Monthly Archives: June 2011

Swiffy from Google Labs – A Dancing Elephant?

There is a bit of buzz this morning about Swiffy. It is a Google Labs project which:

“converts Flash SWF files to HTML5, allowing you to reuse Flash content on devices without a Flash player (such as iPhones and iPads).”

It started as an intern’s summer project but apparently now Google have “formed a team to work on the project”.

Obviously, something like this could have a significant impact on delivering Xcelsius models on iOS but don’t get too excited, just like Wallaby from Adobe this seems (at least for the time being) more suited to converting animated banner adds than anything approaching the sophistication of an Xcelsius model.

Unfortunately, Swiffy only supports ActionScript 2.0 and SWF 8 (both a few years old), so it was inevitable that it would not like an Xcelsius model, nonetheless I couldn’t resist giving it a simple (one gauge and one slider) model to see what it made of it. The results were:

So I suspect it will be a while before we can use this to deliver Xceslius on iOS, more’s the pity. All in all a bit like a dancing Elephant, amazing that it is possible but limited in its usefulness.

Perplexing Xcelsius model – update

Last week I wrote a post about a model which had had been “baffling” the Xcelsius community for years, and got a number of interesting responses, four of which I would like to highlight here.

Andy Bitterer
On Twitter Andy wrote that he thought it was a fake. So it just goes to show that Gartner Analysts are not infallible! Having said that, I did just credit Andy with the “best BI advice I have ever seen in a single tweet” in the recent 5 Unorthodox Principles for Dashboard Success ebook and in a blog post for Everything Xcelsius.

Michael Jones
We know the model is not a fake because Michael (who works for SAP in South Africa) was involved with the original creation of the model and confirmed the technique that they used for the flying dots was an animated transparent XY-chart. If you are interested in how this is done I have put together a short video on the technique (at the foot of the page) or you can download the xlf.

Barry Harmsen
Barry thought that “the more pertinent question is -why- did they do that?”. I suspect his intention was to suggest that they should not have done it. This was also the interpretation Sonny (below) put on it, saying “Barry, great point. What a terrible TERRIBLE dashboard” (note: not just TERRIBLE but “terrible TERRIBLE”). My view is that it is not possible to tell how good or bad this is unless you know what its purpose was. Just because it does not conform to a standard definition of a dashboard does not make it bad (more about definition issues in the eBook).

Sonny Winston
As well as thinking the dashboard was “terrible TERRIBLE”, Sonny also disagreed that it was baffling. While I see his point, and perhaps baffling was a little strong, there is much more to this dashboard than initially meets the eye. As I prepared the video, I realised that not only do the dots fly in a natural arc, they have different end-points, the test-tube increases are all synchronised with the landing of the dots (which disappear on “impact”) and finally there are the area charts at the bottom which track progress. I suspect it would take some time and ingenuity to get all of this working together. All in all, it is a testament to the depth of capabilities in Xcelsius and the creativity of some of the people who use it.

If anyone has any more examples of creative or unusual models created with Xcelsius I would love to hear about them.

A Perplexing Xcelsius Model

The original innovation of Xcelsius was to bring together the flexibility, power and ubiquity of Excel with the visualization of Flash. This meant that a whole new group of people (basically anyone who understood Excel formulas) could produce sophisticated Flash animations. Over the years this has led to an incredible array of Xcelsius models, many of which, I suspect, go well beyond what the creators of Xcelsius intended (or even dreamed of)! However, there is one model which stands alone in that it seems to have baffled pretty much the entire Xcelsius community for the best part of seven years. Every time it is shown (and I must have seen it presented dozens of times) the cry goes up: “How did they do that?!”

Press the play button below and you will see what I mean.

[swfobj src=”” width=”700″ allowfullscreen=”true”]

Can you work out how they did it? Post your ideas in the comments below.

Are you the original author? If so, we’d love to hear from you! You can email me at