On 1/18/12 3:14 AM, "Cosma Colanicchia" wrote:
> It would be great if this framework could be open-sourced and
> donated to Apache as well..
It is on the list for evaluation. No timetable at this time.
--
Alex Harui
Flex SDK Team
Adobe Systems, Inc.
http://blogs.adobe.com/aharui
Yes, I would be interested too.
I've asked the question about server-side dependencies some months ago
to an adobe employee.
This is what he answered to me :
Client Component Framework only comes with ADEP, so you’d need to purchase
ADEP Basic as a minimum.
>From a technical perspective, you co
Hello everybody, this is my first post on the list.
I'm very interested in marshall plan and any other techniques to
obtain modular applications in Flex, I experimented with them in the
past and experienced a lot of headaches..
I was very curious to see Gravity framework in action, I know that it
On 14 Jan 2012, at 11:33, Raju Bitter wrote:
Thanks for the update on Falcon, that sounds promising!
Yeah thanks Gordon/Alex.
Thanks for the update on Falcon, that sounds promising!
On 13 January 2012 23:30, Gordon Smith wrote:
> Falcon has a lot of support already for .mxml, .css, and .properties files
> as well as .as files. It's at the point where it can compile Checkinapp,
> one of the MXML apps than compiles and runs when you do 'ant' in the SDK.
> It compiles the MXML
: Friday, January 13, 2012 1:45 PM
To: flex-dev@incubator.apache.org
Subject: Re: ActionScript and Apache Flex
On 1/13/12 3:47 AM, "Tink" wrote:
> On 13 Jan 2012, at 11:36, Matthew Poole wrote:
>
> We wouldn't be dependent on any particular runtime, which is one of my
&g
On 1/13/12 3:47 AM, "Tink" wrote:
> On 13 Jan 2012, at 11:36, Matthew Poole wrote:
>
> We wouldn't be dependent on any particular runtime, which is one of my
> biggest concerns.
>
> - No Flash Player for mobile, and I think most computing is going to
> become mobile over the next 5 years.
Th
>> We wouldn't be dependent on any particular runtime, which is one of my
biggest concerns.
HaXe compiled to SWFs which runs in the Flash runtime. Therefore we have
project dependencies still with Adobe, but also introduce the HaXe
community as a release dependency / artifact.
Where as if we just
>
> There was a significant effect with this though if we did release a
>
haXe-based SDK. Anyone else using haXe to build upon the SDK would need all
> of the Flex source files, not the SWCs and the compiler would recompile
> Flex
> every time. From crude tests I did with haXe, the need to recompil
On 13 Jan 2012, at 11:36, Matthew Poole wrote:
...but still have dependences on the runtime fom Adobe.
That would make project management significantly more complex.
On 13 January 2012 11:31, Matthew Poole wrote:
We wouldn't be dependent on any particular runtime, which is one of my
bigges
...though we'd then be beholden to Nicolas and his community for updates to
the HaXe compiler, but still have dependences on the runtime fom Adobe.
That would make project management significantly more complex.
On 13 January 2012 11:31, Matthew Poole wrote:
> >> This could be a crazy idea, but h
>> This could be a crazy idea, but how about us moving to Haxe?
Love the leftfield thinking Tink. It would be worth serious consideration
IMO if Falcon JS turns out to be a no-go.
2012/1/13 Frédéric Thomas
> From: David Arno
>> Sent: Friday, January 13, 2012 11:14 AM
>>
>
> The advantages that
From: David Arno
Sent: Friday, January 13, 2012 11:14 AM
The advantages that haXe could bring us (better language; faster
compilation) are now things we can add to the Falcon compiler when we get
our hands on it.
+1
And the dev of these features are unfortunatly out of my skills apparently,
> From: Roland Zwaga [mailto:rol...@stackandheap.com]
> Sent: 13 January 2012 10:49
>
> I also seem to recall from some tweets that HaXe doesn't have support for
libraries, right?
> So if you want to include third-party code, your only option is to include
the sources.
> Or did I understand this w
Hi David,
I also seem to recall from some tweets that HaXe doesn't have support for
libraries, right?
So if you want to include third-party code, your only option is to include
the sources.
