I'm hyped for Webassembly. Maybe Atom will finally be usable.
>assembly
>for web browsers
How far down will this shit go? I guarantee you someone is gonna make a javascript engine in this, or something stupid a webdev would do.
>a faster web
We have fucking CPUs that are 10x more powerful than ones from 1990 in our POCKETS, I refuse to believe the web is that taxing. Fucking webdev.
>>53581927
actually it's about making a binary format that can be used on websites that allows devs to have low-level access and performance in the browser sandbox. More importantly, it would allow people to write in whatever language they want and compile to webassembly. Meaning you wouldn't need to use shitscript anymore.
The problem is not javascript, the problem is shitty people using 2 MB of javascript to do the same thing a properly done TXT file would do.
>>53581927
If you try to make full fledged applications in JavaScript on websites with 10MB of tracking scripts chugging along in the background while your browser crawls through the 15MB of CSS and other shit, then yes.
The web is very taxing.
>>53581966
>using 2 MB of javascript to do the same thing a properly done TXT file would do.
what are you talking about?
>>53581968
10MB of tracking scripts
15MB of CSS and other shit
is hyperbole the recourse of a weak mind?
>>53581964
So instead of a 2MB, poorly coded javascript I can have a 500MB executable file, running on my computer, linked to a bunch of cruft poorly written webdev libraries, all because websites are too lazy to make native apps or write good CSS and JavaScript?
>>53581968
Because yes we very obviously need to have the smooth website image scroll JavaScript that costs 2MB.
I swear sometimes it feels like I'm in bizaro land. Fucking ASSEMBLY on a WEB BROWSER.
>>53581971
>What a shitty developer does
>>Infinite scroll, because apparently that's useful
>>Social sharing buttons, at least twelve of them for hipster pages no one uses
>>A header that never goes away, because you truly need that to know what webpage you are in!
>>A sidebar that is also always there
>>Some kind of annoyance reminding you that you really could use being registered in this website you won't ever enter again
That's the kind of shit I am talking about.
90% webpages could be done in plain HTML with maybe frames and a rss tracker. That's it. Maybe some php if you will update it regulary.
>>53581984
It's the recourse of reality.
>>53582013
Oh, you mean a 2MB javascript file on top of the fairly simple HTML with what could be a txt file?
it sounds like you're collapsing things too much. i agree that there's a trade-off of feature/experience and "weight" of the page, and some things aren't worth the trade-off required, but you're making an argument akin to bitching about all the bandwidth netflix consumes when people could get the same amount of entertainment by reading lightweight ebooks.
>>53581927
It's actually just a bytecode format similar to the code that runs in a Java VM. Basically it's Java again.
Seems like another way to make non-free executable code even harder to reverse, as if it's not hard enough with these JS obfuscators everyones using
>>53582076
Oh yay more vulnerabilities.
>>53582019
any good examples of this then? I just picked a random front page buzzfeed article and the whole page is about 5MB. That's way too high for a long-form piece (this was a "real journalism" something about how weed is a white person thing apparently), but you shouldn't need to exaggerate your point.
If someone is on the fence and sees your hyperbolic bullshit that doesn't even pass the smell test, they're more likely to lean the other way, figuring that only someone on shaky ground would give false data like this.
>>53582082
>as if it's not hard enough with these JS obfuscators everyones using
Wait wait what? Why? Who is idiotic enough to give fucks about "Stolen" javascript?
>>53581988
it's more about finally having a universal platform that performs well and works on all computers with a web browser.
>>53582076
except it's not Java, not affiliated with Oracle, and with people who know what they're doing on it
>>53582089
it's by the same people who made the vms for javascript and NaCl, the latter of which was basically the same idea as this and had no exploits
I just ordered this book, and can't wait until it arrives.
>>53582126
>it's more about finally having a universal platform that performs well and works on all computers with a web browser.
YES BECAUSE THIS HAS WORKED SO WELL ON THE WEB IN THE PAST
>Maybe Atom will finally be useable
Why not just make a native text editor in the first place?
Definitely going to try program a shitty kernel for it.
>>53582126
>it's more about finally having a universal platform that performs well and works on all computers with a web browser.
So all web browsers will be even more bloated and weight two hundred MB's more?
Wow, maybe we will get to 1GB big web browsers! Cool!
>>53581870
>https://hacks.mozilla.org/2015/12/compiling-to-webassembly-its-happening/
>WebAssembly is a new binary format for compilation to the web.
>binary format
do not want
>>53582013
Have you heard of A/B or Bandit testing? All of these UX decisions that you hate so much have been tested by websites with a lot more traffic than your cartoon porn anime tumblr. Do you think you're smarter than the analysts behind Google or Quora? Fuck off.
>>53582148
the problem with java and flash were that they threw away the browser sandbox and implemented their own entire stack. The idea with wasm is to keep the browser sandbox, and the people who are making it have a lot of experience on the subject, considering the javascript sandbox has been pretty good.
>>53582166
You can always use lynx.
>>53582089
Well not really any more vulnerabilities than what we already have, webasm runs in the same sandbox as javascript so it's really just changing the language from javascript to something a little more usable so it's possible to squeeze more performance out of it. The idea is you can write your program in any language(java, C, C#, C++, etc) and then you compile it to webasm. It's extremely similar in concept to a Java VM.
>>53582166
why are you so afraid of progress?
>>53582178
>most people are idiots
>idiots like stupid flashy things
>therefore stupid flashy things are good
The only person your argumentum ad populum is fooling is yourself.
>>53582182
>implemented their own entire stack
And what's gonna stop people from doing that? People are just gonna make frameworks in C# or whatever hipster Starbucks coffee language they like this month for JS and CSS again.
>>53582182
And browsers will fully support every feature of webasm and it'll be comparable performance across all platforms right? RIGHT?
>>53582188
I was talking more about pic related than progress itself.
>>53581870
/thread
>>53582203
I'll assume you aren't trolling. Companies are using the designs that test the best amongst their users. If they could get away with static pages and much less work, then they certainly would. I don't really care what you think is a good practice. I'm suggesting that these things are only in use because they work so well. Creating a site like you suggested:
>90% webpages could be done in plain HTML with maybe frames and a rss tracker. That's it. Maybe some php if you will update it regulary.
isn't as effective as using all of the UI elements that you hate so much. Do you understand that? Does this make sense to you?
>>53582304
This couldn't be further from the truth.
