Live data from Hacker News

Microsoft Edge's JavaScript engine to go open-source

blogs.windows.com

171–180 of 283 posts

Re: Microsoft Edge's JavaScript engine to go open-source

#171
post #123
post #118

Earlier quoted context omitted.

but they are not innovative They've had a seperate research department for over 20 years or so, and if memory serves me well quite some innovative things came (and still come) out of there. So I think they is too general in your phrase - there is innovation (also look at recent Surface line etc) but you're right in saying that it doesn't always come out and doesn't make it to the market, probably because the they you…

Exactly , You correctly rephrased that. I want to note last time I checked people who works at R&D in Microsoft , I literally blown away by their name. Maybe more big name than every other company in whole IT industry. But the fact as you mention remain correct. They are not trying to bring new and innovative stuff to people's life. At least not when they are leader .

I was also surprised to discover Microsoft Research's awesome technology. Actually, they've done a fair bit of licensing that research to other companies who go on to make great things with it. I always thought it was because they were pre-occupied with their own projects: Windows, Office, the Xbox, and of course supporting the vast .NET platform which remains the core of their business model: creating software tools and licensing them out to huge corporations, big big money!

Re: Microsoft Edge's JavaScript engine to go open-source

#172
post #118
post #103

Earlier quoted context omitted.

I think this is perfectly valid point.We can think of it like this, they are pretty competitive and have shitload of resource. but they are not innovative , the moment , they become leader they lose their interest toward growing and growing, although they have huge amount of resource which help them the moment they become underdog. I think this is cultural problem with Microsoft.

but they are not innovative They've had a seperate research department for over 20 years or so, and if memory serves me well quite some innovative things came (and still come) out of there. So I think they is too general in your phrase - there is innovation (also look at recent Surface line etc) but you're right in saying that it doesn't always come out and doesn't make it to the market, probably because the they you…

The papers coming out of Microsoft Research are a goldmine (relatively speaking as far as academia goes) for programming language and graphics research, just to name a couple of areas I'm familiar with.

Re: Microsoft Edge's JavaScript engine to go open-source

#173
post #109
post #19

Wait, Edge does better on ES6 coverage than both Chrome and Firefox? Microsoft have seriously stepped up their game, especially seeing as it's now neck and neck for performance with Chrome: http://venturebeat.com/2015/09/10/browser-benchmark-battle-s...

The next Firefox Nightly build should get 84% on that page, much closer to Edge than Firefox 44 (74%). I work on SpiderMonkey and I'm super excited about this news. All JS engines have added more-or-less similar performance optimizations but often implemented differently and I'm really interested to see what the Chakra team did. I'd be happy to write a blog post on it next month, if people are interested.

yes please

Re: Microsoft Edge's JavaScript engine to go open-source

#174

Earlier quoted context omitted.

One would think Firefox would lead here, since the browser is Mozilla's primary product, imagine how much hype and PR Firefox would get if only they were far ahead on ES6.

Strange thing to think. Companies do not produce browsers - employees do. In fact, not even employees. A handful of talented programmers produce browsers. All the desire in the world from a company won't improve the ability of those handful of programmers. It's actually fairly interesting. These huge mega companies are just support teams for a few programmers. All the corporate vision and endless strategies mean noth…

>Strange thing to think. Companies do not produce browsers - employees do. In fact, not even employees. A handful of talented programmers produce browsers. All the desire in the world from a company won't improve the ability of those handful of programmers.

Strange thing to think. As if the core mission of the company, the motivations of the management, strategic decisions to hire people and structure projects etc, the funding and priority they give to specific products etc, don't determine and affect the end product!

Put that way it's as if a great browser engine can even get out of some accounting software house, if only the right programmers chance to work there.

Re: Microsoft Edge's JavaScript engine to go open-source

#175

Earlier quoted context omitted.

One would think Firefox would lead here, since the browser is Mozilla's primary product, imagine how much hype and PR Firefox would get if only they were far ahead on ES6.

The browser is the primary product to the Chrome team. Does it make any difference?

It's obviously the primary product to ANY browser development team.

The question is if its also the primary product of the COMPANY the development team belongs.

Re: Microsoft Edge's JavaScript engine to go open-source

#176
post #148

"I literally blown away by their name". ^^ I'm so sorry to hear about your accident. I hope your injuries were not serious! :-D https://www.youtube.com/watch?v=9jh4Mpgbi4A

Here, meet a dictionary of the english language:

Literary [lemma 2]: in effect, virtually

It's a use that predates Shakespeare.

Re: Microsoft Edge's JavaScript engine to go open-source

#177

Earlier quoted context omitted.

The browser is the primary product to the Chrome team. Does it make any difference?

Although important for their future, Chrome is not Google's cash cow, Chrome is everyday stealing Firefox's market share, they have the luxury to wait, Chrome's pressure (and now Edge) on Firefox hasn't made Firefox step up their game as one would expect.

It might make one question the underlying assumption that competition somehow can cause people to somehow magically become better (a concept many people have which makes no sense). In reality, competition changes how people allocate resources as they play a strategy game to not lose control over segments of the market they perceive as strategically important. It also causes them to lose their negotiation power in the ecosystem, which can be good (as they can't push around smaller players) but also bad (as they can now be pushed around by larger players or loud users).

Mozilla used to be able to sit around and say "we absolutely refuse to do certain things, and we want to spend our time figuring out how to make the web an interesting place for power users and developers". I respected that Mozilla. It had a lot of clout in the market and used that clout to fight against DRM on behalf of all users while spending their resources building a super-extensible platform (which I think is a better description of Firefox's crazy plug-in oriented nature).

