While ChakraCore is not going anywhere, EdgeHTML will be thrown away, and since MS won't use it anymore, nobody will. So why making it open source then? If it's for curiosity, it would probably be interesting. But in that case the only approach is to make it read-only opensource, just like MS did with .Net Framework. Setting up a project to be fully open source, accept PRs, issues and so on requires big effort even to start, so I doubt MS will do that for, again, component which nobody will use.

@andriysavin I agree with you, making it read-only opensource would probably the only way to go. I have an own opinion about EdgeHTML, so I wouldn't say that nobody will use it... What I've read about in the last one hour in forums and on news sites (comments), that there is a mass interested in open-sourcing this. Also its compatible to WebKit...

@nightwrkr Can you "aggregate" a couple of main reasons why mass would use it? I don't mean EdgeHTML is bad (actually, I was disappointed by that MS decision initially). But why using this engine as a whole if there already are alive and actively developing engines like Chromium? I understand if people want to borrow some code parts to use in their own products, however.

@andriysavin Why do you think "nobody will use it"? It will be picked up by community and ported to other platforms for sure. The key motivations are: freedom of choice and standards compliance. Otherwise we will go back to the times of IE6 monopoly, where there was only a single web browser which forced their own standards.

@niutech I'm skeptical because web rendering engine is a very complicated piece of software, and if the main maintainer of that code (MS) abandons it, I doubt there will be community of engineers of that level of seniority and expertise in the area which will be able to actively develop it. I hope to be wrong. Maybe some crazy guys will decide to make their own "Edge", like it was with Rotor and Mono. But web standards are messy and complicated, and change rapidly, constantly adding new functionality. MS confirms that with all it's resources it has never been able to be "on the edge". This looks different to Mono's story, and will require big team to develop the engine at the speed web evolves.

In my opinion the engine was never the problem. I think that EdgeHTML is very interesting and like @niutech said there will be someone who is going to fork and maintain this thing. I would appreciate and work on that too.

@nightwrkr No doubts it's interesting!

Seeing it even as an archived repo would be gangsta 💯

I wasn't going to post about this because it's technically off-topic, but I would like to see this too. I think a lot of people would be interested for historical/archival reasons at the very least, and also I suspect it's a fairly good, independent, implementation of many web standards up to 2018 or so. It seems a shame to throw the code away when someone else out there might have a use for it. I know a lot of people have been wishing for years that Opera would have open-sourced Presto when they switched to Blink. It's much harder to develop a browser engine from the ground up than to fix up or modify an existing one, and this is one of the few independent implementations of such a thing that still exists. I can imagine it gaining traction among a certain crowd as a browser that implements web standards independently and checks to see how far off Blink and WebKit are from them.

I don't think this is likely to happen, and I hope it doesn't. If you want an engine you have Chromium or Gecko/Servo which are actively maintained and supported. It'd be irresponsible to release the code of an engine that's

a) Still in use and will be for many years as users transition from EdgeHTML
b) Not supported by the developers, where security issues may linger and exploits discovered.

Allowing people to use or examine an unsupported engine has exactly 0 benefits for the community. There are open source engines in active development, use them instead.

I was wrong about ChakraCore, they're throwing it away as well. But at least it's already open source and already used outside of Edge, so may have future.

@thecodingdude you're actually correct, it kinda has zero face value for the community considering that any forks of it won't get mass adoption and will just create noise.

But on the other hand, putting it out could be beneficial for sentimental ones so they get bragging rights and get to tell their juniors / kids "hey kiddo, look at this repo. we built it from the ground up but eventually threw it away because Chromium rocks lol."; and also for the curious ones who like to see how things were crafted in closed doors - one might even discover easter eggs like those found in Apollo 11's source codes.

https://github.com/chrislgarry/Apollo-11/blob/master/Luminary099/BURN_BABY_BURN--MASTER_IGNITION_ROUTINE.agc

Zero business value? err yeah..
But can we at least do it for the lolz.. thats gangsta.
Besides, Microsoft already got the best place to archive source codes: Github.

One of the reasons to open source it, is that it is a very good starting point for creating a new browser that has something different to offer (much like Edge does now, but will not in the future in terms of a web rendering engine). Starting a browser from scratch, while feasible, is mostly not practical and when you have much of the work already done, you have a much better starting point.
There might be companies that could decide to take that code and have a go in the web browser industry, which is a good thing (less monoculture).

But I do not believe Microsoft will open source it.

It could have open source Tasman (the macOS version of Trident, release as part of Internet Explorer 5 for macOS) after they abandoned it and did not (I am not sure anyone suggested that they will, but it was a completely new engine that got thrown away completely). Big companies do not really mind just throwing away code. It is much more (legal?) effort to open source it, I guess.

Per my knowledge EdgeHTML used tons of internal libraries to make that. So they need a major rewrite before it could be open-sourced.

@phistuck I wouldn't say that. Tasman was abandoned in a time where MS was not really into the OSS world. MS could probably look differently on that now...

The best case for open sourcing EdgeHTML is in the readme:

We will offer our Windows platform expertise to improve the experience of all Chromium-based browsers on Windows

By open sourcing EdgeHTML you give others the knowledge of that platform expertise to help port some unique features and performance of Edge over to Chromium and even to Firefox's Quantum. Microsoft has a lot of value to add and while they focus on their top items, others can dig in the source to find other things they could port over.

