My perception as an engineer is that our goal of maximizing Flex adoption is more important than our goal of promoting FlexBuilder as our IDE for it. We want people to be able to write Flex apps with the tools of their choice -- NotePad, Eclipse, whatever --if they prefer them to FlexBuilder. In fact, I still write most of my Flex code with Visual C++, although that will probably change.

 

However, our intention to be to open to other tools doesn't mean that we want to devote engineering resources to continue supporting a Schema file for Flex, since our experience with XSD proved to us that it wasn't designed to describe a language with the flexibility (pun possibly intended) of MXML. As with any Flex feature, this involved a tradeoff of cost vs. benefit and it didn't make the cut in our feature selection for Flex 2.

 

I'm glad to hear that the community may be able to cobble together a "good enough" schema for Flex 2 based on the one we developed for Flex 1.X.

 

- Gordon

 


From: [email protected] [mailto:[email protected]] On Behalf Of Jason
Sent: Friday, June 16, 2006 7:47 AM
To: [email protected]
Subject: [flexcoders] Re: DTD Schema for Flex 2

 

--- In [EMAIL PROTECTED]ups.com, Tom Chiverton <tom.chiverton@...>
wrote:
>
> On Thursday 15 June 2006 22:10, Gordon Smith wrote:
> > For example, one of the rules of MXML is that you can write a scalar
> > property as either an attribute or a child tag. That isn't something
>
> ...that many people do.

I've done it, and I've only been using Flex for a couple weeks.
Instead of using the dataProvider attribute, I defined my dataProvider
as a child component. In fact, if you look at "Creating a DataGrid
control", you'll see they do the same thing.

But it's a moot point. If the language allows it and xsd doesn't,
then the xsd wouldn't always validate the code. So what good is it?

> > that is easy to describe in Schema. Also, you can put controls inside
> > containers, but not containers inside controls... well, except for
>
> It doesn't have to be perfect (esp. if its a best effort labs
project or blog
> posting)·

Why doesn't it have to be perfect? If it's not perfect, it won't
validate mxml. It will validate some subset of mxml. That's hardly
good enough for something you want to use to replace Flex Builder.
And as soon as they changed some components or added new ones, they'd
have to go monkey with the schema again, which would still be less
than complete because of the shortcomings of XSD.

And as soon as you download someone else's piece of code to use -
you're screwed. Because THEY sure didn't write a schema. So now you
either get to leave out their code, mark some areas as
non-validatable, or write their schema for them. Doesn't sound very
appealing to me.

Personally, I'd much rather them spend the same time fixing bugs than
creating and updating a schema that's only good for some certain small
subset of projects. I don't think it's some grand conspiracy to keep
any third-party editor from existing. I think they just build the
best language they could and said if anyone wanted to build an IDE,
that's fine with them. But I'm glad they didn't hobble their IDE to
only allow things that are stricly xml schema-compatible.

__._,_.___

--
Flexcoders Mailing List
FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt
Search Archives: http://www.mail-archive.com/flexcoders%40yahoogroups.com





SPONSORED LINKS
Web site design development Computer software development Software design and development
Macromedia flex Software development best practice


YAHOO! GROUPS LINKS




__,_._,___

Reply via email to