Web pages are 30MB monstrocities with 10,000 hipster.js libraries being loaded on every pageview because webdevs are the literal asshole of the tech industry, they are the lowest possible rung of "developer".
None of them EVER learn how to program, they are taught how to glue together other people's libraries and they NEVER progress from this.
They will import dozens of libraries, using only one or two functions from each one instead of simply implementing the same functionality themselves in javascript.
They don't know how to implement the tools that they use.
They also don't actually know how their fancy effects work when everything is done for them and they simply copy paste shit from web design blogs and tutorial sites.
>>53582209
>>53582218
If everyone thought this way, nobody would ever make new technology.
>>53582386
This. This right here is how web dev got to it's current state. Progressive thinking with no regard for cost or consequence. We should just start killing off humans that aren't genetically perfect right? No? Well then you're just in the way of the advancement of technology.
fuck off.
WILL WE PLEASE STOP REINVENTING THE WHEEL OVER AND OVER AGAIN LIKE SOME INSANE SISYPHUS???
A BROWSER IS FOR PRESENTING HYPERTEXT, STOP USING IT FOR SHIT IT WAS NEITHER INTENDED NOR DESIGNED FOR
USE A FUCKING REMOTE X WINDOW IF YOU WANT TO RUN AN APPLICATION OFF THE MAINFRAME
>>53582408
Did you just compare experimenting with technology to eugenics and genocide? What the fuck kind of mental gymnastics did you just do?
>>53582209
While I agree most websites are disgusting bloated garbage webasm would probably actually help reduce code bloat. If a webasm compiler works like a regular one does worthless code would be optimized out where ever possible and the resulting bytecode would be heavily compressed. But webasm bytecode is also essentially a binary blob, that's the problem I have with it. With webasm there's essentially no way to tell what you're running without reverse engineering. There's already literal botnets that can run inside the javascript sandbox in web browsers.
>>53582445
Well, are you for the total advancement of technology or not? Think about the fucking consequences of what you're doing instead of just blindly going forward. Contrary to your progressive leftist views, the ends don't justify the means. Especially when the ends are shit anyways.
>>53582424
But anon how else are we going to shove our social buttons up everyone's ass and execute our 100MB effects on your side?
>>53582465
Dude, take your lithium.
>>53582424
We passed the point of return some time ago with javascript taking off and plugin support. Now web browsers want to be operating systems and play movies, video games and music directly.
>>53582370
Ok, I can agree to disagree with you. Just wondering, have you worked at a large company like one of these? I have worked at a couple of start ups and I'm currently working for a company which I guess you assume employs idiots to do web development.
In my experience, or at least the interviews that I hold, I have vetted candidates pretty hard. I ask them the same base interview questions I'd ask a core engineer. After they finish that stage, if they made it through each question without help, we move on to a project in which there code is looked at by a senior engineer for style, bloat, and general awareness of standards. What your suggesting
> copy pasting
> they don't know how to implement the tools that they use.
> import dozens of libraries
are all really frowned upon. It's pretty obvious when they have just copy pasted code. The dozens of libraries is especially bad marks.
Finally we'll ask a series of questions designed to make sure they understand the entire stack.
So I don't know where you have worked before, but what you're talking about sounds like a junior engineer who we push on to internal tools and only promote if they get better. Often they don't
>>53582481
obviously by building a new and revolutionary framework that you can install with bower and integrates well with jquery and angular and node and mongodb and mongoose and OH FUCKING HELL BURN IT ALL I HATE THIS PROFESSION
>>53582486
You're the one that needs mental help.
I REPEAT. ASSEMBLY. IN. WEB. BROWSING.
PROPRIETY BINARY BLOBS RUNNING ON UNSECURED BLOAT MACHINES
WHY IS NOBODY THINKING THIS IS A BAD FUCKING IDEA?
>itt: people who don't understand the web
its ok, go whine about "muh new framework every week"
>youre an idiot
my degree is in EE, i do JS now because its the best :^)
>webassembly will fix EVERYTHING
JS will still be the language of the web, and Pajeet can't into memory management so our jobs are safe
>>53582507
woah, homer has a pretty huge dick
>>53582506
Don't worry bro, it's sandboxed :^)
>>53582494
every project I've seen done with mongo could be done with either MySQL (they don't even utilize documents correctly. they could have just used a RDB but they are bad), or Redis (mongo is overkill for cache). Haven't seen a project that utilizes mongo in a way that makes me think
> oh wow, I couldn't have used MySQL or PostGres for this.
So how long before every website is a literal binary blob wasm application running in your browser that prevents you from saving pictures or copy pasting it's text?
>>53582506
You literally have no idea what the subject is and are making assumptions based on its name. And you're spouting batshit insane paranoid schizo shit.
Take your damn lithium.
>>53582520
You and I both know there will be SEVERAL zero day exploits.
>>53582507
finally a sane poster
>>53581870
>Maybe Atom will finally be usable.
That would be fucking hilarious
>write text editor
>use JS instead of C because I'm a node hipster
>editor is too slow
>rewrite editor in C and compile it to JS
>editor is now fast and I retain my hipster cred
>>53582533
BINARY BLOBS.
WEB BROWSING.
In what world do these two things make sense together?
>>53582538
Yup. To be honest I'm surprised the entire web hasn't collapsed yet, but I think it's not long before "it's happening", i.e. the ads dry up and everybody realizes what a scam it's all been. dotcom2.0
>>53582531
Webasm based crytocurrency miners will probably be one of the next forms of revenue for websites.
>>53582580
I can only imagine what a shit hole the web will be in Web 4.0
Hey, web devs. Yes, you!
Why must you keep trying to rube goldberg what was initially a digital research paper into 500 ton application-wannabe monstrosities? Why must you keep trying to take the computer world's equivalent of a Geo Tracker and jerryrig it into pulling full length trailers? Why the fuck can't you idiots just use the right tool for the job?
It's not as if the tools don't exist. No, in fact desktop software development is extremely mature and well-tooled and has been for decades. All the problems you're struggling with now have been solved for years in the desktop realm. And that funny little JS library you just posted on HN and blogged about on Medium? Congratulations on carving a stone wheel! Keep pretending wheels with rims and tires don't exist so you can feel like you actually accomplished something.
It's just makes no sense whatsoever. Why waste time on such dramatically less capable, less purpose-suited, and disgustingly inefficient technologies?
>>53582507
>memory management
in this new format?
>>53582566
it's an Abstract Syntax Tree, and you can still view source. No worse than minified js.
