Mailing-List: contact flex-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: flex-dev@incubator.apache.org
Received-SPF: pass (nike.apache.org: domain of carlos.rovira@gmail.com
 designates 209.85.213.47 as permitted sender)
MIME-Version: 1.0
Sender: carlos.rovira@gmail.com
In-Reply-To: 
 <CAFzDyY31rSODYZHifZAZUtg0cmxpagnDQ3rRBTmdZp38c8BSQA@mail.gmail.com>
References: 
 <CAFQJJxRATHkQxQx3zpEysb7t3-d7FtjKxp2RpLwiKC0z8V-vqw@mail.gmail.com>
	<CAFzDyY0ta0GKbxHpRXtB+-iXVqNAThQc5isNFR_j1iHer7K=ZA@mail.gmail.com>
	<CAJ16SzArHkQ5BfVNWCUE6m_gYyktZjQTm4SP1_B04FRtDZr7+Q@mail.gmail.com>
	<CAMFKFRjBFQA-8SHS73WsXPoT9uji==WWPEGBRahcTg5GYXheUQ@mail.gmail.com>
	<CA+ckY6uGLT3i5ePBdvzfsv3VdqsxaFFK5TpPTOk6ABfnP0ggfA@mail.gmail.com>
	<CAFzDyY31rSODYZHifZAZUtg0cmxpagnDQ3rRBTmdZp38c8BSQA@mail.gmail.com>
Date: Mon, 6 Feb 2012 11:36:34 +0100
Message-ID: 
 <CAMFKFRgz1hoZ7nxpkWarJQD0KOakVGLJ=tqBrNK_JvGRR+yjvA@mail.gmail.com>
Subject: Re: Apache Flex suggestion - dumping SWF support in favor of HTML5 -
 listen to Steve
From: Carlos Rovira <carlos.rovira@codeoscopic.com>
To: flex-dev@incubator.apache.org
Content-Type: multipart/alternative; boundary=20cf303b42299296be04b8493c36

--20cf303b42299296be04b8493c36
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hi Doug,

reading my brief post and your response I think I fell in the problem that
trying to post short answers or sentences makes this kind of communication
problems.

Let me clarify what I tried to express:

I'm aware of the evolution of JS through the years. I like many of the
things that evolution brought: LESS & SASS in css, js frameworks like
jquery or sproutcore, ... and one of the things I would like from a JS
approach is that we are against the browser and we have not to deal with a
plugin, in terms of pure integration.

But IMHO there's lots more problems with JS than benefits right now. The
proof is that we are all here trying to save flex, and right now it's tied
to Flash Player. A Flex HTML5 should be created from scratch,  and I think
Apple, Google and Adobe will bring the key frameworks and tools in the
following years.

What's are the main problems in JS that make me thinks that we have already
something superior (Flex)?

* We continue to have the problems of different browser vendors, and we go
again to the fragmentation way that we all hate. The "one implementation to
rule them all" is crucial in client side application (something that is not
such a problem in server side enviroments). Regarding this the right way
IMHO will be to have a Flash Player open source and managed by a foundation
like Apache. But this is other war and topic to talk in other threads...

* We have lots of things that JS and its evolution does not has already:
Main for me is OOP, but there are many others (DI, Annotations,
AOP, precompilation & bytecode, debuggin, profiling,...and the companion
tools to deal with all of this...should I continue??

* I see lots of great JS code (node.js, jquery, and so on...) but all is
scattered through the web, so to work with all the problems JS would arise
we should work on some "glue framework" or something from scratch that
would bring as a skeleton to build applications. Work with js today
continues to be a nightmare, since is all based on separated scripts. No
packages and no organization at all based in some criteria, only what you
want to impose. All this things are already in our hands with AS3 and Flex
from several years.

JS is evolving greatly, but it continues to be IMHO a poor solution
(compared with what we have today with flex). Moreover if we take into
account that we have already something better that works (aka Flex) (again
why we are all here trying to make Apache Flex sucessful??), we should
trying to make it better.

Another point I want to share, I like the promises that FalconJS could
bring to us, and would love to see it come true. But I think this is a
long-term approach. If this could be taken in few months things will be
very different and maybe Adobe would consider that way.

can we produce some "Flex-sdk-for-HTML5" that allow us to build apps in
MXML+AS3 (or a variant like Dart), using OOP (this is basic)? that will
produce a output based on HTML/JS/CSS scripts? I can elaborate more on this
but is huge task that will take us several months/years to get it done.

Moreover, for the time this effort would bring us something
usable...enterprise projects could benefit from that?. One thing is to make
it successful in little apps, but other thing is to get it work in huge
systems where we're using Flex 4.6 in very advanced ways.

I think as well that is really good that some people work in that JS
approach and others in actual framework evolution (I would contribute on
the later).

My short message was about my desire of see a 4.7 ASAP, since we need
something released soon to avoid any possibility of fail for Apache Flex.
Other works and JS approaches could live in parallel and will be great. I
only want to say that those efforts will take lots of time to see something
working or released, but as many of us (if not all) I would love to see
flex producing HTML5/JS/CSS code and valid high level enterprise apps with
all the benefits we as developers have today with Flex/AS3/Flash/OOP/DI,
and so on...

Hope this long message would make you all understand what I mean to say.

Thanks!

Carlos Rovira


2012/2/5 Stephane Beladaci <adobeflexengineer@gmail.com>

> On Sun, Feb 5, 2012 at 10:02 AM, Doug McCune <doug@dougmccune.com> wrote:
> >
> > Saying nothing changed means you haven't been paying attention. HTML/JS
> is
> > changing faster than almost any other technology stack out there at the
> > moment. It has more momentum and developer interest than almost any oth=
er
> > technology stack. DO NOT write it off as being inferior.
> >
>
> It is not because a lot of people try to hack it and make it work because
> they do not have a choice (Apple banning every single RIA technology from
> iOS gives no choice to developers but to make HTML5 work no matter if it
> means cutting everyone an arm and a leg. I would not applaud it! Facebook
> is a brillant example of what happens when a platform is built on hacks, =
it
> is absolutely not reliable.
>
> HTML5/JS is a scam to me, a scam shove into our troat by Steve Jobs who i=
s
> probably laughing his butt of from the other side seeing so many people
> trying to make something work when in fact the goal from day one was for
> Apple to cripple the web and created an exodus of developers moving from
> the web to proprietary iOS.
>


--=20
Carlos Rovira
Director de Tecnolog=EDa
M: +34 607 22 60 05
F:  +34 912 35 57 77

CODEOSCOPIC S.A. <http://www.codeoscopic.com>
Avd. del General Per=F3n, 32
Planta 10, Puertas P-Q
28020 Madrid

--20cf303b42299296be04b8493c36--