorientationchange events with a bunch more MutationObserver and getBoundingclientRect() sprinkled in plus a generous helping of setInterval(..., 100) for UI polling? cause that's how things basically work today (and how the polyfill works). replacing all that junk with a normal, optimized API is exactly how it should be.
Isn't that changing the goalpost a little bit? I don't think anyone ever called Safari the new IE because of market dominance.
Personally, I compare it to IE because of how weird it can be. Does anyone else remember the time they broke file upload[1], even though those issues were found in beta[2]? Or about the time they changed meta viewport[3]?
Edit: Oh, the last link also remind me of the window.location change that utterly broke Angular/Ionic[4].
[1] https://blog.fineuploader.com/ios8-presents-serious-issues-t...
[2] https://github.com/FineUploader/fine-uploader/issues/1269 (I can't remember the Apple radar bug number(s))
[3] https://www.reddit.com/r/web_design/comments/3la04p/psa_safa...
[4] https://blog.ionicframework.com/ios-9-potential-breaking-cha...
> Can we finally stop saying safari is the new ie, and recognize that chrome is now literally the new IE?
I never said that, and chrome is not literally the new IE, no matter how poorly the word literally is misused.
Chromium is a free (a blend of BSD-style, MIT, GPL, and other generally regarded as free licences apply) piece of software that's performant, cross-platform, receives regular feature updates, rarely breaks backwards compatibility, and doesn't try to lock you into one vendor. Compare and contrast the history of IE.
Microsoft's (new) Edge is using Chromium.
> Seriously, this means that chrome has more or less complete market dominance at this point.
No it doesn't. This is hyperbole.
Why is Safari the new IE?
Safari increasingly doesn't follow standards, implements non-standard features, and updates are tied to OS updates so if users do not update their OS (which is free on macOS but not well advertised in the OS itself), they are stuck running an old browser.
This is incorrect. Safari 12 shipped with support for macOS 10.12 and 10.13 [1].
Huh. That's changed, then, since previous updates were always tied to the OS. Still not great support, though. Chrome supports back to 10.10, Firefox back to 10.9, Vivaldi back to 10.10, Brave back to 10.10, etc.
There are multiple articles online from people claiming that Safari is the new Internet Explorer (IE) in terms of feature parity with other web browsers like Mozilla Firefox and Google Chrome, and specially due to the fact that Apple usually delays the release of certain HTML and JavaScript standards, making web development slightly more difficult for today’s web developers.
• https://www.safari-is-the-new-ie.com
• https://news.ycombinator.com/item?id=12051267
• https://nolanlawson.com/2015/06/30/safari-is-the-new-ie/
• https://www.telerik.com/blogs/safari-is-not-the-new-ie-but
• https://dev.to/nektro/safari-is-the-new-internet-explorer-1d...
• https://www.reddit.com/r/Frontend/comments/6zb73s/safari_is_...
• https://apple.slashdot.org/story/15/06/30/2251253/is-safari-...
• https://arstechnica.com/information-technology/2015/06/op-ed...
• https://medium.com/@matthew.johnson/safari-is-the-new-ie-but...
I am a happy Safari user though :-)
At least on IOS you're stuck with safari/the system webview, so even if you make a web browser app (this includes chrome/firefox) they're using the safari engine.
The issue stems from the slow adoption of certain technologies. That was the issue everyone had with IE 6 - everyone used it but it lacked modern features so it sucked to use, but you had to because of it's market share.
“Being IE” has at least two meanings: a) having a market share so big that others do not care about the other browsers and b) being technically stagnant for years (though this may be looked at just as byproduct of #1).
So Chrome derivarives might fit first criteria but neither them nor Safari fits (yet) the second one. Yes, Safari development may be not as fast but it is still ongoing and the team is not disbanded.
The most evil thing Google is doong with their browser right now may be implementation of non-standard features which only help their business.
To the surprise of no one, its basically chrome. Even my google account came in logged in automatically, same recent sites etc. I wonder if the roadmap will include things like dark mode, I never used the annotations feature so can't vouch much for it. I'm yet to try to make a MS Teams call but looking good so far.
The single engine world domination plan takes one step further today.
> Even my google account came in logged in automatically, same recent sites etc.
LOL, this is hilarious.
Google account integration seems pretty non-strategic for Microsoft - do we have any idea if this is intentional? (I would have assumed they'd implement all the cloud stuff backed by Live accounts.)
If they integrate Google Accounts and Microsoft services, it provides a way to exfiltrate personal data held in Google Accounts to Microsoft; least nefariously, as a way to make MS accounts more useful by, perhaps automatically, importing contacts, etc., but that's not all that is possible.)
