This has been in the works for a while! A number of Microsoft employees have been making contributions to Chromium too. I hope to see Chromium's performance and battery impact on Windows to improve with this decision which is great for those using the OS.
It sounds like Microsoft really just wants a platform to route people into using Bing and service ads on the new tab page, which they're more than capable of doing on a reskin for Chromium. From a cost standpoint it makes sense to use the existing tools available.
This vision is pathetic, unambitious. What if Microsoft is content to reskin and rehome Chrome, submit their battery life PRs, and cede the web to Google? What is the web once Google dominates its access?
The absolute number is not that important, the trend is. That's clear even in Mozilla's own data: https://data.firefox.com/dashboard/user-activity shows they went from 303M monthly active users at the end of Nov. 2017 to 277M this year. This is a serious drop.
Looking at netmarkshet website and web requests it seems this pulls from gator.io for metrics. Looks like 3k-4k websites have this data which lines up with the previous statment.
https://publicwww.com/websites/%22gator.io%22/
But if that engine is open source, isn't that a good thing? There's a single standard but anyone can contribute patches for increased performance, new features, etc?
All commits to the Chrome codebase are gated through Google, an ad company whose driving motivation is in delivering value to shareholders, not to improving the web. Once the web becomes a Google monoculture, I won't be holding my breath to see any features land on the web that could threaten Google's bottom line.
> Aren't changes to many open source projects gated, like the Linux kernel?
All projects gate commits somehow. It's not about the gating, it's about who is allowed to make the big decisions and how much buy-in they need to seek from other stakeholders before they're allowed to proceed. It's also about incentives; if Linus were, say, a Verizon employee and if the Linux Foundation were a Verizon subsidiary, people would feel much differently about the governance of the kernel. Likewise if the kernel were permissively licensed rather than GPL'd.
> On a serious note, doesn't the license allow forks? Couldn't a large company just fork it and make changes without Google's approval?
The thrust of the point here is that forking the codebase is no good if you can't convince people to install the browser and for websites to support the browser. It's a social problem.
> is it not open "enough"?
It's not. Open source gives users the freedom to fork. When forking isn't enough to preserve user freedom, the next step is open governance, which involves delegating decision-making power to users (with many interesting structural varieties to choose from). Amusingly, I gave a speech on this topic at All Things Open just a month ago.
Except that we will probably have to support Edge for another 5 years for the people who refuse to upgrade, so in reality, it will be one extra unsupported browser that we have to write hacky workarounds for.
Honestly, I think it’s a monoculture already and the fact that multiple engines exist puts up an illusion of competition. Google controls where the Web goes today, and that there is an oligopoly is a false symbol. At least Chrome gets more dominance, we can no longer pretend Google doesn’t already own the Web.
Why are you citing Safari's 14% market share when the person to whom you responded was stating that if it's true that Edge will become Chromium-based, then browsers running on Chromium will represent something like 66% of the world's web engine share. Heck, it's already 59%, over half. With Opera, Vivaldi, and Edge (or whatever Anaheim will become), we're talking two thirds of all browsers being based on one engine.
That's the monoculture.
Edit: didn't realise you were responding to someone refuting what you had said, haha. so basically we agree, I just thought you were someone else.
I'm saying that WebKit can't be a monoculture because Safari only has 14% market share. It doesn't matter if other devices and browsers that no one actually uses are WebKit.
Gotcha. Yeah, it seems a lot of people are stuck in the past idea that WebKit dominates. It used to dominate in mobile and it certainly a lot of WebKit-only extensions eventually became standards — but those days are behind us for everything other than iOS.
And while iOS has the greatest market share for mobile, you're right, WebKit doesn't have the greatest market share for general browsers. That's definitely Chromium, now on Windows, macOS, Linux, iOS, Android, your mum's toaster, just about every 'native' web app…
Side note: I guess we should really be saying Blink, that's the name of the engine. But even if, say, you use Google Chrome on iOS, backed by WebKit, that usually makes you a Google Chrome, backed by Blink, user on the desktop.
That person meant that if all the main browsers move to Chromium, then it will dominate and cause a monoculture; the same way that Internet Explorer caused a monoculture in the 90s and 00s.
Pointing out that there are competitors doesn't change that their presence doesn't overthrow the monoculture. After all, Gecko was around for a long time and Firefox did become very popular very quickly after version 2; but realistically, it took until Apple put WebKit on iOS for any real dents to be made.
Also, I think Konqueror still uses KHTML by default. I don't know if Konqueror tracks WebKit, but I think they're rather different these days.
I don't see why monoculture (as long as the browser engine is good) is bad. Unlike the era of Internet Explorer, Chromium-based browsers are actually consistent with official standards.
On websites like "caniuse.com", a site which checks for support of the latest JS/HTML/CSS features, Chrome consistently ranks 1st. These features are all official features of the HTML/JS spec, not arbitrary Google-specific extensions.
Ensuring that more people use a browser with the Chromium-engine will allow web developers to use the latest features.
WebKit is a form of harmful monoculture. I'm ok with Blink's monoculture because it is probably the best rendering engine out there and inferior engines can't compete. But Apple's is creating an artificial monoculture by only allowing WebKit on iOS. This allows Apple to not innovative on WebKit and deliberately cripple new web technologies, so that people are forced to make native apps.
And the end of W3C standard development process. AFAIR it should be 3 independent implementations of the feature in order for Draft to reach Recommendation status.
So technically all that means that web standards will be written by WebKit team alone.
Fair concern. On the other hand nobody uses Edge on Windows 10, do they? In my limited experience users seem to still use IE or go to Chrome or Firefox.
I use it as much as possible. Its user experience on touch windows devices (Surface Go in this case) is several orders of magnitude better than Chrome and Firefox... is not a usable browser, performance-wise in my experience.
Edge is brand new, and foolishly branded to look like IE, this might stop average users who know about getting Chrome from using it. But it shouldn't. It's extremely rare that I'm forced to swap over to Chrome. Mostly I use Chrome to keep my invasive work IT staff from having access to my private browsing info.
The problem with Edge is it still (like IE had been always but at a small number of moments in history when it actually was the original source of an innovation) is the worst of the browsers from the standards (or non-standards, whatever) support perspective. See caniuse.com
Patches is not the only mechanism to keep code in sync.
For two engines forked from the same source (same architecture and even same code tree structure) it is relatively easy to copy/paste feature implementation, no?
In theory maybe, but in practice we rarely do. Sometimes even when the idea is pretty clear, like WebKit's CSS JIT, it's not that easy to copy because of underlying assumptions.
Just curious, what kind of underlying assumptions are required for the CSS JIT (which is completely new to me! I had no idea this even existed)? I read this: https://webkit.org/blog/3271/webkit-css-selector-jit-compile... but I didn't see anything specific (by the way, the spinning animation for the gear is broken on my Safari Technology Preview).
There's this which I'm hopeful for: Android Components [1]. If Mozilla make it easy for Android apps to use Servo as the shell for modern web apps, then the improved performance might make it worth it to "restrict" yourself to using only the features supported by Servo (i.e. without many of the workarounds that make old websites work).
The entire internet and web development community, that's who. While WHATWG has taken over the HTML part, it's the W3C that still gets all the action everywhere else.
The experience of standards bodies has generally agreed that you can't shake out all the problems with a specification until you have at least two independent implementations.
We do still fortunately have three (more or less) independent popular implementations: Google with Blink, Apple (and now Microsoft) with WebKit, and Mozilla with Gecko.
These are still three strong and independent voices; Chrome has substantial control on desktops, Apple has a stranglehold on their highly popular platforms, and Mozilla has both the hearts of the tech community as well as wide regard in bleeding edge technology with Rust, Quantum/Stylo, and Servo.
I'd put the count at 2¾: Blink counts for ½ since it's an ex-fork of WebKit. While a lot of the new feature development of the two are independent of each other, they still largely share a lot of the same architecture, which means there's going to be correlation in things like ease of implementation. The extra ¼ comes from Servo, which ends up having the opposite edge of the dichotomy from Gecko; it's effectively a ground-up re-implementation of the layout engine stack, so its internal architecture is quite different from everybody else, but it's goal isn't to drive new feature development in web standards.
Right, the point being made is that Webkit/Safari are often the slowest of the bunch which is fine when you have three others but now each browser effectively has veto power and the expectation is that Webkit will wield it often.
It's saying "oh no, now Webkit can block things", not "oh no, we don't have enough strong implementations"
or, webkit will become the equivalent of linux for browsers. Considering that there will be less incompatibilities , it's a good thing. The hope is that competition for users will lead to innovations that benefit them, and not wall them off from competing platforms.
As a firefox fax, I hope this isn't true. The dominance of chrome isn't good for anyone. This will lead to (even more) sites not bothering to be made compatible with Firefox. We are headed back to the "designed for IE" days.
My understanding is that they'll be using the Chromium engine underneath, but still their own Edge "browser" and UI on top.
IMHO, I personally think it's a good thing to consolidate on one engine to render HTML cross-platform for these two reasons:
1. Web developers no longer need to worry about supporting CSS and other edge-cases across various browser engines.
2. The Chromium engine itself is open source. There are other browsers (Vivaldi, Opera, etc) that run on top of it.
3. While I think it's great the engine underneath is the same, I think it's equally great that there is a variety of UIs and browsers built on top of the same engine - the innovation happens in the browser space, not the engine, anymore.
Being open source is a red herring in this instance.
Say that Chrome implements a web feature you don't like. You fork the browser and remove that feature. But websites expect Chrome, and they use that feature, so your fork doesn't work with those websites.
Say that Chrome refuses to add a feature you want. You fork the browser and add the feature. But websites expect Chrome, so they don't use your feature so as to not break for their Chrome users, and your fork is no better off.
The insidious part of a web monoculture is allowing Google to dictate the standards of the web platform. Being able to fork the codebase only gives one the power to change things that are strictly client-side.
Your examples have nothing to do with whether you use Chrominum engine, though.