>>53581964
sounds like a whole new level of exploits is coming our way.. what use cases are satisfied by the performance of web assembly but not existing web development platforms? this seems really fucking lame
>>53582584
So now I'll need to forcibly underclock my websites just to prevent them from eating my battery life?
What a time to be alive.
>>53582594
>it's an Abstract Syntax Tree
DUDE, do you have ANY fucking idea what you're talking about. Every language can be made into an AST. It's the intermediate language in compilers.
>>53582590
It's because web development is an entry-tier job. The ones who roll into it are either graphic design majors (who know shit about CS) or dumbasses who just graduated and have nowhere else to go. It's a very youth-driven field with a large drop-off rate (for network engineering, devops, or even regular programming). As a result, nobody ever learns lessons or passes them on. It's also the most fashion-driven, so that's probably related.
>>53582538
>>53582566
Next time try reading literally anything about the subject before posting. If you had, you would already be aware that WASM is nothing more than a binary encoding for asm.js, which is already usable in every browser under the standard sandbox.
>>53582595
The main reasoning behind this is basically "muh games"
>>53582620
the point he's making is that it's not binary output, it's readable ASM.
this thread needs to be deleted ASAP it's very clear this board is filled with nothing but retards. at this point it's just embarrassing to see people like >>53582590 >>53582634 trip over themselves trying to make a point
>>53582636
And I'm sure the source code will be optimized well, open, and changeable.
ASSEMBLY ON BLOATFOX
CAN'T WAIT TO BE A BITCOIN FARMER FOR GAWKER
>>53582620
The binary is an encoded version of the program (converted into wasm, which looks similar to llvm IR) as an AST. It makes the file small and parsing fast. It's a binary in the same way that a .js file is a binary. It's just not human readable without a tool (also part of the spec).
>>53582644
Yeah this.
>OMG CAN YOU BELIEVE THEY WANT TO ADD X TO WEB BROSERS, WHAT A TERRIBLE IDEA
>all major web browsers have already supported X for 3+ years with no major problems
>>53582656
>And I'm sure the source code will be optimized well, open, and changeable.
The source code to what? Blink/V8 and Gecko are as open source as they've always been. And WASM is exactly as unreadable as asm.js already is, given there is a 1-to-1 mapping between the two.
>>53582656
Oh my god, it's not assembly. It's similar to LLVM IR. Please do the bare minimum reading before you go absolutely insane and rub feces on your controllers.
>>53582691
>It's similar to LLVM IR
No it's not, it's based directly on the asm.js spec. There's an LLVM backend that emits asm.js/WASM, but as far as I know they are not actually that similar to LLVM themselves.
>>53582550
It's not compiling to js, it's compiling it to browser bytecode
>>53581870
I'm hyped for some front end web python. I'm fucking tired of having to program js
>lets give web devs full access to the browser
>that will surely not bring about security issues :^)
>>53582706
It's based on asm.js, but it is still similar. I mean it literally reads very similarly.
>>53582721
Jesus Christ, I'm aborting this thread. You people are children.
>>53582687
Yeah, I mean now I'm starting to understand why so many here are out of jobs. They literally struggle to comprehend something prior to forming their opinions on it.
>>53582408
webdev got to this point because they needed to program fucking javascript. Do you think we would be at this point if java, c#, python or any sanely designed language was used for the browser frontend? Coding javascript does shit to your brain, web assembly is the fix. Now people can make their own c#, clojure, python web assembly compiler and never have to use fucking JAVASCRIPT again
>>53582691
>Oh my god, it's not assembly
Why the fuck do they call it that then? Are you telling me my browser is just gonna be decoding bytecode back to JavaScript? I'm gonna spend twice the clock cycles just to execute THE SAME FUCKING JAVASCRIPT?
Jseus fucking Christ, this is just more web devs trying to pretend they're real programmers. They call their little assembly syntax that maps to fucking JAVASCRIPT assembly. I'm gonna fucking faint I'm going insane.
>>53582721
do you even know what you are talking about or are you just retarded?
Does anybody here even know what they are talking about? For fucks sake
>>53582751
Anon don't bother trying to reason with idiots who refuse to understand the premise. You'll only drive yourself insane.
>>53582753
IT DOESN'T DECODE ANYTHING YOU FUCKING MORON. YOU COMPILE THE CODE YOU PROGRAMMED IN TO WEB ASSEMBLY AND YOU SHIP THAT. THAT IS INTERPRETED NORMALLY BY YOUR BROWSER, IT MIGHT EVEN HAVE BETTER PERFORMANCE
>>53582709
...bytecode which is isomorphic to JS
>>53582753
It's bytecode that can be translated into machine code more easily and quickly than javascript can. Nobody cares what you think about the name.
IT'S 2016!
COME ON WE HAVE FAST COMPUTERS!
WHY DO WE NEED THIS PILE OF SHIT THAT WILL BE BLOATED AS FUCK
>>53582789
SO IT'S JUST JAVASCRIPT IN ANOTHER FORM? WHY DON'T WE TRY TO OH I DON'T KNOW IMPROVE THE JAVASCRIPT WE ALREADY MAKE FOR WEBSITES AND THE JAVASCRIPT ENGINE IN OUR BROWSERS? WHY BUILD CRUFT UPON CRUFT?
>>53582751
>>53582780
you're both idiots. JS will remain the language of the web because it runs natively in all browsers.
>>53582808
It's an Assembly interpreter running on a JavaScript backend. You tell me.
>>53582826
you still aren't getting it
you will never get it at this rate
>>53582837
I'm having flashbacks to, what, 2010? When all the hipster web devs were chanting the mantra "Javascript is the assembly language of the web".
Motherfuckers, assembly language is entirely architecture specific! It is *not* portable. C is portable. Those idiots should have been saying, "Javascript is the C of the web", which it is, at least in the specific aspect of portability.
Why isn't this called WebBytecode or WebCode or something? Why do these asinine hipsters have to keep abusing the word "assembly"?
>>53582837
He's right though. What browsers will support WebAssembly when it launches?
>>53581870
What?
Assembly isn't platform independent. That's the whole point of web code...
>>53582797
It's not isomorphic, I don't even know what that means in this context. You have access to lower level resources of the browser + "system" (system in quotes because you are accessing the browser's system) calls that allow you to access the dom.
