Eyes Above The Waves

is it that simple? It's not that simple for me. For me, I actually think Firefox is a great browser. It works well for certain things. Such as Google Hangouts on Mac. Firefox simply runs hangouts without any hiccups, while I often have problems with hangouts on Chrome. Use firefox because it's a good piece of software. Underrated, in my opinion.

what about my gmail?

Yeah. I use GMail too. Fortunately, switching browsers is easier than switching hosted services ... for now ... and I think it also has more impact.

There's always jekyll/octopress (on github pages). or heroku. or openshift. of self-hosting it. There are tons of options that don't rely on google that are free or very cheap.

There's also Wordpress.com. While it isn't self-hosting, it enables you to self-host later on. I believe there's a way to import Blogger data into it, though I don't remember how.

http://en.support.wordpress.com/import/coming-from-blogger/ :)

Weebly is awesome for blog and is a free

Weebly is limited and unprofessional - as is Blogger. It's fine for hyper personal opining but for a more polished, professional credible statement, definitely a self hosted blog on a commercial domain/host is the way to go. It's sad but true - doing business on the web is a huge industry now, even for blogging. To have a weebly.something site sends a message or impression that the owner is too lazy or too cheap to learn how to do something technical or spend 15 bucks on a proper host. That does matter to other people you may want to take you more seriously, even if it doesn't matter to you as the owner. That said, I'm about to leave Chrome for good. I'm setting up Firefox right now and then transferring everything over. I've run into one deal breaker - my bank. I can no longer access it with Chrome. They changed their format and apparently Chrome is incompatible. They suggested IE. I laughed and laughed and laughed. Then downloaded Firefox ;p

Google controls Chromium, and using it contributes to Google's control of the platform almost as much as using Chrome does. Ditto Opera.

What do you mean Opera? Is it Google-controlled?

Opera's browser is based on the Blink engine. Blink is controlled by Google.

But it is in turn a fork of WebKit, which is technically controlled by Apple, which is in turn a fork of KHTML, which is the fine work of the KDE folks.

Fork doesn't mean what you think it means.

With no argument supporting your claim you might as well put a sarcasm tag at the end of your post.

"Should HAVE".

We totally need a sarcasm tag. And a dislike tag. And a slew of other ones as pure standard. Wonder if Firefox devs can get on that and be the leader in something else fun.

That's a fair comment and I strongly agree that privacy is a key way for us to differentiate ourselves. We have put a lot of effort into privacy-related initiatives over the years --- DNT, Lightbeam, Cookie Clearinghouse, restricting fingerprinting via careful Web API design, etc. They've generally been coopted by our competitors, subverted, blocked, or just ineffective. But we haven't stopped trying, and there's another major initiative getting ready to ship soon. Unfortunately, a lot of the ideas bandied about by privacy activists aren't practical for the mass market, so the problems are generally harder to solve than they think. But we're always open to new ideas, especially if people are willing to help implement them.

Lightbeam is cool, but it is really only a gadget that's not even integrated into Firefox by default. Your last paragraph is IMO the line of thinking that hurts Mozilla at the moment: "ideas ... aren't practical for the mass market". That's what Google can't get away from because of what it fundamentally is, but Mozilla can and should. Your trying to appeal to Internet users who care about their privacy, not the mass market. That said, I realize its easier said than done, and one must be practical about how that's being approached. Yet, following planet.mozilla.org and the development of Firefox via mozilla-central pushlog for ages, I feel Mozilla is not being daring enough on that front, and rather seem to want to embrace the status quo. Web standards and superior technology is what got you to take over Internet Explorer back then. I remember Mozilla taking on some pretty hard choice vis-a-vis standards vs compatibility. You ignored the status quo and pushed forward, and you won the users. This time around, the debate is around privacy, not superior technology, and being shy and/or going on with an established status quo needs to stop :)

The problem is that we can't restrict ourselves to users who are willing to make significant sacrifices for improved privacy --- which is what a lot of privacy approaches (e.g. Tor browser) currently require. That's too small a market to be viable (I think). By viable, I mean enough market share that Web devs test their sites in your browser. The minimum to shoot for would be that we're as good as other browsers but with better privacy. I think that's reachable.

Right. Tech-wise, nowadays, Chrome and Firefox are as a whole pretty equal. So yeah, differentiating yourself with privacy is the way to go. And right now, you seem to have both a perception and feature gab to sell yourself as superior. When people talk of a Firefox vs. Chrome privacy comparison, its mostly superficial, revolving around "Google is evil data gatherer, Mozilla is not". That's not sufficient. There needs to be specific _default_ behaviors & features that can be marketed as tangible arguments you guys are superior.

The comment by David (14:52) is actually a pretty good example of what I was referring in the last post as "superficial" :) What does Firefox do and offer that's different from Google on that front? That's what needs to be coded/tweaked, shown up in tech news coverage, on mozilla.org's front page, in the release notes, etc.

Apple increasingly emphasizes privacy and users owning their own data, so that seems to indicate those themes have a mass audience.

I'm an evangelical Christian, I'm very open about it, and most people at Mozilla know it. That viewpoint is disliked by many people at Mozilla, yet here I am after 15 years. Brendan leaving was very very bad. I was very close to him and I was gutted by it. But in fact everyone at Mozilla I know (which is a lot of people!) wanted him to stay. The primary blame for his departure falls on gay marriage activists, not Mozilla.

Thanks, I greatly appreciate your take on the situation, but regardless, as a complete outsider, as a 60s/70s Jewish liberal I take very seriously the threat to speech, to society and to IT that the social justice warrior set poses for us.

The primary blame for his departure falls on gay marriage activists, not Mozilla. So, who runs Mozilla, again?

Same for me. No more Firefox or Mozilla.

What do you mean, boycotting Mozilla because you want tolerance for opinions [they] dislike? I regret the fact that Brendan was pushed into quitting, and yet I'm in favour of gay marriage. I believe that the people who raised the hue and cry against him hadn't known him as well as they ought to have in his capacity as “Mr. JavaScript” and as whatever he was at Mozilla. When he said he left his political opinions on the clotheshanger with his raincoat, I believed him. Robert here is an evangelical Christian; Gerv has a blog titled “Hacking for Christ”; so what? I believe that, according to Jesus of Nazareth, prayers (and religious opinions) are best kept to oneself, as a secret between you and “your Father who sees into the secret”; but their religious opinions, however far from mine they be, don't detract in any way from my admiration to them in their respective capacities at Mozilla. I still believe in the Mozilla Mission, even if I am not sure that Firefox best embodies it. I'm certainly not going to leave a Mozilla product for a Google product, and for my first smartphone (I don't yet have one), I won't buy an Android: I'll rather wait till Firefox OS comes to some place where I can lay hands on it.

Same here, I uninstalled Firefox completely from my systems, I'm sorry, but no Brendan = No Firefox for me.

So Mozilla is crawling with fundies? Another reason to dump Firefox aside from its awful performance.

" The primary blame for his departure falls on gay marriage activists" That includes me and every good person I know.

Anonymous says, with javascript still enabled. People who protested Brandon or Brendon or whatever his name was are uninformed reactionaries. They boycott Firefox but blissfully use all of his other technology that would just be too inconvenient to do without. Sure, they have principles, but let's not get too carried away....

Microsoft and IBM dominated the enterprise. How's that working out for them? Much as I'd like to have resources to pour into enterprise markets, that's where good software goes to die.

"Oh, I use Chrome at the office, so I'll use that at home because that what I know".

Come on roc, you know what I'm talking about and it's not Microsoft/IBM stuff. Google Chrome is making huge strides in the enterprise space. They are taking schools away from Apple. They are making serious headway into government. It's because of one reason - it's easy to manage and deploy. Mozilla has completely missed the boat on that. Managing Firefox is an absolute nightmare.

You must run Windows, if it's that much of a nightmare to maintain. I have maintained several networks. I stick with ESR releases, push the image out when they are updated, and the users never know the difference.

I remember those days. Dark times.

