Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash • Hacker News Board

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#111

Post by TechnicalBonobo » Thu, Sep 15, 2011, 1:53 AM UTC

BrendanEich wrote: post #107

foofighter2 wrote: post #103

Earlier quoted context omitted.

First of all, Javascript appeared in September 1995. A quick glance at the timeline of web browsers reveals that Netscape did NOT have a monopoly at that time. See http://upload.wikimedia.org/wikipedia/commons/7/74/Timeline_... In theory, you could have gotten input from Microsoft, IBM, Mosaic, and so forth before ever releasing Javascript. Did you? No. Even assuming that those other companies had an insignificant sh…

"First of all, Javascript appeared in September 1995. A quick glance at the timeline of web browsers reveals that Netscape did NOT have a monopoly at that time. See http://upload.wikimedia.org/wikipedia/commons/7/74/Timeline_... . First of all, you did not cite market share numbers to show lack of an effective monopoly. You cited a fun wall chart of all the various browsers, most with tiny and non-growing if not rapi…

"1. ["same way"] Netscape had a monopoly, it wasn't injecting JS into a situation where there was already a scripting language widely used on the web and implemented among multiple competing browsers. It was not fragmenting a multi-lateral browser market or web content language ecosystem..."

"First of all, you did not cite market share numbers to show lack of an effective monopoly..."

"Netscape was a monopoly in effect (it's very rare for a real-world monopoly to have 100% of the market)..."

I don't think you know what the word monopoly means. Market-share is entirely irrelevant. What constitutes the definition (the reason the term even exists: what is meant to describe) is not how "big" a company is, but the _exclusivity_ (as in: is anyone else allowed to ENTER that sector of the market). And using the words "effective monopoly" or "in practice/real world" doesn't work as a permission to misuse the term, either. In fact, it does the opposite. Actual real-world "effective" monopolies would be: An entity holding a patent for some invention, the State having exclusive control of force, etc, etc.

Some company being "the only company who is currently doing X" is not a monopoly, as long as anyone else can enter the market.

http://www.merriam-webster.com/dictionary/monopoly

BTW many of us developers are totally against the evil "effective monopoly" (see what I did there :P) that this not-so-pretty JavaScript language has in the world of web development, so the idea of more options doesn't sound bad at all.

Because last time I checked, there was no axiomatic law embedded in the fabric of the universe which stated that "Every desired improvement and/or business endeavor in the realm of browser scripting should be expressed in the form of a proposal for a next version of JavaScript, over at http://wiki.ecmascript.org, or else the oh-so-heavenly multilateral dimension of the interweb net will fragment and spiral down eventually collapsing into a black hole made of kittens"

Dart is probably going to suck, though (and Java-stained ideas polluting its design will probably be the cause). Also, unless Chrome wants to commit suicide, it's gonna keep supporting JS in the current and future versions, so you JS people should put a halt to this soap opera. Pause thy bitchfest.

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#112

Post by juandopazo » Thu, Sep 15, 2011, 2:42 AM UTC

BrendanEich wrote: post #106

foofighter2 wrote: post #105

Earlier quoted context omitted.

I think you are being unfair in your criticism. Microsoft, Apple, and Adobe, just to name a few, have all tried to force web developers into proprietary closed-source platforms. If you want to get angry, get angry at them! Google is just trying to give web developers a choice. They plan to continue fully supporting ECMA's efforts as well, with developers and money. And for this they deserve to be attacked? You keep t…

"If you want to get angry, get angry at them!" What makes you think I have not been "angry" at MS, Apple, and Adobe? You must not follow my writings closely! What's more, my mood is not the point, actual market-facing behavior is. The topic here is Google and its actions. There are good reasons to focus on the big G now: * Microsoft is older, formidable but late to mobile, in some ways in decline. * Apple, we know wh…

Dart will probably work on Chrome Frame, not on its own plugin, which is already a fairly used.

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#113

Post by BrendanEich » Thu, Sep 15, 2011, 6:32 AM UTC

TechnicalBonobo wrote: post #111

BrendanEich wrote: post #107

Earlier quoted context omitted.

"First of all, Javascript appeared in September 1995. A quick glance at the timeline of web browsers reveals that Netscape did NOT have a monopoly at that time. See http://upload.wikimedia.org/wikipedia/commons/7/74/Timeline_... . First of all, you did not cite market share numbers to show lack of an effective monopoly. You cited a fun wall chart of all the various browsers, most with tiny and non-growing if not rapi…

"1. ["same way"] Netscape had a monopoly, it wasn't injecting JS into a situation where there was already a scripting language widely used on the web and implemented among multiple competing browsers. It was not fragmenting a multi-lateral browser market or web content language ecosystem..." "First of all, you did not cite market share numbers to show lack of an effective monopoly..." "Netscape was a monopoly in effe…

I defer to your economic terminology expertise, but Netscape did have 80% of the market during a huge growth phase (people extrapolated exponential growth from a few months or quarters in '95 and '96). What's the term for that position?