Or did I understand this wrong?
cheers,
Roland
On 13 January 2012 11:14, David Arno wrote:
> --Original
--Original Message-
> From: Tink [mailto:f...@tink.ws ]
> Sent: 12 January 2012 19:12
>
> This could be a crazy idea, but how about us moving to Haxe?
> This has an active OS community, compiles to multiple platforms already
meaning
> that Flex wouldn't be dependent on Adobe anymore.
I loo
Alex,
> One of the advantages of AS is that it is forgiving. You can turn off
> strict mode and hack away and crank out an app in a very short time.
Every
> once in a while I try to do that in C or Java and spend lots of time
fixing
> up type errors. I think that is a great benefit to folks l
On 1/12/12 2:16 PM, "Rick Winscot" wrote:
> Alex,
>
> Speaking of marshal plan... any thoughts on how to deal with that going
> forward?
>
> The test cases that Adobe used / uses to validate marshal plan architecture
> and compatibility are pretty important. Do the mustella tests cover that
Alex,
Speaking of marshal plan... any thoughts on how to deal with that going
forward?
The test cases that Adobe used / uses to validate marshal plan architecture and
compatibility are pretty important. Do the mustella tests cover that ground?
--
Rick Winscot
On Thursday, January 12, 2012
On 1/12/12 3:58 AM, "David Arno" wrote:
> Makes sense. As for performance issues, that's easy to predict. If we just
> want compile-time support for enums, private constructors, abstract methods
> and classes and the like, then none of them will have a significant impact
> on runtime performan
On 1/12/12 1:14 AM, "Web DoubleFx" wrote:
> There are many pros and cons using late-binding over early-binding, I don't
> how it is implemented in the Flash VM but and I can beleive to their
> advantages in a constrained environment but what I would like to know is the
> real drawback in term
On 1/12/12 11:12 AM, "Tink" wrote:
> This could be a crazy idea, but how about us moving to Haxe? This has
> an active OS community, compiles to multiple platforms already meaning
> that Flex wouldn't be dependent on Adobe anymore.
There is nothing I know of to stop you or any other committers
Funny i had the same idea today. The language itself seems to have some nice
extra's too
type inference enums etc. see http://ncannasse.fr/blog/haxe_comparison_chart
But i think flex should go/remain enterprise. And for me enterprise also means
solid tools.
For actionscript there is just much mo
On 10 Jan 2012, at 09:21, Raju Bitter wrote:
How does the evolution of ActionScript relate to Apache Flex? If the
ActionScript standard would change - and the language would be
extended or modified - who would be making that decision? Is there an
"official standard" of the language available to
> From: da...@davidarno.org
> To: flex-dev@incubator.apache.org
> Subject: RE: ActionScript and Apache Flex
> Date: Thu, 12 Jan 2012 11:58:57 +
>
> From: Web DoubleFx [mailto:webdoubl...@hotmail.com]
> Sent: 12 January 2012 10:59
>
> > Ok, nothing is perfec
hat we should be open to preproecessing and
performance optimisations at the compiler level though.
On 12 January 2012 10:59, Web DoubleFx wrote:
>
> > From: da...@davidarno.org
> > To: flex-dev@incubator.apache.org
> > Subject: RE: ActionScript and Apache Flex
> > Date
From: Web DoubleFx [mailto:webdoubl...@hotmail.com]
Sent: 12 January 2012 10:59
> Ok, nothing is perfect but in this case, on the AS3, the implementation of
certain
> features would be more than well comed to believe the users including
myself,
> remains to be seen how this would impact on pe
> From: da...@davidarno.org
> To: flex-dev@incubator.apache.org
> Subject: RE: ActionScript and Apache Flex
> Date: Thu, 12 Jan 2012 10:28:34 +
>
> > From: Web DoubleFx [mailto:webdoubl...@hotmail.com]
> > Sent: 12 January 2012 08:28
> >> My first ques
> From: Web DoubleFx [mailto:webdoubl...@hotmail.com]
> Sent: 12 January 2012 08:28
>> My first question would be, why do you absolutely have to have
equivalency with Java?
> In short, to have a state of the art programming language.
