You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@flex.apache.org by Jason Taylor <ja...@dedoose.com> on 2017/04/10 19:17:35 UTC
WebAssembly Flash ByPass
WebAssembly is officially supported in both Chrome and Firefox right now, and in the rest very soon.
WebAssembly can take LLVM and compiles to a binary format that works cross browser at near native speeds without any plugin.
LightSpark is an open source implemention of the flash player written in C++ that can compile to LLVM. Potentionally we can compile LightSpark to LLVM and be able to run all flash apps in all browsers without any plugin and completely bypass the death of the flash player. Thoughts?
~ JT
RE: WebAssembly Flash ByPass
Posted by Jason Taylor <ja...@dedoose.com>.
yes because LLVM to LLVM is sooooo difficult... You have actually have anything constructive to contribute?
-----Original Message-----
From: Gary Yang [mailto:flashflexpro@gmail.com]
Sent: Monday, April 10, 2017 1:09 PM
To: dev@flex.apache.org
Subject: Re: WebAssembly Flash ByPass
Yeah the life in Mars could be very exciting, let me know when you are there ...
On Mon, Apr 10, 2017 at 3:17 PM, Jason Taylor <ja...@dedoose.com> wrote:
> WebAssembly is officially supported in both Chrome and Firefox right
> now, and in the rest very soon.
> WebAssembly can take LLVM and compiles to a binary format that works
> cross browser at near native speeds without any plugin.
> LightSpark is an open source implemention of the flash player written
> in
> C++ that can compile to LLVM. Potentionally we can compile LightSpark
> C++ to
> LLVM and be able to run all flash apps in all browsers without any plugin
> and completely bypass the death of the flash player. Thoughts?
> ~ JT
>
>
Re: WebAssembly Flash ByPass
Posted by Alex Harui <ah...@adobe.com>.
I still don't like these references to aliens. Please, let's focus on the
technical aspects.
IMO, Apache Flex is about components and not about platforms. That is why
FlexJS is attempting to be platform agnostic. It supports SWF and
HTML/JS/CSS today. Maybe some other platform in the future.
Apache projects are about communities of people with similar interests. I
personally would not want to work on any emulation of Flash. The problem
space is too big. There will be expectations that non-Flex content will
run on these emulations. When something goes wrong, you can't just use an
AS-only project or JSFiddle to test whether it is Flex code or the
platform. Once you are responsible for a platform, you have to deal with
lots more configuration issues that are abstracted away from Flex.
But that doesn't mean that folks can't discuss these topics here, although
extended discussion should probably happen in the communities that are
actually building out the code (W3C, skia, Flutter). And those
communities are welcome to discuss their work here in order to draw
attention and recruit for their communities. And I wish all of them luck,
and hope that our components work on their platforms.
My 2 cents,
-Alex
On 4/11/17, 7:53 AM, "Gary Yang" <fl...@gmail.com> wrote:
>Great! when you guy enjoying wherever it is, please do mind people still
>on
>earth can't bypass FP, begging for next release of Flex.
>
>On Tue, Apr 11, 2017 at 6:48 AM, Dev LFM <de...@gmail.com> wrote:
>
>> It's a known fact that Flutter is already doing that, only releases at
>>the
>> moment for ios/android, but I believe that in a 1 year span it will be
>> possible to deploy to desktop too...
>>
>> No comments? too much work? Am I living on Jupiter? ^^
>>
>> 2017-04-11 10:35 GMT+01:00 Dev LFM <de...@gmail.com>:
>>
>> > If drawing on gpu is possible, I would go even further and also add
>>skia
>> > to the discussion.
>> >
>> > My vision is, use skia to render everywhere.. Having then different
>> > "falcon" to compile for webgl/ios/android/win/osx/unix/raspi
>> >
>> > Now.. what would be needed? I think a transpiler of "stage3d" to
>>"skia"
>> > could be a 1st attempt..
>> >
>> > Thoughts?
>> >
>> > 2017-04-11 7:32 GMT+01:00 Alex Harui <ah...@adobe.com>:
>> >
>> >> Gary,
>> >>
>> >> Please try to focus on actual technical issues and let those who
>>want to
>> >> try, try it. Some people think making Flex/FlexJS popular again is a
>> long
>> >> shot as well.
>> >>
>> >> I'm not that familiar with WebAssembly, but from what I've read
>>today, I
>> >> think Josh has hit on the main technical challenge, which is that
>> >> WebAssembly doesn't appear to let you draw directly to the screen.
>>It
>> >> sort of reminds me of Alchemy/FlasCC.
>> >>
>> >> -Alex
>> >>
>> >> On 4/10/17, 7:26 PM, "Gary Yang" <fl...@gmail.com> wrote:
>> >>
>> >> >Thanks, I would love to see how you can get there, W3C started
>> designing
>> >> >HTML5 in 2009, Firefox is still making exception for Flash Player
>>now,
>> >> >maybe you can help them get rid of it and unify the web truly.
>> >> >
>> >> >On Mon, Apr 10, 2017 at 5:24 PM, Jason Taylor <ja...@dedoose.com>
>> wrote:
>> >> >
>> >> >> Gary, do you even have any idea what WebAssembly is? You should
>>check
>> >> >>out
>> >> >> some articles on it:
>> >> >>https://na01.safelinks.protection.outlook.com/?url=https%
>> >> 3A%2F%2Fgithub.c
>> >> >>om%2FWebAssembly%2Fdesign%2Fblob%2F&data=02%7C01%7C%7C9587
>> >> 71644c3444d5448
>> >> >>308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%
>> >> 7C63627474370530
>> >> >>2509&sdata=roF8%2B3Z2KOZAW7NVE8QJ3OSFAraX8ViFl179cVi%
>> >> 2FgJY%3D&reserved=0
>> >> >> master/FAQ.md,
>> >> >>https://na01.safelinks.protection.outlook.com/?url=https%
>> >> 3A%2F%2Fdevelope
>> >> >>r.mozilla.org%2Fen-US%2Fdocs%2FWebAssembly%2F&data=02%
>> >> 7C01%7C%7C958771644
>> >> >>c3444d5448308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1
>> >> %7C0%7C0%7C6362
>> >> >>74743705302509&sdata=sf%2FmlA5b%2FvU79NgnjPu1tdhbF3sJ1FAoQ
>> >> ahTmV06%2FAc%3D
>> >> >>&reserved=0
>> >> >> Concepts
>> >> >>
>> >> >> In a nutshell the browser vendors finally realized that HTML+JS
>>will
>> >> >>never
>> >> >> be efficent enough to implement complex software/games in browser,
>> and
>> >> >>thus
>> >> >> WebAssembly was born. It was designed by the W3C committee that
>> >> included
>> >> >> all the major browser vendors.
>> >> >>
>> >> >> Please do yourself a favor and research your opinion before you
>> spread
>> >> >> mis-information.
>> >> >>
>> >> >>
>> >> >>
>> >> >> -----Original Message-----
>> >> >> From: Gary Yang [mailto:flashflexpro@gmail.com]
>> >> >> Sent: Monday, April 10, 2017 2:21 PM
>> >> >> To: dev@flex.apache.org
>> >> >> Subject: Re: WebAssembly Flash ByPass
>> >> >>
>> >> >> 1) It takes a long time to make a powerful system like Flash
>>Player
>> >> >>stable
>> >> >> enough, Google Map has been using webGL for a long time, improved
>>a
>> lot
>> >> >> still crash all the time.
>> >> >> 2) Politically, none of the major browser vendors would let
>>anything
>> >> our
>> >> >> of their control grow bigger anymore, Flash Player is the only
>> >> >>exception,
>> >> >> it came too early and too important to the web to be killed.
>> >> >>
>> >> >> The only chance to implement high performance application across
>>main
>> >> >> stream browsers is still Flash with stage3D AGAL, time has proved
>> it, I
>> >> >> don't the majority get it, when it comes to tech, this is usually
>>it,
>> >> >>most
>> >> >> just follow what others say.
>> >> >>
>> >> >> Sad haaa? What if spent time in implementing Flex with AGAL from
>>the
>> >> >>very
>> >> >> beginning? Flash platform might not be on the hanging edge...
>> >> >>
>> >> >>
>> >> >> On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala
>><joshtynjala@gmail.com
>> >
>> >> >> wrote:
>> >> >>
>> >> >> > I don't think you can simply compile LightSpark as-is to
>> WebAssembly.
>> >> >> > I suspect that its rendering code will need serious
>>modification to
>> >> be
>> >> >> > able to draw to HTML Canvas or WebGL instead of whatever native
>> APIs
>> >> >> > it uses now. This part probably wouldn't be trivial.
