5 Replies Latest reply on Nov 20, 2010 3:48 AM by MartinHviid

    Would anyone like to help a very experienced, new-to-Flex guy?

    D Baron Level 1

      Hi Folks,

       

      I've been developing GUI apps, on Mac (SuperCard, Omnis, 4D), Windows (VB6), Web (Dot Net C#) and iPhone for a looong time.  I am now considering using Flex/AS to write an in-browser commercial application. Without going into it, let's just say it will have many "screens", will collect both textual data and multimedia content from the user, store it all on a server (dbms), and then output it in multiple ways (back to screen, pdf, flash movie, etc..).

       

      I am trying to understand Flex's "application flow".  I have spent a few hours in a bookstore.  I have spent a bunch of hours in "Tour de Flex" and watched a whole lot of "build an app in an hour" type videos.  I am now coming to you all to ask the basic, but big, question:  How does a sophisticated, multi-screen Flex App "flow".  With standard web stuff, the answser is:  You have a bunch (dozens) of web pages, each their own file; The user "starts out" on one of those pages and then uses navigational widgets to "go to" the other pages.  Don't worry about the back end stuff, I'm not worried about that.   With a standard Mac/Windows app, there are "windows" that you create - each a seperate file.  The user starts out on one of them and then either opens others or "replaces" the current one with another one.  With the iPhone, there are multiple "views", and one or more "view controllers".  The user starts out on one view, and via that view's controller, "goes to" some other view and/or view controller.

       

      How does it work in Flex?  I see only one "stage" thingy.  I understand I can create additional mxml files - are they "other stages"?, or, how does one "go to" them, i.e., replace the current "screen" with the other screen?  I am not interested in using the "States" mechanism - it is great for demos and maybe small-ish apps, but not larger apps, and, I don't like the notion of "mixing" multiple "functional groups of UI/logic" on one "page" thingy.

       

      I have no doubt that I have simply missed some big concept so far...but please, tell me, how does a large, multi-function app "move around"?  How can a big app be "split up" and "dynamically loaded", if and when necessary?  I have no doubt that many of you have done this kind of stuff.  I have no doubt the language and the technology can handle it.  I just haven't "seen" how yet.

       

      Thanks folks.  Based on what I hear back from you all, I'll come back with other questions having to do with the whole skinning controls thing, and questions about "architecture", i.e., seperation of functions (data input/output, business logic, presentation logic).

       

      -David