WebAssembly - A Big Step Towards a Free Web

File: 1478302892554.jpg (113KB, 960x960px) Image search: [Google]

113KB, 960x960px

>>59302971

>Memesters not welcome. Thanks.

Shouldn't you be deleting this thread then?

>>59302998
I don't know how to do that.

>>59302998
Never mind, just found out.

I'll suck any mode who delete this thread!

>>59303022
wtf donut do nat

>>59302971
how the FUG is it BETTER to make it MORE confusing to program and LITERALLY IMPOSSiBLE to read the web code?
fugging yai lets make all websites in FUGGING BINARY SO NO ONE CAN FUGGING READ THE FUGGGING CODE ITS COMPLETELY LOCKED SO THEY CAN PUT THEIR VIRII IN IT

>>59302998

whos this seamen demon

>>59303313
check dis out boi
. e x e

Does this mean the end of big web servers? Will everything be run client side in the browser now?

>>59302971
>The wasm code is distributed in a binary format that can easily be translated to a not so friendly text mode.[2]

>As I see, that is not a big problem, since independently of the license, JS can also be obfuscated to oblivion.
According to the great Stallman, JavaScript is considered non-free if the code is obfuscated and therefore wasm is hereby granted botnet status.

YOU"LL HAVE TO FUGGING RUN A FUGGING DISASSEMBLER ON EVERY FUGGING WEBSITE JUST SO YOU CAN FUGGING MAKE A MINOR COSMETIC CHANGE

File: 1334546485738.jpg (47KB, 193x245px) Image search: [Google]

47KB, 193x245px

>>59302971
>The wasm code is distributed in a binary format that can easily be translated to a not so friendly text mode

>>59303384
YEAH ITS CALLED A FUGGING DISSASSEMBLY

EVER LOOK AT ONE???

THIS IS THE FUTURE OF YOUR WEBSITES
LFF20
lda $01
lsr
bcs LFF34
lsr $03
lsr $03
lsr $03
lsr $03
jmp LFF36

; ----------------
LFF30
lsr $03
lsr $03
LFF34
inc $04
LFF36
lda $0471,y
ora $03
sta $0471,y
inc L0000
inc L0000
ldx $01
inx
cpx #$0D
bcc LFED3

>>59303368
True dat mah nigga

OH LSR $03, OF COURSE!!!!! IT MAKES SO MUCH SENSE I AM GLAD WE ARE USING WEB ASSEMBLY OVER ANYTHING ELSE

>>59303384
>>59303356
>>59303313
I'm happy there are no stupid 4chan users.

Proprietary web pages ARE ALREADY A PROBLEM. WebAssembly WILL NOT make anything possible that wasn't before.

>>BUT<< NOW WE CAN PUSH A STANDARD WAY TO FILTER PROPRIETARY SOFTWARE.
What doesn't exists nowadays. LibreJS doesn't work well.

>>59303427
but what about fast performance? javascript's slow as fuck sometimes

>>59303479
dum tripfig cia nigga shill

>>59303489
get faster computers
we all know intel just delays the processors for increased sales
it's all a big scam

>>59303494
I don't think you understand my post. I'm pushing for free software. That's our chance to be able to create a proprietary filter for the we.

>>59303522
for the web*

This is some Poe's law tier posting, I can only pray it's ironic but it's probably not.

>>59303522
well i don't think you understand my post. ;^)

Can we just use js?

>>59303605
JS is a shit language. It already has obfuscated proprietary software and we have no good way to block it.

>>59303649
It's not that bad, later versions are quite promising.

>>59303722
OK. So let's focus on the fact that there is no easy way to filter JS licenses, ok? WebAssembly is our chance to make it happen.

>>59303807
spin it how u want
i see through ur game

>>59303807
u have to spin and manipulate to try to make your point, when all i gotta do is tell le truth

u r a cia nigga my friend

go honest my friend

>>59303822
>>59303843
Is it summer already?

IF U FUGGING WANTED TO FILTER FUGGING LICENSES U WOULD HAVE A GOOD IDEA NOT THIS CIA NIGGAFIGICATION OF ADDINOGU FI(#FUGGING ASSEMBLY TO THE WEB


THIS IS WHAT YOU WANT
bank7_code51
sty L0000
lda #$00
sta $C8,y
ldy #$03
LF286
lda $4D,x
clc
adc LF274,y
sta L000E
lda $3B,x
adc #$00
sta $0F
lda L000E
cmp $072C
lda $0F
sbc $072A
beq LF2AF
sty $01
lda LF270,y
ldy L0000
ora $C8,y
sta $C8,y
ldy $01


FUG OFF

>>59303862
Adding webassembly to the web was not my idea. It has been pushed forwards for at least two years now. It has google, mozilla and microsoft behind it.
It WILL succeed. I'm trying to make something good with it.