I hope you can see the difference between "one web browser" and "one house". One is a physical structure, and one is essentially an API. A better analogy would be power outlets. All houses in my country have the same kind of power outlet. That makes it easier for engineers. It's hard to imagine how anybody would be better off if power outlets *weren't* standardized on a single form. Yes, there are web standards, but the assumption of this blog post is that web browsers all follow them differently, so they're not "standards" in the same sense that electrical standards are. In the physical world, there are lots of de facto standards in the world, and they work just fine. Bicycle chains seem to be doing OK even though there's no ISO standard that says how big to make them.

Multiple browsers could share a standard api. Standards is what Mozilla is pushing for and standards is what Google dislikes. So if you want one standard api you should be on muzillas side.

Multiple browsers can theoretically share a standard API, but a browser monopoly *necessarily* has a single API to work with. I wouldn't call it a *strong* argument, but dealing with incompatibility is a downside of a competitive browser market.

Avoiding bad futures requires prognostication, which necessarily implies uncertainty. I tried to avoid FUD by mentioning specific examples of Google leveraging lock-in across products. But thanks for your kind words.

Actually, anyone who is paying even a modicum of attention knows this isn't FUD or bullshit. But go on pretending you know better and telling others to shut up because they aren't toeing the party line. When alternative browsers that don't have massive corporate funding are no longer an option, you'll eat your own words, but there are many of us who can see the forest for the trees, just like we did when Microsoft dominated the webspace. The only difference is that Google WANTS to dominate the webspace.

This viewpoint makes no sense whatsoever. Any semi-decent "software engineer" won't be solving the same problem five times if they're writing software for the modern web. Stop pretending otherwise. Would life also be better if there was only one car brand and vendor? Having one engine would only discourage competition and cause stagnation, like it did with Webkit (which is precisely the excuse let Google spin off Blink).

And if you don't like the Chrome interface I guess a person would just have to suck it.

I guess I've never met a semi-decent software engineer, then. This problem is exactly why there are sites like "caniuse.com". Are you one? Can you introduce me to one? I'd love to get the chance to interview one. The problems with saying "for the modern web" are first, that significant numbers of people in the world aren't using modern web browsers (or proxies), and second, that what constitutes the "modern web" is constantly changing. CSS 2 was once cutting-edge, and now it's pretty evenly supported everywhere. The Audio API is still cutting-edge, and inconsistently supported by browsers. What stagnation did Webkit see before Google forked Blink, exactly? As a web developer, it always seemed the most advanced to me. Lots of standards today started as Webkit extensions. It's been groundbreaking in standards compliance, too: Safari was the first browser to pass Acid2. Please stop with the car analogy. It's not the same at all. I guess you think bytes are like gasoline (what's diesel?), and Starbucks is like a gas station (except it's entirely free). There is literally no useful conclusion I can think of that derives from the "web browsers are like cars" analogy.

Obviously op k owe nothing about web design or software engineering

In addition to the difficulty of building a web app that runs on browsers that boast different feature sets, the development teams working on the browsers themselves are also redundant. For every modern browser there is an engineering team that implemented the html5 spec. Those engineers could have been working on improvements for existing browsers, cleaning up the code base, writing more test cases, etc. For the non-technical populace, more browsers means more choices in an area where it takes substantial knowledge to understand the differences between 'products'. The best kind of software solves a problem so thoroughly that you never have to think about it again. I'd rather have one perfect browser that makes web design and experience easy and fun than many browsers which creates confusion, complexity, and redundancy.

People said similar things when Microsoft took over from IBM. And the issue isn't only whether the monopolist continues to innovate or not. Another issue is whether their power suppresses innovation by others. (That was one of the real problems with Microsoft.) An even bigger issue is whether a single organization can be trusted with that much power; I have a dim view of human nature and believe it cannot.

They are no longer innovating as they used to. They have thinned down product lines, reined in their RnD divisions, and no longer want to compete, but control. You can call this sentiment FUD if you'd like, but the only reason the web is growing right now is because Google are being kept partly in check by competitors. They're showing the same signs of stagnation and anti-competitive behavior that Microsoft showed back when it nearly won the browser wars.

Nice observation. It's where observation matches theory: http://www.amazon.com/Wealth-Poverty-Edition-Twenty-First-Century/dp/1596988096/ref=sr_1_3?ie=UTF8&qid=1408201108&sr=8-3&keywords=George+Gilder

That's a good point. I agree.

If you dismissed every person in the world for holding some sort of ridiculous or indefensible idea, you would be left sitting on a desert island all by yourself...

https://yourlogicalfallacyis.com/ad-hominem

I'm not sure what I mean or don't mean by God, but… Please forgive me for ignoring the opinion of an adult who refuses to read anything written about the Internet by authors who happen to believe in God. In my European country I've met many intolerant atheists and many intolerant Christians. I've also met many of both who were respectful of other people's opinions. Belief or non-belief in God are no criteria for being put on my "ignore" list. Rigid intolerance is.

Typical liberal who preaches tolerance but refuses to practice it themselves. Probably voted for Obama.

Well said, Tony.

It'll be great if it's eventually fixed. The question is why it's Chrome-only in the first place. We're seeing something similar with Hangouts. Hangouts was a plugin ... but now you can run it in Chrome without a plugin. Oh wait, it requires NaCl ... but don't worry, they're working on fixing that. Why not just write it the cross-browser way in the first place, or work out the standards extensions up front if you need them? There's a definite approach of "get it working in Chrome now, get it working in other browsers eventually ... maybe." Don't get me wrong, that's totally understandable and not necessarily malicious. But it is very dangerous for a competitive Web.

NaCl is only used for input preprocessing (AKA mustaches), an area that is not yet fully speced, and also performance-critical. Aside from that feature, Hangouts works just fine without NaCl. That said, I can understand asking why Hangouts doesn't work in Firefox. In short, we wanted to transition to WebRTC sooner rather than later, and at the moment there are things holding us back on both our side (e.g. upgrading our ICE implementation) and the Firefox side (e.g. supporting multiple video streams). But overall, surely you're not arguing that transitioning a major Google application from a proprietary plugin to an open web standard somehow demonstrates that Chrome doesn't value web standards.

What standards extensions do you need for "mustaches" and where are they being worked on? > surely you're not arguing that transitioning a major Google application from a proprietary plugin to an open web standard somehow demonstrates that Chrome doesn't value web standards. No, that's not what I'm arguing. I thought what I wrote was pretty clear. Thanks for the information though. I hope it's clear from my original post but the Chrome team, or at least the Blink team, are generally the good guys in my book.

It's not yet clear whether it can be performant enough, given our CPU budget, but canvas + your captureStream ideas (https://dvcs.w3.org/hg/audio/raw-file/tip/streams/StreamProcessing.html#media-element-extensions) might solve this as well as several other interesting problems.

Can you start a thread on public-media-capture or some other list and CC me? My blog isn't the right place to work this out :-). Thanks!

@Jake This answer wasn’t funny already three years ago … https://matej.ceplovi.cz/blog/we-should-stop-even-pretending-google-is-trying-to-do-the-right-thingtm.html

See my first sentence.

Also, do you really think people at Mozilla are sitting around hoping "we're the good guys" is a winning strategy? Of course not. "We have to be the good guys who make things people really want to use" --- everyone understands that. Or almost everyone. I'm not sure why you'd think otherwise. Maybe it's the JS team... Anyway, none of that contradicts what I wrote in my post.

While I've never heard of another one before literally today, I don't think this is the same David Anderson formerly working on JS/e10s. Just today there was a comment on Gerv's blog, also from "David Anderson". While that comment isn't obviously consistent or inconsistent with anything I'd expect from dvander, the URL on the comment certainly is -- taken in light of the URL mentioned on dvander's Twitter page, and in light of its pointing at a non-functional domain. And while I can imagine (but have no particular reason to believe) dvander might concur in some elements of the argument made here (I would myself, for that matter!), at an absolute minimum I don't think he would have expressed them in a manner so reminiscent of a circular firing squad. And no, I don't think the position expressed here is consistent with that of people working on JS, for one. Or at least not fully consistent. I believe we all agree that principles have an important place, and ultimately they're what we're working for. Yet I think things like ES6 feature work, all the debugger and developer tool improvements, games-related work, and so on demonstrate our concomitant interest in also "making things people really want to use", as a necessary prerequisite to implementing those principles.