>>53582814
Are you fucking stupid? YOU CANT JUST IMPROVE JAVASCRIPT. YOU NEED TO GET EVERY FUCKING BROWSER TO ADOPT THE STANDARD. IF YOU JUST HAVE FUCKING WEB ASSEMBLY COMPILER WRITERS CAN MAKE ANY FUCKING LANGUAGE THEY WANT.
FUCK YOU TECH ILLITERATE FUCKS. GO FUCKING COMPILE AND BUILD GENTOO FROM SOURCE YOU RETARDS IT DOESNT MEAN YOU ARE SMART
>>53582853
I'm assuming Chrome, Edge and FireFox as they're supporting it?
>>53582830
>assembly interpreter
wrong
>running on a javascript backend
wrong
>>53582865
everyone is going to support it
>>53582858
>YOU CANT JUST IMPROVE JAVASCRIPT. YOU NEED TO GET EVERY FUCKING BROWSER TO ADOPT THE STANDARD.
He's still right though.
I mean, so what? Fucking revise JavaScript and fix all the fuckery.
Holy shit, automatic declaration of global variables if you forget a 'var'. God, I hate JS.
>>53582865
>At the same time, we knew it was important for WebAssembly to be “of the Web:” it had to access existing Web APIs and integrate tightly with JavaScript by, e.g., allowing calls between WebAssembly and JavaScript. Unlike classic plugin models, this will allow WebAssembly to be more easily integrated into JavaScript applications and libraries, just as asm.js has been able to do.
https://hacks.mozilla.org/2016/03/a-webassembly-milestone/
>>53582858
>WEB ASSEMBLY COMPILER WRITERS CAN MAKE ANY FUCKING LANGUAGE THEY WANT.
I'm just LMAOing imagining those hipster bearded feminist Starbucks fucks making their own language. I can't wait to see whatever abortion of a language they come up with. They fucking use JS for their daily coding LOL.
>>53582876
>He's still right though.
>
>I mean, so what? Fucking revise JavaScript and fix all the fuckery.
He isn't right. You don't get it, you can't "fix" javascript because when you fix it, it will take decades for every browser vendor to implement the fixes. Then they learn that the fixes actually sucked and there needs to be new fixes. Web assembly means you write whatever fucking programming language you want, get a web assembly compiler, compile your shitty code to web assembly and ship that. If you want to fix the language you are using just modify the compiler.
>>53582814
>SO IT'S JUST JAVASCRIPT IN ANOTHER FORM
Basically, yes. A form that's much faster and can be compiled to as a target from other languages without the ridiculous kinds of crap that emscripten has to do. Javascript was designed in 10 days and there's only so much optimizing that can be done.
>>53582837
https://github.com/WebAssembly/design/blob/master/FAQ.md#is-webassembly-trying-to-replace-javascript
>Is WebAssembly trying to replace JavaScript?
>No! WebAssembly is designed to be a complement to, not replacement of, JavaScript. While WebAssembly will, over time, allow many languages to be compiled to the Web, JavaScript has an incredible amount of momentum and will remain the single, privileged (as described above) dynamic language of the Web. Furthermore, it is expected that JavaScript and WebAssembly will be used together in a number of configurations:
>Whole, compiled C++ apps that leverage JavaScript to glue things together.
HTML/CSS/JavaScript UI around a main WebAssembly-controlled center canvas, allowing developers to leverage the power of web frameworks to build accessible, web-native-feeling experiences.
Mostly HTML/CSS/JavaScript app with a few high-performance WebAssembly modules (e.g., graphing, simulation, image/sound/video processing, visualization, animation, compression, etc., examples which we can already see in asm.js today) allowing developers to reuse popular WebAssembly libraries just like JavaScript libraries today.
When WebAssembly gains the ability to access garbage-collected objects, those objects will be shared with JavaScript, and not live in a walled-off world of their own.
blease stob
>>53582745
The reason for the assmad is that shit like webasm is only useful for people wanting to do crazy shit in web browsers. Most people here want LESS crazy shit in their browsers, not more. More 200k-1.5MB webpages, way way way less "web applications".
If the web were doing what it's supposed to do, webasm would've have even made into the scope of consideration.
>>53582814
Because Javascript is a shit language with shit semantics and requires huge amounts of memory for every object, GC, and other garbage. asm.js gets good performance by bypassing all of that fuckery.
>>53582830
>Assembly interpreter
Are you literally retarded?
>>53582853
All the ones you care about. And for the rest, there's a "polyfill" that converts it into asm.js, which itself is 100% compliant javascript code that works in any engine.
>>53582858
Should have said "isomorphic to asm.js". And asm.js is just javascript, written with all the weird extra |0 and shit so the JIT can infer precise type information.
>>53582904
Yeah, should be fucking hilarious. But meanwhile the rest of us will finally be able to use non-shit languages on the web.
Why is it designed to be compatible with javascript? If you can write code in any language why would you purposely subject yourself to javascript? You could even write your code in javascript and just compile it to webasm or whatever I don't get why this compatibility was needed.
>>53582914
>HTML/CSS/JavaScript UI around a main WebAssembly-controlled center canvas, allowing developers to leverage the power of web frameworks to build accessible, web-native-feeling experiences.
>Mostly HTML/CSS/JavaScript app with a few high-performance WebAssembly modules (e.g., graphing, simulation, image/sound/video processing, visualization, animation, compression, etc., examples which we can already see in asm.js today) allowing developers to reuse popular WebAssembly libraries just like JavaScript libraries today.
>When WebAssembly gains the ability to access garbage-collected objects, those objects will be shared with JavaScript, and not live in a walled-off world of their own.
I'm fucking dying. This whole thing is a joke.
>>53582905
>when you fix it, it will take decades for every browser vendor to implement the fixes.
1) Not with that attitude, it won't.
2) I honestly don't even believe that's true. The majority of developers would like to see JS improved. The "browser wars" are over, at least in their original incarnation. Browser makers are pretty proactive about seeking adherence to standards.
>Web assembly means
...another layer between you and the buggy code you wrote. It's not going to help. It's just going to create an explosive proliferation of new hipster languages with unGooglable names.
>>53582917
>would've
*wouldn't have
>>53582904
I think he meant to say "make things IN whatever language they want"
>>53582930
>Why is it designed to be compatible with javascript? If you can write code in any language why would you purposely subject yourself to javascript?
because there's a huge crowd of shitty web devs that know javascript, and they want the biggest possible userbase. It's not just webpage content that degrades to the lowest common denominator, it's the things those websites are built from.