Say that Chrome implements a web feature you don't like. MS use their own MSengine for their Edge browser that doesn't have that feature. But websites expect Chrome, and they use that feature, so MS Edge doesn't work with those websites.
Say that Chrome refuses to add a feature you want. MS use their own MSengine for their Edge browser that has that feature. But websites expect Chrome, so they don't use MS Edge feature so as to not break for their Chrome users, and MS Edge browser is no better off.
Not sure what this is attempting to refute. Your comment is about why monocultures are bad for competition, which I happen to agree with. My comment is about why Chromium being open-source doesn't alleviate any concerns about monoculture.
How is the situation better if there are two browsers with 50% market share? You still can’t add a new feature the other side doesn’t want, because 50% of users won’t be able to use it.
Any feature that requires a site owner to do something to support it isn’t going to be added.
> You still can’t add a new feature the other side doesn’t want
That's not quite true. This happens all the time with things not standardised (yet). With time those either get wide adoption because people want them, or get dropped, or get standardised in a different widely agreed form.
You've just illustrated the advantage of a polyculture: changes to the platform are not unilateral decisions, and therefore require discussion, communication, and documentation. Furthermore, if one actor tries to do something out of blatant self-interest that would be a detriment to the platform as a whole, that action can be blocked. With a Google monoculture, at the end of the day, the web will be whatever the CEO of Google allows it to be. Imagine how little control users have over the Android platform; envision a future where that's the web's model as well.
The difference being now that we're in no danger of IE6-ification. If Google annoys enough people, people can take Webkit and do whatever they want. It happened to Microsoft, arguably to Mozilla, it can happen to Google.
So long as the players are well behaved, having a single dominant browser is beneficial. How much time, energy, productivity, and money has been lost on cross-browser (in)compatibility wrangling?
> If Google annoys enough people, people can take Webkit and do whatever they want.
Not sure if that should have been Blink, or if you're making a subtle point about how Blink was forked from WebKit (which was in turn forked from KHTML).
Personally, while I can see why some may react to this news with concern about a monoculture, I find it hard to feel sorry about the end of the last significant closed source browser engine.
If Microsoft gives up IE and Edge then all the major browsers will have an open source foundation. With Gecko, WebKit, and Blink there remains a healthy range of options, too.
And as the history of KHTML/WebKit/Blink demonstrates, derivatives will appear if there's enough interest. Perhaps someday Microsoft will follow the examples Apple and Google have set by creating their own fork, if the circumstances warrant it.
It also demonstrates that forks are irrelevant and no one cares about them unless they come from major players with enough money and political power to push them.
WebKit was forked from KHTML. Chromium used WebKit, until Blink was forked. All of that was only possible because the browser engines are open source.
As with anything, you always need somebody (or a group of somebodies) to help something new gain traction. There's no magic pixie dust that will let you develop a major new platform without commensurate resources. What open source does do is lower the cost of entry and make it easier to build up to a level where you can make an impact.
> The difference being now that we're in no danger of IE6-ification. If Google annoys enough people, people can take Webkit and do whatever they want.
The entire point of this subthread is that it doesn't matter that you can take Webkit and do whatever you want, because the web is a client/server (browser/site) protocol, and having the ability to fork the client is irrelevant when this makes you incompatible with every server. Technical capital isn't sufficient; it takes social capital to pull off a hard fork.
> So long as the players are well behaved, having a single dominant browser is beneficial.
I've been on HN long enough to remember this tired argument from back before the Blink fork. The gain in efficiency from having a unilateral dictator is not worth the loss of mutual counterbalancing oversight. It is completely without foundation to give Google the benefit of the doubt that they will be well-behaved when there are no consequences for misbehavior. Google is not your friend, nor is Google's mission to make the web better. Google is a corporation whose purpose is to maximize profit by selling advertisements. You might as well hand stewardship of the web over to Comcast.
Your argument is easily proven false by looking at the various CSS properties prefixed by vendors like -moz-* and -webkit-* that eventually made it into the standard.
Microsoft using Chrome only somewhat reduces Microsoft's power in this regard. In no way does that change your ability to project your will on the internet through Chrome forks. Small devs have little power now and that will stay unchanged if Edge goes Chrome.
Web devs will have to build for the lowest common denominator so this change improves that slightly.
An actual downsides is it increases the power of the Chromium team to make defacto changes.
It might also reduces innovation but I can't really see Microsoft deciding to drop a feature because it would take them further from the chrome trunk.
Not even close. IE6 was a browser that was set in stone. It only ran on Windows. It's legacy that we live with today is the ActiveX controls that are basically parts of Windows. MS stopped innovating on the browser part of IE6 and focused on ActiveX e.g. native Windows components that you could use in the browser. You can't talk about IE6 and not mention how embedded it was with Windows. That is what made IE6 bad. You could not use your Mac or Linux environments to browser the internet. Lots of great ideas came out of IE6 XMLHTTPRequest. Access to the native desktop devices. Storage APIs, realtime communication anything you can do on the desktop you should be able to do through the Web. That is what we see happening today with Chromium and Mozilla. These two teams and now with MS being more involved as well. There is a huge amount of collaboration it's nothing at all like IE6 days.
I was doing some webdev back then, and whilst everything you said was true it was the lack of cross browser CSS and Javascript testing which led to the real frustration.
Back in 2002/03 IE6 had over 90% marketshare, and justifying testing in anything else was a constant battle.
As developer back then I had the same experience but with some reflection ie 6 was the better browser in 2002 -
Firefox css got better but that was more like 2005 after many years of zero development on IE6. Reflecting back to that time period the most lasting damages were active x. There are still governments and corporations that can’t migrate from IE 11 because of this legacy.
If open web standards are to matter at all, it is more important for there to be multiple viable implementations of the same standards than it is for the lives of web developers to become easier.
Using the same engine across multiple browsers, open source or not, just doesn’t cut it.
People in web standards community kind of anticipate browsers (there are 4 main browsers now) eventually kind of consolidating down to 2 engines. This seems like a step in that direction. Who knows, Safari might be next :D
Every time I've suggested something along those lines to the Chromium folks I've gotten "we're doing it in C++ ;)"
I mean, it makes sense—Google has an enormous team on Chromium and they don't want to have another language to support. But that is precisely why it's important to not have the entire Web under Alphabet's reporting structure.
Your argument is similar to saying that one party rule in politics is good for the country (because the party is nice, has competent people at the top, and helps people a lot).
Well, it's better than two parties that are putting all their energy into fighting each other, instead of governing a country in hot best way possible for its citizens...
The web works best when there's a European-style balance of power with actual balance [1]. Microsoft, Apple and Google are like France, UK and Germany. Each of them have their own interests. Microsoft had their own interests when they actually had a native developer ecosystem that was thriving, but now that it's gone they have deprioritized Windows as a business [2], made Windows 10 a data sucking "OS as a service" (that doesn't even have QA people anymore other than beta testers), and is now all in on PWAs for their app story. They are in fact more like Google than ever before. That just leaves Apple as the only opposing source of power.
Mozilla is Poland and is destined to get f*cked as their market share drops and any negotiating leverage they have to get Google TAC money disappears (they're more like Saudi Arabia is to the US, a vassal state, rather than a Great Power).
These are dire times for the web. Only Vestager can save us now.
Mozilla is by the way also of the opinion that a monoculture would be horrible for the web. Arguments that I've heard from Mozilla employees:
- Without competitors, innovation is dead. If there was only one web-engine, why bother improving it further? Given that it's open-source, it could theoretically be forked, but in that moment you again have two different web engines or more. Also, Mozilla would have to fork Blink from day one, because many things in it, they do not consider acceptable.
- A monoculture means one security problem makes everyone vulnerable.
- Multiple implementations challenge standards. You still would want a definition of a standard, or an API if you will, to point web developers to. But those are only going to point problems out after you've already implemented it, and likely also after other websites are already using the feature productively. They also can't point out stupid specifications that aren't going to allow you to update your browser engine in the future. So, you'd be much more likely to have to break compatibility, which doesn't play well with the web.
WTF do you mean "lead to sites not bothering to be made compatible with Firefox"?? If Firefox follows the published HTML standards then there wouldn't need to be any special compatibility effort made by us web devs! That is exactly what made I.E. (and Edge) so cr@ppy in the first place!! /facepalm
Firefox seems scummy, trying to shove the service pocket on us. Despite firefox's recent improvements, I find chrome a better experience overall. If you want to browse without google monitoring everything, then Vivaldi seems like a great option.
Yeah, this is just depressing and honestly a little baffling. Seems contrary to Microsoft's interests - why let Google control this aspect of the Web?
(Obviously, MS will maintain their own fork and can permanently fork away whenever they want/need to, I guess, so it's not like they've locked themselves in forever, but still)
I actually wasn't aware of the problems with Edge. This is the first I've heard of them... I hadn't heard much about it from either devs or users, so I assumed lots of people were more or less happily (or at least uneventfully) using it as Windows' default browser.
Never thought I'd be sad to see Internet Explorer (or its descendant) bite the dust, given what a plague upon humanity IE was for so many years, but I'm not sure this is the happy ending we might've wished for.
>Yeah, this is just depressing and honestly a little baffling. Seems contrary to Microsoft's interests - why let Google control this aspect of the Web?
Chromium != Google. Chromium is the FOSS project that Chrome is based off of.
Chromium is just a fork of WebKit, a fork of KHTML. Unless Microsoft is really planning on downsizing their web browser division, they’ll fork it instead of maintaining a rebranded patch set.
Unless MS wants to work on the browser secretly for years (the reason for the Webkit fork) or Google refuses many contributions from MS (the reason for the Blink fork), there's no need for a fork.
Google's employees make up the bulk of Chromium's contributors and leadership.
Microsoft is of course free to diverge from it as much as they like but, like any other Chromium-based browser, they are either going to be joined at the hip to Google's decisions, or they'll be downstream consumers. The bottom line is that while MS will be contributors to Chromium they won't be in the driver's seat.