Thanks for the reply - I didn't meant to imply that people were sitting around. Rather... I can't fault people for not wanting to use Firefox. If someone wants to use Chrome, maybe it is actually better for them. Maybe they really like it. It's hard for me to say "you must use Firefox" when I truly don't know whether they should. I'm not even sure how I could know. The mission is a vague document with vast scope. It means many different things to many people, and that's partly why it resonates so strongly with Mozillians (me included!). My eternal worry is not lounging around, but relying too much on interpretation of the mission to drive the effort.

Man you're living in the past. Firefox has had non-XUL plugins for years, and many popular addons have made the switch. Also, why use Chromium if you're concerned about Chrome? They're the same underlying engine. It's still basically the same browser.

Also for the last couple of years, Mozilla has not been paid by Google (but by Yahoo!).

Google is not an investor in Mozilla. We have a contract where they pay us to deliver search traffic to them. Google offers some services to us that are good for the Web and good for our users, so we use them. Google offers other things that we don't believe are good for the Web or our users --- e.g. WebP, Dart, PNaCl --- and we don't use those; instead we try to come up with better alternatives (mozjpeg, Daala, asm.js).

Perhaps it is no 100% direct influence, but you can not deny that Google offers you "solutions and help" and for many of these suggestions you are thankful and implement them. As i wrote at the bottom, stop accepting and implementing this suggestions, go back to the core what you have been in the past before Google gave you money.. A browser from Geeks, for Geeks!

I wish Mozilla would embrace Dart. It's a great piece of technology, that solves a lot of the crap that is JS. My company is developing a _very_ large web-application using Dart, and it runs *perfectly* in both Firefox and Chrome, which are the targetted platforms for the application. Developing in Dart is a much more pleasant and, dare I say it, mature experience, compared to the utter mess that is JS. If Mozilla embrace Dart, I'm certain the dart2js transpiler will become even better. Please don't dismiss this great language due to a NIH complex.

Dart is part of the problem, Thomas. For all its present flaws, asm.js is the superior solution. It gives browser vendors a reasonable upgrade path, doesn't require a second entire VM to have performance parity, and doesn't lock us into one bloody language that isn't even all that much better than Harmony (which Google is falling behind on in their mad quest to upsell Dart). Frankly speaking, Dart is a dead end already. If one can compile it down to Javascript, then there's no excuse to not improve JS instead of their futile attempts to replace the web stack with tech no one else wants to adopt, like NaCl and Dart.

Accessibility and helping people feel safe online are great areas to focus on. Many people still struggle to understand how to use the web and struggle to understand when visiting certain sites exposes risks for identity/monetary theft.

+1

+1

I moved almost completely away from Firefox because for me Chrome is a much better browser. And I think Google is doing a good job of moving the web forward. I do not think that Mozilla is making good choices by deciding not to sopport WebP, Dart, PNaCl.

WebSQL fails because it ties itself to an existing codebase (a very specific build of sqlite) instead of to a specification.

Truth, not having webSQL puts Firefox at a massive disadvantage to Chrome

Anyway as long as you try to be like Google i only can recommend all people i know to use a different browser. You Mozila guys have practically abandoned your state as Geek browser and have got involved with Google in the hope that their design and limitations of customization - what brings them a massive success - will bring you too a boost. What happens? Your share shrinks, you get desperate! Simple solution, Make the Chrome like parts optional - bring back customization, accept finally that being a so called Geek browser was what kept you healthy! Do yourself a favor and get sane again Mozilla!

SeaMonkey and Palemoon both are Gecko browser. Measured separately, each of them has only a negligible market share. With “inclusive” counting, they contribute to Mozilla's market share. So far so good. What do you mean with “having Google on board”? The default search engine? You don't have to keep it. I have so many search engines installed (all, or almost all, from addons.mozilla.org and mycroft.mozdev.org) that my search pulldown is maybe four times my screen height; yet the only ones I use really often are the English Wikipedia, the Wikipedia in my native language, and the Mozillazine knowledge base. In the rare case where Google would be a better fit than these three, I have Bing, DuckDuckGo, and a few others, which allow me not to be dependent on Google…

They may not be Google, but Mozilla is more then enough influenced by them, so calling against Google is equal with calling for suicide help. And yes, if the Mozilla devs would still go a real independent way, things would be much more brighter as they are right now.... It is just so sad

The entire purpose of this post is to rally us in anticipation of Google's funding review this fall. It's a power ploy with us as the pawns. They are going to use us as bargaining chips to keep Google on board with them. All non-IE users are anti-Microsoft, but that's the only thing they have in common. They diverge into pro-Google and anti-Google camps, and Google realizes this. What Google really wants is to transform anti-Microsoft sentiment into a bi-polar business axis a la the left and right wings in politics. Two major browsers, two bases of uncontested power. The purpose of Mozilla in Google's scheme is to depotentiate the one thing Google fears most: the potential of a truly free and open market, that thing which Firefox (but not necessarily Mozilla) symbolizes. Google will prop Mozilla up for a while longer as they continue the process of consolidating their power, lest those users break for Apple. Make Firefox more and more annoying, more developer and user unfriendly, less and less appealing compared to shiny, polished Chrome. The whole intention is to marginalize the people who can really make a difference. That's what this is all about.

+1

It's worse -- Mozilla is not wholly owned, or else where are my RSUs so I can retire young like my ex-Netscape friends who went to Google in 2003-2004? Mozilla isn't quite tame, it may bite Google now and then, but not hard. So mostly a neutered, not-wholly-owned, not well-fed lapdog.

I've addressed this elsewhere in these comments already but to make it easier to find: PNaCl, Dart and WebP are three examples of technologies that Google would dearly love Mozilla to support, and we have refused to do so because we don't think they're good for the Web. And we're not just exterting stop energy, we're building better alternatives.

... so this idea that Mozilla is Google's lapdog is absurd.

I know this is a troll, but in case it's not obvious, I spend somewhat less than 1% of my time writing blog posts and most of the rest working day and night to improve Firefox.

Only from your perspective. Your partnership with Google has tarnished your product. Next time Buddy Holly sees fit to lecture HTML 5 fans on the importance of high security around CANVAS, remind him that those vulnerabilities didn't even exist until the high res timer was introduced. ;) (That's Google saying, "gotcha!") Not that I really telling you this will make a difference... your logic is transparent: you're just on until Moz no longer has money, and then, you're gone. Pity that the greatest open source project in history is reduced to this... but that's what money does. Gee isn't that the reason why Moz was created as a volunteer project in the first place, to weaken the influence of dollars?

Dude, I don't want to see their innovations. There are builds around which strip out half their "innovations" and reduce the browser bloat by 2/3. Addons were innovative... Google App-ripoffs that can't access the local filesystem... aren't.

Thank you for posting this reply.

Amen... at least Google is working towards something better.

I couldn't care less if it's not as fast as Chrome when doing something with a third-party library. At least it still lets you access the local filesystem in principle.

The user experience is not important when a website use a third party library ? What is the difference between a third party library and a personal library ? In both case the code is written to run as fast as possible on all browsers. But maybe jQuery sucks and is not popular, so let's forget this example. I just don't understand why today Firefox is the only browser that does not support the property "min-height" on tables, see bug 307866, and bug 977929 for an example with jsfiddle, reported this year again. 9 years that it has been reported, and we still need to use a workaround *only* for Firefox. Even in the "Examples" section in mozilla documentation (https://developer.mozilla.org/en-US/docs/Web/CSS/min-height) it suggests since 2007 to use "min-height: 75%;" on a table! But it doesn't work only in Firefox (see http://jsfiddle.net/47n24eqk/). What irony, it's probably so obvious that is should work like other browsers.

