[Sigia-l] The IA in RIAs

[Bill Pawlak]

> The limitations of traditional x-browser HTML has been severely
> internalized to an almost oppressive level on most IAs. Now we as a
> group get to think more richly about the total possibility of
> solutions.

Amen!   I think another problem is that many practicing IAs/IxDers
today have only ever worked on HTML-based applications.  To me, the
whole concept of RIAs is not a "new" thing, but more of a return to
the past, in many ways.

An example: I worked on an intermittently-connected airspace
configuration tool, complete with drag and drop capabilities, rich
data graphs, and "paintable" route canvases (Tcl/Tk & OpenGL for the
UI) back in 1995.  The shift to page-based, HTML apps was staggeringly
restrictive back then and I view the ability to allow users to
manipulate data using a richer interactive environment, without having
to deal with navigation back-and-forth between a "setup page" and a
"view page" as simply a return to where we used to be with respect to
interactive application development.
 
> I would caution suggesting that Java is an RIA environment. The reason
> I would caution that is b/c to me Java is not as easily distributable
> as both Flash or HTML and the runtime engine for Java is not as
> ubiquitous as that of Flash or HTML.

It seems like one of the problems with discussing "RIAs" is the fact
that we're all using different definitions.

Does the definition of RIA necessarily include the
depth-of-distribution of the implementation environment?  Do RIAs have
to be, by default, only partially-connected clients?  Do RIAs have to
have every thing on one screen a-la the Broadmoor Hotel example?

I don't believe that to be the case, so I tend to view RIAs more as
"nothing new here."  If one's definition of RIAs *includes* the
widespread availability of the runtime-environment (JRE vs. Flash, for
example) then of course there's going to be differences of opinion.

bill