Google's Angular 2 Being Built With Microsoft's TypeScript
itwbennett writes Big news for fans of static typing! Google and Microsoft have partnered to both enhance TypeScript and rebuild Angular in the TypeScript language. TypeScript, Microsoft's attempt at improving on JavaScript development, has been out there for a while without a notable use case. Likewise, Dart, Google's attempt at a language which accomplishes many of the same goals, hasn't seen a lot of traction outside of Google. With Google creating the next version of its popular framework Angular 2 using TypeScript, some weight is being thrown behind a single effort. Of course, Angular has its fair share of haters, and a complete re-write in version 2 that breaks compatibility with previous versions isn't going to help matters.
Yes @ script is a superset of Typescript and it will be used in Angular 2. Not really a hot news story.
Democracy Now! - your daily, uncensored, corporate-free
Proprietary company uses proprietary language extension to rewrite proprietary framework. *yawn*
and at the same time. Nothing good can come from this. Fight christian soldiers! Fight like you have never fighted before.
I don't know how to tell whether I should care or not. Is it a language for creating angles? Mathematica can do that.
If they haven't finished it, I don't think would care anyway.
I should use this sig to advertise my book ISBN-13 : 978-1501515132.
Looks like a return to the old "embrace and extend" to me. And we know how that worked out.
=^..^= all your rodent are belong to us
Don't waste time learning this framework. In a month there's a newer and even better framework which will only be supported for a few months before everybody bails to yet another framework.
Next step: make Angular maintainable and non-magic for mortal programmers.
So what exactly is angular?
So why should I start using this, when Google has a history of abandoning their projects after a couple of years?
Every time someone writes "open source language", a pet dies.
Programs can be open source, but languages are not programs. They are specifications of syntax, grammar and semantics. Maybe we have an Open Source compiler for the TypeScript language, but who defines the language itself? Is it a foundation or similarly open group participated by the community and key players of the industry, or is it steered behind closed doors in some meeting room in Redmond, WA? That is what matters.
Compilers... nobody pays for a compiler nowadays, and Microsoft is very aware of that. They are not stupid.
Angular is the shit. My term for web development with JavaScript pre-Angular (and similar tools) was "Web Assembly Language" (WAL). It was so fucking tedious, it took so much work do do simple shit, etc...
Angular isn't for every project, just like sometimes you have to be the poor fucker writing assembly language for some very narrow cases. But for most projects it (and tools like it) are the shit.
Otherwise my client isn't interested
http://saveie6.com/
From the summary (emphasis mine):
Microsoft's attempt at improving on JavaScript development, has been out there for a while without a notable use case.
That is not what 'use case' means (https://en.wikipedia.org/wiki/Use_case). The phrase you're looking for is "without a notable real-world use".
We're done. We can compile C into ASM.js, which is JIT compiled by Chrome & IE, and AOT (ahead of time) compiled and cached by Firefox.
We're done. The snake has ate its tail. All languages, static or dynamic, now can be compiled and ran in the browser. The theoretical proposition that any Turing Complete language can be implemented via another has now been put into legitimate use.
In the future, Hybrid Compilable and Interpretable VM bytecode will tighten this loop and allow seamless on the fly linking between disparate languages across all platforms.
My meta-language can already compile into C, Perl, Python, Lisp, ASM.js, and VM bytecode. I currently have to maintain a "runtime" library for each of these base languages. A standard feature API for windowing, audio, graphics, etc. would take the place of my "API adapter" abstraction layer.
This means ALL NEW LANGUAGES are irrelevant. I simply port my language adapter runtime to the language, then add a module to the meta compiler to output in that language, then I can compile ALL OF MY EXISTING WHEELS down into the new language before my competition can master the new whimsical "expressiveness" of the platform.
Write your code in a meta language, and never reinvent the wheel again.
A year ago they were slowing work on AngularJS to put more effort into a complete rewrite with AngularDart.
Dart is a better language than TypeScript and it's a Google creation... I have no idea why they did this.
I don't care about Angular. It's just another tool for the saps in the web page mines (and one that can get you trapped in those mines as well).
TypeScript, OTOH, is the greatest addition to JavaScript I've seen. No more messy .prototype., and much less "can't read property 'x' of undefined". It's not there yet, I must say. I would like it to add some more transformations instead of just type checking, but if you have to write in JavaScript: do yourself a favor, and check it out.
You remember that? lpStr and things? Where you prefixed some cooked up type notation to the variable name? What can possibly go wrong?
(The reason, at that time btw was that their C compiler was too crappy to do type checking itself).
What major Google product uses Angular?
Anyone can point the cool things that one have and the other doesn't. Sure static typing is nice and all but I rather dislike static typing for big iterative projects, refactoring static typed code is a pain in the ass. Yet at the same time static typing makes a lot easier for a new dev in a big project to start being productive without breaking the whole thing (although test-driven development in dynamic typing languages help a lot in this regard).
Also how are the tools for typescript? Having static typing but no auto-complete IDE is a major drawback. In my opinion the main advantage of using static typing after compile-time errors are the auto-complete IDEs. What about debugging? Coffeescript can be debug rather easily on browser debugger because all variable names remain the same, Clojurescript has the REPL that I believe can push code into the browser, what about typescript?
Visual Studio 2013.
In other new, Angular 3 is planned for next year using MVVM pattern and Google's new HTML5+JS generator that was previously used for converting Flash. Same as with Angular 2, it is not backwards compatible and will be maintained for 2 years.
He really does have the knack for programming language design. I didn't get TypeScript at first, but with 1.4, it clicked. The great news about this is that Angular is a highly visible framework, and with this, more people will look at TypeScript and be willing to use it. Thanks to type definition files and definitelytyped.org, you can use a ton of JS libraries right now; hopefully, more people will officially maintain these files.
Also, this makes it easier to recommend it's use in work projects. Being able to say: "It's good enough for Angular, it's good enough for us" helps a ton.