>> >> >> >
>> >> >> > - Josh
>> >> >> >
>> >> >> > On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor
>><ja...@dedoose.com>
>> >> >>wrote:
>> >> >> >
>> >> >> > > Why would we be implementing anything? I'm talking about
>>taking
>> >> >> > LightSpark
>> >> >> > > as is (with their 252 open bugs) compiling it to LLVM, using
>>that
>> >> >> > > input
>> >> >> > for
>> >> >> > > the WebAssembly compiler and compiling the output binary as a
>> >> >> > > LightSpark WebAssembly drop in flash player. This has
>>nothing to
>> >> do
>> >> >> > > with the
>> >> >> > existing
>> >> >> > > FlexJS work and is merely to allow the existing flash apps to
>>run
>> >> in
>> >> >> > > browser, cross platform without a plug-in (such as Safari on
>> iOS).
>> >> >> Not
>> >> >> > > quite sure what exactly is so crazy here. Sorry, FlexJS will
>> never
>> >> >> > > give
>> >> >> > us
>> >> >> > > the performance we (Dedoose) need, so extending the life
>>cycle of
>> >> >> > > the
>> >> >> > flash
>> >> >> > > player a couple more years to buy us time for a complete
>>rewrite
>> in
>> >> >> > > a performant client technology is pretty important to us.
>> >> >> > > ~ JT
>> >> >> > >
>> >> >> > >
>> >> >> > >
>> >> >> > > -----Original Message-----
>> >> >> > > From: Gary Yang [mailto:flashflexpro@gmail.com]
>> >> >> > > Sent: Monday, April 10, 2017 1:46 PM
>> >> >> > > To: dev@flex.apache.org
>> >> >> > > Subject: Re: WebAssembly Flash ByPass
>> >> >> > >
>> >> >> > > No mean to be offensive, implementing everything in
>>WebAssembly
>> >> >> > > feels
>> >> >> > just
>> >> >> > > like talking about living in Mars, even with
>>HTML/Javascript/CSS
>> >> >> > regardless
>> >> >> > > of performance, after so many years ...
>> >> >> > >
>> >> >> > > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <
>> piotrzarzycki21@gmail.com
>> >> >
>> >> >> > wrote:
>> >> >> > >
>> >> >> > > > Hi Gary,
>> >> >> > > >
>> >> >> > > > Please be tolerant to Jason's opinion and ideas. Apache
>>Flex is
>> >> an
>> >> >> > > > open source project and if Jason would like to bring some
>>idea
>> >> >> > > > here he is very welcome, same as you.
>> >> >> > > >
>> >> >> > > > Piotr
>> >> >> > > >
>> >> >> > > >
>> >> >> > > >
>> >> >> > > > -----
>> >> >> > > > Apache Flex PMC
>> >> >> > > > piotrzarzycki21@gmail.com
>> >> >> > > > --
>> >> >> > > > View this message in context: http://apache-flex-
>> >> >> > > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
>> >> >> > > > status-tp61035p61082.html
>> >> >> > > > Sent from the Apache Flex Development mailing list archive
>>at
>> >> >> > Nabble.com.
>> >> >> > > >
>> >> >> > >
>> >> >> >
>> >> >>
>> >>
>> >>
>> >
>>
Re: WebAssembly Flash ByPass
Posted by Gary Yang <fl...@gmail.com>.
Great! when you guy enjoying wherever it is, please do mind people still on
earth can't bypass FP, begging for next release of Flex.
On Tue, Apr 11, 2017 at 6:48 AM, Dev LFM <de...@gmail.com> wrote:
> It's a known fact that Flutter is already doing that, only releases at the
> moment for ios/android, but I believe that in a 1 year span it will be
> possible to deploy to desktop too...
>
> No comments? too much work? Am I living on Jupiter? ^^
>
> 2017-04-11 10:35 GMT+01:00 Dev LFM <de...@gmail.com>:
>
> > If drawing on gpu is possible, I would go even further and also add skia
> > to the discussion.
> >
> > My vision is, use skia to render everywhere.. Having then different
> > "falcon" to compile for webgl/ios/android/win/osx/unix/raspi
> >
> > Now.. what would be needed? I think a transpiler of "stage3d" to "skia"
> > could be a 1st attempt..
> >
> > Thoughts?
> >
> > 2017-04-11 7:32 GMT+01:00 Alex Harui <ah...@adobe.com>:
> >
> >> Gary,
> >>
> >> Please try to focus on actual technical issues and let those who want to
> >> try, try it. Some people think making Flex/FlexJS popular again is a
> long
> >> shot as well.
> >>
> >> I'm not that familiar with WebAssembly, but from what I've read today, I
> >> think Josh has hit on the main technical challenge, which is that
> >> WebAssembly doesn't appear to let you draw directly to the screen. It
> >> sort of reminds me of Alchemy/FlasCC.
> >>
> >> -Alex
> >>
> >> On 4/10/17, 7:26 PM, "Gary Yang" <fl...@gmail.com> wrote:
> >>
> >> >Thanks, I would love to see how you can get there, W3C started
> designing
> >> >HTML5 in 2009, Firefox is still making exception for Flash Player now,
> >> >maybe you can help them get rid of it and unify the web truly.
> >> >
> >> >On Mon, Apr 10, 2017 at 5:24 PM, Jason Taylor <ja...@dedoose.com>
> wrote:
> >> >
> >> >> Gary, do you even have any idea what WebAssembly is? You should check
> >> >>out
> >> >> some articles on it:
> >> >>https://na01.safelinks.protection.outlook.com/?url=https%
> >> 3A%2F%2Fgithub.c
> >> >>om%2FWebAssembly%2Fdesign%2Fblob%2F&data=02%7C01%7C%7C9587
> >> 71644c3444d5448
> >> >>308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%
> >> 7C63627474370530
> >> >>2509&sdata=roF8%2B3Z2KOZAW7NVE8QJ3OSFAraX8ViFl179cVi%
> >> 2FgJY%3D&reserved=0
> >> >> master/FAQ.md,
> >> >>https://na01.safelinks.protection.outlook.com/?url=https%
> >> 3A%2F%2Fdevelope
> >> >>r.mozilla.org%2Fen-US%2Fdocs%2FWebAssembly%2F&data=02%
> >> 7C01%7C%7C958771644
> >> >>c3444d5448308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1
> >> %7C0%7C0%7C6362
> >> >>74743705302509&sdata=sf%2FmlA5b%2FvU79NgnjPu1tdhbF3sJ1FAoQ
> >> ahTmV06%2FAc%3D
> >> >>&reserved=0
> >> >> Concepts
> >> >>
> >> >> In a nutshell the browser vendors finally realized that HTML+JS will
> >> >>never
> >> >> be efficent enough to implement complex software/games in browser,
> and
> >> >>thus
> >> >> WebAssembly was born. It was designed by the W3C committee that
> >> included
> >> >> all the major browser vendors.
> >> >>
> >> >> Please do yourself a favor and research your opinion before you
> spread
> >> >> mis-information.
> >> >>
> >> >>
> >> >>
> >> >> -----Original Message-----
> >> >> From: Gary Yang [mailto:flashflexpro@gmail.com]
> >> >> Sent: Monday, April 10, 2017 2:21 PM
> >> >> To: dev@flex.apache.org
> >> >> Subject: Re: WebAssembly Flash ByPass
> >> >>
> >> >> 1) It takes a long time to make a powerful system like Flash Player
> >> >>stable
> >> >> enough, Google Map has been using webGL for a long time, improved a
> lot
> >> >> still crash all the time.
> >> >> 2) Politically, none of the major browser vendors would let anything
> >> our
> >> >> of their control grow bigger anymore, Flash Player is the only
> >> >>exception,
> >> >> it came too early and too important to the web to be killed.
> >> >>
> >> >> The only chance to implement high performance application across main
> >> >> stream browsers is still Flash with stage3D AGAL, time has proved
> it, I
> >> >> don't the majority get it, when it comes to tech, this is usually it,
> >> >>most
> >> >> just follow what others say.
> >> >>
> >> >> Sad haaa? What if spent time in implementing Flex with AGAL from the
> >> >>very
> >> >> beginning? Flash platform might not be on the hanging edge...
> >> >>
> >> >>
> >> >> On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala <joshtynjala@gmail.com
> >
> >> >> wrote:
> >> >>
> >> >> > I don't think you can simply compile LightSpark as-is to
> WebAssembly.
> >> >> > I suspect that its rendering code will need serious modification to
> >> be
> >> >> > able to draw to HTML Canvas or WebGL instead of whatever native
> APIs
> >> >> > it uses now. This part probably wouldn't be trivial.