Hi Syn, my name is Vlad, I work on the Firefox performance team. Could I ask you to link to a test case of some kind showing bad Firefox performance with jQuery animate() when resizing an image? I'd like to look into this. Thanks

Hello Vlad, it's not so easy to create a test page. I think that Firefox was lagging and it was visible thanks to jQuery animate. jQuery animate does not lag in an isolated case. When I was testing today, it lagged only sometime, I think that today the garbage collector was the source of the lag (it was a huge lag during 2 seconds). Few months ago, the lag was permanent in our application. I will try to integrate more components from our application to the test page in order to try to reproduce the permanent lag (the slow jquery animation) that was only in Firefox. I will come back to you when it's ok.

Thank you! I also tried reproducing the issue by using jQuery animate() to resize a PNG by manipulating its "width" property and I couldn't find any differences vs Chrome. GC could be to blame -- we're actually currently working on a project to prevent GC interfering with certain types of Firefox animations. Let me know what you find!

This is the very simple test page I used http://people.mozilla.org/~vdjeric/jquery_animate.html

Wow that lags a lot in Firefox and I am using an AVX build of Firefox. IE11 kills here.

You're using googles opposition of prop 8 as a negative? Why would I want to use a browser in the 21st century written and maintained by people still living in the 19th?

If Google is in favor of gay marriage or not is not the issue, the issue is that their CEO haven't been boycotted for having a different opinion, that's the problem with Mozilla, think different and you are out, Go an appeal to those Mozilla employees that attacked Brendan,they were loud attacking him but I don't see them promoting or defending Firefox with the same force. >Joshua McDonal, shut up please you ignorant biggot.

Joshua, I'm not taking a position on Proposition 8. I'm just confused when people against gay marriage boycott Mozilla and not Google.

> If you believe in Javascript as the best technical solution > - then make it win by making it better That's exactly what we are doing. Supporting Dart would force us to divert resources from improving JS. Why fork and have two engines when we can keep making one engine, the one that is standardized and already necessary to run the Web, better? > that means thinking about users and developers first and foremost. That's what we do! Don't mistake this blog post for any kind of significant investment.

Perhaps because the javascript engine is getting close to being as good as it can be, perhaps all the cruft in javascript isn't a stable foundation to build on for the next twenty years, perhaps because sometimes in order to move forward, you have to leave things behind. I think it's fine for Mozilla to make a choice not to work on it now - it may or may not prove itself to be a worthy addition or successor. But the try and block and belittle the experiment from people who are genuinely trying to make the web better is poor in my view. Dart wasn't the main thrust of the post - it was an example of behaviour I'm seeing from some Mozilla people which I don't like - seeing the worse rather than the best in people, being narrow minded rather than brave... Maybe that's unfair, I don't know but it's the perception I'm getting. I hope i'm wrong.

Anonymous, why don't you identify so we can talk on equal footing? I don't care whether you work for Google or just swallow their spin, I'm just looking for real names for a change. I've tangled with socks on HN enough already. Your comments here paint me as *the* big-bad, a role that all the SJWs want to cram me into since late March this year, so I'll bristle briefly. But before that, consider the unreality of your complaint: Big-Bad me kept poor widdle Google from using Mozilla on top of its own market power to do... what, exactly, pray tell? Ram DartVM as well as a JS VM into all browsers, starting with Firefox at very high opportunity cost to a Google competitor? And thereby compell Safari and IE to ram or reverse-engineer DartVM into their engines? You realize this entails a cycle collector or equivalent, something Mozilla has had for many years? See Chrome's OilPan. Big-bad me stopped all this, and poor widdle Google couldn't get DartVM rapidly adopted among well-funded competitors MS and Apple? Really? What rubbish. Concretely, I was negative about Dart when it "debuted" because it had leaked via this two-faced memo: http://markmail.org/thread/uro3jtoitlmq6x7t FYI, I was at the Nov 2010 Apple-hosted Ecma TC39 (JS) standards meeting where I saw Mark Miller turn white as a sheet because he had inadvertently posted this memo to a public Caja google group. He walked over to another Googler and stage-acting tapped him on the shoulder, and they departed for the hallway. Soon other Googlers were leaving the meeting to converse urgently in the hall. Some of us wondered: Has Sergey inadvertently launched the Singularity ahead of schedule? Has Larry canceled free lunches at Google? Only months later did I realize what had happened. The message had instantly forwarded through archives including markmail.org. Mark Miller was not the author of the "Dash memo", just the fall guy. He was skeptical of Dash from the beginning. Another friend of mine at Google apprised me of Dart early on, saying it would become a $100M crater. I was skeptical because I can smell Microsoft-ian spin a mile out, and you should too. The Dash memo was rife with conflicts of interest. JS "Javascript has fundamental flaws that cannot be fixed merely by evolving the language", so poor widdle Google had to do a new language and VM. This set up an obvious "starve V8" path that was reified soon by Google's "no remoties" rule, as Aarhus became at best an "on your own time" V8 shop while the new V8 team in Munich was staffed and set up to climb the learning curves. Double-talk against JS looks very bad in this light: if Google believed in ECMAScript Harmony, they could have invested in JS harder and avoided Dart, or at least made a better dart2js target language. JS could have had bignums by now! This is a real issue for Dart programmers. See https://code.google.com/p/dart/issues/detail?id=1533#c57 and prior comments. Dart2js forked Dart's numeric tower from DartVM, on purpose. It has been four years plus since "Dash" started in 2010 (or ealrier?!). Why has not Google worked diligently to get bignums into ES6? But I see you don't just try to smear me about Dart. I'm the Biggest Big-Bad, who kept poor widdle Google from forcing other stuff, presumably Native Client, down the throats of competing browser implementors. Again, what a crock! PNaCl is now marketed by Google with reference to Emscripten/asm.js and Pepper.js as a parallel second toolchain to use to target other browsers. Even this is failing, as Epic, Unity, and others all prefer one toolchain: Emscripten targeting asm.js. Google is simply failing in slow motion here, and blaming me is the just lame. (If Google wants me to help them standardize PNaCl or Dart, they know how to reach me. I'm unemployed, so it will cost them, and I make no guarantees. But I can do a much better job than they have so far.)

Part 2: The complaints about JS are mostly obvious and fair: no threads, no SIMD, too much mutability so it's hard to optimize. But the solutions are obvious and fair too: SIMD in ES7 (coming fast, thanks to Google's John McCutchan, Intel folks, and Mozilla); threads via confined-data-race array buffer shared among a new kind of web worker that has no DOM access; and ES6 and 7 features in JS that enable optimizations designed into Dart. Ultimately, by splitting JS via V8 from Dart via DartVM, Google perpetrated a doomed game theory setup: DartVM needs JS to stagnate or hit JIT-only walls, while other browser vendors open to AOT as well as JIT optimization, and open to JS evolution via the Harmony agenda, could move the world forward faster, via shorter evolutionary hops. This approach wins for PNaCl too, wherefore the referenced Emscripten/asm.js/Pepper.js path. My advice: stop blindly lashing out at people, and pay attention to the physics governing the bodies in motion. We're all in fields of force whose economics (or thermodynamics if you prefer -- same thing) that favor evolution of JS over "REPLACE" agendas with DartVM and PNaCl/Pepper. Anyway, I'm gone from Mozilla, and still no DartVM or Pepper support. Who to blame now? Talk to Apple and Microsoft people if you need more convincing. Or else take off the sock and come clean, and if you can stand the climb-down, join the Evolution. /be

Thanks for that. I'm sorry if you felt I was lashing about at people - not my intention. In fact it was kinda what I was complaining about - sorry if I inadvertently added to it. I know everybody is working really hard to make the web better, and in fact have done an incredible job. To illustrate what I meant - clearly you think Dart is a waste of talent and that could have better been spent on improving javascript. Time will tell if your right - but there's a difference between Google ( or anyone else ) lacking resources to push everything as fast a possible as everyone one like and there being some 'plot'. You could take exactly the same argument and turn it around - Dart would be so much further on if guys like your were helping. Of course you made the judgement that, that's not the right path - your call - just like it's their call. That's all. Making JS better isn't just about adding new features - in some peoples view it's also about throwing stuff away. This all started because Robert was making dark comments about the motives of large companies - all I was saying was - yes distrust large companies - but at the same time trust the engineers working at the coal face as honest. To reiterate the post wasn't really about Dart per se and certainly not about NaCl - it was about the tone of posts like Roberts. I guess that illustrates how hard it is to communicate your intent right with the written word!! If I put my name to it - you wouldn't recognise it - so there's no point. Maybe I don't like the google aspect of hoovering up all information - but that's a separate argument!

