great, thanks Cliff - just what I needed to know, and again makes sense.

Picking up on this thread rather late, but it addresses my current questions.

Jason - that was just the confirmation I was looking for regarding proxy notification names and makes complete sense for all the reasons you state. My question is - does this apply to Mediators who send their own notifications too?

Just a thought - I guess an alternative way of doing this would be to have one 'SiteDataProxy' which uses delegates to do each of the 3 remote service calls. The test for all data being loaded would be encapsulated into the SiteDataProxy which then sends a notification only once all data has loaded and been cross-referenced.

In this case my question would be - are the delegates also Proxy subclasses? Is it OK to have references to other proxies inside a proxy?

I have 3 proxies running at startup making remote service calls, and once all 3 are returned, I need to cross reference them. By that I mean the following:
The 'interviews' data is an array with each interview having a one-to-many reference to a fundID.
The 'funds' data contains a one-to-many reference to a groupID with each fund having a fundID.
The third data set is 'groups' with each group having a groupID.

After the cross-referencing is done, I would, for example, be able to reference the name of a group associated with a particular interview thus: interview.fund.group.name

I obviously cannot do the cross-referencing until all 3 sets of data are available. How do I flag when all 3 data sets are available within pureMVC?

The only way I can think of is for each proxy to send a 'data loaded' notification which is responded to by a data-checking command. This would retrieve references to the 3 proxies and query them for whether the data is available. If it is, it then either performs the cross-referencing or send a notification which is responded to by a cross-referencing command.

Am I close with this or is there a much nicer / more correct way of doing it within pureMVC?

Many thanks for any help.

Thanks Cliff, sorry but I can't find the demos anywhere - can you give me a pointer as to where I can find them, thanks

Just getting to know pureMVC and already loving it!

This thread seems to tackle the problem I am having - I need to have (in Flex) a 'blank' application file, to which the designers can drag and drop components and have them just work. (The designers have to be kept isolated from any coding). So from the pureMVC point of view, it doesn't know which components may exist on startup - it could be just one, it could be twenty. How would the ApplicationMediator add itself as a listener to the reusable views which should know nothing about the existence of pureMVC? The only way I could see this working would be to wrap all the generic components in a code-behind wrapper which had a reference to the Application Mediator. But this doesn't feel right somehow as the views are then coupled to pureMVC.