> >> >> >
> >> >> > - Josh
> >> >> >
> >> >> > On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com>
> >> >>wrote:
> >> >> >
> >> >> > > Why would we be implementing anything? I'm talking about taking
> >> >> > LightSpark
> >> >> > > as is (with their 252 open bugs) compiling it to LLVM, using that
> >> >> > > input
> >> >> > for
> >> >> > > the WebAssembly compiler and compiling the output binary as a
> >> >> > > LightSpark WebAssembly drop in flash player. This has nothing to
> >> do
> >> >> > > with the
> >> >> > existing
> >> >> > > FlexJS work and is merely to allow the existing flash apps to run
> >> in
> >> >> > > browser, cross platform without a plug-in (such as Safari on
> iOS).
> >> >> Not
> >> >> > > quite sure what exactly is so crazy here. Sorry, FlexJS will
> never
> >> >> > > give
> >> >> > us
> >> >> > > the performance we (Dedoose) need, so extending the life cycle of
> >> >> > > the
> >> >> > flash
> >> >> > > player a couple more years to buy us time for a complete rewrite
> in
> >> >> > > a performant client technology is pretty important to us.
> >> >> > > ~ JT
> >> >> > >
> >> >> > >
> >> >> > >
> >> >> > > -----Original Message-----
> >> >> > > From: Gary Yang [mailto:flashflexpro@gmail.com]
> >> >> > > Sent: Monday, April 10, 2017 1:46 PM
> >> >> > > To: dev@flex.apache.org
> >> >> > > Subject: Re: WebAssembly Flash ByPass
> >> >> > >
> >> >> > > No mean to be offensive, implementing everything in WebAssembly
> >> >> > > feels
> >> >> > just
> >> >> > > like talking about living in Mars, even with HTML/Javascript/CSS
> >> >> > regardless
> >> >> > > of performance, after so many years ...
> >> >> > >
> >> >> > > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <
> piotrzarzycki21@gmail.com
> >> >
> >> >> > wrote:
> >> >> > >
> >> >> > > > Hi Gary,
> >> >> > > >
> >> >> > > > Please be tolerant to Jason's opinion and ideas. Apache Flex is
> >> an
> >> >> > > > open source project and if Jason would like to bring some idea
> >> >> > > > here he is very welcome, same as you.
> >> >> > > >
> >> >> > > > Piotr
> >> >> > > >
> >> >> > > >
> >> >> > > >
> >> >> > > > -----
> >> >> > > > Apache Flex PMC
> >> >> > > > piotrzarzycki21@gmail.com
> >> >> > > > --
> >> >> > > > View this message in context: http://apache-flex-
> >> >> > > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> >> >> > > > status-tp61035p61082.html
> >> >> > > > Sent from the Apache Flex Development mailing list archive at
> >> >> > Nabble.com.
> >> >> > > >
> >> >> > >
> >> >> >
> >> >>
> >>
> >>
> >
>
Re: WebAssembly Flash ByPass
Posted by Dev LFM <de...@gmail.com>.
It's a known fact that Flutter is already doing that, only releases at the
moment for ios/android, but I believe that in a 1 year span it will be
possible to deploy to desktop too...
No comments? too much work? Am I living on Jupiter? ^^
2017-04-11 10:35 GMT+01:00 Dev LFM <de...@gmail.com>:
> If drawing on gpu is possible, I would go even further and also add skia
> to the discussion.
>
> My vision is, use skia to render everywhere.. Having then different
> "falcon" to compile for webgl/ios/android/win/osx/unix/raspi
>
> Now.. what would be needed? I think a transpiler of "stage3d" to "skia"
> could be a 1st attempt..
>
> Thoughts?
>
> 2017-04-11 7:32 GMT+01:00 Alex Harui <ah...@adobe.com>:
>
>> Gary,
>>
>> Please try to focus on actual technical issues and let those who want to
>> try, try it. Some people think making Flex/FlexJS popular again is a long
>> shot as well.
>>
>> I'm not that familiar with WebAssembly, but from what I've read today, I
>> think Josh has hit on the main technical challenge, which is that
>> WebAssembly doesn't appear to let you draw directly to the screen. It
>> sort of reminds me of Alchemy/FlasCC.
>>
>> -Alex
>>
>> On 4/10/17, 7:26 PM, "Gary Yang" <fl...@gmail.com> wrote:
>>
>> >Thanks, I would love to see how you can get there, W3C started designing
>> >HTML5 in 2009, Firefox is still making exception for Flash Player now,
>> >maybe you can help them get rid of it and unify the web truly.
>> >
>> >On Mon, Apr 10, 2017 at 5:24 PM, Jason Taylor <ja...@dedoose.com> wrote:
>> >
>> >> Gary, do you even have any idea what WebAssembly is? You should check
>> >>out
>> >> some articles on it:
>> >>https://na01.safelinks.protection.outlook.com/?url=https%
>> 3A%2F%2Fgithub.c
>> >>om%2FWebAssembly%2Fdesign%2Fblob%2F&data=02%7C01%7C%7C9587
>> 71644c3444d5448
>> >>308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%
>> 7C63627474370530
>> >>2509&sdata=roF8%2B3Z2KOZAW7NVE8QJ3OSFAraX8ViFl179cVi%
>> 2FgJY%3D&reserved=0
>> >> master/FAQ.md,
>> >>https://na01.safelinks.protection.outlook.com/?url=https%
>> 3A%2F%2Fdevelope
>> >>r.mozilla.org%2Fen-US%2Fdocs%2FWebAssembly%2F&data=02%
>> 7C01%7C%7C958771644
>> >>c3444d5448308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1
>> %7C0%7C0%7C6362
>> >>74743705302509&sdata=sf%2FmlA5b%2FvU79NgnjPu1tdhbF3sJ1FAoQ
>> ahTmV06%2FAc%3D
>> >>&reserved=0
>> >> Concepts
>> >>
>> >> In a nutshell the browser vendors finally realized that HTML+JS will
>> >>never
>> >> be efficent enough to implement complex software/games in browser, and
>> >>thus
>> >> WebAssembly was born. It was designed by the W3C committee that
>> included
>> >> all the major browser vendors.
>> >>
>> >> Please do yourself a favor and research your opinion before you spread
>> >> mis-information.
>> >>
>> >>
>> >>
>> >> -----Original Message-----
>> >> From: Gary Yang [mailto:flashflexpro@gmail.com]
>> >> Sent: Monday, April 10, 2017 2:21 PM
>> >> To: dev@flex.apache.org
>> >> Subject: Re: WebAssembly Flash ByPass
>> >>
>> >> 1) It takes a long time to make a powerful system like Flash Player
>> >>stable
>> >> enough, Google Map has been using webGL for a long time, improved a lot
>> >> still crash all the time.
>> >> 2) Politically, none of the major browser vendors would let anything
>> our
>> >> of their control grow bigger anymore, Flash Player is the only
>> >>exception,
>> >> it came too early and too important to the web to be killed.
>> >>
>> >> The only chance to implement high performance application across main
>> >> stream browsers is still Flash with stage3D AGAL, time has proved it, I
>> >> don't the majority get it, when it comes to tech, this is usually it,
>> >>most
>> >> just follow what others say.
>> >>
>> >> Sad haaa? What if spent time in implementing Flex with AGAL from the
>> >>very
>> >> beginning? Flash platform might not be on the hanging edge...
>> >>
>> >>
>> >> On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala <jo...@gmail.com>
>> >> wrote:
>> >>
>> >> > I don't think you can simply compile LightSpark as-is to WebAssembly.
>> >> > I suspect that its rendering code will need serious modification to
>> be
>> >> > able to draw to HTML Canvas or WebGL instead of whatever native APIs
>> >> > it uses now. This part probably wouldn't be trivial.
>> >> >
>> >> > - Josh
>> >> >
>> >> > On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com>
>> >>wrote:
>> >> >
>> >> > > Why would we be implementing anything? I'm talking about taking
>> >> > LightSpark
>> >> > > as is (with their 252 open bugs) compiling it to LLVM, using that
>> >> > > input
>> >> > for
>> >> > > the WebAssembly compiler and compiling the output binary as a
>> >> > > LightSpark WebAssembly drop in flash player. This has nothing to
>> do
>> >> > > with the
>> >> > existing
>> >> > > FlexJS work and is merely to allow the existing flash apps to run
>> in
>> >> > > browser, cross platform without a plug-in (such as Safari on iOS).
>> >> Not
>> >> > > quite sure what exactly is so crazy here. Sorry, FlexJS will never
>> >> > > give
>> >> > us
>> >> > > the performance we (Dedoose) need, so extending the life cycle of
>> >> > > the
>> >> > flash
>> >> > > player a couple more years to buy us time for a complete rewrite in
>> >> > > a performant client technology is pretty important to us.
>> >> > > ~ JT
>> >> > >
>> >> > >
>> >> > >
>> >> > > -----Original Message-----
>> >> > > From: Gary Yang [mailto:flashflexpro@gmail.com]
>> >> > > Sent: Monday, April 10, 2017 1:46 PM
>> >> > > To: dev@flex.apache.org
>> >> > > Subject: Re: WebAssembly Flash ByPass
>> >> > >
>> >> > > No mean to be offensive, implementing everything in WebAssembly
>> >> > > feels
>> >> > just
>> >> > > like talking about living in Mars, even with HTML/Javascript/CSS
>> >> > regardless
>> >> > > of performance, after so many years ...