I can't wait for the rest of the normies to discover adblock to destroy the economy that employs all these shitheads and generates all this crap
>>53582904
luckily you actually have to be intelligent to make a programming language for the most part, even in web assembly
>>53582917
>If the web were doing what it's supposed to do, webasm would've have even made into the scope of consideration.
The web is doing what it's supposed to do, cry more old man. Now we need the ability for web sockets to do udp and I can die happy and rich.
>>53582930
There will always be javascript just like there will always be cobol and prolog
>>53582959
>...another layer between you and the buggy code you wrote. It's not going to help. It's just going to create an explosive proliferation of new hipster languages with unGooglable names.
maybe but who gives a shit. The majority of people will be writing python, ruby, java, C#, C, or C++ like sane humans.
Imagine coding a responsive app in java and compiling to web assembly, android, and ios.
>now instead of the JavaScript being buggy and slow
>we now have bytecode, that's tightly coupled with JavaScript
>for non standard languages built by hipsters
>"When WebAssembly gains the ability to access garbage-collected objects, those objects will be shared with JavaScript, and not live in a walled-off world of their own."
Try and read this without laughing.
>>53582917
Monolithic web apps are dumb as hell, but unfortunately it's the best way to make something that runs on any modern system, is easy for users to get working even if they're complete morons, and doesn't require me to fuck around with portability concerns. I used to write Java (shit tier language) and distribute .jars, now I write Javascript (sub-shit tier language) and distribute URLs.
>>53582917
"the web" is a nebulous concept that has no strict definition anymore, it's expanding to permeate all aspects of computing. What we want "the web" to be isn't want it is, and that's the way it is.
>>53583011
>for non standard languages built by hipsters
Yeah, all those C/C++ programmers on their macs drinking starbucks.
>>53582590
Because the web has the best distribution system. If you can get it to work in the major 3 browsers just publish and it's shipped. No one has to install anything, you can't see the value in that?
>>53583011
>tightly coupled with JavaScript
wrong
>for non standard languages built by hipsters
hipsters aren't smart enough to make their own language. also their languages would never gain adoption so this is literally the dumbest shit to worry about
>"When WebAssembly gains the ability to access garbage-collected objects, those objects will be shared with JavaScript, and not live in a walled-off world of their own."
This is just so the retarded javascript programmers don't have a heart attack because they will be left out of the web assembly revolution
>>53583031
web apps are also good just because you can push out updates without breaking everything on the user's system. they are also secure in general
>>53582975
>because there's a huge crowd of shitty web devs that know javascript
The bigger thing is that these folks are some of the most obstinate "developers" on the planet, refusing to learn anything that's not JS. They'll go any length and make any compromise just so they can write that horrid language because they can't be arsed to understand anything more complex.
>>53583009
>maybe but who gives a shit.
People who care about being able to reliably step through code and debug it. Oh, look, the intermediary layer is acting up again. I guess I'll file a bug with the browser dev team and wait six months for the next major release.
>Imagine coding a responsive app in java and compiling to web assembly, android, and ios.
So, the same limited experience you can get with, what, C# based tool?
No thanks. It's already bad enough dealing with the irritating, nanny-style, compartmentalized, super-limited functionality available to NATIVE mobile apps, I'd be loathe to deal with yet another layer of dumb limitations and even slower mobile performance.
so, this is basically turning all web code into the moral equivalent of nonfree javascript? Count me out
>>53583059
when I said secure I meant on the users system not on the server. That's on the backend programmer
>>53583048
>"the web" is a nebulous concept that has no strict definition anymore,
No it fucking isn't. ANYTHING TO DO WITH NETWORKING IT'S CONSIDERED THE WEB.
>>53583056
>No one has to install anything, you can't see the value in that?
No. If my network goes down, your service is unavailable. It's slower than native apps too.
>>53583059
>they are also secure in general
Yeah good thing the iPhone will never get jailbroken through a web browser.
Oh wait.
>>53583059
>This is just so the retarded javascript programmers don't have a heart attack because they will be left out of the web assembly revolution
No, the GC stuff is legitimately useful. The DOM uses the JS GC, so if you want to interact with the DOM easily, you also need to be able to interact with the GC.
>>53583073
>People who care about being able to reliably step through code and debug it. Oh, look, the intermediary layer is acting up again. I guess I'll file a bug with the browser dev team and wait six months for the next major release.
First of all, the compiler being broken is a problem that could happen with literally any language. Web assembly has nothing to do with it and if web assembly implementation on a browser is broken enough that end users notice it they will update it right away
>So, the same limited experience you can get with, what, C# based tool?
Right now your C# tool is compiling to javascript. Think about that, you are compiling to fucking javascript. The performance of that is not good.
>No thanks. It's already bad enough dealing with the irritating, nanny-style, compartmentalized, super-limited functionality available to NATIVE mobile apps, I'd be loathe to deal with yet another layer of dumb limitations and even slower mobile performance.
There would be no slower mobile performance you idiot, you are compiling to the native language on every device (except ios in this case)
>>53583074
Most JS code is already nonfree, and hard to reverse engineer due to minification. Anyone who cares can compile their free C code to free WASM code, just the same as they compile free JS to free minified JS.
>>53583084
>ANYTHING TO DO WITH NETWORKING IT'S CONSIDERED THE WEB
No, that's the internet. "The Web" was originally intended to be pages on web servers that link to each other to form a web-like graph.
>>53583093
I thought what they were talking about there was that javascript coders could access and use web assembly data structures and code through an api. I don't know how the browser handles the dom though, I thought it would be separate from the JS engine but it seems not.
>>53583009
>>The web is doing what it's supposed to do, cry more old man. Now we need the ability for web sockets to do udp and I can die happy and rich.
But why? Just why? Why do web browsers need to be operating systems unto themselves? Why must otherwise perfectly usable hardware be invalidated because it can't keep up with the web's painfully absurd RAM requirements?
Seriously, I wouldn't hate the web even a quarter as much as I do now if web browsers and the sites they load weren't the bloated bog monsters they are now. Webassembly *can* change that, but I sincerely doubt it will; instead I think webdevs will abuse their newfound efficiency to pile hundreds more layers of pointless shit on.
>>53583084
>No. If my network goes down, your service is unavailable. It's slower than native apps too.
Don't forget that it's many times heavier and will always be stuck in lowest common denominator land, being entirely unable to take advantage of the strengths of any platform.
