Angular Announces AngularDart
news.dartlang.org
news.dartlang.org
However, the hyperbole surrounding anything that even remotely, several years down the line, may threaten the monopoly of JavaScript on the client side web is ridiculous. Language competition is good - it ultimately will make for a better JavaScript.
Personally, I think there are a lot of people deeply invested in Javascript who fear anything that even remotely, one day in the super future, may threaten their skill set. That's understandable - it's the same way MS developers scoffed at learning Objective-C when the iOS ecosystem blew up and the Windows monopoly didn't matter as much.
Let's inject some reality - Dart is no threat to JavaScript for the foreseeable future. AngularJS is not going anywhere. Dart can't be forced down your throat, because no consumer browser even supports it natively yet. You don't need to use it, you don't even need to think about it. Let the people that like it be.
I don't have a problem with that, I just think you're loudly complaining about how loud everyone is. ;)
Hint: you're not in a library, you're in a discussion thread.
Pretty much, I think a large swathe of people who jumped on the Angular JS platform are quite fond of just that -- jumping on the latest technology. They've evangelized for it, knocked on peoples' doors Saturday mornings, wanted to tell them about the great new wonderful framework that will save the world. And now the fear is not that Angular JS will go way. It won't. It will be there sitting on Github accepting pull requests, forks will appear and so on. They fear that there will be something else, newer out there and they will be stuck with 2012 technology. They will be hipsters who are still drinking Pabst Blue Ribbon while all other cooler hipsters are drinking Tecate already.
C# is technically a superior language to Java. Combined with an excellent IDE (VS), the .NET platform is probably technically superior to the Java platform. However, .NET did not destroy Java. And it's because Java is superior to .NET in many non-technical areas (such as a better open source community). And other solutions appeared to help the Java platform with the language stagnation issue (Scala, Clojure, Groovy, etc.). Now Java itself is somewhat catching up (Java 8) and the JVM platform remains a powerful and thriving community.
The situation with Dart and JS seems to be similar. The JS language has stagnated for a long time. Google is trying to be like MS with C# and created Dart to fully replace JS. And Dart is probably technically superior to JS. But just like Java vs C#, I don't think technical superiority will be enough.
The JS platform is addressing the language issue with CoffeeScript, TypeScript, and others, just like Scala and Clojure appeared for the JVM. And JS itself is improving slowly (ECMAScript 6) just like Java 8. And the JS open source community is very very strong (all the web related libraries and frameworks and now additional strength from Node.JS) and it's unlikely Dart will have a comparable community. I think that Dart may achieve some success like .NET/C# did. However, I don't think it will destroy JS like Google hopes it will. The JS platform is just like the JVM platform. Has so many flaws, but is good enough.
Java and C# are both well designed server side languages. C# has the simple advantage of being younger and learn from Java in the real world.
JavaScript in the real world is not even close to Dart. JavaScript is currently trying to solve problems Java never had to begin with.
In almost every category Dart is technically superior. Shouldn't be a surprise, they took their time and learned from more than a decade of dealing with JavaScript. They are also not done yet, they are planning to inject binary code directly into memory.
Just look at Lars Bak's history, he is the right guy to create the Java/JavaScript successor for the web.
Sadly, this is the nature of so-called "Open Source" at Google.
Also, since dart compiles to JS, it could be no back-porting will be necessary.
Secondly, Google is a big company and some teams within it wanted to use Dart, whereas otheres want to use vanila javascript and Angular, so now you have both.
Thirdly, Angular is in the process of being broken apart to modular components for 2.0, it also now cross-pollinates with another Google-y called Web Components.
In future, the equivelent of directives (Your own <x-thingy> tags) will be a native web standard, Mozilla is also working on it.
This all amounts to a sort of polyfill to the future and most of the core of Angular will be thrown away if things work out. You won't need it.
These are all good things. I don't understand your need for FUD.
Edit: [source & details redacted]
This 'FUD' comes directly from the Angular core team.
Edit: still too many details.
Specifically who?
Google employs a LOT of developers. A lot of those developers feel the same way members of HN does, that we don't need a language to replace javascript. Google is using their weight in the web world to try and get a new/alternative language added.
As long as Dart isn't completely shoved down our throats, I'm willing to give it a try. I think there is enough opposition within Google to make sure that Dart will at least have to live in peaceful co-existence with Javascript.
Google is pretty realistic that it'll be many more years before they can even have a chance of getting other browsers to add the DartVM. Heck, Chrome doesn't have it yet (outside of the Dartium build which has some fun memory leaks).
I don't think it was obvious. It may be true but they were uncontroversial comments on a controversial topic. The language and DartVM aren't ready for Chrome, right now. You'd hear the same from the Dart team.