Google is evil

Not even Chrome supports Dart, why should Firefox? Another Anonymous.

"Ultimately, by splitting JS via V8 from Dart via DartVM, Google perpetrated a doomed game theory setup: DartVM needs JS to stagnate or hit JIT-only walls...." I forgot to add that this doesn't just discount JS evolution against other browsers' incentives, it also hobbles dart2js (lack of bignums, performance disparity), which could mean lack of Dart adoption as much as increased Chrome with DartVM adoption. The perf win would have to be 10x, not 10% and not 2x. Even then, Apple and MS would likely work on JS evolution to catch up. Betting against evolution on the web is high risk. I do not think Dart is a complete waste but it has taken years, missed multiple opportunities to co-evolve with web standards, and taken a big-bang integration approach viz landing in any browser. On the up side, it has a nice DOM API. W3C/WHATWG folks please take note. /be

The problem with the Dart project is that the core decisions re syntax and semantics were not done in the open but presented as a done deal. Why should other browser vendors buy into something so non-collegiate.

Anonymous 16 August 2014 22:22 wrote: "To illustrate what I meant - clearly you think Dart is a waste of talent and that could have better been spent on improving javascript." That's inaccurate. Dart may manifest great qualities due to talented people working on it, but (as I've written clearly here and on HN): 1. The effort on DartVM without matching work on JS is misspent because dart2js won't be compatible or fast enough, which limits Dart adoption without a Chrome near-monopoly. 2. The big-bang approach means no other browser implementor can afford scarce staff to hack in Dart and a cycle collector without Dart getting more market share. This cycles back to (1). Obviously if Google had enough market power, they could force DartVM adoption by other browser vendors, but that would require >80% or so, judging by IE vs. Netscape -- and even then, Netscape forced JS by being first. IE never forced JScript on any browser. "Time will tell if your right - but there's a difference between Google ( or anyone else ) lacking resources to push everything as fast a possible as everyone one like and there being some 'plot'." Who said there was some 'plot', as in a much-mocked (but proven all the time in criminal courts, and by historians) "conspiracy theory"? That's a badly-twisted bit of rhetoric when applied to my citing an actual instance of a big company-confidential plan (rather than plot) whose internal memo leaked. The Dash memo did leak, to Google's chagrin. You've read it, right? "You could take exactly the same argument and turn it around - Dart would be so much further on if guys like your were helping. Of course you made the judgement that, that's not the right path - your call - just like it's their call. That's all." This is false parallelism. Google has tons of money for boondoggles, and it wastes it on many trying to "see what sticks". Mozilla has little money by comparison, and too few heads to fool around with costly bets against evolution. If Mozilla were to do any work "on spec", when I was involved and even moreso now with smaller market share, it would be more leveraged and in-demand work. See https://brendaneich.com/brendaneich_content/uploads/Talk.029.png. "Making JS better isn't just about adding new features - in some peoples view it's also about throwing stuff away." No browser vendor, not even Google, can afford to break the Web. This means not only JS, but CSS, HTML, image formats, HTTP headers, and almost everything else grows by addition. The old bits may fall into disuse over decades, at which point browsers do remove support for them. Gopher's dead. "This all started because Robert was making dark comments about the motives of large companies - all I was saying was - yes distrust large companies - but at the same time trust the engineers working at the coal face as honest." I'm not sure what you mean (especially by "at the coal face"). Lots of talent is very well-compensated at Google to work on many things, and at the expense of other things (esp. true of Dart vs. JS due to the Aarhus colocation of talent and the no-remoties rule). That fact doesn't justify any particular choice, unless you are arguing "might makes right". /be

"IE never forced JScript on any browser." s/JScript/VBScript/ /be

Just clear up some misunderstanding first. > clearly you think Dart is a waste of talent I meant here - not that you thought it was technically poor - but that you thought the approach of trying to do something like dart was a waste of talent and better spent on evolving Javascript. Of course that may also be wrong :-) > Who said there was some 'plot', as in a much-mocked (but proven all the time in criminal courts, and by historians) "conspiracy theory"? That's a badly-twisted bit of rhetoric when applied to my citing an actual instance of a big company-confidential plan (rather than plot) whose internal memo leaked. I wasn't referring to anything *you* said - remember me original comment was to the article by Robert. Re: Dart - bottom line you either think Javascript engine will still be the only first class citizen for the web in 20, 40, 100 years time or not. If you think not - then how do you manage change? You have the classic choice - of software development - either add new stuff to the existing thing and deprecate old stuff. Or create a new version 2 in a different name space, leaving version 1 for compatibility. The problem with the first approach is that could take forever and it's already pretty complicated stuff without having to do the old and new in one code base. Google, it appears to me, are trying to take the second approach - though as you point out, they aren't as unencumbered by previous decisions in javascript as they would like because of the need for dart2js. I guess, originally they were hoping to get other browser vendors to allow alternative engines besides javascript. As an outsider - somebody who is just looking at having good developer tools - there is a golden opportunity for the community to leverage the talent assembled at Google and the apparent willingness to give it all away and together help build version 2. Yes they did it in secret to start with, ( yes i've read the dash memo ), yes I'm sure everyone developing version 2 would have made different choices and obviously someone is in charge saying yes or no ( you need that or else it would never get anywhere - concensus in language design??? ) , but once it broke cover they appear to me to be listening for constructive input. My guess is they choose the approach they did because back when they started speed was of the essence - I suspect it wasn't a project born out of fear of Mozilla or hatred particular of JS - but of a fear of Apple - back when they started mobile web was looking like it might have a limited future, looks a little better now ( thanks ironically to Google ), but still worrying if you are an open web person.

> 1. The effort on DartVM without matching work on JS is misspent because dart2js won't be compatible or fast enough, which limits Dart adoption without a Chrome near-monopoly. That depends on developers making a choice. Developer adoption depends on a whole raft of things - as you pointed out yourself - there's a new DOM api etc. Sensible developers choose platforms not languages - note people chose the web - not javascript! Your right about number mismatch being a challenge - but I thought John McCutchan offered to help? > 2. The big-bang approach means no other browser implementor can afford scarce staff to hack in Dart and a cycle collector without Dart getting more market share. This cycles back to (1). As you say developers will decide. So to be clear, if developers clearly loved Dart would you help implement it in Firefox ( you say it's already ahead in terms of managing references )? > I'm not sure what you mean (especially by "at the coal face"). All I was saying was that whether one agrees with the technical choice or not, that when dealing with engineers working on things like Dart it's better to treat them as if they are like 99.9% of everybody - trying to do the best job they do with the best of intentions. ( While always remembering that large companies on the other hand tend to somehow always act in a way most people wouldn't ). Just to be clear, this wasn't aimed at you - it was a general comment around the discussion that Robert started. To sum up - and maybe I'm naive - but I think you, Robert and the people at Google ( whether dart lovers or haters ) are all honestly trying to do the best they can. To quote the meat of Roberts original post: >The potential for lock-in is vast and they're already exploiting it, for example by restricting certain Google Docs features (e.g. offline support) to Chrome users, and by writing contracts with Android OEMs forcing them to make Chrome the default browser. Other bad things are happening that I can't even talk about. Individual people and groups want to do the right thing but the corporation routes around them. I'm suggesting the best way to defeat the lawyers and bean counters in the big companies is not to have those individuals and groups fighting between themselves but working together. For example, if Dart became popular, the way to stop it becoming a tool that Google could wield against other vendors would ironically be to adopt it in a way that meant Google had no advantage!! That choice would be a question of timing ( as it may never get to that point ) - but if you want to be Machiavellian about it, positioning yourself where that would be easy rather than hard if the time came ( ideally getting Google to do the work for you ) would be the ideal.