Sorry to be blunt, but this is a nonsense. Java is not a state-
Yes, and if we can abstract further to allow mix languages would b nice
Sent from Samsung Mobile
Abhilash Krishnan wrote:
Hi,
Whenever I started playing around with Tamarin VM, the AVM2 clone donated
to Mozilla, I really wanted to add some cool features on my own. I know
Flash/AS3 is based o
> From: aha...@adobe.com
>
> On 1/12/12 12:27 AM, "Web DoubleFx" wrote:
>
> > why couldn't we have the same in flash if we really want it, maybe a
> > question
> > of performence,
> Well, I'm not a language expert, but my friends on the VM team constantly
On 1/12/12 12:27 AM, "Web DoubleFx" wrote:
> why couldn't we have the same in flash if we really want it, maybe a question
> of performence,
Well, I'm not a language expert, but my friends on the VM team constantly
remind me about late-binding and its advantages, and dynamic languages and
thei
---> From: aha...@adobe.com> On 1/11/12
10:07 PM, "Web DoubleFx" wrote:>> I haven't done enough research to know what
the costs would be. My first> question would be, why do you absolutely have to
have equivalency with Java?> And, given that the VM is "fixed"
On 1/12/12 12:00 AM, "Abhilash Krishnan" wrote:
> Whenever I started playing around with Tamarin VM, the AVM2 clone donated
> to Mozilla, I really wanted to add some cool features on my own. I know
> Flash/AS3 is based on event-driven programming model. If ever Adobe wanted
> to implement mult
Hi,
Whenever I started playing around with Tamarin VM, the AVM2 clone donated
to Mozilla, I really wanted to add some cool features on my own. I know
Flash/AS3 is based on event-driven programming model. If ever Adobe wanted
to implement multi--threading in Flash or AIR it would be a killer featur
On 1/11/12 10:07 PM, "Web DoubleFx" wrote:
> We do not get something for nothing, I am willing to sacrifice a bit of the
> Falcon compile time gain to get such features, that's a big YES even.
> Before the flexsummit, we did a online meeting for french flex developpers
> about what we wanted f
> If it meant having proper AS3 code annotations, a proper meta data API and>
> runtime performance increases for using said meta data API, then yes, I>
> would trade off some of the Falcon speed increase for features such as a>
> preprocessing phase.>> -omar
+1
> On 1/11/12 12:57 PM, "Matthew
On Wednesday, January 11, 2012, Alex Harui wrote:
>
>
>
> On 1/11/12 2:51 PM, "Omar Gonzalez" wrote:
>
>>
>> If the compile time improvement in Falcon is as good as we're hoping and
>> hearing from people like Deepa then it sounds like Falcon should have the
>> bandwidth for more features such as
>
> Would you be willing to sacrifice some of that for a
> pre-processing phase?
>
I would, assuming this could be turned on and off through additional
compiler flags
- Peter
On 1/11/12 2:51 PM, "Omar Gonzalez" wrote:
>
> If the compile time improvement in Falcon is as good as we're hoping and
> hearing from people like Deepa then it sounds like Falcon should have the
> bandwidth for more features such as preprocessing meta data, or am I
> expecting too much from
>> preprossessing meta data sounds great, especially for DI etc
> Just remember that preprocessing slows down compile times. Are we willing
> to make that trade-off?
>
> --
> Alex Harui
> Flex SDK Team
> Adobe Systems, Inc.
> http://blogs.adobe.com/aharui
>
>
If the compile time improvement in Fa
Maybe not with mxmlc, but falcon has lots of grunt.
On 11 January 2012 21:42, Alex Harui wrote:
>
>
>
> On 1/11/12 12:57 PM, "Matthew Poole" wrote:
>
> > preprossessing meta data sounds great, especially for DI etc
> Just remember that preprocessing slows down compile times. Are we willing
> t
On 1/11/12 12:57 PM, "Matthew Poole" wrote:
> preprossessing meta data sounds great, especially for DI etc
Just remember that preprocessing slows down compile times. Are we willing
to make that trade-off?
--
Alex Harui
Flex SDK Team
Adobe Systems, Inc.
http://blogs.adobe.com/aharui
preprossessing meta data sounds great, especially for DI etc
Parametric polymorophism would be pretty cool too, it would allow for much
more generic abstractions...