I don’t think they intend to let Google control anything — least of all the future of the web. They have been burned too badly by them for that. They are more likely to do what Google did — they used WebKit in the early years of Chrome and then forked it to create Blink once it had gained enough traction. Microsoft is likely to do the same at some point — assuming their browser gains traction. And I, for one, hope it does — and that they make it cross platform. We need an a way to browse the web without Google tracking and Firefox is just not cutting it. And while Safari works, it is not available on other platforms.
Regarding the problems with Edge, mainstream websites don’t face problems — but you hit the edge cases (no pun intended) when you try to use it for smaller sites — and especially sites meant for limited audience — e.g. internal websites built by enterprises. They are often tested for Chrome and Firefox only — and maybe Safari if there are significant number of Macs in use.
I can’t really blame the developers for not wanting to waste their time on a browser with limited usage and — especially since it often presents challenges not posed by other browsers.
What are your objections to modern Firefox? Most of the criticism I've heard is that it's become too much like Chrome. I'm quite happy with it personally.
To be honest, my experience with FF has not be been great in the recent years — although I do acknowledge that it has improved recently. I can’t point my finger at one thing — it was combination of little things that bothered me enough to make me switch to Safari and later Chrome.
That said, my comment was directed towards that FF’s market share is plunging rapidly and unless things change drastically, it might become another Opera — a niche browser at best. We need something that can counter Chrome and by extension, Google — something that developers have to support in addition to Chrome. FF has been playing that role nicely so far, but for how long? FF is not cutting it in terms of adoption — in terms of challenging the might of Chrome.
As someone who switched from Chrome to Firefox a few months ago my main issues are:
* No spellcheck in form fields (without a plugin, and I hate plugins).
* It feels slower. There's something about it that just doesn't feel as fast. Mostly things like opening windows etc.
* There's some annoying defaults.
* Dev tools aren't as good.
* Updates aren't installed automatically.
I mean, it's close to Chrome, which is good (for me), but it just feels about 5% less polished. It's good enough that I'll stick with it, after all there was a driving force for me to leave Chrome too.
I understand this is my fault for having 10-30 tabs open at any point in time, but still, chrome will generally manage that at 2-4gb ram. FF often breaks 10gb.
I understand Jira is a pig of an app (Atlassian software is generally shit), but I unfortunately need to use it all day long. Chrome seems to be able to handle 2-4 each of jira, gmail, google cal, github, aws console, and slack. FF couldn't last time I tried, a couple months after they released Quantum.
This would, of course, be less of a problem if the cheapest macbook with 32g ram didn't cost $3k...
I never see my total RAM usage going over 6G to the whole system if I don't compile anything. I have tens of tabs open and the 32G of RAM is most of the time useless. Mind you the ThinkPad RAM upgrade was about 120€...
edit To clarify, I'm used to Chrome seemingly just updating itself in the background. Firefox needs a restart, and so far the restart behaviour with Firefox has been less reliable than Chrome (losing opened tabs sometimes, for instance).
You really think a Microsoft browser is going to track you less? The same company whose entire OS is deliberately riddled with spyware and tracking, and even managed to sneak telemetry into an open source code editor?
It's a different source of tracking, and frankly speaking, we know why Microsoft wants to track us - the same reason they always have. Microsoft wants to sell us more of their own products, generally speaking. It's not about throwing all private data to the open market for the highest bidder. I'd rather have M$ out there looking to sell their own products and services to me than Google looking to sell me to their advertisers and whoever pays them this week.
> Microsoft wants to sell us more of their own products, generally speaking. It's not about throwing all private data to the open market for the highest bidder.
This is increasingly not true. Between Bing and adding ads[1] to various parts of Windows 10, I expect Microsoft’a long term business strategy will be increasingly ad-heavy. And you can’t make them un-collect the data they’ve already collected on you.
Agreed. As hardware and even operating systems themselves become more or less commoditized, Microsoft sees that the writing is on the wall. They see their future revenue (or at least a significant part of it) coming from advertising and services.
Do we think Microsoft is getting away from maintaining their own browser just so they can go right back into the business of maintaining their own browser?
Maybe someday they'll do that but it's hard for me to believe that's the plan they're starting off with.
Not that I have any insight into Microsoft but it appears to be consistent with their strategy of disengaging from the client. Why bear the cost of maintaining your own browser technology given the extraordinary complexity, a low market share and little to no associated revenues.
I wouldn't be too sad if Firefox was switching to chromium either. Then developers would only need to worry about rendering into one engine, and users would have all their sites always working.
> I wouldn't be too sad if Firefox was switching to chromium either. Then developers would only need to worry about rendering into one engine, and users would have all their sites always working.
That would be a terrible day for the open web. Damn it, somebody inject some IE6 into these people stat.
Ah, I can see you weren't around during the days of IE6, when the web stagnated and suffered under Microsoft's stranglehold and ineptitude.
Keep in mind that, for quite a few years, developers welcomed the oncoming Microsoft browser monopoly; IE3/IE4/IE5 were generally better and faster than their Netscape counterparts and Opera was so niche as to be irrelevant. Then Netscape (the for-profit enterprise) folded and there was no viable competition to IE for a while, and the broken mess known as IE6 became a serious problem for years since it had something like 98% market share.
Trusting any company, even Google, to be the de facto sole steward of the web is insane. They are a for-profit company. That may not make them intrinsically evil, but they sure as hell aren't intrinsically good.
It's interesting, I don't see this argument made against other open source projects that are dominant, such as the Linux kernel.
I rarely see people saying that there need to be competing kernel projects with completely separate code bases but all trying to conform to the same ABI, etc.
(Yes, I know there are other kernels out there, but my point is that people don't complain that Linux is so dominant like they do about Chromium vs Firefox)
Dominant on server or generally? Windows is pretty big generally, and it isn't obscure enough on servers that I don't have to deal with vendors daily who require me to inter-operate with windows.
And the OS kernel monoculture is to the open source world's detriment. I think it's fantastic that Fuchsia is going to give Linux a run for its money: Fuchsia is fundamentally better architected in numerous ways than Linux is.
Linux is not dominated by a single company. Developers working for many companies contribute heavily to it, but as long as Linus Torvalds is at the helm, it is unlikely to be dominated by any one of them. So you don’t have to worry about the Linux Kernel tracking every keystroke to harvest data for adverts.
It comes down to who's in charge, and what their motivations are.
Google's motivations include learning everything about you, and how to serve you more ads.
Linus' motivations include writing solid code and flipping people the bird.
It's not just about the current situation, either.
When Linus finally goes to the bitbucket in the sky, there are so many companies, organizations and private individuals with an incentive to keep Linux operating and free from backdoors, we can somewhat rest easy at night.
But with Chromium/Chrome - there are only a handful of organizations in the world that can keep up with the rate of patches coming out of Google. Google de-facto controls the direction of Chromium.
I'm not saying that's worse than the alternative. Maintaining something like that is a hell of a lot of work. Unfortunately, things have become so complex and monolithic that there are only a handful of viable browsers.
At least two of them have source available, so we can start over if we absolutely have to.
Perhaps Microsoft adopting Chromium will be healthy for Chromium too then. I don't expect Microsoft to immediately keep pace with Google's browser teams but they should be able to contribute significantly to Chromium and help balance out the direction of the project.
Maybe, yeah. It would be great if there were multiple heavy hitters involved in such an important project. I think a lot of us on here are assuming that they'll fork and stop submitting their patches upstream, once they get comfortable with the codebase though, but who knows what will happen these days!
The kernel isn't trying to adhere to an open protocol. It's defining it's own to match ever evolving hardware.
Ultimately it's about the client driving the protocol. We don't want that. It means there will be only one client.
We have an ISO C standard and many different C compilers. We have a standard for C++, and many different C++ compilers. With things like Matlab and Go, you have an implementation. There is no standard. The implementation defines the standard. And there is only one implementation. That's not good for anybody.
Headed? We've been there for quite some time and this was piloted by Google itself. It's been years since they started to do the "works better in Chrome" and blocking features if you are not in Chrome (or, to be more specific, if you are not using chrome UA).
A large percentage of enterprises still use windows and some sort of Microsoft browser. For the ones that’ll switch, I wonder if they’ll switch to chrome or Firefox?
I don't know, I am typing this post on Edge, and HTML rendering is way down on my list of complaints about it. Is switching to Chromium going to fix the lack of a "Paste As Plain Text" option, for instance? (Not the biggest issue, just the one that's freshest in my mind.) And I hope that whatever they do retains their focus on not burning battery, which is a big reason I use Edge as much as I do.
If Microsoft really does this and makes the Chromium based browser the default in Windows, what is the value proposition of installing Chrome? Presumably all the websites will work exactly as they do in Chrome and it will be no faster or slower?
It's value proposition is integration with the Google ecosystem. If one rejects Google then there isn't much of a reason to use Chrome over another Chromium-based web browser.
I try to avoid Google as much as possible these days, thus it's funny because I never thought I'd consider using a Microsoft browser instead of a Google one. I think this is a potentially good move on Microsoft's part(the EdgeHTML thing turned out not to be).
Also: Data collection by Microsoft instead of Google. Since Microsoft is already collecting data from the Windows OS, they are already part of your trust chain.
Interesting, but what I really, really want them to do is EOL IE 11 so that the web can move forward without feeling guilty about the X% of users still on it. IE 11 has unfortunately become the new IE 6.
As it is right now I believe the IE 11 EOL is tied to the Windows 10 EOL...which is not happening anytime soon. Many frameworks are dropping support for it anyway so I guess it will end up being defacto desupported.
As someone who has (luckily) only had to support Firefox/Edge/Chrome, can you go into why IE11 is so bad? In the very small amount of testing I've done, everything seems to render just fine (especially compared to IE6/7/8)
I've been on multi year projects to modernize old IE applications, and one of the things that IE11 misses is "standard" xpath evaluation. A lot of old "dhtml" apps were all in on soap/xml. There's some wierd IE only hold over css in it like cursor hand vs pointer. On the whole it's not too bad really.