>>53583074
Well it's more like turning them into java .class file. Webasm is bytecode that runs in the javascript engines that already exist in web browsers.
>>53583133
Okay then, Mr Pedantic, anything that displays like a webpage and uses a native app to view another shitty, slow, hipster JS ladden app is "The Web" then. I know, you want to program in C and C++ too, Fox and the Grapes and all, but "The Web" isn't computing. Sorry.
>>53583156
>But why? Just why? Why do web browsers need to be operating systems unto themselves? Why must otherwise perfectly usable hardware be invalidated because it can't keep up with the web's painfully absurd RAM requirements?
Because we can, the web is the most secure way of viewing information and running applications and it's mostly instantaneous.
Honestly, I don't know why people complain about ram usage. I'm on 8 gigs of ram and have 80% usage with 3 browser windows open with 15 tabs on each.
Web browsers will keep evolving until they eventually become proper operating systems. Then some new network or technology will be developed and devs will develop "web browsers" for that thing and the cycle will continue once more.
>>53583199
>Because we can, the web is the most secure way of viewing information and running applications and it's mostly instantaneous.
I'm now convinced you're a troll. This has got to be a fucking joke.
>>53583218
I can assure you I'm not a troll. This isn't the fucking 00s. I don't need to worry about viruses from going on a web page anymore. If I want to play agar.io I don't need to download a program, I just go to the website and play. No thinking about where to store it or anything.
I'm not talking about security of information you put on the web. I'm talking about security of the computer form malware.
>>53583185
Okay? Nobody cares what you think is and isn't computing.
>>53583185
> "The Web" isn't computing. Sorry.
guess what? it is and it will be. No enjoy reading this from your browser with nojs you fucking hipster
>all the web devs in the thread instantly get ass blasted and try to rationalize webdev as the end all be all of computing
Kek
>>53583241
Yeah I'll just log into my FPS from my browser and have them stream the game frame by frame in JavaScript, hope I can still play if my internet goes out!
>>53583199
>Honestly, I don't know why people complain about ram usage. I'm on 8 gigs of ram and have 80% usage with 3 browser windows open with 15 tabs on each.
My main machine is a 2015 quad i7 rMBP w/16GB of RAM, so my reason is not because I'm running a machine that can't handle the modern web.
I simply think it's hopelessly wasteful. A Pentium 4 or PPC G4 machine can handle just about any modern task, save for the web due to the crazy amounts of RAM required. Core 2 Duo machines are comparable to smartphones and once again are beginning to have trouble with handling the web not due to lack of power, but due to lack of RAM.
This is really sad. There are thousands and thousands of machines out there that could be humming along serving people daily instead of sitting in a dump or being picked through for gold by some third world kid. Hundreds of thousands of machines' worth of materials that wouldn't have to be expended until many years later.
But web devs don't give a fuck.
http://everyfuckingwebsite.com
http://motherfuckingwebsite.com/
http://bettermotherfuckingwebsite.com/
>proprietary web sites will now really be a thing
Whoever supports this is retarded
Enjoy your built in ads that can't be removed
>>53583241
>I don't need to worry about viruses from going on a web page anymore
Ajajajajajajajajaj
Good Troll
So how long before every website starts executing actual virus payloads on your computer?
>>53581870
this thread:
>I don't know what webassembly is
>instead of trying to find out, I am going to get incredibly angry about what I incorrectly think webassembly is, and post about it
>>53583343
I'M NOT UNINFORMED, YOU'RE UNINFORMED!
>>53583317
>I have never bothered trying to look at real world js: the post
>>53583342
I'm seriously keking outloud thinking about how websites will start running all kinds of low level shit on our browsers.
inb4 all the webdev hipsters rush in and say "WELL IT'S NOT ACTUALLY LOW LEVEL HURRR, IT'S BYTECODE HURR"
>>53583371
Enjoy your built in ads that can't be removed
>>53583380
It's still the DOM and you can presumably still block resources.
>>53583380
>bytecode means no one could POSSIBLY block urls
>>53583402
>>53583403
Enjoy your built in ads that can't be removed
>>53583410
You don't actually know how ad blockers work do you?
>>53583420
NO ANON DON'T
>just abstracting away the hardware as a web browser, treating it as a single platform, with none of the benefits and all of the cons
I don't get it.
>>53583342
It would require knowledge of the browser's javascript sandbox to break free of it and start executing code on the host. Then it would need higher privileges to actually execute code outside of the user's own account. If the browser is run with very few privileges and under a jail or other kind of sandbox then most of the threat would be eliminated.
>>53583420
Cool, what if they embed the ads in the webasm that is required for the website to work properly?
And no, you won't be able to just use a JS alternative, that is only for a transition period until webasm compatible browsers are not as widespread jet
Enjoy your built in ads that can't be removed
>>53583449
>>53583449
just run it on windows, dummy
>>53583454
WHAT IF THEY EMBED THE ADS IN THE HTML ANON
>websites can now have bitcoin miners and other trash without you knowing
This will be great
>>53583464
Muh element hiding helper which wont work for webasm
>>53583466
google 'webassembly' anon
>>53583442
The entire OS basically becomes the web browser's "kernel."
>>53583464
Well you can block it then
But if it's in WebAsm you can't, because "BYTECODE"
Unless you think an addon going in between a:
>JavaScript engine
>JavaScript and WebAssembly comparability layer
>the WebAssembly interpreter or VM (you fucking hipsters keep changing it up)
>HTML
>CSS
>AND recompiling the same bytecode with ads removed, re injecting it properly into all of the above
Is efficient or is going to work.
>>53583479
>>53583478
>>53583502
you both really don't know how adblock works do you?
>>53583508
Have you noticed that it actually doesn't involve executing native binaries on client computers?
>>53583508
>in-browser client side scripting
yawn
>>53583529
>Have you noticed that it actually doesn't involve executing native binaries on client computers?
Not him, but yes, it involves executing native binaries on a web browser. Notorious for being secure, with no exploits, efficient, and standard conforming. This TOTALLY won't be another Java thing.
>>53583547
Well, no. I'm not the guy you're replying to, but you clearly haven't read anything.
>>53583547
wtf is this meme of browsers being insecure? there hasn't been a major browser exploit for the last 5 years
are all you niggas living in 2005?
>>53583561
Well then what IS WebAssembly, other than some new form of JavaScript targetting through LLVM?
If it's not powerful like a native executable, what's the point?
If it's not portable like JavaScript, what's the point?
