www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - WebAssembly design is done?

reply Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
«WebAssembly CG members representing four browsers, Chrome, Edge, 
Firefox,
and WebKit, have reached consensus that the design of the initial 
(MVP [1])
WebAssembly API and binary format is complete to the extent that 
no further
design work is possible without implementation experience and 
significant
usage. This marks the end of the Browser Preview and signals that 
browsers
can begin shipping WebAssembly on-by-default. From this point 
forward,
future features will be designed to ensure backwards 
compatibility.»

https://lists.w3.org/Archives/Public/public-webassembly/2017Feb/0002.html

Cool.
Feb 28
parent reply Jack Stouffer <jack jackstouffer.com> writes:
On Tuesday, 28 February 2017 at 18:04:39 UTC, Ola Fosheim Grøstad 
wrote:
 «WebAssembly CG members representing four browsers, Chrome, 
 Edge, Firefox,
 and WebKit, have reached consensus that the design of the 
 initial (MVP [1])
 WebAssembly API and binary format is complete to the extent 
 that no further
 design work is possible without implementation experience and 
 significant
 usage. This marks the end of the Browser Preview and signals 
 that browsers
 can begin shipping WebAssembly on-by-default. From this point 
 forward,
 future features will be designed to ensure backwards 
 compatibility.»

 https://lists.w3.org/Archives/Public/public-webassembly/2017Feb/0002.html

 Cool.
What a long, roundabout path we've taken to end up back where we started: shipping binaries in a sandboxed environment.
Feb 28
next sibling parent reply Joakim <dlang joakim.fea.st> writes:
On Tuesday, 28 February 2017 at 21:50:11 UTC, Jack Stouffer wrote:
 On Tuesday, 28 February 2017 at 18:04:39 UTC, Ola Fosheim 
 Grøstad wrote:
 [...]
What a long, roundabout path we've taken to end up back where we started: shipping binaries in a sandboxed environment.
What is the alternative you prefer: javascript and its myriad vulnerabilities?
Feb 28
parent reply "Nick Sabalausky (Abscissa)" <SeeWebsiteToContactMe semitwist.com> writes:
On 02/28/2017 10:28 PM, Joakim wrote:
 On Tuesday, 28 February 2017 at 21:50:11 UTC, Jack Stouffer wrote:
 On Tuesday, 28 February 2017 at 18:04:39 UTC, Ola Fosheim Grøstad wrote:
 [...]
What a long, roundabout path we've taken to end up back where we started: shipping binaries in a sandboxed environment.
My new favorite quote :)
 What is the alternative you prefer: javascript and its myriad
 vulnerabilities?
Probably shipping binaries in a sandboxed environment...but, you know, *without* the browser community badly reinventing the entire software stack.
Feb 28
parent reply "Nick Sabalausky (Abscissa)" <SeeWebsiteToContactMe semitwist.com> writes:
On 02/28/2017 11:26 PM, Nick Sabalausky (Abscissa) wrote:
 On 02/28/2017 10:28 PM, Joakim wrote:
 What is the alternative you prefer: javascript and its myriad
 vulnerabilities?
Probably shipping binaries in a sandboxed environment...but, you know, *without* the browser community badly reinventing the entire software stack.
For that matter, wasn't one of the two key points to webapps in the first place the promise of thwarting vulnerabilities by using the browser as a sandbox? Oh! Why yes, it was!
Feb 28
parent reply Saurabh Das <saurabh.das gmail.com> writes:
I'm not so up-to-date about the mechanics of WebAssembly, but it 
would be pretty exciting to run D code in the browser.

Is this now possible or have I completely misunderstood what 
WebAssembly allows for?
Mar 01
parent reply Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
On Wednesday, 1 March 2017 at 08:01:41 UTC, Saurabh Das wrote:
 I'm not so up-to-date about the mechanics of WebAssembly, but 
 it would be pretty exciting to run D code in the browser.

 Is this now possible or have I completely misunderstood what 
 WebAssembly allows for?
It should be possible, at least if you give up on the garbage collector.
Mar 01
parent reply "Nick Sabalausky (Abscissa)" <SeeWebsiteToContactMe semitwist.com> writes:
On 03/01/2017 03:47 AM, Ola Fosheim Grøstad wrote:
 On Wednesday, 1 March 2017 at 08:01:41 UTC, Saurabh Das wrote:
 I'm not so up-to-date about the mechanics of WebAssembly, but it would
 be pretty exciting to run D code in the browser.

