Now now, I don’t think he said anything outrageous. Don’t let that Adobe evangelist position go to your head, Ryan

1. I think there IS a lot of frustration with AIR. That doesn’t mean it’s a bad product, but it does mean there’s room to grow, and there will be things that it won’t satisfy developers with (that Silverlight might do better for Windows users). OpenGL, access to peripheral devices, native OS-level libraries, control over OS stuff like taskbar notifications, shell execute… these are all things I’ve heard people complaining about already. Some of them will be solved in future releases, some of them it doesn’t seem like Adobe plans on supporting at all. Recently I’ve been really frustrated with the primitive drag and drop support from AIR to desktop (fairly happy with the other direction). AIR is clearly an early release, and a lot of the stuff that we need to make good desktop apps just isn’t there yet.

2. Regarding better tools and a richer offering, I’m with you on this one. Silverlight ships today with no controls? WTF kind of rich offering is that? I can’t make apps quickly if I have to reinvent every damn UI control for my first project. Flex has a framework that’s been around for a few years and already undergone a few iterations. The kind of shit I can build in a single day because of the rich framework that I start with is insane. I just don’t see Silverlight really working for developers until there’s a solid set of framework controls to build on top of.

I think we’ve tried to be up front about what AIR would and wouldn’t support. And you’re right, it is young, just like Silverlight. I think AIR is filling a niche right now and hopefully that’s going to expand as the product grows. Part of the frustration may also be just because it’s in beta.

Chris, I think the .NET population is significant, but we’re seeing a lot of .NET shops use Flex, so I think they like the idea of a mature RIA platform.

Who said we can’t have a little title baiting around here?

I agree with Scott on the tooling stuff too. I think Flex Builder is great, but Visual Studio has a very, very impassioned user base. And the deployment with the cost means that people are more than willing to pay for it, so I think that’s a sign. Just like designers are willing to pay for our kick ass design tools.

I’ve read both sides points, I in short kind of shrugged both off? what’s new here is probably the question? what made you pause and post?

Ryan, I think you know that when ScottGu makes an on the record comment on something it’s 99.9% done so with respect and keeps a point(s) in full check (if it was a guy like me, you have a case to argue heh).

I think personally AIR needs a better story behind it, not in terms of feature prowess but more .. what’s the story.. show me a story around how it can solve the world’s problems and cure cancer, talk more about that space instead of what it can “potentially” do as that song’s been played (not to be critical, but ever wonder why Silverlight gets massive headlines? – story sounds good, we just need to work harder to put “super duper” features in it.. to be continued.. heh)… that for me is still the missing point that makes me hesitate to agree with AIR’s existence ( I get the features, I can think of a million ways to use it.. but I’m not selling it upstream to decision makers).

I’m still hanging in for AIR’s story, what’s the next turn going to do for me.. I read the various blogs / mailing lists etc (like you do) and with the developer base you have today, there’s some interesting ideas on where they want to take it! – fuel that fire, tell me the story around that path… get me excited Ryan (oh dear, that came out wrong) heh… show me the vision / roadmap of its future beyond 1.0. How does Flex 3, Flex Builder 3, AIR, Open Source, LiveCycle the whole gang.. what’s the intersection story here, how does it all come together and where is it likely to end up?

Any who, Passion is what got you into the Evangelist role and it’s why you reacted the way you did to Scott’s comments (Andrew knows better than most – so I was surprised by his comment: “…Adobe’s view is not to tie you into a particular server language or particular stack of technology…” I hope he was taken out of context heh).

–
Scott Barnes
RIA Evangelist
Microsoft.

but ever wonder why Silverlight gets massive headlines?

Because it’s from Microsoft? Because Microsoft is putting big bucks behind it’s marketing? You think Silverlight is making headlines based on it’s own merits? None of this new tech, with the exception of RoR, is making headlines based on it’s merits. It’s the large corporations pushing it down our throat, telling us how we *need* to learn it or else we’ll be “left behind”.

The “story” for Silverlight is, “Hey, we do that too but with our OWN plugin”. Remember ActiveX vs. Java? Remember how that went? Massive security holes. Deployment issues. COM programming wrapped up in an ugly interface? Of COURSE ActiveX is a better technology that Java applets right? It’s from MICROSOFT.

jd: only .NET? have you not seen the DLR or JScript approach to Silverlight? I disagree that Silverlight is a 1xTrick .NET pony… as examples are the bluedragon folks are looking into enabling coldfusiion developers to write Silverlight code via cfml..