Also, it uses a text format that is much more friedly than what you posted.

>>59303807
We could make a requirement to serve sourcemaps for GPL compliance. As for libraries, npm and bower both support license field in package description. Most libraries use some sort of permissive license such as MIT or ISC.

>>59303907
>We could make a requirement to serve sourcemaps for GPL compliance
How would you enforce this on everyone using JS?

>As for libraries, npm and bower both support license field in package description.
Yes, but how would the end user see this within the browser and automatically filter certain licenses.

>>59303906
post friendly text

>>59303951
https://github.com/WebAssembly/design/blob/master/TextFormat.md

>>59303411
H4CK TH3 PL4N3T

>>59303906
>I'm trying to make something good with it.
You're living in denial by trying to spin a negative thing into a posative one, more over you're trying to convert others to think the same way, this is harmful, intentional or not.

Do not spin or warp the truth, accept that it is bad and cope with it or figure out a means with which to deal with it, do not try and convince yourself that it is alright, especially do not try to impose that belief on others.

>>59303965
get_local 0
i64.const 0
i64.eq
if i64
i64.const 1
else
get_local 0
get_local 0
i64.const 1
i64.sub
call 0
i64.mul
end

FUG YOU

ONCE AGAIN A TRIPFIG PROVES HIS IGNORANCE

>>59303950
Oh, it's simple.
If script is minified or generated and there are no sourcemaps, source is closed.

watch how easy it is for me
what do you want to read

<tbody><tr>
<th>C++</th>
<th>Binary</th>
<th>Text</th>
</tr>
<tr>
<td rip-style-borderwidth-backup="" rip-style-borderstyle-backup="" style="" rip-style-bordercolor-backup=""><pre>int factorial(int n) {
if (n == 0)
return 1;
else
return n * factorial(n-1);
}</pre></td>
<td><pre>

OR

get_local 0
i64.const 0
i64.eq
if i64
i64.const 1
else
get_local 0
get_local 0
i64.const 1
i64.sub
call 0
i64.mul
end

>>59303987
>figure out a means with which to deal with it
That's literally what I'm doing.

The fact that it's distributed in binary format is not bad. Free software is usually distributed in binary format.

JS, even being interpreted can be obfuscated to oblivion. The amount of code available under a certain type of license will not change with webassembly.

We have the chance to give the user the power to choose. A browser extension coul pop up and show a message: This is botnet, proceed?

>>59304011
Ok, but how do we enable the user to chose which license to use and which not without a analysis of every js library?

>>59303991
I said it was more friendly not that brainlets could grasp it.

>>59303950
How would you enforce it with assembly when everything will be merged into one blob?

or even better
WEBASSEMBLY WANT TO GO FROM THIS
int factorial(int n) {
if (n == 0)
return 1;
else
return n * factorial(n-1);
}

TO THIS

get_local 0
i64.const 0
i64.eq
if i64
i64.const 1
else
get_local 0
get_local 0
i64.const 1
i64.sub
call 0
i64.mul
end

>>59304051
u have no arguement

>>59304045
Dear imbecile,

Your legible code could be obfuscated until it's harder to read than the assembly.
There already is proprietary JS on the web. The fact that you can open a file and see symbols you are familiar with, even if they don't form a single word, does not mean it's open source or free software.

>>59304051
You're not explaining how you are dealing with anything and the things you are saying are no different than the current alternatives, there are plenty of means with which to filter javascript already, even if you manage to convince people this will be done on the same granular level with web assembly it still will not be a "good" thing, at best it is the same, not better in any way and potentially worse since there are no guarantees it will be on par with the current solutions.

You make no point in any direction, your posts are useless fluff and "what if"'s.

>>59304078
use ALL your logical fallacies

you have already been exposed, my cia nigga

>>59304053
I'm pushing for a extension on the file that would specify the license used.

Something like, the first 1kb, if present, of every file is the license name and source url. This would be part of the language. The user would be able to decide which licenses to accept and which not.

>>59304100
>I'm pushing for a extension
What does this mean? You've made an official proposition or developed something yourself or you just wish one existed?