This new Edge seems to reuse the cookie jar, history, etc. of the real Edge, so emp_ is probably already logged in to his Google account on the real Edge. (Corrections welcome, but it's what I've noticed after trying it.)
I'm not logged in anything google on vanilla Edge at all, this is not coming from the Edge history.
Wouldnt this be expected that your Edge cookies continue to work? Or are you saying it uses Chrome's cookies?
it says on the first launch page that dark mode is coming.
There's also an experimental option to turn on in edge:flags.
I liked Edge because of its fantastic battery life (for browsing, Edge was like having a 25% larger battery.) I was annoyed that many sites seemed to be Chrome-specific, including some I absolutely need to rely on. I switched to Brave, and it's amazing. Chrome-based, but somehow results in far better battery life than Chrome.
> I liked Edge because of its fantastic battery life (for browsing, Edge was like having a 25% larger battery.)
For what it's worth, I suspect this is the result of using the system DirectComposition. If you look at the more recent additions to that API, they're essentially just CSS 3. Basically, Windows integrates portions of CSS deeply into its OS stack, like macOS does with Core Animation. It's sad that moving to Chromium will regress this elegant architecture, as saving the extra blits that come from doing everything at the application level as Chromium mostly does (except for video playback IIRC) is a nice power and performance improvement.
> Chrome-based, but somehow results in far better battery life than Chrome.
Most likely because of ad blocking.
Neat, I didn't even know that existed.
> Compositions and animations created by DirectComposition are passed to a built-in component of Windows called Desktop Window Manager (DWM) for rendering to the screen. Therefore, no special rendering components or UI frameworks are required on the computer.
Wow. Everything old is new again: https://nnc3.com/mags/LM10/issue/04/Berlin.pdf.
(As an aside, Berlin has somehow almost completely disappeared from Google. I had trouble remembering the name, and I searched "display server scene graph corba" and it didn't come up in the first few pages. Finally found this slashdot announcement: https://tech.slashdot.org/story/02/11/24/189226/fresco-m1-re...)
Yeah, it's neat. The long-term goal is to move Firefox to use DirectComposition as much as it can, by the way, by integrating it into WebRender. I've done some work on this with the "planeshift" library. I wouldn't be surprised if Chromium eventually does the same (as it already uses Core Animation on macOS).
I think there's a tongue-in-cheek joke about IE/Edge being so bloated as is, that the only route they could take was to slap electron on top for even more memory usage.
All jokes aside, are they still using Chakra or switching to V8? I know there was a ton of work in node.js to make it work with chakra, not to mention edge was one of the early adopters of many ES2015/ES6 features, it seems odd to me to abandon all of that momentum.
They're using V8. https://github.com/Microsoft/ChakraCore/issues/5865#issue-38...
Which begs the question why would anyone use Edge when they can use Chrome?
I _believe_ they are doing this because they are heavily investing in Electron (Skype, Visual Studio Code, Teams, more). Having both Chromium and V8 be part of Edge (and the OS) means they can remove them as bundled dependencies from Electron on Windows and lighten the payload of those apps. I wouldn't be surprised if we see future Windows builds of Electron becoming smaller because of this.
There is not proof of this, this is just a hunch I have.
Eletrino[1] and Quark[2] tried tackling this idea of using native OS engines but they seem to just be side projects and/or experiments.
Porting VS Code to Chrome OS, without need for an electron-based GTK app inside a 'Linux container'?
The alternative would be Microsoft ships windows with an outdated browser, or they include Chrome/Firefox, or they don't include any browser.
For me, I plan to use Edge to install Chrome, business as usual :P
Not sure that's the question when you already have edge... I think it turns into...
question is, why use chrome when edge is basically chrome but already installed on windows out of the box?
Because it is the windows default. From the Microsoft perspective they maintain the branding and can still pit features on top when they need to and direct people towards bing.
They trust Microsoft more than Google?
That's an interesting point. 10 years ago I would, today with Windows data mining gone mad I wouldn't. They are on par.
/* Looks to other top HN Posts */
A majority of consumers do not expect Google to track their activities (niemanlab.org)
Ironic?
It’s truly remarkable that a company that was once nearly broken up by the government over its browser bundling, is now barely implementing a browser.
> Support for Mac and all supported versions of Windows will also come over time.
Is anyone actually planning to use Edge on macOS? What reasons would you have for doing so?
This reminds me of the 5 year period from 2007 to 2012 when Apple made Safari for Windows. I got curious and used it for a few weeks, but moved back to Firefox for the extensions.
Related: are there still any WebKit browsers for Windows?