The reason why IE11 will live for a long, long time is it's compatability modes. You can run apps all the way back to IE7 on it. And yes these still exist.
>There's some wierd IE only hold over css in it like cursor hand vs pointer
Wow, that would really frustrate me. One of the first things I do (or check to make sure it's done by whatever css library I'm using) is add
a {
cursor:pointer;
}
to my css file. The first time I tested a website in IE11, I would be pulling my hair out!! Or more likely I would google it, but still, it would be frustrating.
It's utterly depressing how many red "No"s there are in the left column next to green "Yes"s -- all the features we could use if we could merely drop support for IE11.
IE 11 is the last browser with any substantial share that is no longer being actively developed (and automatically updated). As such the are a number of new features that get broad support in "evergreen" browsers that IE will never support. Most frameworks and developers end up bending over backwards to polyfill these features leading to more complex code and larger file sizes. Here are a few examples:
Every other major browser supports this natively aside from IE, but most people still transpile their modern code down to ES5 + polyfills for compatibility
Again, widely supported, but instead we use tools like SASS/LESS
Web Components[1] (Shadow DOM[2] and Custom Elements[3])
Newer frameworks like Ionic 4 rely heavily and web components and see it as the future of all UI frameworks and the end of framework churn[4]. Once again IE 11 holds back the pack and has to be pollyfilled[5]
Vuejs is completely rewriting its observation mechanism[6] to be proxy-based in version 3, however it appears that cannot be polyfilled and so they will be providing a second, optional, not-fully-compatible build, specifically for IE 11[7]
> Most of the ES2015 features used can be transpiled / polyfilled for IE11, with the exception for Proxies. Our plan is to implement an alternative observer with the same API, but using the good old ES5 Object.defineProperty API. A separate build of Vue 3.x will be distributed using this observer implementation. However, this build will be subject to the same change detection caveats of Vue 2.x and thus not fully compatible with the “modern” build of 3.x. We are aware that this imposes some inconvenience for library authors as they will need to be aware of compatibility for two different builds
SharePoint Online’s user interface has been rewritten in recent years using an open source stack (node.js, React etc) and certainly works far better using Chrome and not IE. This is true of SharePoint on-prem too, at least the recent versions. Even the tools we use to make web parts now are React, Yeoman, webpack, Yarn, Sass etc. Microsoft are heavily investing in these and most of the training materials they produce for developers are pushing us more and more in that direction.
In my understanding, Corporate is tied to IE for three reasons: it is easy to manage via central policies, it is reliable at rendering legacy (esp. intranet) apps, and Edge isn't even available for some Windows editions (Enterprise LTSB).
Especially the second point isn't new. So in the past, MS has built compatibility modes into IE, where a modern rendering engine is used by default (or at least when a certain X-UA-Compatible header/meta tag is present) and a fallback engine is used upon certain triggers (explicitly via compatibility lists and X-UA-Compatible, or via heuristics, e.g. for intranet sites).
I don't understand why they couldn't do a similar thing now: use EdgeHTML in IE by default, and fall back to Trident via certain triggers.
Yes, it would have required using two different rendering engines, while IE 10 and earlier can be emulated by IE 11 Trident. I don't know how IE is built, but heck, in the early Firefox days, I used to have an extension that could seamlessly switch to IE rendering within the Firefox window. So I would like to believe that a better solution than making IE the new Walking Dead among browsers should be realistic for MS.
What really pains me is the X% of users still on Safari 9. Not only does it not support a lot of new features, but how do you even test it? At least IE11 I can easily spin up a free VM to test.
It just depends on how far you're trying to push these browsers. Chrome has a lot of crazy cutting edge APIs which allow amazing developer and therefore user experiences. For example, IntersectionObserver, which allows developers to know when a part of a page is active, and respond accordingly; Safari doesn't even implement IntersectionObserver, even though it's been a standardized spec since like 2013.
It might seem like minutiae, but, safari also doesn't pay any attention to SVG specs. This makes the pages I build for beaver [1] like, impossible. They work beautiful in chrome; but they don't work at all in safari; and because of that, I can't really explore what potential experiences pushing the limits on SVG might have for my clients. Whatever you build has to work in all clients. It's a sad state of affairs, because if safari wasn't so far dilapidated, I would have been able to build some really impressive experiences; but, without safari support it's a no-go.
I think if you're trying to browse sites like cnn or huff post it doesn't matter. It's really when you get into the metaweb and find developers trying to push the browsing experience you really yearn for more since non-chromium browsers are just so far behind when it comes to staying compatible with web specs. Safari has implemented many of the most important APIs, Edge, is missing many that have been defined and cemented for like a decade.
The latest version of Edge broke the "FormData" constructor. So you can't manually create FormData to post from a set of elements not already in a form element together.
And the failure is insidious, it would wipe the value, but still submit the key. Overwriting any previously saved data.
Reported it three months ago, assigned to someone, no progress.
Safari is broken on purpose in a lot of ways when it comes to security, for better or for worse. I've had to do a lot of workarounds for Safari to compensate for the restrictions on cross-origin frames even though the host allows for the origin in its CSP, just as an example. Then of course there's the lack of support for WebRTC and virtually no support for PWAs. As Apple doesn't really view itself as a software company, and because they value user privacy to some extent, I don't think they care about having Safari compete with the rest of the browser market.
If Edge is broken, I can't really tell beyond what I think were some quirks still present when I was trying to target it last year. It wasn't that bad, but it didn't seem like Microsoft was making as big an effort towards it as they could have.
What disappoints me is that, for reasons unknown, Microsoft didn't choose Gecko as its rendering engine, as it(and Firefox) could benefit from the extra attention and money from Microsoft.
EDIT: I initially said "Spidermonkey" when I meant to say Gecko.
Safari is broken in ways that don't make sense if you don't live in apple land, so I can't bother with their mess for 5 or 6% of the total browser market
most of that 2 billion is concentrated in the US and the wealthier european nation states. places with people who are educated enough to understand "this website works best on chrome" or "this browser is not supported" error messages. it's pretty easy to see where the wind is blowing
i'm living in india right now and basic websites like dominoes ordering websites glitch out on safari/iphone prompting me to switch. complaining does nothing,i've had lengthy chats with dominoes support - it's just not a priority. apple is going to pay for ignoring developing markets imo
Then of course there's the lack of support for WebRTC and virtually no support for PWAs.
From Webkit.org, June 7, 2017 [1]:
>Today we are thrilled to announce WebKit support for WebRTC, available on Safari on macOS High Sierra, iOS 11, and Safari Technology Preview 32. In this post, we will go through an overview of our implementation. We will have future posts that cover more best practices for developers.
Android Browser, Chrome, and Firefox all support this feature at this point but not Safari or anything that uses WebKit. So while service workers are supported on iOS Safari, PWAs are relegated to being mere cached webpages. The manifest is probably more important than service workers because without it all you have is a webpage that is treated no differently from other web pages. A progressive web app is supposed to share characteristics with normal apps by definition and be recognized as such by the browser, which is still not the case in Safari.
So yes, there is still virtually no support for PWAs. There would be partial support if some manifest features were implemented, but currently it's not supported at all.
Very sad. I'm not a Microsoft fan but I thought this was one of the things they were doing right.
The performance was really good, compatibility with standards was actually the best IMHO. Every thing (SVG) I tried that got past Firefox and Chrome worked on Edge without modification.
And yeah, we need the competition.
It's not like Microsoft is running out of money. Guess they need their most talented people doing something else.
> compatibility with standards was actually the best IMHO
Unfortunately it wasn't. See caniuse.com.
> And yeah, we need the competition.
This is the key point. Indeed. Google becoming the ultimate web monopolist seems scary. I just hope Mozilla is going to be able to keep competing (but I'm not sure how long will it stand, a neighbour post says "Firefox desktop market share now below 9%"). I wish MS could choose the Mozilla engine instead...
It's a good thing more talented people can focus on development of an open source browser. That means more people can share in a consistent and reliable way. It also means much like Mozilla has been innovating with Servo we're seeing amazing things from that and it doesn't have to be X vs Y. MS vs Google. It can be a collective working towards a better common platform for everyone.
… and it will always use a severely outdated version of the Blink engine with nothing but some of vulnerability fixes back-ported occasionally (after unreasonable pauses of course).
"fuck you" is not same as stupid. The comment is good actually- edge has potential, why not open source it since you're not going to do much with it anyway in terms of making money. this also brings devs to windows eco system and helps them explore windows internals.
Sad news for me. I liked Edge. It missed few essential features for me (no search in history, really?), so after a while I switched to Chrome, but I was checking it from time to time and I never wanted it to have another engine.
Microsoft seems to lost an ability to fight and make successful projects. There’s absolutely no reason for me to use their browser now.
A number of the comments mention monoculture. But where does monculture end and a solid reliable and honest standard begin? For exampke, at some point the power grid and gasoline formulations were standardized, yes? TV screen sizes, tin cans and shipping containers. Is the browser not the shipping container of our time?
To call W3C a standard and then have different implementations is no standard at all. That is the nature of guidelines. The fact that said "standard" so often led to a suboptional UX only poured salt on the wound.
Maybe this really is bad news? But there doea seem to be some upside, monoculture or not.
Standard is an interface with multiple implementations. Shipping container standard with certain dimensions is good. Shipping container standard “from evil Google” is bad.
Standard can also be a ubiquitous implementation: OOXML, PDF, CUDA, Linux, glibc, pulseaudo, ffmpeg, imagemagick, tensorflow, systemd, Windows, MySQL, React, Java, QT, Electron.
Since standards bodies these days are really documentation effots it seems silly to say that implementation defined standards aren't. It's practically where any 'proper' standard comes from these days where the market leader is essentially the reference implementation.