> we don't need a language to replace javascript
Absolutely... in the same way we didn't need Python, Ruby, C++, Java, Clojure, Haskell, etc. In fact, I'd say we needed those less considering the enormous choice on non-web platforms.
j/k. I doubt even Microsoft would want to bring that abomination back.
Or would they??
As a fairly knowledgeable user of Angular & sometime contributor, this is a disturbing rumor.
Dart encourages a very standard frontend: HTML, CSS and JavaScript, so I don't get that point. Dart compiles to JS like Haxe, TypeScript, ClojureScript, CoffeeScript, etc.
I'm not sure what "rumor" you're referring to. AngularDart is a thing and it reached beta today. That's not a rumor. AngularJS 1.2 is due out soon, and AngularJS 2.0 will be heavily influenced by the work done on AngularDart to modularize and use more web standards. That's also not a rumor.
I don't understand why did you put "Open Source" in quotes there.
Did Google take Angular of the internet? Did they not put AngularJS out to the world, open sourced? Did they forbid you from forking Angular, working on it, and create a refuge for all "abandoned" developers?
Yay... I was hoping this would happen :-)
> This is part of Google's strategy to push Dart in the browser.
I know, great, right.
> Sadly, this is the nature of so-called "Open Source" at Google.
I criticize Google often but yeah this is great. Good for Google. It is more than what Oracle or Microsoft or others do.
My opinion is this -- JS is a shitty language. It was a shitty language, no matter how many "good parts" books will be published or how many frameworks one builds on top of it -- it is still an ugly language that makes me want to tear my hair out every time I use it. You love it, great. I hate it. I was hoping from day one someone, some company, would find a good replacement. I can tolerate compiling to it and treating it like assembler. It is even better if it gets bypassed altogether.
There were other posts here about JS's warts, so I am not the only one who welcome any other sane, normal, language to replace it.
Anyway, great work Google, keep releasing new versions of Dart, Dart UI, Dart IDE and so on.
That won't happen. With the amount of JavaScript on the Web, JS is immortal. All you can do is to add another language alongside it.
> There were other posts here about JS's warts, so I am not the only one who welcome any other sane, normal, language to replace it.
But it won't be replaced.
One thing that would help is a couple of killer websites or applications that would do something much better than the rest. If it used Dart and ran on the latest Chrome, that would help its adoption.
It would help Google too because it would increase Chrome install instances running.
What's up with those scare quotes? Angular is open source (MIT), not "open source".
That is the only explanation I can come up with. Scouring the comments filled with hate and vitriol I couldn't find much substance as to exactly why the are upset. There is no indication Google is going to sue them for patent infringement for using AngularJS, they are not closing the project. It is open source, if they love and think it is great, they are free to fork it and contribute to it.
Also, Chrome apps seem to be a great use for this even if no other browsers embed a dart VM (if it gets to that). I'm just not seeing why this is so bad
One thing I noticed though: transpiling a .dart file with just the import statement for AngularDart (import 'package:angular/angular.dart') generates a non-minified 5MB .js file with almost 150 kloc. Every code change will create a new version of it, invalidating previously cached versions. :/
Tracking the size of AngularDart: https://github.com/jbdeboer/angular-dart-js-size
pubspec.yaml:
name: test1 dependencies: browser: any angular: git: https://github.com/angular/angular.dart.git
test1.dart:
import 'package:angular/angular.dart'; main() => ngBootstrap();
dart2js launch => builds test1.dart.js, about 5MB, 142,749 loc.
You should get ~173KB gzipped, minified dart2js output.
Whoever thinks this is part of some grand plan to force Dart down our throats clearly has no clue whatsoever about how Google works.
Standard disclaimer: I'm a Google engineer and these opinions are my own and don't reflect those of my employer.
Google is a really large company with >10,000 engineers. A lot of what we do starts as grassroots efforts. Honestly, getting 10,000 engineers to act in a coordinated fashion is probably like trying to herd 10,000 cats.
AngularJS started as a side project that became Google-sponsored and it grew organically. Being Google-sponsored simply means some team somewhere has dedicated some engineers to working on it. That's it.
Likewise a group of engineers separately started Dart due to dissatisfaction with JavaScript being the lingua franca of the Web. The idea was that Dart can be implemented as a VM within a browser and get faster performance that way. Likewise, it should compile to JavaScript for browsers that don't support Dart in the browser (which is all of them except Chrome). Also, Dart shares a lot of the philosophies of GWT in sharing code between client and server.
Now my personal view is that I honestly don't see the point of Dart. I mean no disrespect to the team or their efforts but I don't agree with the premise or the design philosophy of Dart.