On 11 January 2012 19:00, Alex Harui wrote:
>
>
>
> On 1/11/12 2:38 AM, "Dirk Eismann" wrote:
>
> >
> > I think I saw some JIRA i
On 1/11/12 2:38 AM, "Dirk Eismann" wrote:
>
> I think I saw some JIRA issues in the Adobe JIRA some time ago where
> folks talked about how they enhanced the mxmlc / asc compilers to
> evaluate metadata at pre-compile time to emit code. The same could
> happen with custom language additions.
2012/1/11 David Arno :
> The compiler turns the language into bytecode. If we have control of a
> AVM-targeting compiler, then we are free to define the specification of the
> language that it compiles.
exactly my point, same as Scala does for the JVM (see my previous post)
I think I saw some JIR
+lots to exciting times
On 10 January 2012 23:50, Alex Harui wrote:
>
>
>
> On 1/10/12 3:26 PM, "David Arno" wrote:
>
> > I assume you mean here that the mxmlc compiler uses code also found in
> the
> > asc compiler?
> Yes. Actually, these days, it shortcuts that a bit, but that was
> essentia
On 1/10/12 3:26 PM, "David Arno" wrote:
> I assume you mean here that the mxmlc compiler uses code also found in the
> asc compiler?
Yes. Actually, these days, it shortcuts that a bit, but that was
essentially the original architecture.
> I didn't think asc existed as a separate executable t
> From: Alex Harui [mailto:aha...@adobe.com]
> Sent: 10 January 2012 18:04
> MXMLC and COMPC currently leverage ASC. I believe that Flash Authoring
> leverages the same ASC. Flash Authoring has no use for MXMLC
> (at least, the MXML compilation).
I assume you mean here that the mxmlc compiler
On 1/10/12 10:17 AM, "Matthew Poole" wrote:
> Presumably the community could know ahead of time when language features
> where proposed.
I wouldn't count on lots of advanced notice of new language features. The
player will likely continue being backward compatible so existing technology
shoul
The most interesting thing that happened to Java (or more specifically
to the JVM) in the last few yeras is how the community came up with
solutions on how to improve the platform from the "outside" by not
waiting for Sun or Oracle to deliver something. Examples are Groovy
and more recently Scala.
>> I don't understand how the compiler can affect the language
Am I correct in thinking if the bytecode generated was understood by the
runime you could make changes to how the language looked before it
compiled? I dont see any usecases for this and I dont see it as a good idea
either so non-starte
> IMHO, language evolution is not within the domain of responsibility of this
> project. If you are wondering what will happen if Adobe changes
> ActionScript, I assume this project will just react to it.
>
+1, likewise since Adobe continues development at Flash Builder at one time
it will proba
MXMLC and COMPC currently leverage ASC. I believe that Flash Authoring
leverages the same ASC. Flash Authoring has no use for MXMLC (at least, the
MXML compilation).
I don't understand how the compiler can affect the language. ActionScript
and the VM are not being contributed to Apache. It is
Have Adobe said it would continue to use compc / mxmlc internally?
Either way the runtime would still have to support the languages features
added so that limits whats possible.
On 10 January 2012 16:18, Raju Bitter wrote:
> 2012/1/10 Matthew Poole :
> > I dont think it necessarily follows that
2012/1/10 Matthew Poole :
> I dont think it necessarily follows that because the compiler is donated
> that control over the language is assumed. Though it would be fantastic to
> add some of the languges features that have been added to HaXe...
I'm not sure if there are other Adobe products genera
Very interesting.
I dont think it necessarily follows that because the compiler is donated
that control over the language is assumed. Though it would be fantastic to
add some of the languges features that have been added to HaXe...
On 10 January 2012 10:51, David Arno wrote:
> I think you raise
I think you raise some really important points here Raju. Since the compilers
are going to be contributed to Apache too, then the Apache Flex contributors
get to define the future nature of the language (within the constraints of
complying with the needs of the flash players of course.) However
How does the evolution of ActionScript relate to Apache Flex? If the
ActionScript standard would change - and the language would be
extended or modified - who would be making that decision? Is there an
"official standard" of the language available to the public, and what
are Adobe's plans for Actio
60 matches
Mail list logo