W3C has had only figurehead status for a number of years now. Browser standards have been defacto agreed by a consortium of the major browser vendors, and W3C have rubber stamped them.
“This isn’t about Chrome. This is about ElectronJS. Microsoft thinks EdgeHTML cannot get to drop-in feature-parity with Chromium to replace it in Electron apps, whose duplication is becoming a significant performance drain. They want to single-instance Electron with their own fork”
The irony. They had HTA apps (running on IE engine which still work more or less, I replaced most default Win10 apps like the image viewer and the "groove" music player with a version of them built with HTA since they were slow as hell to start up) for 20 years already, they did barely anything with it, they didn't really invest seriously in a good browser engine (Edge isn't) and now all they can do is use Google's? typical Microsoft... reminds me the whole ES4/Jscript.net debacle only to then reinvent the wheel (ineffectively) with Typescript... those who don't know the story should read about it, how many years wasted? 10, 15 years?...
What MS is really doing is replacing HTA,WinStore apps, whatever with Chromium, because their own tech can't compete with Chromium as they didn't seriously invest in it. I'm actually starting to believe the people who are claiming that MS is rebuilding Office in JS from scratch...
Their desktop strategy is really confusing. I think even to themselves. It seems since around 2000 or when .NET came out they lost direction, repeatedly tried something, didn't really finish and dropped it. Mobile strategy wasn't much better.
They are fragmented and not under a single focus and that is the problem. While the executives at the top are focused on cloud those in charge of the OS and other crucial products are making some oddball decisions for whatever reason. If Microsoft could push all their teams towards a consistent vision it would be great. Otherwise we will continue to see this giant love-hate of Microsoft vs the appreciation they deserve for the things they do get right. I am company neutral so I try to appreciate companies that do some things right.
Yup. Alas, that's not a feature, that's a gaping security hole. (Had the misfortune of working on a HTA e-mail client. Turns out, some of the VBS viruses of the age were executable in HTA context...hilarity ensued.)
That's an interesting video. Office has developed an incredible toolset for OS independent development and also very powerful UI libraries. I always wonder how Windows desktop development could look if they had released office components as libraries for developers. I think Windows development went downhill when office introduced the ribbon and started to deviate from the standard look and feel of the OS.
Honestly having trouble making a coherent strategy out of those words, despite knowing what all of them mean. Seems like at the least it is written with way more confidence than could possibly be appropriate.
Perhaps Microsoft developers, with access to Windows' source code, can better optimize Google's upstream Chromium on Windows for the benefit of Electron apps using Google's Chromium? Otherwise, porting Edge to Chromium isn't going to anything for Electron apps.
I think it's referring to Microsoft's own Electron-powered apps, like VSCode. There have been rumblings in the past about MS wanting to fork Electron and replace Chromium with EdgeHTML in their fork, but it sounds like the strategy has changed to forking Chromium and using that in their future Electron fork instead.
I have no idea if this correctly interprets the SwiftOnSecurity tweet, but that's the best I can make of it.
That sounds fine, it just sounds unrelated to a decision to kill off Edge. You can maintain Edge as a web browser while using Chromium Electron. If you're Microsoft, your decision on whether or not to maintain Edge as a web engine is unlikely to completely pivotally depend upon whether you end up choosing Chromium Electron or Edge Electron. So the tweet doesn't make sense to me.
Off topic but I'll be that guy: I am continually blown away by just how good VSCode is and how it very much overcame two huge obstacles: being an election app, and its poisonous name.
I don’t know any anti-MS folks anymore. Those who were that generally stopped hating MS when MS stopped being a threat to the existence of their preferred platform, or to open source in general.
I think its more than just them using it for their apps. I think they want to turn Electron into a reusable Desktop UI framework for Windows and given the fact they now own GitHub they can force that for the Windows rendition of Electron. It all seems like almost a bad idea if it makes Edge too reliant on Chrome. I really was hoping Microsoft would instead have fully open sourced Edge. Edge could use some community love to make it a sweet out of the box Windows browser.
"drop-in feature-parity with Chromium" == They want to be able to replace Chromium with their own engine, but it must be very compatible.
"duplication is becoming a significant performance drain" == Every electorn app bundles its own browser engine (as opposed to a single browser engine for the whole OS, like a shared library), which increases system RAM usage and hinders performance.
"They want to single-instance Electron with their own fork" == MS wants to support electron apps but they want to use a shared browser instance for performance.
They rightly believe that doing so with their own browser engine is a huge pain in the ass.
Might as well just use the real Chromium for all browser needs throughout the OS including electron apps instead of trying to reimplement it.
> which increases system RAM usage and hinders performance.
I can see that it increases disk space, but everything on top of that is questionable. I believe DLLs will share/reuse memory pages even if they were downloaded in separate bundles.
But how would that situation be any different in the presence of some kind of system Chromium install provided by MS?
Chromium updates often, so you don't want to run your client-side JS against any other Electron version than the one you wrote it for because they change APIs frequently, so you'll still have one set of DLLs for each needed Electron version.
Given that it's Chromium (and thus effort to get things running should be minimal), and most Electron app writers aren't pushing Chromium to the bleeding edge, why wouldn't they test on the one that ships with the desktop OS with the largest installed base out there?
Another thought: If the same browser shared library can be used for every process, the same RAM pages should be executed for each process, so the CPU should be able to keep more of the code in CPU cache, along with probably branch prediction statistics as well. To the CPU it is the same code running which has many performance impacts. Less code = faster.
The problem with electron is that every electron app is shipping with chromium, which means a lot of resources are being duplicated.
A way to solve this is to make the JavaScript/html layer part of the OS, so that electron apps would ship as thin layers on top of a the system’s engine (The same way that java applications are thin layers on top of Java’s virtual machine).
Apparently, the way Edge is built doesn’t allow for replacing chromium in electron apps. So they need their own version of chromium instead.
I don't even see why it would have to be a full fork, could just be a drop-in replacement for whatever npm modules builds them into Windows installers/EXEs. But it would definitely make sense. It would be another great way for MS to emphasize cross-platform, because the Linux/Mac support would already be there and be left completely alone.
That makes sense. I might sound a little absurd at first, but recently I came across the documentation about submitting PWAs to the Microsoft Store [1], which completes the idea that Microsoft wants to embrace apps built with web technologies. After all, Electron apps are just the makeshift solution until we have all required APIs available within the browser(s).
I can see how this is creating a monoculture on the web which could have some really negative implications going forward. However, chromium is open source; so I don't really think this is comparable to the IE days.
Also, as a web dev who is trying to push the limits of what is possible given current web APIs, being shackled by edge's lack of compatibility is really a hindrance and makes really cutting-edge stuff impossible. So, it will be nice to not have to worry about that as much.
It didn't seem like microsoft was ever serious about advancing the development of the web with edge; they were just always trying to catch up (and doing so poorly). Microsoft is probably gauging the state of their browser now and coming to the conclusion that they're too far behind to make a realistic comeback without totally revamping their approach; lots of firings/organizational reshuffling, etc.
At least google is serious about the web APIs, even if its' only because it aligns with their financial interests—at least it does end up being a good user experience; i think that's what matters.
--
edit:
Also, not trying to belabor the point, but this subject on the whole is especially important to me. Chrome has allowed me to do crazy-amazing things with SVGs for my dev agency (1) since, well, they actually follow most of the SVG spec. I think most browser vendors see the SVG spec as superfluous and don't follow the spec verbatim, and that stops people like me from doing more awesome things with it; I'll literally have clients pitch me awesome ideas and my response is; sorry, we can do that but it just won't work in safari! So it becomes a no-go for all.
If other vendors really were able to dedicate serious resources towards their browser implementations then yeah, I would also be unhappy about microsoft's decision here. Optimally they would shell out more resources to their Edge division; but since they don't care about web experience the way google does—I agree that deferring to the experts is the best case scenario for everyone (i.e. developers like me and then users). At least, until microsoft redefines their priorities.
wouldn't you expect though; if the experience got so bad, that you'd have a sort of 'hard-fork' like we see happening all the time in the crypto space; with devs moving to whatever platform is best? That would have just been nigh impossible with close sourced software; at least there's a potential path should we ever have to confront that situation with chromium.
Practically, I wouldn't guess that we'd see that, since chromium ends up being the closest implementation to true web standards; well, since they define the standards practically, anyways.
"Hard fork" à la blockchains is a very good analogy. It's also a chilling one, considering that no hard fork of Bitcoin has ever seriously challenged it for dominance of the cryptocurrency space, and that the web has orders of magnitude more inertia than the Bitcoin blockchain.
>I can see how this is creating a monoculture on the web
It's already here, sad to see so many engineers I work with lately complaining so much at having to support anything other than Chrome and they're certainly not doing cutting edge work.
Find it pretty sad when really cross browser development is the easiest it has ever been in many ways, we're certainly not in the IE 5.5 days anymore yet there is less desire than ever to support anything outside their comfort zone and when real customers bring up something doesn't work in Safari their response it "Can't they just use Chrome?".
My first job in the industry was HTML/CSS for sites that needed to work in IE 5.5/6/7, Firefox, Safari and Crome had only just been released so maybe this bothers me more than it bothers most.
773 comments
[ 2.6 ms ] story [ 389 ms ] threadIt seems that with more interest in that, it's likely to improve.
> It probably won’t be as tied to the OS APIs as Edge was
What makes you say that? Seems like they could integrate it more or less exactly the same way.
It sounds like Microsoft really just wants a platform to route people into using Bing and service ads on the new tab page, which they're more than capable of doing on a reskin for Chromium. From a cost standpoint it makes sense to use the existing tools available.
For developers, it's one less platform to target.
if you look at desktop traffic only:
Desktop IE in Nov 2018 was 11.37%, Edge 15.65%, Chrome 52.74% and Safari 11.65%.if you include mobile traffic (where Firefox is basically non-existent), then Nov 2018 is closer to 3.5%.
talk about bleak :(
They say: "...widely distributed over thousands of websites." https://netmarketshare.com/methodology
Looking at netmarkshet website and web requests it seems this pulls from gator.io for metrics. Looks like 3k-4k websites have this data which lines up with the previous statment. https://publicwww.com/websites/%22gator.io%22/
Aren't changes to many open source projects gated, like the Linux kernel?