>> >> > >
>> >> > > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <piotrzarzycki21@gmail.com
>> >
>> >> > wrote:
>> >> > >
>> >> > > > Hi Gary,
>> >> > > >
>> >> > > > Please be tolerant to Jason's opinion and ideas. Apache Flex is
>> an
>> >> > > > open source project and if Jason would like to bring some idea
>> >> > > > here he is very welcome, same as you.
>> >> > > >
>> >> > > > Piotr
>> >> > > >
>> >> > > >
>> >> > > >
>> >> > > > -----
>> >> > > > Apache Flex PMC
>> >> > > > piotrzarzycki21@gmail.com
>> >> > > > --
>> >> > > > View this message in context: http://apache-flex-
>> >> > > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
>> >> > > > status-tp61035p61082.html
>> >> > > > Sent from the Apache Flex Development mailing list archive at
>> >> > Nabble.com.
>> >> > > >
>> >> > >
>> >> >
>> >>
>>
>>
>
Re: WebAssembly Flash ByPass
Posted by Alex Harui <ah...@adobe.com>.
On 4/12/17, 11:37 AM, "Jason Taylor" <ja...@dedoose.com> wrote:
>Would be really nice if somehow we could get a lot of votes on this
>issue, the flash player team would be the best group to work on this, and
>this would hugely benefit a all flash/flex/air/flex js people. Alex,
>any chance you could help this vote get more attention?
I do not have influence in that arena.
-Alex
RE: WebAssembly Flash ByPass
Posted by Jason Taylor <ja...@dedoose.com>.
Would be really nice if somehow we could get a lot of votes on this issue, the flash player team would be the best group to work on this, and this would hugely benefit a all flash/flex/air/flex js people. Alex, any chance you could help this vote get more attention?
~ JT
-----Original Message-----
From: Bernhard Schleicher [mailto:bernhard.schleicher@emmgt.at]
Sent: Wednesday, April 12, 2017 3:23 AM
To: dev@flex.apache.org
Subject: RE: WebAssembly Flash ByPass
There is an issue by adobe to implement a HTML Target for Air (WebGL /
WebAssembly)
https://tracker.adobe.com/#/view/AIR-4178275
Please vote for this issue if you are interested.
Regards,
Bernhard
-----Original Message-----
From: Tom Chiverton [mailto:tc@extravision.com]
Sent: Mittwoch, 12. April 2017 09:53
To: dev@flex.apache.org
Subject: Re: WebAssembly Flash ByPass
I certainly look forward to seeing what comes out of it.
Sounds like you relish the challenge !
Tom
On 11/04/17 20:09, Jason Taylor wrote:
> problem area is obviously the rendering engine, but I do have quite a bit of
> experience in the area.
RE: WebAssembly Flash ByPass
Posted by Bernhard Schleicher <be...@emmgt.at>.
There is an issue by adobe to implement a HTML Target for Air (WebGL /
WebAssembly)
https://tracker.adobe.com/#/view/AIR-4178275
Please vote for this issue if you are interested.
Regards,
Bernhard
-----Original Message-----
From: Tom Chiverton [mailto:tc@extravision.com]
Sent: Mittwoch, 12. April 2017 09:53
To: dev@flex.apache.org
Subject: Re: WebAssembly Flash ByPass
I certainly look forward to seeing what comes out of it.
Sounds like you relish the challenge !
Tom
On 11/04/17 20:09, Jason Taylor wrote:
> problem area is obviously the rendering engine, but I do have quite a bit of
> experience in the area.
Re: WebAssembly Flash ByPass
Posted by Dev LFM <de...@gmail.com>.
About Skia, there are a lot of port's on github..
2017-04-12 8:52 GMT+01:00 Tom Chiverton <tc...@extravision.com>:
> I certainly look forward to seeing what comes out of it.
>
> Sounds like you relish the challenge !
>
> Tom
>
>
>
> On 11/04/17 20:09, Jason Taylor wrote:
>
>> problem area is obviously the rendering engine, but I do have quite a bit
>> of experience in the area.
>>
>
>
RE: WebAssembly Flash ByPass
Posted by Jason Taylor <ja...@dedoose.com>.
lol yes, currently WebAssembly dosen't include the bindings to WebGL, so still have to hack things around a bit, but yes essentially we would have to rewrite the main rendering engine to WebGL, still haven't made it deep into LightSpark's rendering engine, but assuming they are mimicking the native flash renderer which is an absolutely beast
-----Original Message-----
From: Tom Chiverton [mailto:tc@extravision.com]
Sent: Wednesday, April 12, 2017 12:53 AM
To: dev@flex.apache.org
Subject: Re: WebAssembly Flash ByPass
I certainly look forward to seeing what comes out of it.
Sounds like you relish the challenge !
Tom
On 11/04/17 20:09, Jason Taylor wrote:
> problem area is obviously the rendering engine, but I do have quite a bit of experience in the area.
Re: WebAssembly Flash ByPass
Posted by Tom Chiverton <tc...@extravision.com>.
I certainly look forward to seeing what comes out of it.
Sounds like you relish the challenge !
Tom
On 11/04/17 20:09, Jason Taylor wrote:
> problem area is obviously the rendering engine, but I do have quite a bit of experience in the area.
RE: WebAssembly Flash ByPass
Posted by Jason Taylor <ja...@dedoose.com>.
That's a really great idea! Just starting to get LightSpark and WebAssembly building separately, so a lot of stuff to digest and tear through before I'll be able to even wrap my head around it, but yes the problem area is obviously the rendering engine, but I do have quite a bit of experience in the area. Hopefully I'll have more information in a month or so. I'm not quite convinced this idea is as crazy as others are making it out to be when many others have done most of the difficult work already. In an ideal world it would allow the Falcon swf output to be bundled into Lightspark compiled to WebAssembly and run in browser with massive performance gains over the html/js version.
Gary, long shots and out of the box thinking have never scared me, and I wouldn't be nearly as successful if I had blindly followed the masses.
~ JT
-----Original Message-----
From: Dev LFM [mailto:developerlfm@gmail.com]
Sent: Tuesday, April 11, 2017 2:35 AM
To: dev@flex.apache.org
Subject: Re: WebAssembly Flash ByPass
If drawing on gpu is possible, I would go even further and also add skia to the discussion.
My vision is, use skia to render everywhere.. Having then different "falcon" to compile for webgl/ios/android/win/osx/unix/raspi
Now.. what would be needed? I think a transpiler of "stage3d" to "skia"
could be a 1st attempt..
Thoughts?
2017-04-11 7:32 GMT+01:00 Alex Harui <ah...@adobe.com>:
> Gary,
>
> Please try to focus on actual technical issues and let those who want
> to try, try it. Some people think making Flex/FlexJS popular again is
> a long shot as well.
>
> I'm not that familiar with WebAssembly, but from what I've read today,
> I think Josh has hit on the main technical challenge, which is that
> WebAssembly doesn't appear to let you draw directly to the screen. It
> sort of reminds me of Alchemy/FlasCC.
>
> -Alex
>
> On 4/10/17, 7:26 PM, "Gary Yang" <fl...@gmail.com> wrote:
>
> >Thanks, I would love to see how you can get there, W3C started
> >designing
> >HTML5 in 2009, Firefox is still making exception for Flash Player
> >now, maybe you can help them get rid of it and unify the web truly.
> >
> >On Mon, Apr 10, 2017 at 5:24 PM, Jason Taylor <ja...@dedoose.com> wrote:
> >
> >> Gary, do you even have any idea what WebAssembly is? You should
> >>check out some articles on it:
> >>https://na01.safelinks.protection.outlook.com/?url=
> https%3A%2F%2Fgithub.c
> >>om%2FWebAssembly%2Fdesign%2Fblob%2F&data=02%7C01%7C%
> 7C958771644c3444d5448
> >>308d480821d4b%7Cfa7b1b5a7b34438794aed2c178de
> cee1%7C0%7C0%7C63627474370530
> >>2509&sdata=roF8%2B3Z2KOZAW7NVE8QJ3OSFAraX8ViFl179cVi%2FgJY%3D&reserv
> >>ed=0
> >> master/FAQ.md,
> >>https://na01.safelinks.protection.outlook.com/?url=
> https%3A%2F%2Fdevelope
> >>r.mozilla.org%2Fen-US%2Fdocs%2FWebAssembly%2F&data=
> 02%7C01%7C%7C958771644
> >>c3444d5448308d480821d4b%7Cfa7b1b5a7b34438794aed2c178de
> cee1%7C0%7C0%7C6362
> >>74743705302509&sdata=sf%2FmlA5b%2FvU79NgnjPu1tdhbF3sJ1FAoQahTm
> V06%2FAc%3D
> >>&reserved=0
> >> Concepts
> >>
> >> In a nutshell the browser vendors finally realized that HTML+JS
> >>will never be efficent enough to implement complex software/games
> >>in browser, and thus WebAssembly was born. It was designed by the
> >>W3C committee that included all the major browser vendors.