 Is this now possible or have I completely misunderstood what
 WebAssembly allows for?
It should be possible, at least if you give up on the garbage collector.
Huh? Isn't webasm GC'ed? Ir is it like a D-vs-wasm CG incompatibility?
Mar 01
parent reply Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
On Wednesday, 1 March 2017 at 08:56:37 UTC, Nick Sabalausky 
(Abscissa) wrote:
 On 03/01/2017 03:47 AM, Ola Fosheim Grøstad wrote:
 On Wednesday, 1 March 2017 at 08:01:41 UTC, Saurabh Das wrote:
 I'm not so up-to-date about the mechanics of WebAssembly, but 
 it would
 be pretty exciting to run D code in the browser.

 Is this now possible or have I completely misunderstood what
 WebAssembly allows for?
It should be possible, at least if you give up on the garbage collector.
Huh? Isn't webasm GC'ed? Ir is it like a D-vs-wasm CG incompatibility?
Nope, no GC. It is a "minimum viable product", so also no threading or SIMD. This is a good strategy, make sure that all implementations are fully compliant and collect experience before adding more features into the mix. http://webassembly.org/docs/gc/ I believe both Rust and C++ compilers will target WebAssembly. In the beginning I suspect the best approach is to the core application in a set of WebAssembly modules and tie it together with the UI in JavaScript/TypeScript/JSX/Angular2 etc
Mar 01
parent Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
This looks like a nice tool for those wanting to learn more about 
WebAssembly:

