Googling I found this interesting presentation about Flex 4 (aka Gumbo) next plan :
Flex 4 SDK next plan by Matt Chotin
Flex 4 SDK presentation in PDF
These are the 3 Primary Themes for Flex 4 SDK release:
Design in Mind
Developer Productivity
Framework Evolution
Flex 4 - Design in Mind
Adjust the MXML language to support more toolability and make it easier to
describe experience-oriented features such as states and transitions (MXML
2009)
Create a format that promotes the features available in the Flash Player and
is usable by tools to describe various graphic assets, including skins (FXG)
Provide a component and skinning architecture that invites easy tool
participation (code name Gumbo)
Improve existing experience-oriented features such as states, effects, and
layout
Integrate new functionality (Gumbo) into existing component model (Halo)
Create new design-friendly components
Initial set of ~10 required by new target use-cases
Additional set to prove out architecture
Size of resulting applications within same order of magnitude as Flex 3
Low-hanging size and performance improvements
Alternative approach considered: Lightweight, design-friendly framework
built ground-up
Customer research indicated a near-term requirement for either functional parity or interop
with Halo, and even then much harder to adopt incrementally
Unable to deliver in desired timeframe
Much harder on the tools
Flex 4 - Target Application
All apps supported in Flex 3; e.g.,
Data visualization
Line of business
Simple event-driven site, e.g., portfolio
Selector / Simple Item Browser
Interface for medium-complexity RIA/product selector
Interactive widget
Flex 4 - Developer Productivity
Compiler Performance
2-way binding
Automation support for Flex in AIR
CSS Improvements
Multiple styleNames (space delimited)
Descendant and child selectors
id selectors
Framework evolution with Flash Player 10
Leverage new text features available from Flash Player 10
Bidirectional layouts (mirroring) and text support
New Video Component
Based on work being done by FMS team to make underlying video logic much more robust
Comments