> >>
> >> Please do yourself a favor and research your opinion before you
> >> spread mis-information.
> >>
> >>
> >>
> >> -----Original Message-----
> >> From: Gary Yang [mailto:flashflexpro@gmail.com]
> >> Sent: Monday, April 10, 2017 2:21 PM
> >> To: dev@flex.apache.org
> >> Subject: Re: WebAssembly Flash ByPass
> >>
> >> 1) It takes a long time to make a powerful system like Flash Player
> >>stable enough, Google Map has been using webGL for a long time,
> >>improved a lot still crash all the time.
> >> 2) Politically, none of the major browser vendors would let
> >>anything our of their control grow bigger anymore, Flash Player is
> >>the only exception, it came too early and too important to the web
> >>to be killed.
> >>
> >> The only chance to implement high performance application across
> >>main stream browsers is still Flash with stage3D AGAL, time has
> >>proved it, I don't the majority get it, when it comes to tech, this
> >>is usually it, most just follow what others say.
> >>
> >> Sad haaa? What if spent time in implementing Flex with AGAL from
> >>the very beginning? Flash platform might not be on the hanging
> >>edge...
> >>
> >>
> >> On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala
> >> <jo...@gmail.com>
> >> wrote:
> >>
> >> > I don't think you can simply compile LightSpark as-is to WebAssembly.
> >> > I suspect that its rendering code will need serious modification
> >> > to be able to draw to HTML Canvas or WebGL instead of whatever
> >> > native APIs it uses now. This part probably wouldn't be trivial.
> >> >
> >> > - Josh
> >> >
> >> > On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com>
> >>wrote:
> >> >
> >> > > Why would we be implementing anything? I'm talking about taking
> >> > LightSpark
> >> > > as is (with their 252 open bugs) compiling it to LLVM, using
> >> > > that input
> >> > for
> >> > > the WebAssembly compiler and compiling the output binary as a
> >> > > LightSpark WebAssembly drop in flash player. This has nothing
> >> > > to do with the
> >> > existing
> >> > > FlexJS work and is merely to allow the existing flash apps to
> >> > > run in browser, cross platform without a plug-in (such as Safari on iOS).
> >> Not
> >> > > quite sure what exactly is so crazy here. Sorry, FlexJS will
> >> > > never give
> >> > us
> >> > > the performance we (Dedoose) need, so extending the life cycle
> >> > > of the
> >> > flash
> >> > > player a couple more years to buy us time for a complete
> >> > > rewrite in a performant client technology is pretty important to us.
> >> > > ~ JT
> >> > >
> >> > >
> >> > >
> >> > > -----Original Message-----
> >> > > From: Gary Yang [mailto:flashflexpro@gmail.com]
> >> > > Sent: Monday, April 10, 2017 1:46 PM
> >> > > To: dev@flex.apache.org
> >> > > Subject: Re: WebAssembly Flash ByPass
> >> > >
> >> > > No mean to be offensive, implementing everything in WebAssembly
> >> > > feels
> >> > just
> >> > > like talking about living in Mars, even with
> >> > > HTML/Javascript/CSS
> >> > regardless
> >> > > of performance, after so many years ...
> >> > >
> >> > > On Mon, Apr 10, 2017 at 4:18 PM, piotrz
> >> > > <pi...@gmail.com>
> >> > wrote:
> >> > >
> >> > > > Hi Gary,
> >> > > >
> >> > > > Please be tolerant to Jason's opinion and ideas. Apache Flex
> >> > > > is an open source project and if Jason would like to bring
> >> > > > some idea here he is very welcome, same as you.
> >> > > >
> >> > > > Piotr
> >> > > >
> >> > > >
> >> > > >
> >> > > > -----
> >> > > > Apache Flex PMC
> >> > > > piotrzarzycki21@gmail.com
> >> > > > --
> >> > > > View this message in context: http://apache-flex-
> >> > > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> >> > > > status-tp61035p61082.html
> >> > > > Sent from the Apache Flex Development mailing list archive at
> >> > Nabble.com.
> >> > > >
> >> > >
> >> >
> >>
>
>
Re: WebAssembly Flash ByPass
Posted by Dev LFM <de...@gmail.com>.
If drawing on gpu is possible, I would go even further and also add skia to
the discussion.
My vision is, use skia to render everywhere.. Having then different
"falcon" to compile for webgl/ios/android/win/osx/unix/raspi
Now.. what would be needed? I think a transpiler of "stage3d" to "skia"
could be a 1st attempt..
Thoughts?
2017-04-11 7:32 GMT+01:00 Alex Harui <ah...@adobe.com>:
> Gary,
>
> Please try to focus on actual technical issues and let those who want to
> try, try it. Some people think making Flex/FlexJS popular again is a long
> shot as well.
>
> I'm not that familiar with WebAssembly, but from what I've read today, I
> think Josh has hit on the main technical challenge, which is that
> WebAssembly doesn't appear to let you draw directly to the screen. It
> sort of reminds me of Alchemy/FlasCC.
>
> -Alex
>
> On 4/10/17, 7:26 PM, "Gary Yang" <fl...@gmail.com> wrote:
>
> >Thanks, I would love to see how you can get there, W3C started designing
> >HTML5 in 2009, Firefox is still making exception for Flash Player now,
> >maybe you can help them get rid of it and unify the web truly.
> >
> >On Mon, Apr 10, 2017 at 5:24 PM, Jason Taylor <ja...@dedoose.com> wrote:
> >
> >> Gary, do you even have any idea what WebAssembly is? You should check
> >>out
> >> some articles on it:
> >>https://na01.safelinks.protection.outlook.com/?url=
> https%3A%2F%2Fgithub.c
> >>om%2FWebAssembly%2Fdesign%2Fblob%2F&data=02%7C01%7C%
> 7C958771644c3444d5448
> >>308d480821d4b%7Cfa7b1b5a7b34438794aed2c178de
> cee1%7C0%7C0%7C63627474370530
> >>2509&sdata=roF8%2B3Z2KOZAW7NVE8QJ3OSFAraX8ViFl179cVi%2FgJY%3D&reserved=0
> >> master/FAQ.md,
> >>https://na01.safelinks.protection.outlook.com/?url=
> https%3A%2F%2Fdevelope
> >>r.mozilla.org%2Fen-US%2Fdocs%2FWebAssembly%2F&data=
> 02%7C01%7C%7C958771644
> >>c3444d5448308d480821d4b%7Cfa7b1b5a7b34438794aed2c178de
> cee1%7C0%7C0%7C6362
> >>74743705302509&sdata=sf%2FmlA5b%2FvU79NgnjPu1tdhbF3sJ1FAoQahTm
> V06%2FAc%3D
> >>&reserved=0
> >> Concepts
> >>
> >> In a nutshell the browser vendors finally realized that HTML+JS will
> >>never
> >> be efficent enough to implement complex software/games in browser, and
> >>thus
> >> WebAssembly was born. It was designed by the W3C committee that included
> >> all the major browser vendors.
> >>
> >> Please do yourself a favor and research your opinion before you spread
> >> mis-information.
> >>
> >>
> >>
> >> -----Original Message-----
> >> From: Gary Yang [mailto:flashflexpro@gmail.com]
> >> Sent: Monday, April 10, 2017 2:21 PM
> >> To: dev@flex.apache.org
> >> Subject: Re: WebAssembly Flash ByPass
> >>
> >> 1) It takes a long time to make a powerful system like Flash Player
> >>stable
> >> enough, Google Map has been using webGL for a long time, improved a lot
> >> still crash all the time.
> >> 2) Politically, none of the major browser vendors would let anything our
> >> of their control grow bigger anymore, Flash Player is the only
> >>exception,
> >> it came too early and too important to the web to be killed.
> >>
> >> The only chance to implement high performance application across main
> >> stream browsers is still Flash with stage3D AGAL, time has proved it, I
> >> don't the majority get it, when it comes to tech, this is usually it,
> >>most
> >> just follow what others say.
> >>
> >> Sad haaa? What if spent time in implementing Flex with AGAL from the
> >>very
> >> beginning? Flash platform might not be on the hanging edge...
> >>
> >>
> >> On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala <jo...@gmail.com>
> >> wrote:
> >>
> >> > I don't think you can simply compile LightSpark as-is to WebAssembly.
> >> > I suspect that its rendering code will need serious modification to be
> >> > able to draw to HTML Canvas or WebGL instead of whatever native APIs
> >> > it uses now. This part probably wouldn't be trivial.