Whatever you call it, if Google had that now or very soon, it could indeed ship new stuff and "make it stick". Since it doesn't have that power, I repeat that Dart is fragmenting.

No one is obligated to work on extending existing standards only, not try injecting new ones. Doing both without market power to make the new ones stick is going to make a mess in my view. I keep saying this, do you disagree?

I'm the last person who wants to save JS from extinction. If it were cheap enough to kill, I'd do the deed myself. It's not cheap to kill -- quite the reverse -- and the leaked memo's assertions about it being unfixable are exaggerations at best, and betray a significant conflict within Google.

(BTW I agree there's a smell of "Java-stained ideas polluting [Dart's] design.")

If TC39 had a crack at standardizing Dart or putting ideas from it into ES6, everyone would be better off -- even if Google then launched Dart anyway.

Instead, while we in TC39 were working in the open on ES6 (which is past new-proposal freeze), we knew nothing. That is not just missed opportunity, I call it poor stewardship on Google's part.

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#114

Post by equark » Thu, Sep 15, 2011, 10:04 PM UTC

KamilTrebunia wrote: post #108

Dart will not be 2x (or more) faster than JS. Dart will not be so much more expressive and elegant than JS is. Google would have to replace the whole stack (from HTTP to DOM to Dart to IDEs) to make it really compelling, but I'm pretty sure they will give it a shot. Fortunately there is a competition: Nobody will dedicate himself to Chrome Web Store and Google Web technologies and miss out on future Facebook WebApp S…

> Dart will not be 2x (or more) faster than JS.

Why are you so sure of this? The people involved appear to be big believers that dynamic languages can be extremely performant provided they are designed with the VM in mind. LuaJit proves this out. A simple 1000x1000 matrix multiply in LuaJit is already 20x (not 2x!) faster than V8 and on par with C.

http://attractivechaos.wordpress.com/2011/01/23/amazed-by-lu...

http://lua-users.org/lists/lua-l/2009-06/msg00071.html

I think far too much emphasis is being placed on whether Dart replaces JS in the browser. Google would gain a huge amount by having a performant server-side language that can be tooled and compiled to JS for the client.

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#115

Post by mraleph » Fri, Sep 16, 2011, 7:51 AM UTC

equark wrote: post #114

KamilTrebunia wrote: post #108

Dart will not be 2x (or more) faster than JS. Dart will not be so much more expressive and elegant than JS is. Google would have to replace the whole stack (from HTTP to DOM to Dart to IDEs) to make it really compelling, but I'm pretty sure they will give it a shot. Fortunately there is a competition: Nobody will dedicate himself to Chrome Web Store and Google Web technologies and miss out on future Facebook WebApp S…

> Dart will not be 2x (or more) faster than JS. Why are you so sure of this? The people involved appear to be big believers that dynamic languages can be extremely performant provided they are designed with the VM in mind. LuaJit proves this out. A simple 1000x1000 matrix multiply in LuaJit is already 20x (not 2x!) faster than V8 and on par with C. http://attractivechaos.wordpress.com/2011/01/23/amazed-by-lu... http:…

> A simple 1000x1000 matrix multiply in LuaJit is already 20x (not 2x!) faster than V8 and on par with C.

You are using outdated measurements. Up to date results:

http://attractivechaos.github.com/plb/plb-lang.png

http://attractivechaos.github.com/plb/

The difference between V8 and LuaJIT is nowhere near 2x on matmul and comes from differences in language features (no debugger interruption support for generated code in LuaJIT2) and lack of certain optimizations (e.g. array bounds check elimination) in V8 [matmul_v2.lua used for measurements relies on ffi arrays which do not include bounds checks at all]. Code generated by both compilers is roughly equivalent (invariants hosted out of the loop, temporary double values are on xmm registers, etc). A bit more details: http://blog.mrale.ph/post/5436474765/dangers-of-cross-langua...

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#116

Post by mmullany » Fri, Sep 16, 2011, 11:10 PM UTC

BrendanEich wrote: post #106

foofighter2 wrote: post #105

Earlier quoted context omitted.

I think you are being unfair in your criticism. Microsoft, Apple, and Adobe, just to name a few, have all tried to force web developers into proprietary closed-source platforms. If you want to get angry, get angry at them! Google is just trying to give web developers a choice. They plan to continue fully supporting ECMA's efforts as well, with developers and money. And for this they deserve to be attacked? You keep t…

"If you want to get angry, get angry at them!" What makes you think I have not been "angry" at MS, Apple, and Adobe? You must not follow my writings closely! What's more, my mood is not the point, actual market-facing behavior is. The topic here is Google and its actions. There are good reasons to focus on the big G now: * Microsoft is older, formidable but late to mobile, in some ways in decline. * Apple, we know wh…

Just a small historical revision - Netscape was losing a point of market share per month in mid 1997 - even before IE4 came out, because of Microsoft enforcing IE as the OEM browser on most desktop PCs

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#117

Post by BrendanEich » Sat, Sep 17, 2011, 2:29 AM UTC

mmullany wrote: post #116

BrendanEich wrote: post #106

Earlier quoted context omitted.

