I now have the ASP.NET AJAX (Atlas) client-side foundation object model, the forms object model and loading sequence mapped out. I haven't chased down every last detail but I believe I have a pretty solid understanding of the type system and event hooking model.
In the 20+ years I've been a developer, I spent most of those years as an application developer and systems architect for non-software businesses (brokerage mostly). I've also spent a few years with software companies. In general I think that business app developers and software company developers approach hooking into an existing system somewhat differently. As a business app developer, one's responsibility is to support the core business which often means that you gain just enough understanding of a system to do what you have to do - you expand your knowledge incrementally as necessary. When you're building software as the end goal (like when working at a software company), one tends to need a fairly thorough understand of the any system with which they integrate.
I have to say that when I first approached the Atlas/WM problem, I was trying to work as quickly as I could and focus on only those things I absolutely needed to know. Once I became aware how significant the challenge of this problem is, I realized that I need to get a much deeper understanding of Atlas' client-side implementation. Now that I've mapped out the base & forms object models, I believe I now have the foundation of that understanding.
So the net of this dissertation is that I think that I should now be able to get at least a rough solution worked out in just a couple of more days. I'm taking off Wednesday through Friday this week to visit my sister who’s been ill so it may be next week before I have something significant to report. That said, I'm pretty sure that I'm just about to the point where I'll have a 1st-pass rough solution worked out.
Fingers crossed. J
Posted
Sep 23 2007, 06:42 PM
by
jim-wilson