>>59304091
>there are plenty of means with which to filter javascript
Which would those be?

Also note that I didn't invent webassembly. It is going to happen you like it or not. And even if there was any means to license filter JS, there will be none in wasm, because it's note distributed as a text file.

>>59304121
I did. I posted the link in the OP. I opened a issue in the repo and I'm discussing it with the main developers.

>>59304142
>I'm discussing it with the main developers.
Maybe a exaggerated, I'm discussing with some developers. I posted some hours ago, few people saw it.

Ayo 'hol up
So what you be telling me
Is that I can write a C-to-WebAssembly transpiler, ship that to the end-user, and operate my entire website using C instead of godawful fucking Javascript?

>>59304142
>Which would those be?
Most browsers have a built in settings which can disable it altogether, every modern browser including ie has a plugin system that allows you to run user made plugins, many of which exist for various levels of filtering javascript based on usermade rules, a popular one is ublock, another being noscript, but there are more than that. I know of no webassembly filter systems, you haven't suggested any that give at least the same functionality.

>>59304051
Again, it's simple, and I think modern browser extensions even let you do it. You disable all scripts and then add them to white list after a code review. It's not convenient, but it's more useful than obeying a "I'm a good script" marker that you otherwise suggest.

>create proprietary wasm
>claim it's ISC licensed
>bypass filter
lel

>>59304205
Oh god.

There is no easy way to filter JS based on license. Because there is no standard way to inform license in JS, You can use filter to try to guess the licenses, but it's not 100%, in fact it's much less than that. ublock and noscript were not meant to filter licenses and they do not do it well. The best solution is libreJS but it's awful.

I'm pushing for the wasm SPEC to have license field, then any bowser can filter in a simple manner.

To make it more clear. Wasm is going to happen. It has major companies behind it.

>>59304230
Now you can sue them and get the source for the code.

>>59304228
It's not simple. Far from that. If the developer doesn't follow the license, you sue them.

>>59304228
Also, wasm will be a thing, you like it or not.

>>59304291
You never specified license with me, and neither did I. Why this arbitrary filter rule now? More importantly how does your proposal make any difference when this is possible >>59304230

You argue that filtering by license in JS is not 100% but it would be just as inaccurate in wasm too, no? I see no advantage, in addition how are you supposed to deal with multiple licenses when compiling multiple licensed components into a single blob?

>>59304291
License or not it does nothing to prevent malicious code executing on your machine, it shouldn't be assumed that all GPL, MIT et al. licensed software is non-harmful.

>>59304375
1. Yes, they help to prevent that;
2. That's not the point. The point is give the user the ability to choose what licenses to use.

>>59304344
Ok, I cant tell who is who.

Yes my proposal makes some difference because now it's easier to sue people who infringe the license.

>but it would be just as inaccurate in wasm too, no?
No. There is no standard way to specify the license in JS. I'm proposing a standard way to do so in wasm.

>I see no advantage, in addition how are you supposed to deal with multiple licenses when compiling multiple licensed components into a single blob?
You link them together and distribute many binaries. Exactly like it's done with JS, or any other language.

>>59304456
>Yes, they help to prevent that
Please elaborate.

>No. There is no standard way to specify the license in JS. I'm proposing a standard way to do so in wasm.
I don't see how this helps at all, just because you have a field for people to fill in doesn't mean they will or will do so correctly. What prevents a malicious person from filling in the wrong license specifically to dodge filters? People have been using fowl tactics and trying to break conventions for a long time which is why people dislike these technologies in the first place, you can't just assume everyone has good intentions and will comply just because something is standard, the license field may as well be useless if there's no way for users to automatically validate such a thing.

>>59304569
>Please elaborate.
If you have access to the source you can study it. It's not 100% but is much better than nothing.

>What prevents a malicious person from filling in the wrong license specifically to dodge filters?
The same thing that prevent them from doing so in any other scenario. If anyone finds out, you sue them.

It also means your browser can automatically download and compile the source if you want.

If they don't use the field, the user is free to decide what to do, being absolutely sure that no license was defined and there is no misunderstanding.

I don't really see what your point is. What benefits would not including the license field bring? Now proprietary software would work the same way as in JS but there would be no easy way to link the source for FOSS.

>>59304665
You really are trying to trick people aren't you? That kind of response is just terrible.

>>59302971
OP fighting the good fight

>>59302998
I don't dislike 3D anymore.