Optional typing seems to me to combine the worst of both worlds. If typing is useful (as the static typers will argue) then be statically typed. If it's not be dynamically typed. Dart's type system is basically just the Closure JsDoc system used by the Closure compiler. That's it. To me, there's no real innovation there.
Dart is really just another Coffeescript and I don't see the point of Coffeescript either.
One of the benefits of development in JavaScript is you can simply refresh your browser and it just works. I'm a huge fan of this fast iteration cycle in development. It's the same reason I'm a fan of Python, Ruby and even PHP.
Additionally, as soon as you add a translation step from something to JavaScript, you introduce potential bugs in that translation (eg I've seen bugs with the Closure compiler incorrectly removed code it thought was dead and that was JS to JS let alone the issues with, say, GWT's Java to JS). Even if there aren't bugs, tracing issues back to the source becomes more difficult.
Rob Pike through Go seems to believe this too (not that I claim to speak for him) because he's repeatedly mentioned that the simple syntax of Go and fast compilation is a deliberate feature, a philosophy I agree with 100%. It's also one reason I detest C++: slow compilation kills productivity (IMHO).
So AngularDart is nothing more than another group of engineers (perhaps there's a crossover with the Dart folks) wants a modern Web framework for client-side Dart so they created one.
The idea that this is part of some grand strategy is quite frankly ludicrous.
Google puts their effort behind a lot of experimental projects. Some work out, many don't. Will Dart succeed? I don't presume to know. But while I don't agree with the premise/philosophy it makes me happy that the company I work for is still prepared to experiment with such efforts and devote resources to them.
If you work at Google you'll often see three teams working on roughly the same thing, typically without knowing about the existence of the others. They may consolidate or compete (if they have incompatible philosophies that the teams feel strongly about) or something in between.
This happens because so many efforts just start out as a single engineer or a small group of engineers just trying something out or they want to solve a problem they see (and in an organization this large it's rare you're the only one who sees a particular problem and wants to do something about it).
So, please, can we discuss this on its merits rather than the new norm of seeing vast conspiracies or simply seeing things as some collusion with the NSA. It's really tiresome.
EDIT: added a paragraph about the issues of bugs and translating something to javaScript.
But this is HN, it is easier to rant about how evil Dart Wader will destroy the beautiful Javascriptalderaan.
It's being designed as a javascript replacement. Google's overbearing intention is obvious.
You're right: Dart is, functionally, another CoffeeScript, but that's despite Google's plan--ok, 'experiment'--not by design.
>So, please, can we discuss this on its merits
Your criticisms are shared. That's what causes the friction: resentment that a pretty flawed language is being pushed. Dart does take time away from other more worthwhile projects.
None of the cynicism is over wishing that Google would stop experimenting with technology, it's over Google's waste of time.
Edit: Actually, apparently there might be a formalized 'grand plan,' reading nateabele.
Open? Dart has been open-sourced for 2 years now - only having its proof-of-concept done in-house before the announcement.
Cheerful? Not sure how one language that compiles to JS is more "cheerful" than an other, but we're really a cheerful bunch in person. I'm sure you could have a beer with us if you'd like :)
As to the your and the OP's point about Dart being another CoffeeScript, this is really far from reality. CoffeeScript is just syntax changes, and that's not sufficient for fixing JavaScript major problems - which come form its semantics, not it's syntax.
Dart and CoffeeScript are actually quite far apart on the spectrum of compile-to-JS languages: Dart tries to keep a similar syntax and fix the semantics, CoffeeScript tries to keep the semantics and "fix" the syntax (I personally find CoffeeScript very hard to read). Claiming that Dart is another CoffeeScript, to me, shows a real lack of knowledge about one or both of the languages.
You might not be involved enough in the politics of programmer mindshare, but plainly: Dart is attempting a takeover.
Takeovers can be smiles and open source, but they're still takeovers, and this one is doomed to fail.
If you don't see that, you're focused on the work in front of you and (again) not on the politics of technology. Which is fine. I'm sure it would be lovely to have a beer with you, and talk languages and compiler design well into the next day's hangover.
But if you're going to get involved with a conversation on politics, you should think about Google's goals with Dart and whether they're reasonable or beneficial.
General sentiment here? They're neither.
I joined the Dart team because I thought is was a worthy and important project that could beneficially impact a very large number of people. I _do_ think about these things, they're quite important to me. The goals of more sane semantics, more optimizable structure, and better toolability and tools are to me, and a lot of other engineers, incredibly reasonable and beneficial. That's actually the entire point of the project is to be more reasonable and beneficial than the alternatives :)
If fact, I think I'll use that as my informal Dart tagline: "Dart, reasonable and beneficial!"