Why must we reinvent the wheel?
>>53583571
https://www.microsoft.com/security/portal/threat/encyclopedia/entry.aspx?Name=Exploit%3aJava%2fCVE-2013-0422
>>53583547
>native
No anon, not native. That's important
web assembly is literally the fucking same as javascript, but lower level
wasm will execute in the browser on a sandboxed vm. It is no more or less secure than javascript
>>53583571
https://www.mozilla.org/en-US/security/advisories/
https://blog.mozilla.org/security/2015/08/06/firefox-exploit-found-in-the-wild/
http://www.theregister.co.uk/2015/09/04/mozilla_firefox_bugzilla_leak/
>>53583571
0/10 b8
>>53583571
https://www.cvedetails.com/vulnerability-list/vendor_id-26/product_id-9900/Microsoft-Internet-Explorer.html
https://www.cvedetails.com/vulnerability-list/vendor_id-452/product_id-3264/cvssscoremin-7/cvssscoremax-7.99/Mozilla-Firefox.html
https://www.cvedetails.com/vulnerability-list/vendor_id-49/product_id-2935/Apple-Safari.html
https://www.cvedetails.com/vulnerability-list/vendor_id-1224/product_id-15031/opec-1/Google-Chrome.html
>>53583589
THEN IT IS A NATIVE EXECUTABLE TO THE WEBASM VM
HOLY SHIT
YOU CAN'T ESCAPE JAVASCRIPT'S SANDBOX BECAUSE IT'S OLD AND HIGH LEVEL AS SHIT
REMEMBER JAVA DRIVE BYS?
>>53583589
You could say the same thing about java too. Even though you can theoretically debug the bytecode, nobody fucking does
>>53583583
Imagine javascript translated into something that can be mapped to machine code faster and more directly with more optimization than regular javascript can. Also, with the intention that other languages can have compilers that compile C or whatever to this bytecode.
Also, Java is a wholly separate stack from the browser. Exploits in that are very different.
>>53582109
>Who is idiotic enough to give fucks about "Stolen" javascript?
People who sell javascript based web apps, and the web devs who get a shoestring budget that does not allow them to buy a $150 responsive slider with 3d effects, but they have to implement it because the client point at it and told them "I want the site to do that".
>>53583628
So, how is this BETTER? This just seems SLOWER to me, not FASTER.
You mention "machine code" are you talking about the VM, or my actual CPU? There is a BIG difference.
>>53583583
>Well then what IS WebAssembly, other than some new form of JavaScript targetting through LLVM?
It's pretty much just ahead of time compilation for browser scripts.
>>53583646
It doesn't matter what it seems like to you.
And if you want to know more, you can read the extensive documentation or watch any of a number of presentations on youtube. But really I don't care whether or not you understand it.
>>53583668
So you just can;t answer my simple question then? Because all of the reading I've done so far makes it seem like it generates "machine code" for the VM. So now, instead of JavaScript just directly affecting the website, it has to go through a VM, and THEN does the same thing as JavaScript. How is this faster? Especially since it includes support for interfacing with JavaScript directly.
>>53583667
AOT compilation for hipster monstrosities that take up dozens of MBs of RAM. Welcome to the future.
>>53583701
It doesn't matter whether or not you think it's faster, because it's been tested and shown to be faster.
>>53583736
So is this bytecode for the VM or my CPU?
>>53582506
>PROPRIETY BINARY BLOBS RUNNING ON UNSECURED BLOAT MACHINESWHY IS NOBODY THINKING THIS IS A BAD FUCKING IDEA?
You don't have a problem when you do the same thing on your phone by downloading fart apps on it. So why would it matter for browsing?
>>53583748
What does this post have to do with anything?
>>53583747
The VM turns bytecode into machine code that runs on your computer in the same way that it turns javascript into machine code that runs on your computer, except faster.
>>53581927
>I refuse to believe the web is that taxing.
You better belive it, because it is.
Blame webdevs for it.
>>53583765
Okay, so what optimizations have they done that make it faster? It has to have some sort of reliance on the JavaScript engine since you can apparently import JavaScript elements in WebAssembly and vice-versa.
>>53583701
Runs in the same sandbox as javascript so the performance should at least be the same. In reality webasm is designed to be easier to interpret by the browser's scripting engine so it should always run faster than javascript. Javascript is just an awful language and runs like shit by default basically. Scripting engines have had to work around this problem by including compilers that compile particular parts of code during execution but compilation is resource intensive and really compiling every time the script runs is a waste of time. Webasm allows you to just ship and run the compiled code directly rather than compiling it each time.
>>53582590
>Hey, web devs. Yes, you!Why must you keep trying to rube goldberg what was initially a digital research paper into 500 ton application-wannabe monstrosities? Why must you keep trying to take the computer world's equivalent of a Geo Tracker and jerryrig it into pulling full length trailers?
Web dev here
Because we are paid to do it. Believe me, I hate those type of designs too, but if this is what I'm paid to build, then this is what I have to build.
>>53583785
So it's just a compiled version of a new JavaScript? And you guys are praising it like the second coming of Christ? Why? Why is the web dev world so behind?
>>53583773
>Blame webdevs for it.
and believe you me they will find a way to fuck this up too.
if that time machine i ordered arrives, mccool and berners-lee are biting it
>>53583803
Basically yeah. It's actually an intermediate language though, you can theoretically compile to it from any language including other interpreted languages. There's no actual requirement that you use javascript and I would highly recommend against it.
>>53583795
>Web dev here
>
>Because we are paid to do it. Believe me, I hate those type of designs too, but if this is what I'm paid to build, then this is what I have to build.
Why not move into iOS or Android dev or something?
Pays better, gives you better prospects for moving into other programming niches, and is generally far more sanity-preserving (UIKit is a fucking dreamworld compared to the hell of the front end web).
(I'm an iOS dev)
>>53583861
cause with html/css I got to the point that I can build sites that look and behave like native apps, and I only need Notepad for it, and perhaps an FTP client.
For Android/ios development I need a 5gb IDE just so I can start, and I'd be bumbling with the basics for months, plus I need a Mac and/or an iPhone/Android devics and register to whatever the fuck app store and apple ID I need just so I can test my app.
Cost of entry is too high, the system is locked down in retarded ways, and I have to give up 14 years of knowledge and experience and start from scratch.
>>53583803
It's more like shoe horning nodejs C bindings into the browser.