What about IronPython, IronRuby? even SmallTalk? http://vistasmalltalk.wordpress.com/2007/06/05/smalltalk-in-firefox/

Scott: There is an old saying… You can lead a horse to water, but you can’t make it drink. Marketing hype is one thing, getting developers to use it and embrace it.. thats a totally different thing.

-
Scott Barnes
RIA Evangelist
Microsoft.

I’m not sure why everyone keeps saying that existing .NET developers will be able to develop Silverlight like it’s nothing. There’s going to be a lot of learning, especially since the .NET developers aren’t going to be able to just drag-and-drop a text box onto a canvas. And .NET developers definitely aren’t used to creating rich applications like the Flash guys have been doing for a decade. Those former .NET developers still need to learn to design apps that aren’t part of your standard 1990′s crappy MSFT MDI standards (I know this has been disappearing for awhile), and all the same techniques and thought that a Flex developer needs to know for creating the rich applications. Not to mention, I’m a .NET C# developer and it took me all of five minutes to learn ActionScript 3.0. Another couple of hours to learn some of the more interesting features like e4x and all it’s dynamic functionality. Use the right tool for the right job. Want the richest browser experience? Use Flex. Period.
Another argument that Microsoft keeps making about AIR is that the applications don’t look native to the OS. I think it’s in the quote above, too, by Gutherie. Well, Office 2007 no longer looks like a native application windows application. It’s so different from all the other apps running on my system. Long gone are the standards on how to create a windows application. Now, as MSFT proved with Office 2007, each application is going to have it’s own look and feel determined on what the user’s supposed to accomplish. Just look at how rich web applications have been designed. That’s where user interfaces are going, or have gone.
One great thing about Adobe’s Flex offering is that they’ve done a kick-ass job of making their client development platform very agnostic about the backend it connects to. It’s just as easy to connect to a MSFT IIS server with SOAP as it is a Tomcat Java Server serving up RESTful services. This is the big advantage of Adobe, I think, it’s totally backend agnostic. With MSFT, I know they’re building their stuff to tie into their backends primarily, and then any interoperability secondary. They want to sell all those expensive server licenses. (I’ve even had to prototype a server with and entity model written with LINQ, accessing SQL Server, and then sending JSON objects back to a Flex application.)
What I can do with Flex TODAY is WAAAAY ahead of what one can do with Silverlight. Sure, MSFT will one day get there…but it ain’t there, which is why me, as a .NET guy, came over to FLEX. I wanted to create the future now.
Another example of how far MSFT lags behind cutting edge is looking at it’s adoption of AJAX. ASP.NET 3.0 offering finally included AJAX. All the .NET developers were at least 2 years behind on creating AJAX applications — unless they were free thinkers and integrated some third party tools.
I’m sure I’ll be working with Silverlight at some point, like when I can use Python, and when the runtime is proven to be cross-browser/cross-platform, and their’s controls, and there’s an upgrade path to release my consumer applications to OSX. Or, when I’m working inside some corporations firewall and because of some golf-course handshake deal between the CEO and Microsoft Sales, I’ll have to use the Silverlight platform. Silverlight is only a natural pathway for corporations already bought into MSDN subscriptions whose development tools will just include the proper Silverlight tools. Nothing new to see here.
Microsoft doesn’t have a STORY that AIR meets. Their evangelists are claiming one isn’t needed. But watch in three years when people are using rich applications that are both connected and disconnected and that don’t come from MICROSOFT.

Robert said what I was trying to say, in much fewer words. It’s all going to be a commodity in a few years regardless, just look at the database market. The same will be true for RIA.

So Visual Studio 2008 launched this week. Silverlight controls and all. .Net 3.5 framework and Linq. For a programmer, this is a dream come true –an all in one package.

You can program, link (linq) back to just about any type of data source and have an increadably rich media front end for your UI that a graphic designer can work on while you are developing…

Hats of to MS.

Flex, well what can I say. Still a long way to go and no strong clear roadmap.

Absolutely fantastic. If you had done your research, you would have realized that Scott has also worked on the VS project… if only you had have waited as he did have insight as to what was coming.