On a serious note, doesn't the license allow forks? Couldn't a large company just fork it and make changes without Google's approval?
Once it becomes "de facto", couldn't one argue to setup a foundation outside of Google's control as is done in many other open source projects?
I get that people hate Google but why bash on the project being open source; is it not open "enough"?
All projects gate commits somehow. It's not about the gating, it's about who is allowed to make the big decisions and how much buy-in they need to seek from other stakeholders before they're allowed to proceed. It's also about incentives; if Linus were, say, a Verizon employee and if the Linux Foundation were a Verizon subsidiary, people would feel much differently about the governance of the kernel. Likewise if the kernel were permissively licensed rather than GPL'd.
> On a serious note, doesn't the license allow forks? Couldn't a large company just fork it and make changes without Google's approval?
The thrust of the point here is that forking the codebase is no good if you can't convince people to install the browser and for websites to support the browser. It's a social problem.
> is it not open "enough"?
It's not. Open source gives users the freedom to fork. When forking isn't enough to preserve user freedom, the next step is open governance, which involves delegating decision-making power to users (with many interesting structural varieties to choose from). Amusingly, I gave a speech on this topic at All Things Open just a month ago.
edit: Guess the pro-monoculture folks flagged this.
https://en.wikipedia.org/wiki/Usage_share_of_web_browsers
That's the monoculture.
Edit: didn't realise you were responding to someone refuting what you had said, haha. so basically we agree, I just thought you were someone else.
And while iOS has the greatest market share for mobile, you're right, WebKit doesn't have the greatest market share for general browsers. That's definitely Chromium, now on Windows, macOS, Linux, iOS, Android, your mum's toaster, just about every 'native' web app…
Side note: I guess we should really be saying Blink, that's the name of the engine. But even if, say, you use Google Chrome on iOS, backed by WebKit, that usually makes you a Google Chrome, backed by Blink, user on the desktop.
Pointing out that there are competitors doesn't change that their presence doesn't overthrow the monoculture. After all, Gecko was around for a long time and Firefox did become very popular very quickly after version 2; but realistically, it took until Apple put WebKit on iOS for any real dents to be made.
Also, I think Konqueror still uses KHTML by default. I don't know if Konqueror tracks WebKit, but I think they're rather different these days.
KHTML was the only community driven browser engine and thank god for it. God knows what the browse landscape would've been without it.
On websites like "caniuse.com", a site which checks for support of the latest JS/HTML/CSS features, Chrome consistently ranks 1st. These features are all official features of the HTML/JS spec, not arbitrary Google-specific extensions.
Ensuring that more people use a browser with the Chromium-engine will allow web developers to use the latest features.
WebKit is a form of harmful monoculture. I'm ok with Blink's monoculture because it is probably the best rendering engine out there and inferior engines can't compete. But Apple's is creating an artificial monoculture by only allowing WebKit on iOS. This allows Apple to not innovative on WebKit and deliberately cripple new web technologies, so that people are forced to make native apps.
https://bugs.chromium.org/p/chromium/issues/detail?id=893460
Is there anything else to dig into to validate?
Abandoning EdgeHTML on the basis of that bug report would be drawing a long bow.
And the end of W3C standard development process. AFAIR it should be 3 independent implementations of the feature in order for Draft to reach Recommendation status.
So technically all that means that web standards will be written by WebKit team alone.
Sic transit Gloria mundi, sigh.
Edge is brand new, and foolishly branded to look like IE, this might stop average users who know about getting Chrome from using it. But it shouldn't. It's extremely rare that I'm forced to swap over to Chrome. Mostly I use Chrome to keep my invasive work IT staff from having access to my private browsing info.
[1] sorry for the URL: https://netmarketshare.com/browser-market-share.aspx?options...
Anyone is willing to invest in that?
For two engines forked from the same source (same architecture and even same code tree structure) it is relatively easy to copy/paste feature implementation, no?
[1] https://github.com/mozilla-mobile/android-components/
Also, Blink is only the renderer. Safari does not use V8.
These are still three strong and independent voices; Chrome has substantial control on desktops, Apple has a stranglehold on their highly popular platforms, and Mozilla has both the hearts of the tech community as well as wide regard in bleeding edge technology with Rust, Quantum/Stylo, and Servo.
It's saying "oh no, now Webkit can block things", not "oh no, we don't have enough strong implementations"
WebKit's JavaScriptCore is the best performant of the big 4.
https://twitter.com/webkit/status/728643624464883712?s=21
or, webkit will become the equivalent of linux for browsers. Considering that there will be less incompatibilities , it's a good thing. The hope is that competition for users will lead to innovations that benefit them, and not wall them off from competing platforms.
IMHO, I personally think it's a good thing to consolidate on one engine to render HTML cross-platform for these two reasons:
1. Web developers no longer need to worry about supporting CSS and other edge-cases across various browser engines.
2. The Chromium engine itself is open source. There are other browsers (Vivaldi, Opera, etc) that run on top of it.
3. While I think it's great the engine underneath is the same, I think it's equally great that there is a variety of UIs and browsers built on top of the same engine - the innovation happens in the browser space, not the engine, anymore.
Being open source is a red herring in this instance.
Say that Chrome implements a web feature you don't like. You fork the browser and remove that feature. But websites expect Chrome, and they use that feature, so your fork doesn't work with those websites.
Say that Chrome refuses to add a feature you want. You fork the browser and add the feature. But websites expect Chrome, so they don't use your feature so as to not break for their Chrome users, and your fork is no better off.
The insidious part of a web monoculture is allowing Google to dictate the standards of the web platform. Being able to fork the codebase only gives one the power to change things that are strictly client-side.
Say that Chrome implements a web feature you don't like. MS use their own MSengine for their Edge browser that doesn't have that feature. But websites expect Chrome, and they use that feature, so MS Edge doesn't work with those websites.
Say that Chrome refuses to add a feature you want. MS use their own MSengine for their Edge browser that has that feature. But websites expect Chrome, so they don't use MS Edge feature so as to not break for their Chrome users, and MS Edge browser is no better off.
Any feature that requires a site owner to do something to support it isn’t going to be added.
That's not quite true. This happens all the time with things not standardised (yet). With time those either get wide adoption because people want them, or get dropped, or get standardised in a different widely agreed form.
So long as the players are well behaved, having a single dominant browser is beneficial. How much time, energy, productivity, and money has been lost on cross-browser (in)compatibility wrangling?
Not sure if that should have been Blink, or if you're making a subtle point about how Blink was forked from WebKit (which was in turn forked from KHTML).
Personally, while I can see why some may react to this news with concern about a monoculture, I find it hard to feel sorry about the end of the last significant closed source browser engine.
If Microsoft gives up IE and Edge then all the major browsers will have an open source foundation. With Gecko, WebKit, and Blink there remains a healthy range of options, too.
And as the history of KHTML/WebKit/Blink demonstrates, derivatives will appear if there's enough interest. Perhaps someday Microsoft will follow the examples Apple and Google have set by creating their own fork, if the circumstances warrant it.
As with anything, you always need somebody (or a group of somebodies) to help something new gain traction. There's no magic pixie dust that will let you develop a major new platform without commensurate resources. What open source does do is lower the cost of entry and make it easier to build up to a level where you can make an impact.
The entire point of this subthread is that it doesn't matter that you can take Webkit and do whatever you want, because the web is a client/server (browser/site) protocol, and having the ability to fork the client is irrelevant when this makes you incompatible with every server. Technical capital isn't sufficient; it takes social capital to pull off a hard fork.
> So long as the players are well behaved, having a single dominant browser is beneficial.
I've been on HN long enough to remember this tired argument from back before the Blink fork. The gain in efficiency from having a unilateral dictator is not worth the loss of mutual counterbalancing oversight. It is completely without foundation to give Google the benefit of the doubt that they will be well-behaved when there are no consequences for misbehavior. Google is not your friend, nor is Google's mission to make the web better. Google is a corporation whose purpose is to maximize profit by selling advertisements. You might as well hand stewardship of the web over to Comcast.
Web devs will have to build for the lowest common denominator so this change improves that slightly.
An actual downsides is it increases the power of the Chromium team to make defacto changes.
It might also reduces innovation but I can't really see Microsoft deciding to drop a feature because it would take them further from the chrome trunk.
So, as the OP said - back to the bad old days of IE6 dominance.
I was doing some webdev back then, and whilst everything you said was true it was the lack of cross browser CSS and Javascript testing which led to the real frustration.
Back in 2002/03 IE6 had over 90% marketshare, and justifying testing in anything else was a constant battle.
That's exactly the same situation with Chrome. It's faster than Firefox, has less CPU issues etc (and I say that as a long time Firefox user).
The mono-culture is the problem, and it's embarrassing that we've forgotten this.
Using the same engine across multiple browsers, open source or not, just doesn’t cut it.
Sometimes innovation is at odds with whatever Google's Chrome team feels like doing. That is why engine diversity is a good thing.
> Suggest it to them?? If it's really faster, why wouldn't they merge it?
How many major software projects do you know of that will casually add a programming language?
I mean, it makes sense—Google has an enormous team on Chromium and they don't want to have another language to support. But that is precisely why it's important to not have the entire Web under Alphabet's reporting structure.
Take your pick.
I look forward to this...
Mozilla is Poland and is destined to get f*cked as their market share drops and any negotiating leverage they have to get Google TAC money disappears (they're more like Saudi Arabia is to the US, a vassal state, rather than a Great Power).
These are dire times for the web. Only Vestager can save us now.
[1] https://en.wikipedia.org/wiki/European_balance_of_power
[2] https://stratechery.com/2018/the-end-of-windows/
Not that I know that well, my work that forces my Windows usage still makes me use IE.