> >> >
> >> > - Josh
> >> >
> >> > On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com>
> >>wrote:
> >> >
> >> > > Why would we be implementing anything? I'm talking about taking
> >> > LightSpark
> >> > > as is (with their 252 open bugs) compiling it to LLVM, using that
> >> > > input
> >> > for
> >> > > the WebAssembly compiler and compiling the output binary as a
> >> > > LightSpark WebAssembly drop in flash player. This has nothing to do
> >> > > with the
> >> > existing
> >> > > FlexJS work and is merely to allow the existing flash apps to run in
> >> > > browser, cross platform without a plug-in (such as Safari on iOS).
> >> Not
> >> > > quite sure what exactly is so crazy here. Sorry, FlexJS will never
> >> > > give
> >> > us
> >> > > the performance we (Dedoose) need, so extending the life cycle of
> >> > > the
> >> > flash
> >> > > player a couple more years to buy us time for a complete rewrite in
> >> > > a performant client technology is pretty important to us.
> >> > > ~ JT
> >> > >
> >> > >
> >> > >
> >> > > -----Original Message-----
> >> > > From: Gary Yang [mailto:flashflexpro@gmail.com]
> >> > > Sent: Monday, April 10, 2017 1:46 PM
> >> > > To: dev@flex.apache.org
> >> > > Subject: Re: WebAssembly Flash ByPass
> >> > >
> >> > > No mean to be offensive, implementing everything in WebAssembly
> >> > > feels
> >> > just
> >> > > like talking about living in Mars, even with HTML/Javascript/CSS
> >> > regardless
> >> > > of performance, after so many years ...
> >> > >
> >> > > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com>
> >> > wrote:
> >> > >
> >> > > > Hi Gary,
> >> > > >
> >> > > > Please be tolerant to Jason's opinion and ideas. Apache Flex is an
> >> > > > open source project and if Jason would like to bring some idea
> >> > > > here he is very welcome, same as you.
> >> > > >
> >> > > > Piotr
> >> > > >
> >> > > >
> >> > > >
> >> > > > -----
> >> > > > Apache Flex PMC
> >> > > > piotrzarzycki21@gmail.com
> >> > > > --
> >> > > > View this message in context: http://apache-flex-
> >> > > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> >> > > > status-tp61035p61082.html
> >> > > > Sent from the Apache Flex Development mailing list archive at
> >> > Nabble.com.
> >> > > >
> >> > >
> >> >
> >>
>
>
Re: WebAssembly Flash ByPass
Posted by Alex Harui <ah...@adobe.com>.
Gary,
Please try to focus on actual technical issues and let those who want to
try, try it. Some people think making Flex/FlexJS popular again is a long
shot as well.
I'm not that familiar with WebAssembly, but from what I've read today, I
think Josh has hit on the main technical challenge, which is that
WebAssembly doesn't appear to let you draw directly to the screen. It
sort of reminds me of Alchemy/FlasCC.
-Alex
On 4/10/17, 7:26 PM, "Gary Yang" <fl...@gmail.com> wrote:
>Thanks, I would love to see how you can get there, W3C started designing
>HTML5 in 2009, Firefox is still making exception for Flash Player now,
>maybe you can help them get rid of it and unify the web truly.
>
>On Mon, Apr 10, 2017 at 5:24 PM, Jason Taylor <ja...@dedoose.com> wrote:
>
>> Gary, do you even have any idea what WebAssembly is? You should check
>>out
>> some articles on it:
>>https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.c
>>om%2FWebAssembly%2Fdesign%2Fblob%2F&data=02%7C01%7C%7C958771644c3444d5448
>>308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C63627474370530
>>2509&sdata=roF8%2B3Z2KOZAW7NVE8QJ3OSFAraX8ViFl179cVi%2FgJY%3D&reserved=0
>> master/FAQ.md,
>>https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdevelope
>>r.mozilla.org%2Fen-US%2Fdocs%2FWebAssembly%2F&data=02%7C01%7C%7C958771644
>>c3444d5448308d480821d4b%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C6362
>>74743705302509&sdata=sf%2FmlA5b%2FvU79NgnjPu1tdhbF3sJ1FAoQahTmV06%2FAc%3D
>>&reserved=0
>> Concepts
>>
>> In a nutshell the browser vendors finally realized that HTML+JS will
>>never
>> be efficent enough to implement complex software/games in browser, and
>>thus
>> WebAssembly was born. It was designed by the W3C committee that included
>> all the major browser vendors.
>>
>> Please do yourself a favor and research your opinion before you spread
>> mis-information.
>>
>>
>>
>> -----Original Message-----
>> From: Gary Yang [mailto:flashflexpro@gmail.com]
>> Sent: Monday, April 10, 2017 2:21 PM
>> To: dev@flex.apache.org
>> Subject: Re: WebAssembly Flash ByPass
>>
>> 1) It takes a long time to make a powerful system like Flash Player
>>stable
>> enough, Google Map has been using webGL for a long time, improved a lot
>> still crash all the time.
>> 2) Politically, none of the major browser vendors would let anything our
>> of their control grow bigger anymore, Flash Player is the only
>>exception,
>> it came too early and too important to the web to be killed.
>>
>> The only chance to implement high performance application across main
>> stream browsers is still Flash with stage3D AGAL, time has proved it, I
>> don't the majority get it, when it comes to tech, this is usually it,
>>most
>> just follow what others say.
>>
>> Sad haaa? What if spent time in implementing Flex with AGAL from the
>>very
>> beginning? Flash platform might not be on the hanging edge...
>>
>>
>> On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala <jo...@gmail.com>
>> wrote:
>>
>> > I don't think you can simply compile LightSpark as-is to WebAssembly.
>> > I suspect that its rendering code will need serious modification to be
>> > able to draw to HTML Canvas or WebGL instead of whatever native APIs
>> > it uses now. This part probably wouldn't be trivial.
>> >
>> > - Josh
>> >
>> > On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com>
>>wrote:
>> >
>> > > Why would we be implementing anything? I'm talking about taking
>> > LightSpark
>> > > as is (with their 252 open bugs) compiling it to LLVM, using that
>> > > input
>> > for
>> > > the WebAssembly compiler and compiling the output binary as a
>> > > LightSpark WebAssembly drop in flash player. This has nothing to do
>> > > with the
>> > existing
>> > > FlexJS work and is merely to allow the existing flash apps to run in
>> > > browser, cross platform without a plug-in (such as Safari on iOS).
>> Not
>> > > quite sure what exactly is so crazy here. Sorry, FlexJS will never
>> > > give
>> > us
>> > > the performance we (Dedoose) need, so extending the life cycle of
>> > > the
>> > flash
>> > > player a couple more years to buy us time for a complete rewrite in
>> > > a performant client technology is pretty important to us.
>> > > ~ JT
>> > >
>> > >
>> > >
>> > > -----Original Message-----
>> > > From: Gary Yang [mailto:flashflexpro@gmail.com]
>> > > Sent: Monday, April 10, 2017 1:46 PM
>> > > To: dev@flex.apache.org
>> > > Subject: Re: WebAssembly Flash ByPass
>> > >
>> > > No mean to be offensive, implementing everything in WebAssembly
>> > > feels
>> > just
>> > > like talking about living in Mars, even with HTML/Javascript/CSS
>> > regardless
>> > > of performance, after so many years ...
>> > >
>> > > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com>
>> > wrote:
>> > >
>> > > > Hi Gary,
>> > > >
>> > > > Please be tolerant to Jason's opinion and ideas. Apache Flex is an
>> > > > open source project and if Jason would like to bring some idea
>> > > > here he is very welcome, same as you.
>> > > >
>> > > > Piotr
>> > > >
>> > > >
>> > > >
>> > > > -----
>> > > > Apache Flex PMC
>> > > > piotrzarzycki21@gmail.com
>> > > > --
>> > > > View this message in context: http://apache-flex-
>> > > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
>> > > > status-tp61035p61082.html
>> > > > Sent from the Apache Flex Development mailing list archive at
>> > Nabble.com.
>> > > >
>> > >
>> >
>>
Re: WebAssembly Flash ByPass
Posted by Gary Yang <fl...@gmail.com>.
Thanks, I would love to see how you can get there, W3C started designing
HTML5 in 2009, Firefox is still making exception for Flash Player now,
maybe you can help them get rid of it and unify the web truly.
On Mon, Apr 10, 2017 at 5:24 PM, Jason Taylor <ja...@dedoose.com> wrote:
> Gary, do you even have any idea what WebAssembly is? You should check out
> some articles on it: https://github.com/WebAssembly/design/blob/
> master/FAQ.md, https://developer.mozilla.org/en-US/docs/WebAssembly/
> Concepts
>
> In a nutshell the browser vendors finally realized that HTML+JS will never
> be efficent enough to implement complex software/games in browser, and thus
> WebAssembly was born. It was designed by the W3C committee that included
> all the major browser vendors.