"... but that you thought the approach of trying to do something like dart was a waste of talent and better spent on evolving Javascript. Of course that may also be wrong :-)" You keep revising what you wrote, with changes of meaning. Stabilize, please. I've been clear: see my points (1) and (2). Dart and JS should have co-evolved more incrementally, as TypeScript is doing. Then there wouldn't be that bad bug of bignums in DartVM yielding observably different numeric results from what dart2js gives. Co-evolution does not mean "no Dart". It would mean no silly stealth mode starting in 1H2010 if not earlier, no "REPLACE" agenda in the leaked memo, no refusal to engage with TC39 and the V8 team (which had to be recreated in Munich), etc. etc. Both Dart and JS would be better off, and Dart would be closer to being supported (whether by compiling to JS or to a new VM) in other browsers. "I wasn't referring to anything *you* said - remember me original comment was to the article by Robert." You were replying to me, though. Please mind the context. Roc didn't allege any plot, but I'll say here what I said on twitter: I have multiple reliable sources attesting that Google is optimizing its web properties, and (sometimes) testing them, and (less often; this part is public) developing parts of them (NaCl-based plugins), *only* for Chrome. This guarantees other browsers perform poorly by comparison. "Re: Dart - bottom line you either think Javascript engine will still be the only first class citizen for the web in 20, 40, 100 years time or not. If you think not - then how do you manage change?" My blog and twitter cover this frequently: through evolutionary adaptation and growth. Web compat is required by all browsers trying to keep or grow market share, so JS adds to fill semantic gaps and increase usability where syntax isn't sweet enough (until the http://sweetjs.org macro-pocalypse). In the very long run (10 years is enough; 100? Don't expect any of this to be around), JS may mutate into the end-state of being a common language transport format and runtime for the Web. That's my goal, shared by many on TC39. "You have the classic choice - of software development - either add new stuff to the existing thing and deprecate old stuff. Or create a new version 2 in a different name space, leaving version 1 for compatibility." Please catch up at http://esdiscuss.org/, search for "1JS". "The problem with the first approach is that could take forever and it's already pretty complicated stuff without having to do the old and new in one code base." It's not that complicated. V8 is drowning in conflicting JIT optimizations, but other engines are using AOT techniques based on type systems (asm.js; others are possible; SpiderMonkey does a hybrid/online type inference too). Google hasn't gone any faster by doing two things instead of one, especially when the second thing (DartVM) requires a compile-to-JS workalike (dart2js). Again see my point (1): for adoption, dart2js quasi-parity is required. See the complaints in that dart issue from users who will not use Dart until bignums (Dart int) work via dart2js. That requires bignums in JS, which are on the Harmony agenda, but to which Google contributed zero work till John McCutchan chipped in to help SIMD and other "value types". [continued in next combox]

"I guess, originally they were hoping to get other browser vendors to allow alternative engines besides javascript." In a pig's eye! They knew odds were extremely long of DartVM and something like Gecko's cycle collector or chromium's OilPan being created in Safari and IE. Huge taxes self-levied by competitors, just because Dart looked hawt? Never gonna happen! Were they trying to pressure other browser implementors indirectly, by putting their adopters in pain via dart2js, so those adopters would complain to Apple, Microsoft, and Mozilla to adopt DartVM? Again that was extremely unlikely to work, and in fact it has failed. "... there is a golden opportunity for the community to leverage the talent assembled at Google and the apparent willingness to give it all away and together help build version 2." Version 2 of what? The Web? The Web isn't owned by anyone. Google cannot do a better "Web version 2" in a closed environment (as Dart was designed and mostly implemented in). They'll fail if they try. Arguably they are failing already with Dart and (P)NaCl. Jury's out on SWs and WCs, but they look oversized and they do not seem to be converging quickly enough. "... ( you need that or else it would never get anywhere - concensus in language design??? )" Read about the champions model TC39 uses, observe other analogues in WHATWG. No one does design by committee on the Web. "but once it broke cover they appear to me to be listening for constructive input." Not really: look at that bignum issue. It was up to others to go work on bignums for JS. "Not my department" or worse: pressure browser implementors indirectly via forked dart2js semantics hurting early Dart adopters. Big-picture, Dart leads started with "REPLACE" and, after the leak-botched launch and years passing, ended up saying they <3 JS, but it's clear they don't. Yet Google is only to happy to keep funding. My friend inside who predicted a $100M crater may prove a prophet. >60 people work on Dart (>100 by now, I think). That's a huge cost! "I suspect it wasn't a project born out of fear of Mozilla" Certainly not. " or hatred particular of JS" Au contraire -- Lars Bak and others hate JS with the fire of 1000 suns. "but of a fear of Apple" As many noted when the Dash memo leaked, JS was far from the worst problem the Web had in competing with iOS native. With asm.js, it's even farther from the top hit-list, which leads off with things like custom-view 60fps scrolling, as roc blogged about the other month. I have no idea why you keep ascribing too-pure intentions to Dart principals. Do you know more than others? I don't judge intentions, and the road to hell is paved with good ones. Look at effects, results. Judge those against designed and expressed intent. /be

From your reply Anonymous 17 August 2014 11:47: "Your right about number mismatch being a challenge - but I thought John McCutchan offered to help?" He did, after getting involved in SIMD for JS based on his Dart work, after I tweeted at him. Good for John, pretty bad for whoever came before him. Again, 1H2010 was 4.5 years ago! "So to be clear, if developers clearly loved Dart would you help implement it in Firefox ( you say it's already ahead in terms of managing references )?" I'm not working on Firefox code, or at Mozilla -- you may have heard. Arguments about what developers love are both fruitless (opinions vary) and easy to bend via money. Google can do a push-poll showing Dart is in demand. Let's see dart2js usage rising first. "... trying to do the best job they do with the best of intentions." Intentions are not enough, and everyone thinks theirs are good to great. "To sum up - and maybe I'm naive - but I think you, Robert and the people at Google ( whether dart lovers or haters ) are all honestly trying to do the best they can." This seriously misstates the market power and wealth-effect compensation power that Google has, compared to Mozilla. "I'm suggesting the best way to defeat the lawyers and bean counters in the big companies is not to have those individuals and groups fighting between themselves but working together." You can't get rainbows and unicorns among hundreds of people at four companies, sorry. Competition and carefully championed innovations that aim to be standardized incrementally are the only way to win. "For example, if Dart became popular, the way to stop it becoming a tool that Google could wield against other vendors would ironically be to adopt it in a way that meant Google had no advantage!!" This is very naive. Google has dominated Dart from the beginning, and while they'd surely share power in the Ecma TC52 group once they'd won the market, at that point the latecomers are all arguing about the window dressing. Or as Rutherford said, "All science is either physics or stamp collecting". Dart's physics was created by a few gods at Google, and vanquished competitors won't stamp-collect over much that matters "( ideally getting Google to do the work for you ) would be the ideal." This is naive too. No free lunch, even if Google adds mythical-man-month "free" labor to jam DartVM into other open source browsers. Chrome will always be first and best. The jamming will make bug farms with direct and opportunity costs on the competitors. No thanks! Really, there's zero demand for DartVM anyway. This is all a sideshow. 60fps custom view scrolling, more complete (Pepper.js without stubs) asm.js, WebGL2 and 3 (ARB_compute_shader) -- these are some of the prizes to keep your eyes on. /be

