MindCanvas: Reimagining a software category as a Rich Internet Application

AJAX and Rich Internet Applications offer the chance to re-imagine product categories that were previously thought of as mature, finished, and mundane. The first example of this was GMail, which succeeded in re-inventing the moribund category of web mail. More recently, a number of products have re-imagined the word processor as a web application.
I think that, sooner or later, pretty much every application category is going to be rebuilt from the ground up as a rich internet application. It is in this spirit that my crew (<a href="hey guys) and I built MindCanvas. It is a web survey research service, re-imagined as a rich internet application. The process of re-imagining changes the application so much that I’m reluctant to even call what we’ve built survey software. But that’s the easiest way to explain what it is. Now that we’ve released a beta version of the system, I’m finally free to talk about it a little!

Having gone through an entire development cycle of a rich internet application, I can say with confidence that making an application with AJAX or Flash changes EVERYTHING about the application. The opportunity to run “real” software within the browser opens up the potential to do things that will change the application you’re building in unpredictable ways. Game-changing ways.
For example, the survey experience in MindCanvas is way more engrossing than a typical survey. A typical question in one of our surveys will involve the user making hundreds of decisions, contributing dozens of snippets of text, and in general doing high-level creative/cognitive design work. This is the kind of complex task that you wouldn’t even think of implementing with a vanilla html interface. But it’s a lot more valuable to invite your users to participate in your design process then it is to barrage them with a bunch of dumb multiple-choice questions. The rich interface opens up possibilities that weren’t even considered before. We call the questions “Game-Like Elicitation Methods”, because they have a lot more in common with Flash games on the internet then static, boring customer satisfaction surveys.
(Note: Information Architect types will recognize these questions as being various types of card-sorting. Bingo! We’re going to implement more and more types of questions, but for now, card sorting is our strongest single question type.)
(Note 2: MindCanvas does support multiple-choice questions and stuff like that. But if that’s the main thing you want to do you’re much better served by a system like surveymonkey, which is really good for that kind of thing, and is free to boot!).
Similarly, a typical survey report contains a lot of bar charts and spreadsheets. MindCanvas does something totally different: the reports from our surveys are animated visualizations that allow the client to drill down into the “work” done by all of the users who took the survey, seeing the results of the creative efforts of hundreds of people in aggregate. And because they are Flash files, they can be embedded into PowerPoint or emailed as attachments.
Rashmi Sinha is the methods geek for MindCanvas, so her blog is the best place to learn more about Game-like Elicitation Methods. Check out this post Rashmi wrote, it explaines the GEM concept in better detail, and also gives more background on why we built this service.
Now that we’ve gone out of “stealth” mode, I’ll be blogging more about what I learned in the process of building a large RIA (tens of thousands of lines of java, tens of thousands of line of actionscript).
In the coming weeks I’ll be writing about some of the decisions we made, stuff I wish we’d done differently, and lessons learned. As a teaser, I will say I’ve definitely learned where to use Flash and when not to use Flash. ;->