>
> Please do yourself a favor and research your opinion before you spread
> mis-information.
>
>
>
> -----Original Message-----
> From: Gary Yang [mailto:flashflexpro@gmail.com]
> Sent: Monday, April 10, 2017 2:21 PM
> To: dev@flex.apache.org
> Subject: Re: WebAssembly Flash ByPass
>
> 1) It takes a long time to make a powerful system like Flash Player stable
> enough, Google Map has been using webGL for a long time, improved a lot
> still crash all the time.
> 2) Politically, none of the major browser vendors would let anything our
> of their control grow bigger anymore, Flash Player is the only exception,
> it came too early and too important to the web to be killed.
>
> The only chance to implement high performance application across main
> stream browsers is still Flash with stage3D AGAL, time has proved it, I
> don't the majority get it, when it comes to tech, this is usually it, most
> just follow what others say.
>
> Sad haaa? What if spent time in implementing Flex with AGAL from the very
> beginning? Flash platform might not be on the hanging edge...
>
>
> On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala <jo...@gmail.com>
> wrote:
>
> > I don't think you can simply compile LightSpark as-is to WebAssembly.
> > I suspect that its rendering code will need serious modification to be
> > able to draw to HTML Canvas or WebGL instead of whatever native APIs
> > it uses now. This part probably wouldn't be trivial.
> >
> > - Josh
> >
> > On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com> wrote:
> >
> > > Why would we be implementing anything? I'm talking about taking
> > LightSpark
> > > as is (with their 252 open bugs) compiling it to LLVM, using that
> > > input
> > for
> > > the WebAssembly compiler and compiling the output binary as a
> > > LightSpark WebAssembly drop in flash player. This has nothing to do
> > > with the
> > existing
> > > FlexJS work and is merely to allow the existing flash apps to run in
> > > browser, cross platform without a plug-in (such as Safari on iOS).
> Not
> > > quite sure what exactly is so crazy here. Sorry, FlexJS will never
> > > give
> > us
> > > the performance we (Dedoose) need, so extending the life cycle of
> > > the
> > flash
> > > player a couple more years to buy us time for a complete rewrite in
> > > a performant client technology is pretty important to us.
> > > ~ JT
> > >
> > >
> > >
> > > -----Original Message-----
> > > From: Gary Yang [mailto:flashflexpro@gmail.com]
> > > Sent: Monday, April 10, 2017 1:46 PM
> > > To: dev@flex.apache.org
> > > Subject: Re: WebAssembly Flash ByPass
> > >
> > > No mean to be offensive, implementing everything in WebAssembly
> > > feels
> > just
> > > like talking about living in Mars, even with HTML/Javascript/CSS
> > regardless
> > > of performance, after so many years ...
> > >
> > > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com>
> > wrote:
> > >
> > > > Hi Gary,
> > > >
> > > > Please be tolerant to Jason's opinion and ideas. Apache Flex is an
> > > > open source project and if Jason would like to bring some idea
> > > > here he is very welcome, same as you.
> > > >
> > > > Piotr
> > > >
> > > >
> > > >
> > > > -----
> > > > Apache Flex PMC
> > > > piotrzarzycki21@gmail.com
> > > > --
> > > > View this message in context: http://apache-flex-
> > > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> > > > status-tp61035p61082.html
> > > > Sent from the Apache Flex Development mailing list archive at
> > Nabble.com.
> > > >
> > >
> >
>
RE: WebAssembly Flash ByPass
Posted by Jason Taylor <ja...@dedoose.com>.
Gary, do you even have any idea what WebAssembly is? You should check out some articles on it: https://github.com/WebAssembly/design/blob/master/FAQ.md, https://developer.mozilla.org/en-US/docs/WebAssembly/Concepts
In a nutshell the browser vendors finally realized that HTML+JS will never be efficent enough to implement complex software/games in browser, and thus WebAssembly was born. It was designed by the W3C committee that included all the major browser vendors.
Please do yourself a favor and research your opinion before you spread mis-information.
-----Original Message-----
From: Gary Yang [mailto:flashflexpro@gmail.com]
Sent: Monday, April 10, 2017 2:21 PM
To: dev@flex.apache.org
Subject: Re: WebAssembly Flash ByPass
1) It takes a long time to make a powerful system like Flash Player stable enough, Google Map has been using webGL for a long time, improved a lot still crash all the time.
2) Politically, none of the major browser vendors would let anything our of their control grow bigger anymore, Flash Player is the only exception, it came too early and too important to the web to be killed.
The only chance to implement high performance application across main stream browsers is still Flash with stage3D AGAL, time has proved it, I don't the majority get it, when it comes to tech, this is usually it, most just follow what others say.
Sad haaa? What if spent time in implementing Flex with AGAL from the very beginning? Flash platform might not be on the hanging edge...
On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala <jo...@gmail.com> wrote:
> I don't think you can simply compile LightSpark as-is to WebAssembly.
> I suspect that its rendering code will need serious modification to be
> able to draw to HTML Canvas or WebGL instead of whatever native APIs
> it uses now. This part probably wouldn't be trivial.
>
> - Josh
>
> On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com> wrote:
>
> > Why would we be implementing anything? I'm talking about taking
> LightSpark
> > as is (with their 252 open bugs) compiling it to LLVM, using that
> > input
> for
> > the WebAssembly compiler and compiling the output binary as a
> > LightSpark WebAssembly drop in flash player. This has nothing to do
> > with the
> existing
> > FlexJS work and is merely to allow the existing flash apps to run in
> > browser, cross platform without a plug-in (such as Safari on iOS). Not
> > quite sure what exactly is so crazy here. Sorry, FlexJS will never
> > give
> us
> > the performance we (Dedoose) need, so extending the life cycle of
> > the
> flash
> > player a couple more years to buy us time for a complete rewrite in
> > a performant client technology is pretty important to us.
> > ~ JT
> >
> >
> >
> > -----Original Message-----
> > From: Gary Yang [mailto:flashflexpro@gmail.com]
> > Sent: Monday, April 10, 2017 1:46 PM
> > To: dev@flex.apache.org
> > Subject: Re: WebAssembly Flash ByPass
> >
> > No mean to be offensive, implementing everything in WebAssembly
> > feels
> just
> > like talking about living in Mars, even with HTML/Javascript/CSS
> regardless
> > of performance, after so many years ...
> >
> > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com>
> wrote:
> >
> > > Hi Gary,
> > >
> > > Please be tolerant to Jason's opinion and ideas. Apache Flex is an
> > > open source project and if Jason would like to bring some idea
> > > here he is very welcome, same as you.
> > >
> > > Piotr
> > >
> > >
> > >
> > > -----
> > > Apache Flex PMC
> > > piotrzarzycki21@gmail.com
> > > --
> > > View this message in context: http://apache-flex-
> > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> > > status-tp61035p61082.html
> > > Sent from the Apache Flex Development mailing list archive at
> Nabble.com.
> > >
> >
>
Re: WebAssembly Flash ByPass
Posted by Gary Yang <fl...@gmail.com>.
1) It takes a long time to make a powerful system like Flash Player stable
enough, Google Map has been using webGL for a long time, improved a lot
still crash all the time.
2) Politically, none of the major browser vendors would let anything our of
their control grow bigger anymore, Flash Player is the only exception, it
came too early and too important to the web to be killed.
The only chance to implement high performance application across main
stream browsers is still Flash with stage3D AGAL, time has proved it, I
don't the majority get it, when it comes to tech, this is usually it, most
just follow what others say.
Sad haaa? What if spent time in implementing Flex with AGAL from the very
beginning? Flash platform might not be on the hanging edge...
On Mon, Apr 10, 2017 at 5:16 PM, Josh Tynjala <jo...@gmail.com> wrote:
> I don't think you can simply compile LightSpark as-is to WebAssembly. I
> suspect that its rendering code will need serious modification to be able
> to draw to HTML Canvas or WebGL instead of whatever native APIs it uses
> now. This part probably wouldn't be trivial.
>
> - Josh
>
> On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com> wrote:
>
> > Why would we be implementing anything? I'm talking about taking
> LightSpark
> > as is (with their 252 open bugs) compiling it to LLVM, using that input
> for
> > the WebAssembly compiler and compiling the output binary as a LightSpark
> > WebAssembly drop in flash player. This has nothing to do with the
> existing
> > FlexJS work and is merely to allow the existing flash apps to run in
> > browser, cross platform without a plug-in (such as Safari on iOS). Not
> > quite sure what exactly is so crazy here. Sorry, FlexJS will never give
> us
> > the performance we (Dedoose) need, so extending the life cycle of the
> flash
> > player a couple more years to buy us time for a complete rewrite in a
> > performant client technology is pretty important to us.