"If you want to get angry, get angry at them!" What makes you think I have not been "angry" at MS, Apple, and Adobe? You must not follow my writings closely! What's more, my mood is not the point, actual market-facing behavior is. The topic here is Google and its actions. There are good reasons to focus on the big G now: * Microsoft is older, formidable but late to mobile, in some ways in decline. * Apple, we know wh…

Just a small historical revision - Netscape was losing a point of market share per month in mid 1997 - even before IE4 came out, because of Microsoft enforcing IE as the OEM browser on most desktop PCs

Right, Win95 bundling. How could I forget?

I cited some market share links from wikipedia elsewhere in this thread, and TechnicalBonobo set me straight on monopoly vs. (let's say) dominant competitor or market leader (whatever that term is).

The point for Dart (vs. JS) is that Google doesn't have 80% or even 50% share.

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#118

Post by gruseom » Sat, Sep 17, 2011, 4:48 PM UTC

bzbarsky wrote: post #110

gruseom wrote: post #80

Earlier quoted context omitted.

"So you would be OK with requiring everyone to use Windows as their operating system, in perpetuity?" Of course I don't think that. I meant by "ubiquitous" to imply the opposite. I fear we're talking past each other. The risks of Google behaving like a Microsoft-style monopolist here are low because (a) they have no monopoly in browsers, VMs, or OSes to abuse, (b) their culture is the most hacker-driven of any large…

> I meant by "ubiquitous" to imply the opposite. In that case, I have no idea how you think ActiveX could ever have become "ubiquitous". > they have no monopoly in browsers, VMs, or OSes to abuse They have a monopoly in search engines (and some would argue webmail clients), and are trying hard to work on browsers and OSes, especially on mobile... Give them a few more years, and see how things look. > their interests…

You know what, scratch everything I said about ActiveX. I didn't think about it very much (it was so awful, I can't bear to), and you're probably right.

Really? To my mind the industry is more hacker-centric than it used to be. For example, there's a spectrum of how open and sharing Google may turn out to be with Dart. Some points on that spectrum are better than others, but nowhere on it is the possibility of an old-school closed-source implementation. That's a major change from how things used to be.

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#119

Post by BrendanEich » Sat, Sep 17, 2011, 6:13 PM UTC

gruseom wrote: post #118

bzbarsky wrote: post #110

Earlier quoted context omitted.

> I meant by "ubiquitous" to imply the opposite. In that case, I have no idea how you think ActiveX could ever have become "ubiquitous". > they have no monopoly in browsers, VMs, or OSes to abuse They have a monopoly in search engines (and some would argue webmail clients), and are trying hard to work on browsers and OSes, especially on mobile... Give them a few more years, and see how things look. > their interests…

You know what, scratch everything I said about ActiveX. I didn't think about it very much (it was so awful, I can't bear to), and you're probably right. Really? To my mind the industry is more hacker-centric than it used to be. For example, there's a spectrum of how open and sharing Google may turn out to be with Dart. Some points on that spectrum are better than others, but nowhere on it is the possibility of an old…

You're right, open source and the Web have both done a lot to take down old-school, straight-ahead proprietary ploys such as closed source, market-power-based de-facto standards that competitors have to reverse-engineer at very high cost. Even on Windows (itself still closed source).

The two technical/cultural shifts, open source and the open web, make for a good trend.

That's why counter-trend action such as delayed-open (Dart), delayed/partly-open (Android, e.g., but other examples are easy to find) raise hackers' hackles. At least for some hackers.

And hackers aside, the competing vendors meeting in existing standards bodies get left out. That clouds the prospects for future standardization, unless (again) based on market power. Which is not there, if the topic is Google Dart (and it is :-|).

Re: Brendan Eich on Ecma TC39, JavaScript.next, and Google Dash

#120

Post by equark » Sat, Sep 17, 2011, 7:05 PM UTC

mraleph wrote: post #115

equark wrote: post #114

Earlier quoted context omitted.

> Dart will not be 2x (or more) faster than JS. Why are you so sure of this? The people involved appear to be big believers that dynamic languages can be extremely performant provided they are designed with the VM in mind. LuaJit proves this out. A simple 1000x1000 matrix multiply in LuaJit is already 20x (not 2x!) faster than V8 and on par with C. http://attractivechaos.wordpress.com/2011/01/23/amazed-by-lu... http:…

> A simple 1000x1000 matrix multiply in LuaJit is already 20x (not 2x!) faster than V8 and on par with C. You are using outdated measurements. Up to date results: http://attractivechaos.github.com/plb/plb-lang.png http://attractivechaos.github.com/plb/ The difference between V8 and LuaJIT is nowhere near 2x on matmul and comes from differences in language features (no debugger interruption support for generated code…

Interesting, thanks. Do you have a guess on what the Dart means by "Dash is designed with performance characteristics in mind, so that it is possible to create VMs that do not have the performance problems that all EcmaScript VMs must have." Where will a redesign likely see gains?