JD on EP

I think Adobe's plans are interesting, but speaking as someone currently developing in Flex, I think the benefits will be more long-term than short-term, and more of these benefits will be strategic wins for Adobe than for individual developers.

I do think that the piece of the Flex pie that would benefit the most (especially in a short-term view) from being open sourced is Flex Builder. I mean, it's a $500 IDE and it doesn't even offer custom color selections for syntax highlighting! And it's not exactly the most responsive IDE I've used. I bought it because it was the best thing available for working with Flex, however I felt like I was being overcharged due to vendor lock-in. Adobe's mantra is "we have a FREE SDK!", but that's like a choice between using a canoe and a motorboat.

I understand the monetization factor, but Adobe would be better off making their money on kick-ass components like the Charting stuff. (monetized downloadable content is the new big thing on game consoles, right?) Open source Flex Builder: give them a free pizza, and they will come back to you for your delicious toppings. It's the best thing for customer adoption of the framework, the best thing for the developer community, and probably the best revenue model for Adobe from a big picture view.

The other main issue that Adobe needs to address is the Data Services package. They've talked a bit about being interested in signing up ISPs licensing deals, but I've seen no traction on this. I'm not arguing for open sourcing of Data Services; it's probably something that should stay commercial for financial viability. However, Macromedia/Adobe is always coming up with these ridiculous server licensing schemes that appeal to only a narrow base of large-scale user. However Flex Data Services is such a core part of the WOW factor in Flex that this needs to change. We need to have licensing schemes that are reasonable for small developers to implement. This needs to happen sooner than later. Data Services Express does not cut the mustard.

Sorry for the somewhat tangential post.

Wake me up when Flex has a native XAML import/export.

You'll be asleep for a while then since XAML is still vaporware, but hey; Microsoft has never tried anything like this before, right?(via
Brandon Ellis)

"You'll be asleep for a while then since XAML is still vaporware"

XAML shipped in the WPF library which is in .NET 3.0

Not sure what you mean. You can judge/rate the technology, but you cannot say XAML is vaporware.

What I was saying is that the faster Adobe/Macr expose a decent XAML import/export functionality, the faster and broader Flex will get adopted.

I don't believe that open source is game changing, especially when only a fraction of it is. Remember RealNetworks "Helix" open source projet? Where is RealNetworks now? (to make up for the lost money the last few years, they settled with Microsoft in the EU. Talk about visionaries).

"Not sure what you mean. You can judge/rate the technology, but you cannot say XAML is vaporware."

You're right, I should have been more specific. I was referring specifically to Silverlight (WPFe) as the WPFe plugin enables cross platform web experiences using among other things, XAML. XAML in the WPF library may have indeed shipped as you said, but Silverlight is still extremely new and as I mentioned in my above post, one of Microsoft's many attempts to (until now, unsuccessfully) compete. Since I don't develop Windows software I don't keep up to date on the latest and greatest for the Windows platform, though I hear its not getting any easier. XAML may be the next big thing in Windows development, and thats wonderful. But in terms of cross platform RIA development, calling Microsoft's stuff vapor ware would be a complement.

Why would Adobe, who is developing its own web and desktop deployment platforms, race to support a competitor's unproven technology? Like most other web developers who create primarily in AJAX and Flash, I'm waiting to see whether Silverlight sticks. If it does then great for Microsoft and maybe the idea of Flex supporting XAML can be revisited. Until then I feel like it would be a waste of time when there are so many other things that Adobe could focus its attention on in regards to Flex.

"I was referring specifically to Silverlight (WPFe) as the WPFe plugin enables cross platform web experiences using among other things, XAML."

It's not accurate either. The Microsoft marketing brochure (i.e. MSDN blogs) says that WPF/E is a subset of XAML, but it isn't. The reality is that whatever is done using a XAML design tool will have to be done twice (once for apps running WPF, the other for apps running WPF/E), debugged twice, tested twice and deployed twice.
It's Microsoft fault here, but I wonder if this is not a strategy to make it harder for third-parties to provide those import/export filters (i.e. interoperability tools) in the first place. In practice it means third parties will need both XAML import/export filters and XAML/E import/export filters.

And of course, WPF/E is not cross-platform. Remember it's Microsoft. At best, it's bi-platform, and if Microsoft history serves something, we already know the Mac version will be bastardized compared to a better optimized Windows version (using DirectX+hardware and all that).

Still, any attempt from Adobe/Macr, any third-party or better yet, the community, at tearing down the walls and making this stuff accessible to Flex/Flash and vice versa is GOOD for everybody.

You cannot say it's not GOOD and that even Adobe/Macr should not invest in that, while the entire history of successful software proves otherwise.

@Alan, why would Adobe spend time and other resources to play nice with a product family that may or may not go anywhere and, quite frankly, sucks right now. That's like saying you'll be interested in Ferrari when they include interoperability with Ford Focus engines.

Microsoft is betting a lot on XAML.

Improving interoperability with XAML is evidence of goodwill. It inclines others to join a community of goodwillers. It's better than putting non-essential parts open source : again, remember what happened to RealNetworks's "open source" helix project.