More seriously, Dart is created by people who think that JS is broken and we need an alternative. I hope you're not claiming that that's really an unreasonable opinion, and that the only reasonable point of view is that JavaScript should always be the only language used to develop web applications. I'm sure reasonable people can disagree about the degree of brokenness of JavaScript, and about different solutions to the problems.
Developers aren't flocking to Dart. Many are resentful of it.
You can ignore the resentment and the reasoning behind it if you want, but that won't make Dart more reasonable or beneficial.
In all of your condescending[1] comments, you have forgotten to actually articulate what your reasoning is. These are as close as you got:
Dart doesn't have the open, cheerful intentions
Google's overbearing intention is obvious
flawed language
this one is doomed to fail
forced down our throat
---
I think my reasoning throughout this page has been more or less articulated, but I suppose I might try for a compact representation.
What exactly do you think is going to happen with Dart?
Is Google going to add Dart to every Chrome install? Sure, maybe. Will so many developers cry out that Safari and IE will also be forced to implement Dart runtimes?
Is every single browser going to run Dart because Dart will have made such a compelling case that it and it alone is preferable to javascript?
Why not python? Why not C, Java, Scala? We could be running those languages in browsers just as easily as we might run Dart.
Do you think Dart is that amazing? Do you think Dart is just so wonderful that it'll unite the programming world as never before?
No.
Even Google's pet framework is divided and recoiling.
The problem isn't that Dart has to compete with javascript, it's that it has to compete with every other language out there, and it's just not that special. Why Dart? If we're going to be serious about ripping up browsers, why Dart?
Now, I maybe should make it more clear: I don't have any problem with Dart as an experiment, Dart as a transpiled language, or even Angular.dart.
But Google doesn't get to decide the future of the web just because it has a bunch of fancy language designers. Google is trying to shove Dart down our throats, and it's not working, and it's not going to work.
Dart will remain a transpiled-to-js language for a time, and then fall off the radar when it's clear that it doesn't transpile to js very well. The alternative is a fantasy.
> Do you think Dart is that amazing? Do you think Dart is just so wonderful that it'll unite the programming world as never before?
The world isn't black and white. Dart doesn't have to be perfect to be a viable alternative to JS.
> Even Google's pet framework is divided and recoiling.
No, they both thrive and benefit from the other. It is a feature. The same thing is happening with Polymer.
> Why Dart?
Take a cursory look and you'll find some arguments for Dart but I would think you would have covered this before having such a strong opinion.
> Google is trying to shove Dart down our throats
Said multiple times. Substantiate please.
> fall off the radar when it's clear that it doesn't transpile to js very well. The alternative is a fantasy.
So you know more about dart2js capabilities than the creators of V8 and Dart? This is the condescending and arrogant attitude I spoke of.
I was echoing OP (the one who was bemoaning snark and cynicism). Read his paragraphs. Flawed doesn't mean crippled. There are things about Dart I like.
A language can't be perfect.
>No, they both thrive and benefit from the other.
Pay attention to negative feedback. Some substantial portion of AngularJS devs are unhappy.
>Take a cursory look and you'll find some arguments for Dart
I'm not arguing for Dart's existence here, I'm saying Dart has absolutely no chance of securing the browser's runtime over javascript. Why Dart, why not Go?
>So you know more about dart2js capabilities than the creators of V8 and Dart?
https://github.com/jbdeboer/angular-dart-js-size/blob/master...
173KB. This is ridiculous.
Say whatever you want about benchmarks, Angular.dart is either a terrible port (quite possible) or Dart's transpiler isn't on the level of CoffeeScript or TypeScript.
>This is the condescending and arrogant attitude I spoke of.
I'm asserting something grand and sweeping that I think to be true. Get over it or walk away.
Define "substantial". I've seen some complaints here that AngularDart has, and will, take away from AngularJS, which, again, is unsubstantiated. It's counter to the evidence and the likelihood that the groups working together will make both better.
> I'm not arguing for Dart's existence here, I'm saying Dart has absolutely no chance of securing the browser's runtime over javascript.
Uh, you've expressed "cynicism [...] over Google's waste of time" and called Dart "pretty flawed", which you haven't substantiated yet. If your were merely expressing skepticism over Dart "securing the browser's runtime over javascript", there wouldn't be a discussion here. Brendan Eich did that for you a long time ago. I'd contend that it doesn't really matter as long as dart2js can fulfill its goals; however, I'd still maintain a wait-and-see attitude about how we'll see DartVM used. There are some very interesting possibilities.
> 173KB. This is ridiculous.
Substance! Thank you. Dart is pre-1.0. It's being worked on. Also, notice at the bottom of that page:
We want your help, especially if you can make update
@MirrorsUsed in main.dart to generate a smaller binary.