>>"... but that you thought the approach of trying to do something like dart was a waste of talent and better spent on evolving Javascript. Of course that may also be wrong :-)" >> You keep revising what you wrote, with changes of meaning. Stabilize, please. Really? You know better than me, what I meant by that? You might want to reflect on that for a moment on how that comes across. > "For example, if Dart became popular, the way to stop it becoming a tool that Google could wield against other vendors would ironically be to adopt it in a way that meant Google had no advantage!!" > This is very naive. Google has dominated Dart from the beginning, and while they'd surely share power in the Ecma TC52 group once they'd won the market, at that point the latecomers are all arguing about the window dressing. If that were true then why doesn't Mozilla still kick everyones ass at Javascript.... ? As you alluded to above it's all a matter of investment - while Mozilla is admittedly at a disadvantage to Google, MS and Apple - each of those can choose to compete on implementations if they wish to. If you are refering to design decisions already been made for better or worse, leaving no space for intellectual input - that's true - but doesn't that happen with everything? I suspect there are lots of people that wish they had been their at the start of JS - some of which weren't born at the time! I think we are almost getting to the crux here - do we have a battle of ideas for the future amongst the elite in a group ( however constructed ) or do we have a battle of ideas in the market? Back to Roberts original comment about the importance of choice. >Really, there's zero demand for DartVM anyway. This is all a sideshow. 60fps custom view scrolling, more complete (Pepper.js without stubs) asm.js, WebGL2 and 3 (ARB_compute_shader) -- these are some of the prizes to keep your eyes on. If you really think the attraction of Dart is equivalent to benefits that asm.js & WebGL bring ( good that they are ) then I'm not sure you really understand the mass of developers out there .. and if that's the case .. how can you be so sure you are making the right decision for them?

Corrections to last comment: * I meant "underinvesting" not "malinvesting" before "in it [V8] and TC39". * I was CTO since 2005, legit typo. /be

[repeating last one-line graf:] So what? That [JS getting "on first" in 1995] happened, and we can't recall it. I wish I could! But don't pick on JS alone. The HTML, CSS, and JS first-mover advantages on the Web led to terabytes of content that represent a lot of human capital. This capital, and more to come soundly based on it, constitutes the real value on which to build, not "REPLACE", the Web. The world-wide sustained technical value of the Web as a shared artifact and asset is the reason all of us work on JS, not because we love it and think it's superior to any particular wannabe. (Wannabe replacements have a lot to learn from JS as it has been embedded and developed by its users, though. Dart is needed becamse gmail loads too much JS at once, so VM snapshot undump is the fix? LOL. Try shrinking gmail's bloat first.) You seem to think it's Google's turn, as if to say: "Why not? Labor is free, has no direct or opportunity costs, adding 2nd VMs with complex inter-heap cycle collectors doesn't make bug farms or cost otherwise." Sorry, you sound like a Google sock again. If you're arguing that *because* Google, for path-dependent reasons similar to JS's getting on first and dominating, has enough money to fund a new language, *even if* Mozilla doesn't, therefore Mozilla should jump on the 2nd bandwagon, then I reply: Even if the costs were bearable (they're not), the demand for Dart _per se_ is too scattered and weak, a non-issue to most developers. Browser vendors evidently agree (I'm not working for one atm). It's not Google's turn just because they can afford Dart, sorry. They don't have Netscape's 80+% share, so they can't force the issue, either. OTOH, 60fps custom-view scrolling, now that's a hot topic in the Web vs. Native marathon! [continued in next comment]

[Continued] > There is no other realistic path to Dart adoption than ongoing and much better-supported-by-Google JS uplift. I know many Googlers who concede this. How about you? Well there is the path you pointed out before - Dart VM only in Chrome with adoption there. I think adoption is much more likely than you think for two reasons: 1. Not all development is done on the public web. Pay for web apps, or inhouse apps care less about the widest possible compatibility and more about either ultimate performance and/or cost to develop and maintain. 2. Your under estimate the attraction of Dart for developers - I know Google made a big splash about performance at the start, but it really isn't about that - it's the software development experience ( I can't see debugging stuff compiled down to asm.js being great without a lot of work on extra tools, for example ). > Sorry, you sound like a Google sock again. Not - just an end user developer. I'm going to stop here ( but feel free to reply - I will read ) - because I originally started this because I wanted to make the point that, while you shouldn't trust big corps ( you never know who is going to be in charge tomorrow ), that shouldn't lead to second guessing everybody motives as Machiavellian. As I keep getting accused of being that myself I can see it's not reducing the emotion. If you read the rest of the comments to the whole post I'd agree with one thing above all others - privacy, privacy, privacy is Mozilla's opportunity to create a strong niche for itself. Brendan - I have enormous respect for your work - and don't worry you won't of heard of mine - different field altogether.

> So in the end, it doesn't matter so much that X number of specific people designed Dart, what matters is, is it good? Now I completely accept it's more *likely* to be better with more eyes and vigourous debate, but to use an argument around people not being involved isn't a stand alone argument - what matters is the output. There's been a developer mindshare shift to functional/immutable/type inference/async/reactive/AOT. See Clojurescript, elm, mori etc. ES6/7 seem more aware of this shift despite the cruft baggage. Whereas Dart seems to have a very OO perspective. And the attractive part of dyanmic OO is obscured by Dart's syntax choices. So if developers aren't biting it's not because of Mozilla/MS/Apple.

"what matters is, is it good?" Dart's fine -- although young and undertested -- but we've lots of good languages. Again you seem to think there's some rational and fairly abstract process for debating and agreeing on languages to burn into browser multi-VM silicon. There is no such process! Compile-to-JS (more below) is the likely evolutionary selection filter for evolution of many languages and of JS. "Now I completely accept it's more *likely* to be better with more eyes and vigourous debate, but to use an argument around people not being involved isn't a stand alone argument - what matters is the output. Though their approach is obviously not good politics." You can't separate the two, any more than we could have just evaluated "output" apart from politics and control of Microsoft's VBScript or ActiveX back in 1998-2003 at Mozilla. The better way to separate tech from politics is to find the common gaps in JS that dart2js needs. That exercise, even if it resulted in the conclusion that JS could not be evolved, would bear fruit. But it's obvious that JS *can* be evolved -- it is being evolved via ES6 and 7 right now. So the likely fruit of Google realism over idealism is better JS, and that contradicts the crappy politics on parade in the Dash memo. "Note I didn't know whether you were using that argument or not - which is why I said 'if you mean this then...'." Noted, thanks. "Don't think it's their turn - just think that's it's a honest technical approach to trying to improve the web." "Honesty" like "good intentions" is self-attributed all the time. Let's not talk about the duplicity of the secret Dash memo, the conflict between investing in Dart and in JS (V8/TC39), the hand on the very real scale of what talent at Google is paid to work on, and where. I've beaten that to death (yet you still seem not to see it). Let's talk about "improve the web". Is doing a big new codebase, with new requirements (inter-VM cycle collection, imposing new write barrier costs), really going to "improve the web"? It might not even "improve Chrome", due to opportunity costs and new bug habitat, but suppose it does. Then there's a long march through the market and through other browsers' codebases, ultimately winning Dart adoption (if not DartVM -- perhaps we'd see multiple implementations). How is this an improvement to the Web in the foreseeable future -- even the next decade? It's not. It is a giant bet by Google, bootstrapped on bogus assertions such as "JS can't be fixed", mainly because Google could afford it and the principals wanted it. My two-year old wants things all the time, things that I can afford. Doesn't mean she gets them! Think about what's good for the Web in less abstract, over-the-horizon, all-or-nothing terms. Otherwise I argue you're doing something other than "good". "You go on about the costs and opportunity costs of Dart - of course it has costs - that applies to everything!" You are not responding to my very concrete argument that for Dart to be adopted in any browser market not about to be cornered by Chrome, dart2js must pretty-much match DartVM. Therefore JS must improve. The all-or-nothing, decades-long nature of the Dart bet makes it a Web-standards "social ill" in my view. Please respond, stop rehashing my own mention of opportunity cost or flattening it via abstraction as if to say all trade-offs are the same. [continued next comment]

