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 aharui@adobe.com designates
 64.18.1.183 as permitted sender)
From: Alex Harui <aharui@adobe.com>
To: "flex-dev@incubator.apache.org" <flex-dev@incubator.apache.org>
Date: Sun, 21 Oct 2012 10:34:38 -0700
Subject: Re: Starling + Flex
Thread-Topic: Starling + Flex
Thread-Index: Ac2vnK469+AKQg7jSQumk3AkoDDssQAFanX8
Message-ID: <CCA9803E.4776C%aharui@adobe.com>
In-Reply-To: <50840DB2.9020701@gmail.com>
Accept-Language: en-US
Content-Language: en-US
user-agent: Microsoft-Entourage/13.13.0.120411
acceptlanguage: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

The current Flex Framework is a very large code base that contains 50% code
written for AS2 and then ported to AS3 and 50% code written with extra
overhead to support Flash Catalyst.

All of the code is written to work against the Sprite/DisplayList API.

A straight up port will bring all of that baggage with us, and, because of
the interdependencies in the code that is a vestige of what was required to
get AS2 performance, it will be hard to do such a port in phases.

I've been saying from the beginning of Apache Flex that it is time for a
next-generation framework designed to be ported/transcoded/whatever to
different targets, which could include AVMNext/AS4 if it proves worthwhile,
but again, AVMNext doesn't have a good story around accessibility, so while
you could write some simple UI to get someone to pay for your game, I'm not
clear it upscales to true internet applications.

What will that look like?  I have some ideas, but I'm not totally sure yet,
I hope to have some discussions on this at 360Min before bringing the ideas
to the list.

On 10/21/12 7:58 AM, "s=E9bastien Paturel" <sebpatu.flex@gmail.com> wrote:

> Thanks for the update,
> but why would it require a "fairly massive rewrite" exactly?
> when you say "I would look at that" about feathersui, you mean as an
> alternative to flex?
>=20
> its quite a bad news IMO. if flex can't "easely" target a new rendering
> layer like starling, which stay in an Adobe runtime,
> what is the future of flex as a multi platform framework, and especially
> as a mobile SDK?
>=20
>=20
>> After discussing with Thibault and spending time working on this I've
>> determined that a fairly massive rewrite would be required. Shortly afte=
r
>> they released http://feathersui.com/ and I will say that it is a nice bi=
t
>> of work. I would look at that.
>>=20
>> On Sat, Oct 20, 2012 at 8:40 AM, s=E9bastien Paturel
>> <sebpatu.flex@gmail.com>wrote:
>>=20
>>> Hi jonathan,
>>> What is the state of this very promising project?
>>>=20
>>> Le 14/06/2012 18:46, Jonathan Campos a =E9crit :
>>>=20
>>>   Recently I've been getting aquatinted with Starling to see if it coul=
d
>>>> work
>>>> with Flex. After a few days of playing I think that it is possible but=
 I
>>>> do
>>>> see the issues now and there is plenty of work that is necessary to ma=
ke
>>>> it
>>>> happen.
>>>>=20
>>>> To get things going I basically cut down the UIComponent (to the parts
>>>> that
>>>> I cared about), made a copy of the entire Flex framework (as many
>>>> interfaces rely on DisplayObject, etc), replaced them with Starling
>>>> classes, rebuilt some of the Spark primitives, and "adjusted" some of =
the
>>>> starling classes to fit some of the Flash interfaces.
>>>>=20
>>>> At this point I'm definitely going to wait to get the 4.8 release out
>>>> before giving this more time but I think it is possible. I'm sure othe=
r
>>>> developers are already aware of it but if we could make some new
>>>> interfaces
>>>> such as an IDisplayObject it would be much easier switching out
>>>> DisplayObject for a Starling DisplayObject.
>>>>=20
>>>> Just thinking out loud at this point.
>>>>=20
>>>>=20
>>=20
>=20

--=20
Alex Harui
Flex SDK Team
Adobe Systems, Inc.
http://blogs.adobe.com/aharui