- Without competitors, innovation is dead. If there was only one web-engine, why bother improving it further? Given that it's open-source, it could theoretically be forked, but in that moment you again have two different web engines or more. Also, Mozilla would have to fork Blink from day one, because many things in it, they do not consider acceptable.
- A monoculture means one security problem makes everyone vulnerable.
- Multiple implementations challenge standards. You still would want a definition of a standard, or an API if you will, to point web developers to. But those are only going to point problems out after you've already implemented it, and likely also after other websites are already using the feature productively. They also can't point out stupid specifications that aren't going to allow you to update your browser engine in the future. So, you'd be much more likely to have to break compatibility, which doesn't play well with the web.
Right now I have perfectly valid SVG and HTML5 that straight up does not work correctly in Chrome.
There is a reason why people regard the MDN docs to be the best documentation for web technologies.
(Obviously, MS will maintain their own fork and can permanently fork away whenever they want/need to, I guess, so it's not like they've locked themselves in forever, but still)
I actually wasn't aware of the problems with Edge. This is the first I've heard of them... I hadn't heard much about it from either devs or users, so I assumed lots of people were more or less happily (or at least uneventfully) using it as Windows' default browser.
Never thought I'd be sad to see Internet Explorer (or its descendant) bite the dust, given what a plague upon humanity IE was for so many years, but I'm not sure this is the happy ending we might've wished for.
Same, but neither of us actually use Edge. Just like nearly everyone reading this comment.
Chromium != Google. Chromium is the FOSS project that Chrome is based off of.
Microsoft is of course free to diverge from it as much as they like but, like any other Chromium-based browser, they are either going to be joined at the hip to Google's decisions, or they'll be downstream consumers. The bottom line is that while MS will be contributors to Chromium they won't be in the driver's seat.
Regarding the problems with Edge, mainstream websites don’t face problems — but you hit the edge cases (no pun intended) when you try to use it for smaller sites — and especially sites meant for limited audience — e.g. internal websites built by enterprises. They are often tested for Chrome and Firefox only — and maybe Safari if there are significant number of Macs in use.
I can’t really blame the developers for not wanting to waste their time on a browser with limited usage and — especially since it often presents challenges not posed by other browsers.
What are your objections to modern Firefox? Most of the criticism I've heard is that it's become too much like Chrome. I'm quite happy with it personally.
That said, my comment was directed towards that FF’s market share is plunging rapidly and unless things change drastically, it might become another Opera — a niche browser at best. We need something that can counter Chrome and by extension, Google — something that developers have to support in addition to Chrome. FF has been playing that role nicely so far, but for how long? FF is not cutting it in terms of adoption — in terms of challenging the might of Chrome.
* No spellcheck in form fields (without a plugin, and I hate plugins).
* It feels slower. There's something about it that just doesn't feel as fast. Mostly things like opening windows etc.
* There's some annoying defaults.
* Dev tools aren't as good.
* Updates aren't installed automatically.
I mean, it's close to Chrome, which is good (for me), but it just feels about 5% less polished. It's good enough that I'll stick with it, after all there was a driving force for me to leave Chrome too.
It seems to me that spell checking of form fields should be a basic feature integrated into the browser.
1. Why not Firefox?
2. Why do I have to jump through some undocumented hoops to get something that should work out of the box?
I understand this is my fault for having 10-30 tabs open at any point in time, but still, chrome will generally manage that at 2-4gb ram. FF often breaks 10gb.
I understand Jira is a pig of an app (Atlassian software is generally shit), but I unfortunately need to use it all day long. Chrome seems to be able to handle 2-4 each of jira, gmail, google cal, github, aws console, and slack. FF couldn't last time I tried, a couple months after they released Quantum.
This would, of course, be less of a problem if the cheapest macbook with 32g ram didn't cost $3k...
And Google sites are actually slower on Firefox, which I assume is not their fault but still doesn't work in their favor.
edit To clarify, I'm used to Chrome seemingly just updating itself in the background. Firefox needs a restart, and so far the restart behaviour with Firefox has been less reliable than Chrome (losing opened tabs sometimes, for instance).
This is increasingly not true. Between Bing and adding ads[1] to various parts of Windows 10, I expect Microsoft’a long term business strategy will be increasingly ad-heavy. And you can’t make them un-collect the data they’ve already collected on you.
[1] https://www.thurrott.com/windows/windows-10/192251/microsoft...
Maybe someday they'll do that but it's hard for me to believe that's the plan they're starting off with.
I wouldn't be too sad if Firefox was switching to chromium either. Then developers would only need to worry about rendering into one engine, and users would have all their sites always working.
Apparently I called for this over four years ago:
https://news.ycombinator.com/item?id=7909383
That would be a terrible day for the open web. Damn it, somebody inject some IE6 into these people stat.
Keep in mind that, for quite a few years, developers welcomed the oncoming Microsoft browser monopoly; IE3/IE4/IE5 were generally better and faster than their Netscape counterparts and Opera was so niche as to be irrelevant. Then Netscape (the for-profit enterprise) folded and there was no viable competition to IE for a while, and the broken mess known as IE6 became a serious problem for years since it had something like 98% market share.
Trusting any company, even Google, to be the de facto sole steward of the web is insane. They are a for-profit company. That may not make them intrinsically evil, but they sure as hell aren't intrinsically good.
I rarely see people saying that there need to be competing kernel projects with completely separate code bases but all trying to conform to the same ABI, etc.
(Yes, I know there are other kernels out there, but my point is that people don't complain that Linux is so dominant like they do about Chromium vs Firefox)
Google's motivations include learning everything about you, and how to serve you more ads.
Linus' motivations include writing solid code and flipping people the bird.
It's not just about the current situation, either.
When Linus finally goes to the bitbucket in the sky, there are so many companies, organizations and private individuals with an incentive to keep Linux operating and free from backdoors, we can somewhat rest easy at night.
But with Chromium/Chrome - there are only a handful of organizations in the world that can keep up with the rate of patches coming out of Google. Google de-facto controls the direction of Chromium.
I'm not saying that's worse than the alternative. Maintaining something like that is a hell of a lot of work. Unfortunately, things have become so complex and monolithic that there are only a handful of viable browsers.
At least two of them have source available, so we can start over if we absolutely have to.
Ultimately it's about the client driving the protocol. We don't want that. It means there will be only one client.
We have an ISO C standard and many different C compilers. We have a standard for C++, and many different C++ compilers. With things like Matlab and Go, you have an implementation. There is no standard. The implementation defines the standard. And there is only one implementation. That's not good for anybody.
If I had my way I'd test on Chromium and Firefox only. And only the latest versions.
I already don't bother with Egde. Safari is the only pain I still have to bother with. Apple should just make a Chromium browser as well.
I try to avoid Google as much as possible these days, thus it's funny because I never thought I'd consider using a Microsoft browser instead of a Google one. I think this is a potentially good move on Microsoft's part(the EdgeHTML thing turned out not to be).
As it is right now I believe the IE 11 EOL is tied to the Windows 10 EOL...which is not happening anytime soon. Many frameworks are dropping support for it anyway so I guess it will end up being defacto desupported.
The reason why IE11 will live for a long, long time is it's compatability modes. You can run apps all the way back to IE7 on it. And yes these still exist.
Wow, that would really frustrate me. One of the first things I do (or check to make sure it's done by whatever css library I'm using) is add
a { cursor:pointer; }
to my css file. The first time I tested a website in IE11, I would be pulling my hair out!! Or more likely I would google it, but still, it would be frustrating.
It's utterly depressing how many red "No"s there are in the left column next to green "Yes"s -- all the features we could use if we could merely drop support for IE11.
ES6/ES2015 - https://caniuse.com/#search=ES6
Every other major browser supports this natively aside from IE, but most people still transpile their modern code down to ES5 + polyfills for compatibility
CSS Variables - https://caniuse.com/#feat=css-variables
Again, widely supported, but instead we use tools like SASS/LESS
Web Components[1] (Shadow DOM[2] and Custom Elements[3])
Newer frameworks like Ionic 4 rely heavily and web components and see it as the future of all UI frameworks and the end of framework churn[4]. Once again IE 11 holds back the pack and has to be pollyfilled[5]
1. https://developer.mozilla.org/en-US/docs/Web/Web_Components
2. https://caniuse.com/#feat=shadowdomv1
3. https://caniuse.com/#feat=custom-elementsv1
4. https://blog.ionicframework.com/the-end-of-framework-churn/
5. https://blog.ionicframework.com/october-2018-a-big-month-for...
Proxy - https://caniuse.com/#feat=proxy
Vuejs is completely rewriting its observation mechanism[6] to be proxy-based in version 3, however it appears that cannot be polyfilled and so they will be providing a second, optional, not-fully-compatible build, specifically for IE 11[7]
> Most of the ES2015 features used can be transpiled / polyfilled for IE11, with the exception for Proxies. Our plan is to implement an alternative observer with the same API, but using the good old ES5 Object.defineProperty API. A separate build of Vue 3.x will be distributed using this observer implementation. However, this build will be subject to the same change detection caveats of Vue 2.x and thus not fully compatible with the “modern” build of 3.x. We are aware that this imposes some inconvenience for library authors as they will need to be aware of compatibility for two different builds
6. https://medium.com/the-vue-point/plans-for-the-next-iteratio...
7. https://medium.com/the-vue-point/plans-for-the-next-iteratio...
Imagine the number of hours that could be saved globally if all this work and testing didn't have to happen anymore.
I feel ya. I keep having to stop myself from doing things the right way so that I can support my IE11 users.
Unfortunately, even after EOL, the ghost of IE11 will be around for a very long time. Heck, my November stats show 10% WinXP users.
Especially the second point isn't new. So in the past, MS has built compatibility modes into IE, where a modern rendering engine is used by default (or at least when a certain X-UA-Compatible header/meta tag is present) and a fallback engine is used upon certain triggers (explicitly via compatibility lists and X-UA-Compatible, or via heuristics, e.g. for intranet sites).