No, I won't. The attitude isn't conducive to civil discourse and not welcome here so I'll continue calling it out where I see it. I'll take the upvotes as confirmation that others agree with me.
You've got your unsubstantiated opinions and will continue making these "sweeping" claims so there's no point in continuing here. It leaves little room for meaningful discussion, unfortunately.
Typescript is pre-1.0.
It's telling, to me, that the first argument of mine you take seriously you handwave over by vaguely saying it will get better and it can be fixed.
I'd place a bet that Dart is about as good as it is ever going to be.
>I'll take the upvotes as confirmation that others agree with me.
Sorry to disappoint you, but I've got upvotes, too.
>The attitude isn't conducive to civil discourse and not welcome here so I'll continue calling it out where I see it.
Whatever floats your pompous self-righteous boat.
I may be curt, strident, and critical, but you don't really have high ground when it comes to condescension. You went straight to childish insults. https://news.ycombinator.com/item?id=6679901
When I disagree with someone I don't directly call them stupid or immature. (Well, usually.) I may speak and act as if something I believe strongly is obvious, and I may not mask my disdain for the ideas expressed. There's nothing wrong with being wrong, but I'm not going to hide the fact that I think you're wrong.
You seem to have a problem that someone a) disagrees with you, and b) says so confidently and forcefully. Again, get over it or walk away. If I'm so wrong and so boorish, you are under no contract to continue.
You can't change me and I am not your responsibility.
Fortunately I think you've finally clued in ("so there's no point in continuing here"). I've certainly spent enough time on this. Later.
> I may be curt, strident, and critical, but you don't really have high ground when it comes to condescension. You went straight to childish insults. https://news.ycombinator.com/item?id=6679901
Not me. Try again.
Edit: Mind you, you're still self-righteous and pompous, but you're not hypocritical about it. :P
Dart is a great language, I like it. It has consistent typing rules, very fast, has a good IDE. I for one, would like to thank Google for working on it and open sourcing it.
But I don't have any problem with Dart as an experiment in language design. I don't even really have any problem with Angular.dart.
We can always use another language that transpiles down to javascript. People are free to use what they want. (Unfortunately, my understanding is the dart2js process is horrifically inefficient, but that's no skin off my nose.)
I will, however, drive home the point that Dart cannot out-compete CoffeeScript or TypeScript; Dart will not replace javascript and will not be the native language of the web.
There's nothing to distinguish it from any other language, and it has to deal with backlash like this.
How are you driving it home besides just repeating it without any other supporting statements? Saying something 3 times doesn't make it truer.
> Dart will not replace javascript and will not be the native language of the web
Still don't get why it wouldn't. What is qualitatively different about DartVM vs v8. It could be 2-3 development cycles away from inclusion should Google decide to do it. What exactly code-wise is preventing that to happen?
> Unfortunately, my understanding is the dart2js process is horrifically inefficient, but that's no skin off my nose.)
Looking at their benchmarks their dart2js is not too far off v8 numbers. Now add types plus a nice IDE and I can see how a larger project might choose instead of JS.
What is the pathway for Dart adoption? Google puts it in Chrome. Sure. Dart programmers get to deploy to the Chrome browser, so they do some fun side projects and they run on Chrome only. Fine.
Show me how Dart can then move to every browser on every device.
Then show me how whatever hypothetical process you came up with doesn't apply to any other language that currently exists.
If we're going to put a new runtime in the browser, Dart isn't the only option.
dart2js: https://www.dartlang.org/docs/dart-up-and-running/contents/c...
I'm personally happy that there's still hope we have a future option for a re-invented language optimized for web development rather relying on the glacial pace of continually patching JS to make it more maintainable.
It's a small yet powerfully expressive language that's easy to learn, it's very well thought-out and balanced providing greater consistency in both language and library behavior which is a nice contrast to the dump of incompatible JS fx's, it's optional typing adds a lot less friction and is a lot more productive that statically-typed langs, but also provides time-saving compiler feedback alerting errors as you type. It's support for source-maps are best-in-class whilst Dartium provides strong tool-ability and debug support and very fast live-reload cycles. Dart is an awesome option even without considering it's ~10x faster start-up times with Snapshots or its fast run-time performance which sees it can solve the Norvig's top95 Sudoku grids 200x times faster than Ruby: https://github.com/dartist/sudoku_solver#side-by-side-result...
Dart's not perfect and there are some language features I wish it had but over-all I think it's a magnitude better option than vanilla JS. It currently has less momentum than JS, but I hope to see that change as a JS-only future ensures web development will be forever stuck in the middle-ages.
