It might be possible to make Web Components "just work" in MXML. Not
sure. Definitely worth exploring. It is certainly possible to have HTML
markup be the presentation layer for code written in AS. That's what
Randori is essentially doing.
IMO, early versions of FlexJS need to support IE8. But we will probably
end up with an additional HTML5-only set of components. As those get built
up, we definitely want to try to conform to the Web Components spec.
-Alex
On 9/10/13 8:22 AM, "Peter Ent" <p...@adobe.com> wrote:
>I just looked at Web Components, albeit briefly. I'm under the impression
>that a) it is geared for HTML 5 and b) some parts are still not
>well-defined (e.g., Decorators). Please correct me if I'm wrong about
>that.
>
>One of the nice things about FlexJS is that if you want to use something
>like Web Components, you can. There is already a few HTML 5 components in
>the project, but they are really just starters. It would be great if
>someone could build those out, too.
>
>The current set up might not work as seamlessly as you describe, but
>that's today; anything is possible in the future.
>
>Peter Ent
>Flex SDK Team
>Adobe Systems
>
>On 9/10/13 11:09 AM, "Nick Collins" <ndcoll...@gmail.com> wrote:
>
>>I realize this may be a tad premature, and I don't wish to discredit any
>>of
>>the hard work that guys like Alex have put into FlexJS, but a thought
>>occurred to me that perhaps the Web Components spec could/should be a
>>part
>>of our future considerations for FlexJS. With Web Components, we could
>>create Flex "MXML" tags that would already be backed by JavaScript so you
>>would not even need the compiler step. We would even be able to leverage
>>the {} data-binding just like Angular and other JS frameworks do already.
>>
>>The way I envision it is you write your code in the IDE of your choice,
>>and
>>when you export the release build, you would choose your target just like
>>you do for mobile, and the compiler would only have to translate the AS3
>>=>
>>JS, all the MXML tags would have Web Component analogs that would
>>translate
>>over at least close to 1:1. This may or may not be doable, but it was
>>just
>>a thought that occurred to me.
>>
>>Nick
>