"So you are saying JS is like IPv4 and Dart is like IPv6? Or have I got that wrong - is the DOM actually the fixed interface that hard to evolve?" JS is part of the evo-kernel, very hard to replace and not likely to be supplemented by 2nd, 3rd, Nth VMs-in-browsers in the next decade. JS and DOM are entangled of course, so I'm not sure it matters. Both are in the kernel. DOM has lots of bad old APIs that some people keep imitating when adding new APIs, and I see soi-disant designers on the case who really are not skilled at language or API design, who cut their teeth paving cowpaths (spec'ing de-facto-standard APIs -- quite different from good de-novo API design). [Brendan wrote > Nice try changing the subject. In referencing 60fps, etc., I was talking about making needed in-browser ("Kernel") investments, without which 60fps isn't happening -- even for Dart in Chrome.] "Lost me there" Please reread. I was replying to the bogus Dart-because-iOS-Native-kills-Web interpretation of the Dash memo as reason for Dart. I wrote that there's zero demand for DartVM per se and it wouldn't help real Web vs. Native problems. iOS native can do 60fps custom view scrolling if you are willing to write Obj-C and use IOKit. But there's no rocket science there that the Web could not add to fill its kernel gap. "so these are investments outside either JS or Dart?" Yup. "I'd imagine the possibilities are almost endless there - not so much me changing the subject as me not noticing that you had." No, you swerved back to your assumption, that DartVM in particular achieves some important good in the near term for the Web. It does not. The subject, which I addressed directly, was what Web needs to compete with Native. "And to do that don't you need to throw away stuff as well as add?" See my first reply where I wrote "No browser vendor, not even Google, can afford to break the Web". Why are you again going in circles? Dart does not enable Google to throw away JS in any foreseeable future. Do you disagree? "Well there is the path you pointed out before - Dart VM only in Chrome with adoption there." DartVM isn't even in Chrome yet, so you have only "hopium" fueling the belief that it would somehow increase Chrome's share. "I think adoption is much more likely than you think for two reasons: ..." Show me the numbers. Dart is not even on Tiobe's Enterprise-skewed language index: http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html Are you using Dart in a firewalled setting? Your point (2) about ease of development is fair, but TypeScript and many other compile-to-JS languages win there too. Why is Dart the sole solution you're pushing in all these comment-replies? "As I keep getting accused of being that myself I can see it's not reducing the emotion." Your very lengthy replies assume some kind of "fair play" model that doesn't exist, ignore Google's own intentions in the Dash memo that have been demonstrated by underinvesting in JS to the point of the starved bignum bug, etc. etc. You push Dart only, ignoring TypeScript from Microsoft. Why is that? In short, your comments show a pro-Dart, pro-Google bias. Perhaps unconscious; we all suffer unconscious biases. I try to surface mine and make sure they're conclusions, post-judices rather than prejudices. I'm not against either Dart or Google. Google's too big to be all-for or all-against. I work well with more than a few Googlers on standards of common internest. By now it's clear Dart won't "REPLACE" JS, so debating not just what's the end-game but how to get there from here, and revisiting underinvestment in JS seem appropriate. /be

Hi Anonymous, I'm karl, I'm working on Web Compatibility issues, aka Web sites targeting some browsers more than others and how it impacts users in their daily usage. About what you said on services being developed only by chrome. It's a bit more complex than that. But there is a big load of sites which could work with Firefox on Mobile and do not because of development choices. This is just some examples but it might help you to have a better understanding of what is happening. Because everything is public. # Not receiving tier1 * Gmail (FxOS, FxAnd) https://bugzilla.mozilla.org/show_bug.cgi?id=668275 * Google Calendar (FxOS, FxAnd) https://bugzilla.mozilla.org/show_bug.cgi?id=914252 * Google Docs/Drive (FxAnd) https://bugzilla.mozilla.org/show_bug.cgi?id=754754 * Google search cards (FxAnd, FxOS) https://bugzilla.mozilla.org/show_bug.cgi?id=975444 * Google search advanced features (FxAnd, FxOS) https://bugzilla.mozilla.org/show_bug.cgi?id=1047854 * Google search images - no infinite loading (FxAnd, FxOS) https://bugzilla.mozilla.org/show_bug.cgi?id=921532 * Google search images on tablet - carrousel (FxOS) https://bugzilla.mozilla.org/show_bug.cgi?id=793216 * Google Top Menu - basically everywhere (FxOS, FxAnd) https://bugzilla.mozilla.org/show_bug.cgi?id=920466 https://bugzilla.mozilla.org/show_bug.cgi?id=755590 * Google News (FxOS) https://bugzilla.mozilla.org/show_bug.cgi?id=865043 * Google Finance (FxOS, FxAnd) https://bugzilla.mozilla.org/show_bug.cgi?id=679025 # receiving desktop while chrome/safari receives mobile * Google Books (FxOS, FxAnd) https://bugzilla.mozilla.org/show_bug.cgi?id=754752 * Google Docs/Drive (FxOS) https://bugzilla.mozilla.org/show_bug.cgi?id=1053648

Look at microsoft google fight over youtube for windows phone and you'll know what lock in is. What google is doing right now is a long con for PR. They will lock in as soon as they get the chance.

Thanks. What platform are you on? I thought some form of print preview existed on all platforms. Windows and Linux have built-in print preview, and on Mac print-preview launches the PDF viewer.

Mac. I use preview to select only relevant pages for print. Mac workflow for this is too long: print to pdf, switch to pdf viewer, print again, select relevant pages. Chrome prints look better to me, they use smaller fonts, less paper: http://totic.org/chromeprint.pdf http://totic.org/ffprint.pdf You should not be the one worrying about this. Hope there a guy at Moz whose job title is "Beat Chrome", and who tracks all the reasons people switch.

Do you have "ignore scaling and shrink to fit page width" set in Firefox? That looks like the difference to me.

Wow. This has bugged me for years, and I've never noticed the checkbox. Thanks for starting this thread. I hope FF thrives.

Glad I saw this. (FWIW I'm sympathetic to the anxieties being expressed about future lock-in.) Chrome's excellent Print preview has been the sticking point for me as far as going to FF fulltime. Printing two sheets of paper when the second page is just some footer garbage is just spectacularly irritating. This is a very solvable problem I'm on a Mac, and FF has wasted a couple medium sized trees for me over the years. Chrome's print preview was a godsend of sorts when it debuted. Still, today I don't see a print-preview in FF when I command-P. I just checked FF Nightly now to make sure I'm not behind on something. What are you referring to exactly? Do you mean, from the FF print modal, to select PDF>Open PDF in Preview? This works well, but I agree with Aleks that exiting the browser just to see a print preview feels like an extra step. I don't see what "ignore scaling" has to do with this. Sorry, this isn't your area, and I don't mean to stray, but am I missing something?

Sure, google has better services, absolutely. But just becaues our ovrelord is being nice to us now so that he can convert us doesn't mean he wouldn't turn bad. (And he would because end of the day, both MS and google are public firms created to give revenue to investors (same for the most part). So avoid them now, Use alternatives wherever possible.

No they didn't. MS was all over them, and they didn't have a monopoly in any of the areas where disruption is difficult. Now they do, they have Android and chrome now and as more people invest in their platforms the more powerful their monopoly becomes. Btw, there is no such thing as google and Microsoft, it's all Wall Street honchos who control them, so if you think google is diferent from MS, you're wrong.

One important reason why some things works better in proprietary browsers is that the content is implemented in the same proprietary technologies. If I deliver the content my way, I'll have it easy tailoring my browser to a perfect fit. Google and Microsoft can dictate standards through the share force of their market share, just by using their tech in all their content, and they can exclude competitors by proividing poor APIs and lacking documentation. This way, guess who has a harder time bringing proprietary content to the user. Don't you see this is how lock-in w works?

fastmail.fm (and even better is the paid-for version).