> > ~ JT
> >
> >
> >
> > -----Original Message-----
> > From: Gary Yang [mailto:flashflexpro@gmail.com]
> > Sent: Monday, April 10, 2017 1:46 PM
> > To: dev@flex.apache.org
> > Subject: Re: WebAssembly Flash ByPass
> >
> > No mean to be offensive, implementing everything in WebAssembly feels
> just
> > like talking about living in Mars, even with HTML/Javascript/CSS
> regardless
> > of performance, after so many years ...
> >
> > On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com>
> wrote:
> >
> > > Hi Gary,
> > >
> > > Please be tolerant to Jason's opinion and ideas. Apache Flex is an
> > > open source project and if Jason would like to bring some idea here he
> > > is very welcome, same as you.
> > >
> > > Piotr
> > >
> > >
> > >
> > > -----
> > > Apache Flex PMC
> > > piotrzarzycki21@gmail.com
> > > --
> > > View this message in context: http://apache-flex-
> > > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> > > status-tp61035p61082.html
> > > Sent from the Apache Flex Development mailing list archive at
> Nabble.com.
> > >
> >
>
Re: WebAssembly Flash ByPass
Posted by Josh Tynjala <jo...@gmail.com>.
I don't think you can simply compile LightSpark as-is to WebAssembly. I
suspect that its rendering code will need serious modification to be able
to draw to HTML Canvas or WebGL instead of whatever native APIs it uses
now. This part probably wouldn't be trivial.
- Josh
On Mon, Apr 10, 2017 at 1:52 PM, Jason Taylor <ja...@dedoose.com> wrote:
> Why would we be implementing anything? I'm talking about taking LightSpark
> as is (with their 252 open bugs) compiling it to LLVM, using that input for
> the WebAssembly compiler and compiling the output binary as a LightSpark
> WebAssembly drop in flash player. This has nothing to do with the existing
> FlexJS work and is merely to allow the existing flash apps to run in
> browser, cross platform without a plug-in (such as Safari on iOS). Not
> quite sure what exactly is so crazy here. Sorry, FlexJS will never give us
> the performance we (Dedoose) need, so extending the life cycle of the flash
> player a couple more years to buy us time for a complete rewrite in a
> performant client technology is pretty important to us.
> ~ JT
>
>
>
> -----Original Message-----
> From: Gary Yang [mailto:flashflexpro@gmail.com]
> Sent: Monday, April 10, 2017 1:46 PM
> To: dev@flex.apache.org
> Subject: Re: WebAssembly Flash ByPass
>
> No mean to be offensive, implementing everything in WebAssembly feels just
> like talking about living in Mars, even with HTML/Javascript/CSS regardless
> of performance, after so many years ...
>
> On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com> wrote:
>
> > Hi Gary,
> >
> > Please be tolerant to Jason's opinion and ideas. Apache Flex is an
> > open source project and if Jason would like to bring some idea here he
> > is very welcome, same as you.
> >
> > Piotr
> >
> >
> >
> > -----
> > Apache Flex PMC
> > piotrzarzycki21@gmail.com
> > --
> > View this message in context: http://apache-flex-
> > development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> > status-tp61035p61082.html
> > Sent from the Apache Flex Development mailing list archive at Nabble.com.
> >
>
Re: WebAssembly Flash ByPass
Posted by Harbs <ha...@gmail.com>.
I would be very interested in hearing how an experiment like that would work out.
> On Apr 10, 2017, at 4:52 PM, Jason Taylor <ja...@dedoose.com> wrote:
>
> Why would we be implementing anything? I'm talking about taking LightSpark as is (with their 252 open bugs) compiling it to LLVM, using that input for the WebAssembly compiler and compiling the output binary as a LightSpark WebAssembly drop in flash player. This has nothing to do with the existing FlexJS work and is merely to allow the existing flash apps to run in browser, cross platform without a plug-in (such as Safari on iOS). Not quite sure what exactly is so crazy here. Sorry, FlexJS will never give us the performance we (Dedoose) need, so extending the life cycle of the flash player a couple more years to buy us time for a complete rewrite in a performant client technology is pretty important to us.
> ~ JT
>
>
>
> -----Original Message-----
> From: Gary Yang [mailto:flashflexpro@gmail.com]
> Sent: Monday, April 10, 2017 1:46 PM
> To: dev@flex.apache.org
> Subject: Re: WebAssembly Flash ByPass
>
> No mean to be offensive, implementing everything in WebAssembly feels just like talking about living in Mars, even with HTML/Javascript/CSS regardless of performance, after so many years ...
>
> On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com> wrote:
>
>> Hi Gary,
>>
>> Please be tolerant to Jason's opinion and ideas. Apache Flex is an
>> open source project and if Jason would like to bring some idea here he
>> is very welcome, same as you.
>>
>> Piotr
>>
>>
>>
>> -----
>> Apache Flex PMC
>> piotrzarzycki21@gmail.com
>> --
>> View this message in context: http://apache-flex-
>> development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
>> status-tp61035p61082.html
>> Sent from the Apache Flex Development mailing list archive at Nabble.com.
>>
RE: WebAssembly Flash ByPass
Posted by Jason Taylor <ja...@dedoose.com>.
Why would we be implementing anything? I'm talking about taking LightSpark as is (with their 252 open bugs) compiling it to LLVM, using that input for the WebAssembly compiler and compiling the output binary as a LightSpark WebAssembly drop in flash player. This has nothing to do with the existing FlexJS work and is merely to allow the existing flash apps to run in browser, cross platform without a plug-in (such as Safari on iOS). Not quite sure what exactly is so crazy here. Sorry, FlexJS will never give us the performance we (Dedoose) need, so extending the life cycle of the flash player a couple more years to buy us time for a complete rewrite in a performant client technology is pretty important to us.
~ JT
-----Original Message-----
From: Gary Yang [mailto:flashflexpro@gmail.com]
Sent: Monday, April 10, 2017 1:46 PM
To: dev@flex.apache.org
Subject: Re: WebAssembly Flash ByPass
No mean to be offensive, implementing everything in WebAssembly feels just like talking about living in Mars, even with HTML/Javascript/CSS regardless of performance, after so many years ...
On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com> wrote:
> Hi Gary,
>
> Please be tolerant to Jason's opinion and ideas. Apache Flex is an
> open source project and if Jason would like to bring some idea here he
> is very welcome, same as you.
>
> Piotr
>
>
>
> -----
> Apache Flex PMC
> piotrzarzycki21@gmail.com
> --
> View this message in context: http://apache-flex-
> development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> status-tp61035p61082.html
> Sent from the Apache Flex Development mailing list archive at Nabble.com.
>
Re: WebAssembly Flash ByPass
Posted by Gary Yang <fl...@gmail.com>.
No mean to be offensive, implementing everything in WebAssembly feels just
like talking about living in Mars, even with HTML/Javascript/CSS regardless
of performance, after so many years ...
On Mon, Apr 10, 2017 at 4:18 PM, piotrz <pi...@gmail.com> wrote:
> Hi Gary,
>
> Please be tolerant to Jason's opinion and ideas. Apache Flex is an open
> source project and if Jason would like to bring some idea here he is very
> welcome, same as you.
>
> Piotr
>
>
>
> -----
> Apache Flex PMC
> piotrzarzycki21@gmail.com
> --
> View this message in context: http://apache-flex-
> development.2333347.n4.nabble.com/FlexJS-feature-chart-work-
> status-tp61035p61082.html
> Sent from the Apache Flex Development mailing list archive at Nabble.com.
>
Re: WebAssembly Flash ByPass
Posted by piotrz <pi...@gmail.com>.
Hi Gary,
Please be tolerant to Jason's opinion and ideas. Apache Flex is an open
source project and if Jason would like to bring some idea here he is very
welcome, same as you.
Piotr
-----
Apache Flex PMC
piotrzarzycki21@gmail.com
--
View this message in context: http://apache-flex-development.2333347.n4.nabble.com/FlexJS-feature-chart-work-status-tp61035p61082.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.
Re: WebAssembly Flash ByPass
Posted by Gary Yang <fl...@gmail.com>.
Yeah the life in Mars could be very exciting, let me know when you are
there ...
On Mon, Apr 10, 2017 at 3:17 PM, Jason Taylor <ja...@dedoose.com> wrote:
> WebAssembly is officially supported in both Chrome and Firefox right now,
> and in the rest very soon.
> WebAssembly can take LLVM and compiles to a binary format that works cross
> browser at near native speeds without any plugin.
> LightSpark is an open source implemention of the flash player written in
> C++ that can compile to LLVM. Potentionally we can compile LightSpark to
> LLVM and be able to run all flash apps in all browsers without any plugin
> and completely bypass the death of the flash player. Thoughts?
> ~ JT
>
>