So, people are naturally suspicious. CoffeeScript doesn't have this vibe, to me at least. It's made by a team of people across companies (although I think it is sponsored partially by The New York Times Company?) and there is a lot of community which lives outside a specific company's ecosystem.
That said, it could just be my own personal bias, having used coffee script extensively and not used Dart much at all.
[1] http://arstechnica.com/gadgets/2013/11/google-just-pulled-a-...
Yet, CoffeeScript is not just a different syntax - for one, it changes declarations and scoping rules. That's a pretty big semantic change (comparison operators etc. follow).
Please.
> That's what causes the friction: resentment that a pretty flawed language is being pushed.
"flawed language" you mean [] + [] doesn't equal '' ?
Oh look it throws a "TypeError: Object #<JSArray> has no method '$add'", yes please bring more such flawed languages around Google, I love them!
Yes, it is. But I'm not sure why you're viewing that as something negative.
It addresses the kind of problems everyone encounters when they try to write bigger applications in JavaScript.
GWT and the Closure Compiler also did this kind of thing. Dart does the same, but in a far more elegant manner. It's terser and the workflow is far more streamlined.
Having proper tooling is really great. Also, thanks to the type annotations you'll need fewer tests, you get better documentation, and it also acts as a safety net for refactoring/maintenance work.
It will enable more people (myself included) to write larger high-quality web applications.
To the much more important point, the development of important features stalled for AngularJS as the team was busy working on Dart version. This is the main problem.
Not true. Google grew the team. Take a look at the activity on Github. The velocity has been consistently high throughout the development of the Dart fork: https://github.com/angular/angular.js/pulse
If there was some important feature you wanted, you are always able to submit a PR yourself. That's how open source works.
Much like android dart is controlled by a small group of engineers in Google. Where's the source code to the current HEAD of dart? I'm pretty sure it's not on GitHub. What features are planned for the next release? I'm fairly sure dart-discuss doesn't exist publicly.
Dark doesn't belong to the web developer community. No conspiracy theory needed, dart is built to suit googles purposes, which is fine. The only thing is expecting people to use it and think its a successor to JS, which, for all the flaws of JS, it is not.
Google know what a community developed successor to JS looks like. Optional strong typing in even MS typescript is more popular than dart. Get together with the community or dart will be he next GWT.
Most of the work is done completely in the open. See some of the language discussion writeups: https://groups.google.com/a/dartlang.org/forum/#!searchin/mi...
We went public two years ago when the language was still in its infancy so that developers could provide their feedback.
Edit: contrary to your claim we did get feedback and acted on it. Some changes were rejected (like putting the types behind the variables), but it's normal that not every request can be accepted.
If you have suggestions on how we can make Dart even more open-source I would be genuinely interested in hearing them.
The most popular feedback from HN https://news.ycombinator.com/item?id=3092558 and Twitter was was less verbose syntax. AFAICT dart remains as chatty as the day it was announced.
Syntax is almost like a religion. You can't make everyone happy. So yes: although we changed the syntax in some areas, we kept the C-like syntax.
Fwiw you can't satisfy every feedback. There are always people that want to make Dart look and behave like their favorite language (be it Java, C#, JS, CS, Python or Ruby), then there are others that just want to boycott Dart, and finally there are simply conflicting views.
While we listened to every feedback we got, we found that feedback from people that had actually tried and used Dart to be the most helpful. They weren't always the loudest, but helped to make Dart a better language.
Nobody is asking you to make everyone happy. You could have made (and still have the opportunity to make) dart look like any modern web language - coffeescript, or Python, or ruby, and you'd have both users and a measurable impact on web development.
It's pretty damn concise and a far sight from Java. It's more concise in many cases than JavaScript (and only more verbose around things like type coercion where being explicit is a much safer choice). I admit though, that Dart is more verbose than Perl :)
C-style syntax doesn't have to mean verbose when you have features like short lambdas, optional types, string interpolation, operator overriding, getters and setters, setting instance vars directly from constructor parameters, fewer keywords, mixins, etc. C-style syntax also makes the language pretty much instantly readable, and if you're targeting web developers, being familiar to JavaScript, Java, C#, and ActionScript, is really important.
Honestly, Dart is the most readable language I've used, and that includes many years writing Python.
[disclaimer: I work on the Dart team]
This is simply not true! I've been following the language for a long time and they have made a lot of changes, they have listened to feedback. Just search 'BREAKING CHANGE' on the mailing lists. During the first year there were tons of changes to the language and libraries. Now they trying to stabilize the APIs
The designers have also made it very clear that the types next to a variable are basically useless at runtime (unless you turn on strict mode, which kills performance). Types are for tooling only, which seems silly.
Not only that, implement it right away in the JS engines..