@be5invis do the internal library issues matter anymore? By moving to Chromium, open sourcing WPF, XAML, WinForms, and UWP UI controls, we are closer to a fully open source Windows. The code they need to port over to Chromium I suspect will still depend on many internal libraries that they will in affect open source by porting it over (unless they are writing new APIs to handle it).

The value is not in keeping EdgeHTML alive, historical, or curiosity. The value is what it can bring to the Chromium project without Microsoft being the only ones with the knowledge to do it.

You know, you can just replace the names with [redacted].

Kidding aside that's actually fucked up, would be extra hassle to open-source it.

@dragonwolf83 @davalapar
EdgeHTML is not like ChakraCore, which is pretty an independent library.
EdgeHTML is built over many Microsoft key IPs, and some of them are not even from Windows department.
They are not publicly available so I call them "internal libraries".
So if you want to open source that you have to rewrite pretty a lot of the code...

Opensource edgehtml please. I want edge compete with chrome by Edgehtml itself, not Webkit-clone.

Also the UI components like the default progress bar are awesome.

Time to pay our respects, let's close this issue.

If it's true that EdgeHTML uses lots of internal libraries, then that's pretty unfortunate :(
Otherwise, I'd see no harm in releasing the source, even if you completely abandon it immediately after. Just because the source is out there doesn't mean you need to maintain it.

I'm interested in this being open-sourced because it(appears to be, I haven't done any tests) lighter weight than Chromium. It would be really sweet if we could, for example, use EdgeHTML inside Electron instead of Chromium.

This, Preferably under a Permissive license, like BSD, MIT, Apache, Public Domain, etc.

deep integration and exploitation of internal Windows features and APIs that shouldn't fall into the public consciousness.

So that means you don't believe in the open-sourcing of Windows itself then. Well many of us do believe the NT kernel should be open-sourced (they've basically already open-sourced everything else and all their patents are with OIN now).

IBM is a member of the OIN: https://www.openinventionnetwork.com/about-us/members/ so technically they shouldn't be opposing it but I guess it depends on if that particular code, if it's still there in existence, had it's patent moved to OIN. MS is doing a lot of work to remove a lot of old code other than ensuring certain compatibilities still exist for their business partners so there should be a low likelihood of any code from IBM still there. EdgeHTML probably has a bit more than just IBM code in there.

It would indeed be interesting to open source edge html, because it has some kind of optimizations and some very useful features (e.g http live streaming) that blink doesn't have.
Microsoft developers should make an effort to bring every optimizations and every useful features that edge has, to blink.
Especially webassembly performance.
That would be beautiful.

I don't want fate of Presto to befell EdgeHTML. But I know, that @microsoft opens source codes of non-profit or abandoned products, such as WinFile or MS-DOS.

Unfortunately, EdgeHTML cannot be open source. If anything, it's because EdgeHTML is based on Spyglass Mosaic's source code.

Sure, there is probably nothing left out of Spyglass Mosaic's source code, but it's almost impossible to prove that this is the case. Currently, Kudelski Group owns the copyrights to Spyglass Mosaic's source code, not MIcrosoft.

Unfortunately, EdgeHTML cannot be open source. If anything, it's because EdgeHTML is based on Spyglass Mosaic's source code.

Sure, there is probably nothing left out of Spyglass Mosaic's source code, but it's almost impossible to prove that this is the case. Currently, Kudelski Group owns the copyrights to Spyglass Mosaic's source code, not MIcrosoft.

How does Microsoft end up using some third party companies source code in Microsoft's product and NOT end up with the right to do whatever they want with the source code?

I know it's kinda off topic, but Microsoft had similar issues with open sourcing their STL too, and I just don't understand how that happens.

How does Microsoft end up using some third party companies source code in Microsoft's product and NOT end up with the right to do whatever they want with the source code?

In short, the licensing agreement allowed them to do a lot. However, it did involve percentage-based royalty (based on the sales from Internet Explorer), which obligated Microsoft to pay for percentage of sold Internet Explorer copies. Microsoft avoided the cost by making Internet Explorer gratis (making revenue zero).

This however means Microsoft cannot really make EdgeHTML free software, as free software licenses allow for selling software, and Microsoft would be obligated to pay royalties for sold copies.

If EdgeHTML will be thrown away, then there is no harm in giving it away to open source community.
All what has to be done is to upload the source code to GitHub. Who knows? it might turn into something useful.

Also if Microsoft says they love open source they will have to show it by open sourcing edge.

I would be very interested in seeing EdgeHTML become open source. As someone who is interested in how browsers are being developed, seeing this stuff is not only true to Microsoft, considering they have open sourced even MS-DOS in the past, but it's also a huge help to see what Microsoft Edge was, and what it could've been.

Well, should this come to fruition, it'll be quite a few more years, as Microsoft will be supporting IE11/MS Edge (using EdgeHTML), especially in Enterprise scenarios for many years to come.

Thanks for the great discussion. I'm closing this PR as we won't be taking the proposed text into the document ;).

@travisleithead
Will the EPub reader in the old MS Edge be open sourced?

I don't believe the EPub reader would be considered for open-sourcing independently of EdgeHTML as it relies on many of the platform features provided by EdgeHTML such as CSS Regions.