The post-Chrome reality is that Firefox no longer has an automatic dominating position in the "alternative" (non-IE) browser space, and so they have had to start caving to loud user demand and start fighting for the end-user market segment. They don't have the ability to fight against DRM anymore, so they have been forced to include Adobe DRM by default. They don't have the ability to waste a lot of time on power users anymore, so they are dropping all the complex-to-maintain parts of their platform spec and have started dumbing down the UI.

The one thing that Chrome did that was truly important was not to compete against Firefox: it was to prove to the world that something--specifically high-performance JavaScript--was both possible and desirable. This is the one positive aspect of "competition", and it is something that frankly should never need to happen in the world of open source, as one can do that in the context of the other project: I can't imagine a scenario where Firefox would have turned down performance patches :/.

But of course, Google isn't going to want to do that, because Google is a company with a strategic vision that happens to benefit from owning the web browser and being able to unilaterally make major decisions and perform weird experiments and crazy product integrations through it, which is all the easier for them to bootstrap as they can use their position as "the place almost all people both search and advertise" to push Chrome on people. This means that Chrome has no reason to collaborate with anyone, and even the one alliance they sort of had (with Apple on WebKit) they broke off when they decided they didn't have enough unilateral control: rather than collaborate as part of a community, Google just wants to own the product.

They also happen to be the primary customer of Firefox, so Mozilla is being forced to operate on smaller budgets. Note that this is the usual effect of competition and should be the obvious one: the idea of someone "stepping up their game" makes no sense when you are now operating on smaller margins (as competition means you can't demand as much share of the profit on any particular transaction) of a smaller market (as competition means that some customers will be using your competition). You only get to "step up your game" momentarily, often towards frustrating ends (such as giving up the DRM battle or trying to dumb down your UI as fast as possible), until your resources start to wither. (Yes: in a small initial market, competition can cause greater customer awareness leading to more pie for everyone; but that obviously isn't the case here: that is only true near the beginning of a new concept, when no one even believes the thing you are doing is relevant or valuable.)

In this case, it is even worse, as the primary customer to Mozilla's product was Google... and so they are essentially screwed in that negotiation. Firefox has had to switch to Yahoo as the default search engine and start making content deals to bundle marketing and software with their product, something they were morally opposed to doing in the past but have been forced into doing due to competition. This also doesn't come cheap with respect to executive time: rather than working out their product and platform vision, they are having to spend time negotiating and having painful conversations about how to keep their company from being destroyed and what morals they are willing to compromise for how long in order to maintain that fight. I don't particularly love Mozilla (as someone who has been paying attention to the open Internet since the beginning, I frankly found Netscape's business model of selling web browsers bundled with ISP contracts terrifying), but I have great sympathy for them these last few years, and absolutely do not see Chrome as being a positive force for anything at all in this ecosystem, except maybe security :/.

Re: Microsoft Edge's JavaScript engine to go open-source

#178
post #109

Earlier quoted context omitted.

The next Firefox Nightly build should get 84% on that page, much closer to Edge than Firefox 44 (74%). I work on SpiderMonkey and I'm super excited about this news. All JS engines have added more-or-less similar performance optimizations but often implemented differently and I'm really interested to see what the Chakra team did. I'd be happy to write a blog post on it next month, if people are interested.

One would think Firefox would lead here, since the browser is Mozilla's primary product, imagine how much hype and PR Firefox would get if only they were far ahead on ES6.

Microsoft has about 120 times the number of employees that Mozilla has. That's an insane number. If they consider the browser remotely important, they can put much more resources behind it than Mozilla ever can.

Only things like the Mythical Man Month save Mozilla a bit here.

Re: Microsoft Edge's JavaScript engine to go open-source

#179
post #33

Earlier quoted context omitted.

No wonder: Microsoft needs "class" syntax in JS, they are a driving for behind TypeScript and ES6. Especially interesting, as Javascript has prototype inheritance. The reason: All of MS code is class-based, and they want it to look similar to C#.

Developers of large JavaScript codebases have come to realize that class-based OO is exactly what they need. Prototype OO sounds interesting in theory, but in practice it tends to not be what's needed. Projects that use it end up trying to poorly imitate class OO. This wouldn't necessarily be a problem, except it isn't done consistently. There are multiple approaches, some of them which don't mesh well with others. C…

"Those people will go to their graves, never knowing how miserable they are" :-)

See 5:30 into video (on ES6): https://www.youtube.com/watch?v=PSGEjv3Tqo0

I have been doing a fair amount of JS work in an app using Angular this year. I don't use much inheritance of any kind, mostly FP techniques and "objects" (associative arrays) as aggregates.

If you use "closure" variables instead of "this", it's pretty easy to graft functions from one aggregate to another for code reuse.

Re: Microsoft Edge's JavaScript engine to go open-source

#180

Earlier quoted context omitted.

The browser is the primary product to the Chrome team. Does it make any difference?

It's obviously the primary product to ANY browser development team. The question is if its also the primary product of the COMPANY the development team belongs.

What is actually relevant here is whether the percentage that Google considers Chrome to be valuable to their company times the resources of the company is competitive against the resources Mozilla has (times the percentage that Mozilla considers the browser their focus, as they clearly spend a lot of money on side projects people sometimes seem to enjoy, such as Persona). Mozilla is small enough (at least in comparison to Google) that I think comparing their entire company to one division at Google is probably the correct strategy.
Post reply on HN