Javascript must move faster, because even that people are being forced to like it(because they dont have anywhere to hide); everything has limits.. everybody knows that we cant go on with the javascript the way it is right know and pretends everything its just fine and its all roses and unicorns..
Im trying to create something new here that will need an embedded VM, and deciding between JS, Lua or Dart is just killing me.. if that was an option, and the next javascript was just something like typescript.. this would be a easy decision
It did not happen BECAUSE of Microsoft (Adobe pushed FOR it, Mozilla got cold feet).
All but Dartium. Even Chrome doesn't support it natively yet.
> Dart's type system is basically just the Closure JsDoc system used by the Closure compiler.
Yes, it's basically something like that. The difference is that it's terser and far more convenient to use. Since it's baked into the language, it's standardized and properly supported by all tools.
Another difference is that you immediately benefit from putting those annotations at the boundaries. You get the checks and call-tips right off the bat.
> Dart is really just another Coffeescript
No, Dart and TypeScript offer proper tooling. CS doesn't.
Dart also offers much nicer semantics.
> One of the benefits of development in JavaScript is you can simply refresh your browser and it just works.
Same with Dart. You just use Dartium during development.
At some point, I believe I will come to a position not to touch anything that has even a remote connection to Google. Just a couple of years ago I worshipped Google, now I feel they are always taking my toys away.
Javascript needs some challenge, just like Flash forced HTML to evolve.
So this is a good thing. Congratulations. And devs should know this, nothing is set in stone on the web.
Check out the dart:js library, which is pretty new, and the article that _ak linked (which needs some updating).
Hello Ember.
That's quite a sure statement. Why don't you think Dart will ever be natively adopted by Firefox at some point in the future?
Google wants to do it, sure, but that doesn't mean they'll succeed, even if they try something as stubborn as running Dart alongside Javascript in Chrome.
There simply isn't a path to ubiquity for Dart, and if there were it's not the only language on it.
Check tag count for questions on SO; 4500 for CoffeeScript, 1500 for Dart.
Not to mention the violent revulsion many people (myself included) are expressing in this topic.
On that note, Java has 508,338 questions on StackOverflow, which crushes CoffeeScript, /especially/ with those 15,000 GWT questions ;)
As someone else pointed out, SO isn't a great metric, but finding github activity by language seemed to take more time than I was willing to invest.
Immaturity seems like a good explanation of that effect
:P
Dart is fundamentally different in that it was designed for speed (as well as toolability and readability) and by directly supporting Dart a VM can make many more optimizations than a JS VM can. The Dart VM is already 1.5-2x faster than JS in some cases, while being _much_ younger.
When it might matter is if there are apps (like games) that run faster in a Dart VM than in javascript. Anyway though, even Chrome doesn't include a Dart VM yet by default, so no big deal yet.
As well as, forcing developers to infer the way Dart compiler generates JavaScript code, in the absence of source maps.
I rather use JavaScript, thank you.
For the record, I'm not worried about JS being replaced by a better language. I am working to do that within Ecma TC39, by evolving JS aggressively.
The leaked Google doc's assertion that this is impossible and that a "clean break" is required to make significant improvements is nonsense, a thin rationale for going it alone rather than cooperating fully.
The big issue I have with Dart, which you seem to consider inconsequential, is whether Google forks the web developer community, not just its own paid developers, with Dart, and thereby fragments web content.
A Dart to JS compiler will never be "decent" compared to having the Dart VM in the browser. Yet I guarantee you that Apple and Microsoft (and Opera and Mozilla, but the first two are enough) will never embed the Dart VM.
So "Works best in Chrome" and even "Works only in Chrome" are new norms promulgated intentionally by Google. We see more of this fragmentation every day. As a user of Chrome and Firefox (and Safari), I find it painful to experience, never mind the political bad taste.
Ok, counter-arguments. What's wrong with playing hardball to advance the web, you say? As my blog tries to explain, the standards process requires good social relations and philosophical balance among the participating competitors.
Google's approach with Dart is thus pretty much all wrong and doomed to leave Dart in excellent yet non-standardized and non-interoperable implementation status. Dart is GBScript to NaCl/Pepper's ActiveG.
Could Google, unlike Microsoft ten or so years ago, prevail? Only by becoming the new monopoly power on the web. We know how that story ends.
https://news.ycombinator.com/item?id=2982949
EDIT: updated to the long form quote, for additional context.
I'm new to javascript & dart, so honest question here: why?
He might have spoken too soon though (or I missed the point), as there are projects out now that I think could very well make a decent runtime for Dart. Using tech such as asm.js[1] -- which Firefox now includes natively (drawing a parallel to Google's Dart) -- along with new versions of JS, and modern JS browser optimizations....it can make for a very efficient script.