I don't understand why they couldn't do a similar thing now: use EdgeHTML in IE by default, and fall back to Trident via certain triggers.
Yes, it would have required using two different rendering engines, while IE 10 and earlier can be emulated by IE 11 Trident. I don't know how IE is built, but heck, in the early Firefox days, I used to have an extension that could seamlessly switch to IE rendering within the Firefox window. So I would like to believe that a better solution than making IE the new Walking Dead among browsers should be realistic for MS.
Honestly asking...
It might seem like minutiae, but, safari also doesn't pay any attention to SVG specs. This makes the pages I build for beaver [1] like, impossible. They work beautiful in chrome; but they don't work at all in safari; and because of that, I can't really explore what potential experiences pushing the limits on SVG might have for my clients. Whatever you build has to work in all clients. It's a sad state of affairs, because if safari wasn't so far dilapidated, I would have been able to build some really impressive experiences; but, without safari support it's a no-go.
(1) www.beaver.digital
The latest version of Edge broke the "FormData" constructor. So you can't manually create FormData to post from a set of elements not already in a form element together.
And the failure is insidious, it would wipe the value, but still submit the key. Overwriting any previously saved data.
Reported it three months ago, assigned to someone, no progress.
If Edge is broken, I can't really tell beyond what I think were some quirks still present when I was trying to target it last year. It wasn't that bad, but it didn't seem like Microsoft was making as big an effort towards it as they could have.
What disappoints me is that, for reasons unknown, Microsoft didn't choose Gecko as its rendering engine, as it(and Firefox) could benefit from the extra attention and money from Microsoft.
EDIT: I initially said "Spidermonkey" when I meant to say Gecko.
Moreover, it's irrelevant to non apple people
Are you equating "apple land" with increased privacy for your users, then disregarding it?
This ain’t your father’s Safari web browser: https://webkit.org/status/#?status=under%20consideration,in%...
There’s a lot of stuff that’s already supported in the released Safari and lots more stuff in development and being considered.
i'm living in india right now and basic websites like dominoes ordering websites glitch out on safari/iphone prompting me to switch. complaining does nothing,i've had lengthy chats with dominoes support - it's just not a priority. apple is going to pay for ignoring developing markets imo
From Webkit.org, June 7, 2017 [1]:
>Today we are thrilled to announce WebKit support for WebRTC, available on Safari on macOS High Sierra, iOS 11, and Safari Technology Preview 32. In this post, we will go through an overview of our implementation. We will have future posts that cover more best practices for developers.
[1] https://webkit.org/blog/7726/announcing-webrtc-and-media-cap...
For PWAs, you need Service Workers and some accompanying APIs and Safari seems to have those as well: https://webkit.org/blog/8090/workers-at-your-service/
Safari supports service workers but without support for the Web App Manifest, hence no "Add to Homescreen", PWAs are basically crippled on iOS.
https://webkit.org/status/#specification-web-app-manifest https://developer.mozilla.org/en-US/docs/Web/Manifest
Android Browser, Chrome, and Firefox all support this feature at this point but not Safari or anything that uses WebKit. So while service workers are supported on iOS Safari, PWAs are relegated to being mere cached webpages. The manifest is probably more important than service workers because without it all you have is a webpage that is treated no differently from other web pages. A progressive web app is supposed to share characteristics with normal apps by definition and be recognized as such by the browser, which is still not the case in Safari.
So yes, there is still virtually no support for PWAs. There would be partial support if some manifest features were implemented, but currently it's not supported at all.
There are over 2 billion iOS devices and every browser on iOS—Firefox, Chrome, Brave—uses WebKit.
The performance was really good, compatibility with standards was actually the best IMHO. Every thing (SVG) I tried that got past Firefox and Chrome worked on Edge without modification.
And yeah, we need the competition.
It's not like Microsoft is running out of money. Guess they need their most talented people doing something else.
Unfortunately it wasn't. See caniuse.com.
> And yeah, we need the competition.
This is the key point. Indeed. Google becoming the ultimate web monopolist seems scary. I just hope Mozilla is going to be able to keep competing (but I'm not sure how long will it stand, a neighbour post says "Firefox desktop market share now below 9%"). I wish MS could choose the Mozilla engine instead...
At least they could send some money Mozilla's way to ensure there is an independent competitor around.
If they wanted competition they should released it under Mac and Linux.
2018. “Fuck you, Microsoft, for bundling Firefox or Chrome.”
Hilarious.
To call W3C a standard and then have different implementations is no standard at all. That is the nature of guidelines. The fact that said "standard" so often led to a suboptional UX only poured salt on the wound.
Maybe this really is bad news? But there doea seem to be some upside, monoculture or not.
Since standards bodies these days are really documentation effots it seems silly to say that implementation defined standards aren't. It's practically where any 'proper' standard comes from these days where the market leader is essentially the reference implementation.
Precisely false. A standard must have different implementations to ensure it is robust.
https://twitter.com/SwiftOnSecurity/status/10697763353362923...
“This isn’t about Chrome. This is about ElectronJS. Microsoft thinks EdgeHTML cannot get to drop-in feature-parity with Chromium to replace it in Electron apps, whose duplication is becoming a significant performance drain. They want to single-instance Electron with their own fork”
What MS is really doing is replacing HTA,WinStore apps, whatever with Chromium, because their own tech can't compete with Chromium as they didn't seriously invest in it. I'm actually starting to believe the people who are claiming that MS is rebuilding Office in JS from scratch...
dHTML FTW!
I am sure upgrading static linked programs is easy. :D
Wasn't WebKit Apple's first? And before that KDE's KHTML and KJS. Funny how the narrative has changed.
Yes, but while Blink forked from WebKit, it isn't WebKit.
> Funny how the narrative has changed.
That's because the facts changed.
Office is C++, with 85-95% code shared between Windows, OS X, Android, iOS and Web; with a very tiny presentation layer on top.
Source: https://www.youtube.com/watch?v=zFmByeQ_ulo
Well if it’s going to run on Electron, that’s a given.
I have no idea if this correctly interprets the SwiftOnSecurity tweet, but that's the best I can make of it.
What, like https://en.wikipedia.org/wiki/Active_Desktop ?
"duplication is becoming a significant performance drain" == Every electorn app bundles its own browser engine (as opposed to a single browser engine for the whole OS, like a shared library), which increases system RAM usage and hinders performance.
"They want to single-instance Electron with their own fork" == MS wants to support electron apps but they want to use a shared browser instance for performance.
They rightly believe that doing so with their own browser engine is a huge pain in the ass.
Might as well just use the real Chromium for all browser needs throughout the OS including electron apps instead of trying to reimplement it.
I can see that it increases disk space, but everything on top of that is questionable. I believe DLLs will share/reuse memory pages even if they were downloaded in separate bundles.
Chromium updates often, so you don't want to run your client-side JS against any other Electron version than the one you wrote it for because they change APIs frequently, so you'll still have one set of DLLs for each needed Electron version.
A way to solve this is to make the JavaScript/html layer part of the OS, so that electron apps would ship as thin layers on top of a the system’s engine (The same way that java applications are thin layers on top of Java’s virtual machine).
Apparently, the way Edge is built doesn’t allow for replacing chromium in electron apps. So they need their own version of chromium instead.
I think it’s a better alternative to Apple's Marzipan.
It’s a tragedy for the Web but makes sense for Windows.
Anyway, as others have noted, this seems more and more prophetical every year:
https://www.destroyallsoftware.com/talks/the-birth-and-death...
https://twitter.com/garybernhardt/status/1069787375247622144
[1]: https://docs.microsoft.com/en-us/microsoft-edge/progressive-...
Also, as a web dev who is trying to push the limits of what is possible given current web APIs, being shackled by edge's lack of compatibility is really a hindrance and makes really cutting-edge stuff impossible. So, it will be nice to not have to worry about that as much.
It didn't seem like microsoft was ever serious about advancing the development of the web with edge; they were just always trying to catch up (and doing so poorly). Microsoft is probably gauging the state of their browser now and coming to the conclusion that they're too far behind to make a realistic comeback without totally revamping their approach; lots of firings/organizational reshuffling, etc.
At least google is serious about the web APIs, even if its' only because it aligns with their financial interests—at least it does end up being a good user experience; i think that's what matters.
--
edit: Also, not trying to belabor the point, but this subject on the whole is especially important to me. Chrome has allowed me to do crazy-amazing things with SVGs for my dev agency (1) since, well, they actually follow most of the SVG spec. I think most browser vendors see the SVG spec as superfluous and don't follow the spec verbatim, and that stops people like me from doing more awesome things with it; I'll literally have clients pitch me awesome ideas and my response is; sorry, we can do that but it just won't work in safari! So it becomes a no-go for all.
If other vendors really were able to dedicate serious resources towards their browser implementations then yeah, I would also be unhappy about microsoft's decision here. Optimally they would shell out more resources to their Edge division; but since they don't care about web experience the way google does—I agree that deferring to the experts is the best case scenario for everyone (i.e. developers like me and then users). At least, until microsoft redefines their priorities.
[1] www.beaver.digital
On Chromium being open source, and why that doesn't matter: https://news.ycombinator.com/item?id=18595978
Practically, I wouldn't guess that we'd see that, since chromium ends up being the closest implementation to true web standards; well, since they define the standards practically, anyways.
It's already here, sad to see so many engineers I work with lately complaining so much at having to support anything other than Chrome and they're certainly not doing cutting edge work.
Find it pretty sad when really cross browser development is the easiest it has ever been in many ways, we're certainly not in the IE 5.5 days anymore yet there is less desire than ever to support anything outside their comfort zone and when real customers bring up something doesn't work in Safari their response it "Can't they just use Chrome?".
My first job in the industry was HTML/CSS for sites that needed to work in IE 5.5/6/7, Firefox, Safari and Crome had only just been released so maybe this bothers me more than it bothers most.