Chrome and Opera are technically blink but are essentially Webkit AFAIK (Google forked webkit and Opera followed). At least I haven't heard of any huge disparities between them.
There's a fair bit in way of difference now, and it's definitely long since passed a point where plenty of fixes couldn't be copied between them.
Making browser cross-platform gives a bigger user base and instills confidence in developers to take a browser into consideration when designing websites, which in turn popularizes the browser as sites work properly.
I would use it for browser testing websites if it's renders identically to the Windows version.
Seems unlikely to ever render identically. At the very least I believe chrome uses the platform specific font shapers (coretext on mac, cleartype on win).
So in the end, a mac version of edgeium just adds yet another browser to the test matrix columns....
Its Chrome but not Google? Sort of the same usecase as Brave?
…but it’s Microsoft. Is that any better?
Brave was made originally on Electron fork Muon which was recently abandoned in favor of Chromium.
I hope Microsoft makes a point of prioritizing efficiency over features in Chrome-Edge. Google has seemingly placed low resource consumption and low battery life impact as tertiary concerns (at best), opting to woo developers with a constant stream of hastily developed shiny features instead.
As it is the only browser that seems to care at all about not sucking your battery life down a black hole is Safari…
The Edge team has repeatedly said that Performance/Battery Life has always been important to them. One of the reasons they've mentioned for why they are going Chromium-based is that it gives them more time to focus on Performance/Battery rather than web platform bugs, and they've definitely listed Performance/Battery as a continued "Focus Area" for the team.
It seems like they do not have planned to support Linux. It simply says "Not supported" and on the download page it is not even mentioned: https://www.microsoftedgeinsider.com/en-us/download/
It looks nice so far, but Microsoft really needs to drop the E logo, and maybe even the Edge name. Both already have negative connotations; and I don't think Microsoft can now claim that it makes it easier to find access to the internet. Especially when Chrome and Safari on tablets and phones look nothing like the iconic blue E, and yet late age boomers can still figure out the devices.
I'm not so sure that all old consumers can find another browser icon. Most in my experience look for the E.
That's what they said when they announced the Edge name – if they didn't choose an E- name (and logo), literally millions of people wouldn't be able to get on the web.
Now try to imagine providing support for one of these users and explaining no, you're clicking the wrong shade of blue 'e'.
I agree with this sentiment. Edge was an extremely fast browser and I think it lost before it was even released simply by using similar branding to Explorer.
Edge lost because, even in April 2019, it is a buggy pile of shit that lacks fundamental features it's competitors have and chokes on web pages constantly. (I'm forced to use it for work.)
In truth I didn't use edge much myself, but for the slim number of uses the speed was nice.
Seriously, why is MS doing this? Internet Explorer died years ago, Edge never took off, and Google isn't about to give up their market share for spyware without a fight. It's like MS thinks this is Browser Wars 3.0. Heaven help me if it turns out that MS seems to be doing something comparatively good.
They need a browser they control but they only need to control look/feel, interaction with windows settings etc. They don’t want to spend a fortune maintaining a js runtime and renderer.
So assuming an OS does need to ship with a browser, skinning a chromium one seems very sensible.
I think you answered your own question. Edge never took off, why continue investing money into it?
This offloads a lot of work to the open-source Chromium project, allowing Microsoft to just skin it and add whatever else they need.
> This offloads a lot of work to the open-source Chromium project, allowing Microsoft to just skin it and add whatever else they need.
Plus, it means any work MS does at implementing web features they view as strategic in their browser will naturally easily port back to Chrome, eliminating technical barriers for them becoming common user-agent features (there might still be political/strategic reasons for Chrome not accepting them, of course.)
Curiously, had they made Edge compatible with Windows 7 it would have had a much bigger chance of it actually being used.
The greedy strategy of wanting to have both Windows 10 and browser market share is what doomed Edge.
URL contains tracking device.
chrome://flags/#enable-experimental-web-platform-features works but redirects to edge:// :)
This wasn't necessary, they just had to get rid of the slow, ugly UWP UI framework. The Edge rendering engine was fine.
It looks like this new Edge uses Win32 instead of UWP, so people will actually use it, and they'll pat their own backs thinking the problem was the rendering engine. No, it was the stupid UI framework. >:(
99% of people probably never even noticed Edge is UWP based. I don't really think MS switched over because they were losing out on performance either. Seems more like they were just tired of maintaining a first class browser stack that wasn't very popular when using Google's gets them everything just about the same.
I mostly use Windows on Dell Venue 8 Pro tablet.
TL;DR - no 32bit builds. currently only 64bit builds.