flex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Hordur Thordarson <hor...@lausn.is>
Subject Re: Flex 5 in haxe
Date Wed, 21 Nov 2012 11:08:17 GMT
On 21.11.2012, at 10:35, sébastien Paturel wrote:

> its more JS that need some improvements than HTML5. The dream solution i'd love to see
happen would be to see JS evolve to something very close to what AS3 is. (AS3 is ECMAScript
and is supposed to be the future of JS)

All the effort on JS currently is about making it faster, not so much about improving the
language.  Which is why Google created Dart.  Also, even if an effort to modernize JS started
now, it would take 5-10 years for it to go through development, standardization and browser
implementation in such a way that it was available in a decent percentage of used browsers.
 This is just an enormously slow process.

> HTML5 is not mature enough to compete with flash player right now, but we have to think
about future, and it will improve quite quick with all the hype around it.

Hmm, quick isn't the first word that comes to my mind...

> And even if you prefer flash player, there is our dreams, and there is reallity. And
reallity is that Adobe is pushing HTML5 for web and giving up flash player for that. And we
have to be prepared to a time where there will not be flash player only HTML5 everywhere.
(in several years maybe, but will happen unless something drastically change)

Adobe has absolutely not given up Flash player for web allthough their focus has narrows.
 They may give up on FP eventually but as far as I know they are currently pouring resources
into a AS4, a new VM and Flash player updats.

I do agree that the possible eventuality of no Flash player needs to be considered, I'm just
advocating that the current excellent platform we have isn't scrapped just because this and
that could happen at some point in the future.

> You talk about JavaFX and the discussion among them for the need of a plugin for what
they want. But to achieve that they need Oracle. And they will conclude the same than Flex
community: its only dream and we don't have the power to make it change. You can create a
perfect VM in a plugin, but if you can't install it on devices because of political issues,
then it worse nothing.

No, they don't want a plugin, they want smth like AIR, easy deployment of desktop and mobile
apps on top of a single runtime.

I think we can all agree that there will never be another plugin like Flash player, that is
with both the functionality and enormous reach.

> "This is of course still possible, we just don't know how long it is going to last :-(
But while it works, I hope Apache Flex will continue to be maintained/improved in it's current
> It seems that we all say that as well. Theres no contradiction to maintain/improve the
sdk in its current shape, AND start to prepare its future (in another shape if we don't have
the choice to do differently)
> We need to figure out RIGHT NOW a path for the future of the SDK because :
> 1- it will take time to realise it and get back the same level of features we have today.
especially if the only available solution is to rewrite it.
> 2- we need to give visibility for the future so that IT decision makers can choose Flex
for long term projects. If we don't give this visibility, only short term projects will be
able to use Flex. And it will be seens as "no future" technology in IT world.


> Le 21/11/2012 10:33, Hordur Thordarson a écrit :
>> On 20.11.2012, at 22:14, Kevin Newman wrote:
>>> Mark Zuckerberg also said very publicly that Facebook "burned" (his word) 2 years
of development with HTML5, "We burned two years. That's really painful. Probably we will look
back saying that is one of the *biggest mistakes* if not *the biggest strategic mistake* that
we made." It was less of a "cave" and more of a fundamental shift in understanding (and a
correct one).
>> Agreed and that's really what I ment by "caved in", they just realized it was never
going to be as good as a native app.  The problem with HTML5/JS as an app mechanism is that
it just wasn't designed for that.  Some changes have been made to it in order to make it easier
to write applications (as opposed to web sites which is a totally different thing) but it
really isn't very good for that at all except maybe for small apps.  The JavaFX crowd is having
the exact same discussion the Flex crowd is, except pretty much no one in the JavaFX crowd
wants to deploy to HTML/JS.  They want JavaFX runtimes for mobile so that they can have one
set of code and the same or very similar runtime everywhere (sound familiar ?).  And the community
is actually working towards a solution that gives them that.  But Oracle, like Adobe, seems
to have given into the "HTML5 for everything" rhetoric so they are at least currently not
backing this.
>>> This is where Adobe has an opportunity with AIR, that they seem intent on failing
to capitalize on (at least in their marketing narratives, and the signals the decision makers
are sending out into the market place - the Flash engineers are doing pretty cool stuff with
stage3D and whatnot).
>> Yep, very frustrating that Adobe gave up on this vision because they had by far the
strongest dev/deployment story out there with almost the best of everything. with Flash player
or AIR for the desktop and AIR for mobile and almost single source for all the platforms (UI
tweaks/diffs for phones/tablets obviously).  This is of course still possible, we just don't
know how long it is going to last :-(  But while it works, I hope Apache Flex will continue
to be maintained/improved in it's current shape.
>>> Anyway, Apache Flex doesn't need to wait for Adobe's higher-ups to figure it
out - Flex can go HaXe, and have a multi-platform ubiquity story and an open source story
to boot.
>> Sure.  I have to say though that my clients don't really care if the tools I use
are open source or not or whether the language I write in is ActionScript or Haxe or smth
else.  They care about functionality, usability, cross-platformness and ease of deployment/updating
of the resulting product, and they also want development to cost as little as possible, hence
the less problems I have during dev and the less testing I have to do in multiple browsers
or with multiple runtimes, the better.
>>> Kevin N.
>>> On 11/17/12 5:25 AM, Hordur Thordarson wrote:
>>>> Eventually FB caved in and created a fully native app.

View raw message