Think of it as the native C/Java sdks for Android. Most people won't use it, but it adds heaps for people that need to take advantage of hardware that Java/JavaScript (interpreted languages) cannot, while still maintaining the platform for those that do not need the power of native who opt for an easier (I.e quicker to market) language.
If you cannot objectively compare the trade-offs between using a powerful language like C++ and the simplicity of a language like JavaScript, you probably shouldn't be a dev. There isn't one language that solves EVERY problem under EVERY constraint, and you're gunna need a bigger toolbox to deal with issues in the future
I don't understand the hate really, just looks like C/C++ compiler for scripts.
It's like taking JavaScript and making it better by default.
>>53583999
Java is not an interpreted language.
>>53584100
>what is the JVM
Umad fanboy?
>>53584126
No, I'm not mad at all. I'm just pointing out that people don't interpret java code. You can make the argument that java bytecode is interpreted (which also isn't true on most jvm implementations) but no one uses an interpreter on the java language.
It's a common misunderstanding.
>>53584151
Righto m80, far from the point though
>>53584189
I agree. The post had a valid point, I just didn't agree with that statement.
Holy shit I didn't realise we have such high profile programmers on /g/. I mean guys who literally know better than the Google, Apple, Microsoft heads plus Eich. Holy shit, it puts everything in a different light. This is some high profile board here people. These guys must have built a lot of language standards in their lives.
kek
>>53582852
This
>tfw you'll finally get to post "This feature is not available for your Browser. Please upgrade to one of the following" messages
Can't wait for client side libraries to bootstrap in less than a second on mobile devices, too.
And developing in non-shit languages and having zero cost abstractions around the terrible browser APIs...
This is going to be delicious.
>WebAssembly-binary-blob
>transmitted encrypted via EME
adblockers btfo
>>53582152
There are native editors available, but Atom is one of the easiest ones to edit.
>>53582260
Microsoft, Google and Mozilla are working on it, I hope everybody else follows.
>>53582370
Webassembly will let you play pirated PC games online. It's not just about bootstrap.js and meme.js
>>53582594
>and you can still view source
You mean I can see the fucking assembly?
It's pretty much the same as running Java Applets, and that turned out to be a great idea, right?
We've finally gone past those, have a language that is readable by everyone viewing the site and we want to throw all that out, and go for a format that's unreadable.
For what purpose? What do we need the extra performance for, when we give up the freedom to read what's actually being run on your computer?
>>53582424
Too late. The web is now a universal application platform. And JavaScript will not let it.
>>53586919
You can view Assembly "pseudo code", not binary data.
If you don't give a crap about speed, stop complaining when a page takes more than 2 seconds to load, and don't visit WebGL websites.
>>53586939
>Assembly "pseudo code"
well that's just marvellous! Don't you love reading bytecode?
shadertoy.com runs just fine in my browser as of now, what'd anyone need more for?
>>53586956
Well, if you cannot understand it, nobody can help you with that.
https://github.com/WebAssembly/spec/tree/master/ml-proto/test
And why do you keep on hating on something you haven't bothered to investigate about?
2 years from now we're going to have to run fucking 3rd party anti malware software alongside our poor mangled browsers to vet the massive 3rd party blobs(try read an AST yourself you mongoloid webdev shithead) that achieve in the most horrific manner imaginable what the stack they are precariously built on top of was capable of for years.
Calling it now, browser AVs and all the stupidity associated with that business model.
>>53582590
Dude, developers are not the people planning projects.
>>53582637
i agree
>>53587341
6 months developing the ultimate malware!
4 seconds to block it based on the md5 signature.
Yes anon, Flash was created for malware.
>>53582506
>>53582566
In the real world, where most people live.
>this thread
OK here's the actual story so hopefully you morons will quit being enraged by your own ignorance.
There are three layers here:
>1) Javascript
Shit-tier language, but it's currently the only way to write standard (HTML+CSS+JS) webapps. Being a shitty scripting language, it requires massive effort on the part of the runtime system to achieve even vaguely tolerable performance. It's usually executed using a JIT compiler, which compiles JS to native machine code for execution.
JS runs in a sandbox, so it can't install viruses and shit onto your actual computer. (Though of course there's nothing to stop you from writing a pure-JS bitcoin miner, if you want.) The sandbox works pretty well, these days.
>2) asm.js
A subset of JS, designed for high-performance JIT or ahead-of-time (AOT) compilation. Compiling asm.js with the normal JIT asm.js produces fast native code because asm.js is written in a really ugly style, with lots of weird useless constructs that act as type hints to the JIT. AOT compilation is even better for performance, since it bypasses the normal JS interpreter and JIT, and just produces native code directly, making asm.js a sort of intermediate representation between LLVM IR and x86/ARM/etc. machine code.
asm.js is still JS, and is still sandboxed the same way. The only difference is it uses a different (simpler) code path to produce the native code. So the only possible asm.js-specific exploits would be ones due to bugs in the native code generator. (And historically, there have not been many vulnerabilities in the JIT native code generator, which is similar.)
>3) WASM
Literally nothing but a binary encoding for asm.js. Go read the asm.js section again, everything there applies to WASM as well. The only thing that's really changing is more browsers adding support for AOT compilation.
TL;DR nothing is fucking changing, chill the fuck out
>>53588555
>TL;DR nothing is fucking changing, chill the fuck out
But it is, instead of plaintext JS, websites are now sending me binary blobs.
>>53586838
>This feature is not available for your Browser
Unlikely to happen. There's already a polyfill that converts WASM -> asm.js, and asm.js runs everywhere because it's just funny-looking JS.
>>53586862
>Javascript-text-blob
>transmitted encrypted via EME
adblockers btfo
I mean come on, we already had this discussion while EME was getting railroaded through the W3C.
>>53586919
>You mean I can see the fucking assembly?
Yeah, you can see the fucking assembly. Just the same way you can see the minified asm.js code (completely fucking unreadable), or minified javascript (also practically unreadable).
The kinds of sites that currently use unminified javascript, will most likely continue to use unminified javascript.
>>53587341
Please consider reading literally anything about the subject before you start posting. WASM runs in the same sandbox as javascript, the only difference is speed.
>>53588649
Have you ever looked at that "plaintext" JS? As bad as that is, asm.js is far worse, and that's what WASM is replacing. If anything, it looks like disassembled WASM will be slightly easier to read than normal asm.js, if only because all the +(...), (...)|0, HEAP32[...] noise will be gone.
japs have been doing this shit for years