As efficient as Dart with the VM? Probably not, but I'm not sure just how much of the market you're going to corner by such an advantage. Any JS outside of a browser will undoubtedly run on NodeJS, which is going to be more efficient than Dart for most (if not all) tasks, and so it just comes to down to in-browser usage. For 99.99% of current web apps, I don't think you'd never notice a difference.
Dart has a nice IDE, good clean types and language design. People are upset because they jumped on the wrong band-wagon and now something shiner and newer came around so they are not the cool hipsters anymore. Kind of like in the kindergarten the kid with the shiny new bike is boasting then some other kid gets the later, better, shinier model, the first gets upset and throws a tantrum.
Look at Dart's performance benchmarks. Its DartVM is about 2x as fast as v8 (JS) and its compiled to JS version is on par with v8.
I have a serious problem with Dart's design: the type system is deliberately unsound around generics because the designers apparently think that supporting variance (which is not that difficult!) is worse than having types tell the truth. I would much rather there be no type system at all than there to be an unsound one, as I don't think compilers and IDE's should lie to you.
I think that's a mistake that the new language of the Web should not make, but Google seems unwilling to change it. Unlike JavaScript's warts, which be worked around for the most part, I don't see any way to fix a flaw that serious in the type system once Dart is stable.
But the tooling and documentation aren't telling you what will happen at runtime.
> At run time you can run it in checked mode and it will always tell the truth.
It doesn't tell you the truth. A static type system is designed to rule out errors are compile time. That's what makes a static type system different from type assertions or guards.
Besides, I don't really know what "telling you the truth" means if, for example, function arguments unify under covariance instead of contravariance like they should. Whether that happens at runtime or at compile time makes no difference—it's equally wrong either way.
> This means that if you cover your code at all, manually or automatically then it will inform you of type errors.
This is always true in any dynamic language.
> How is this worse than a dynamic language that offers you no optional typing with full checking at runtime?
In JS's case we can still add a static type system that's actually sound, whereas Dart has already crossed the bridge into unsoundness.
> This means that if you cover your code at all, manually or automatically then it will inform you of type errors. >> This is always true in any dynamic language.
No, there's a difference. I will get the error in dart as soon as I try to use the object in any way. In a dynamic language I could call a few methods on it before calling a method it does not support. i.e. it behaves like a compiler would except at run-time. This makes a big difference in how quickly you can debug it and again this is only for the minority of scenarios that are not covered at compile time.
I honestly don't even care if Dart doesn't take off, maybe clojurescript or something similar will instead. But my god something needs to happen in this space because the JS standards body has spent almost 2 decades aggresively doing nothing about the many gaping, egregious flaws in Javascript. Its 2013 and we don't have lexical scoping as standard, but hey let's dick around in committee for another year packing in generators, etc, etc, etc.
I would honestly put a lot more weight in Brenden Eich's words if the state of Javascript and web dev wasn't so absolutely dismal and if he wasn't trying to crap on a project actually working to do something about it.
Uhhhh TC39 meets bi-monthly. If you meant ECMASCRIPT 6, it's targeting finalization next month. Furthermore, future releases are set to be much, much faster than previous ones.
I would be somewhat happy if ES6 was finalized next month, but if it slipped (again) I would not be surprised. You'll also have to forgive me if I am skeptical of claims about the increased speed of future standards. The burden is on TC39 to prove this to me and I reserve the right be skeptical based on prior evidence until I see otherwise.
Also think if anything or anyone is going to make a change it would have to be Google. Can't image a small company out there coming up with a new browser with a new language. It would have to come in Firefox or Chrome. So I am excited for Dart.
Brendan Eich does not have a right to anybody else's effort. It is totally unreasonable to use artificial means to trap people on languages they have problems with, just because you want control of more of the programming world for yourself.
Angular's Object.observe like digest mechanism is superior.
Like someone said, you/we can always decide to fork project, but again... let people who care about Dart try to do it, I think this is legitimate prof of concept. I don't think it will catch up, if there is genuine work outside Google on Angular.
I am grateful for good work done on Angular project and look forward for new features. I will not use Dart that is for sure.
Dart is an open-source project and compiles to JS. Not unlike CoffeeScript or TypeScript. The cool thing that makes me excited about dart2js is that Dart is a much better language than JS itself (not a high bar, but still) and that dart2js as of late is producing more efficient code than hand-coded JS because they leverage Dart's more ubiquitous typing as well as other tricks.
Coincidentally, this is how I feel about AngularJS. By using it, you're developing apps the way Google wants you (or it's own developers) to. It could be good, it could be bad. it depends on the app, but you still have to do it their way.
You don't want to be locked into Google's walled garden, but, by using Angular, you're already inside...