https://mbebenita.github.io/WasmExplorer/
Mar 07
prev sibling parent reply Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
On Tuesday, 28 February 2017 at 21:50:11 UTC, Jack Stouffer wrote:
 What a long, roundabout path we've taken to end up back where 
 we started: shipping binaries in a sandboxed environment.
Portable object-files is something that that people have tried to achieve since the 1980s, but vendors have always undermined it. If this goes through then it will be the first time we have a portable object files with backing of the major vendors. Even Microsoft. The Java VM essentially ended up as a single vendor solution and didn't live up to it's hype in the browser.
Mar 01
parent reply "Nick Sabalausky (Abscissa)" <SeeWebsiteToContactMe semitwist.com> writes:
On 03/01/2017 03:46 AM, Ola Fosheim Grøstad wrote:
 On Tuesday, 28 February 2017 at 21:50:11 UTC, Jack Stouffer wrote:
 What a long, roundabout path we've taken to end up back where we
 started: shipping binaries in a sandboxed environment.
Portable object-files is something that that people have tried to achieve since the 1980s, but vendors have always undermined it. If this goes through then it will be the first time we have a portable object files with backing of the major vendors. Even Microsoft.
If we could invent a technical way to screw over, underminine, and completely replace corporate interests, it would be the single greatest achievement in computing (not to mention economic theory), EVER. Believe it it not, I'm still holding my breath for that, impossible though it may be. Call me a dreamer. Or a nutjob ;) Same thing, I suspect ;)
Mar 01
parent reply Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
On Wednesday, 1 March 2017 at 09:00:42 UTC, Nick Sabalausky 
(Abscissa) wrote:
 If we could invent a technical way to screw over, underminine, 
 and completely replace corporate interests, it would be the 
 single greatest achievement in computing (not to mention 
 economic theory), EVER.
I bet Intel would have undermined WebAssembly if they were given a chance. Now other CPU vendors (e.g. mobile) can strengthen their position by making WebAssembly applications a target for testing. So, maybe the conclusion is: don't let corporations create technology standards in their main domain, because then they try to create monopolies and start misbehaving. Browsers is a side-line tech, so... this might have a chance..?
Mar 01
parent reply "Nick Sabalausky (Abscissa)" <SeeWebsiteToContactMe semitwist.com> writes:
On 03/01/2017 05:15 AM, Ola Fosheim Grøstad wrote:
 So, maybe the conclusion is: don't let corporations create technology
 standards in their main domain, because then they try to create
 monopolies and start misbehaving. Browsers is  a side-line tech, so...
 this might have a chance..?
Well, it's Google's main domain, and they've kinda already settled into a pattern of making decisions more on self-serving grounds than for the good of the product/users. "Don't be evil" doesn't exactly set a very high bar.
Mar 01
parent reply Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
On Wednesday, 1 March 2017 at 18:28:00 UTC, Nick Sabalausky 
(Abscissa) wrote:
 Well, it's Google's main domain, and they've kinda already 
 settled into a pattern of making decisions more on self-serving 
 grounds than for the good of the product/users. "Don't be evil" 
 doesn't exactly set a very high bar.
I don't know, I have more trouble with Safari and Edge than Chrome and Firefox. Google's main domain is advertising, same with Facebook. Software is just a way to keep competition away and collecting information about users. The Internet is getting very 1984ish, but this goes way beyond Google, which I find to be better than average. Have you noticed how you receive advertising all over the Internet for the same product you looked at a few days before in a webshop and how Facebook and Linked In lists suggestions based on people you have only had peripheral interaction with? Extremely annoying. It makes me rank those companies as shady. At some point there will be a resistance movement, forking one of the main browsers and building in collaborative blacklisting etc. Unfortunately, non-tech people are quite oblivious to what is going on. Maybe the new collaborative website annotation standards will turn into something. Firefox could take a lead there.
Mar 02
parent reply "Nick Sabalausky (Abscissa)" <SeeWebsiteToContactMe semitwist.com> writes:
On 03/02/2017 04:18 AM, Ola Fosheim Grøstad wrote:
 On Wednesday, 1 March 2017 at 18:28:00 UTC, Nick Sabalausky (Abscissa)
 wrote:
 Well, it's Google's main domain, and they've kinda already settled
 into a pattern of making decisions more on self-serving grounds than
 for the good of the product/users. "Don't be evil" doesn't exactly set
 a very high bar.
I don't know, I have more trouble with Safari and Edge than Chrome and Firefox.
Well, web devs cater to Chrome and FF, and tend to ignore Safari/Edge. And I wasn't really just talking about browsers with that.
 The Internet is getting very 1984ish, but this goes way beyond Google,
 which I find to be better than average. Have you noticed how you receive
 advertising all over the Internet for the same product you looked at a
 few days before in a webshop and how Facebook and Linked In lists
 suggestions based on people you have only had peripheral interaction
 with? Extremely annoying. It makes me rank those companies as shady.
Maybe it's ADD-related or something, but my brain literally isn't capable of reading a page of text if there's something animating on the page (usually ads). Even the blinking cursor in a code editor can break my focus. So I've had to install Adblock Edge and NoScript[1] just to be *able* to use the web at all (basic, honest-to-goodness accessibility). So, no, I honestly haven't noticed that phenomenon (although I have heard about it once before). [1] Back when I was using Windows more, that Adblock Edge/NoScript combo also had the additional benefit of keeping my machine much safer from drive-by malware, even when other people around me were far more careful about good antivirus, never turned off auto-updates, and were still having their machines taken over by ransomware - which never touched any of my machines.
 At some point there will be a resistance movement, forking one of the
 main browsers and building in collaborative blacklisting etc.
I hope, but I'm skeptical. Big business is definitely headed very 1984, but that's happening less because of Orwellian control, and more because of mass apathy and widespread short-sighted self-interest (more Huxley than Orwell, from what I gather). I also see certain issues with collaborative rankings - they can only be as intelligent as the average user, which often doesn't seem to be very much. And then there's other difficulties like this: https://xkcd.com/937/
Mar 02
parent reply Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= writes:
On Thursday, 2 March 2017 at 19:52:58 UTC, Nick Sabalausky 
(Abscissa) wrote:
 At some point there will be a resistance movement, forking one 
 of the
 main browsers and building in collaborative blacklisting etc.
I hope, but I'm skeptical. Big business is definitely headed very 1984, but that's happening less because of Orwellian control, and more because of mass apathy and widespread short-sighted self-interest (more Huxley than Orwell, from what I gather).
The general public has become more ignorant. I guess to a large extent because of information overflow and the downfall of real journalism (e.g. the old payment model is failing which means media is converging on click-bait-ad-sales). A news story has to run for several weeks now for anyone to take notice, otherwise it will just drown in all the irrelevant noise (celebrity news and what not).
 I also see certain issues with collaborative rankings - they 
 can only be as intelligent as the average user, which often 
 doesn't seem to be very much. And then there's other 
 difficulties like this: https://xkcd.com/937/
I think it would be more like organizations like Amnesty International and consumer rights organizations having the ability to annotate websites and webpages with vetted information about the content.
Mar 03
parent "H. S. Teoh via Digitalmars-d" <digitalmars-d puremagic.com> writes:
On Fri, Mar 03, 2017 at 08:06:00AM +0000, Ola Fosheim Grstad via Digitalmars-d
wrote:
[...]
 The general public has become more ignorant. I guess to a large extent
 because of information overflow and the downfall of real journalism
 (e.g.  the old payment model is failing which means media is
 converging on click-bait-ad-sales). A news story has to run for
 several weeks now for anyone to take notice, otherwise it will just
 drown in all the irrelevant noise (celebrity news and what not).
[...] Ahahaha... "celebrity news"... an oxymoron, if there ever was one. T -- Programming is not just an act of telling a computer what to do: it is also an act of telling other programmers what you wished the computer to do. Both are important, and the latter deserves care. -- Andrew Morton
Mar 03