Blog

I don’t get around to saying nice things about Microsoft very often, which I’ll admit isn’t very fair.  They’ve been good to Iris over the years, and we’ve sold our share of Microsoft licenses, too. To give them credit, they are always working on interesting things, and occasionally really interesting work comes out of Redmond. The subject of this blog entry, Pivot, is one of those interesting things.

So, what is Pivot?

It’s kinda hard to describe. One way to start is to list the things it isn’t. It isn’t a replacement for Internet Explorer, Firefox, Safari, or other browsers, although you can use it that way. It isn’t a Digital Asset Management system, although it makes it much easier to find things buried in a huge data haystack. It also isn’t a typical Business Intelligence tool, but it’s easy to see where it could be used to spot trends or patterns in business data, do drill-down analyses, KPI dashboards, or similar functions.

Maybe we should start is by quoting the people that put Pivot together:

Pivot makes it easier to interact with massive amounts of data in ways that are powerful, informative, and fun. We tried to step back and design an interaction model that accommodates the complexity and scale of information rather than the traditional structure of the Web.

The first sentence there describes a benefit of using Pivot, rather than a feature or a functional description, but you can get the meaning pretty easily. It gives you a way to play with information, move it around, “see what comes out of it” kind of experience. You could say the same thing about Business Intelligence tools or DAM’s, too. I don’t know that I’d classify the experience as “fun”, but you can do all kinds of sorts and filters in those tools, too. What Pivot adds is the stuff in the next sentence, which is a way of saying that web search tools are good, but not necessarily good at handling complexities on the fly.

What’s the catch?

There are two big things that have to be dealt with before you can use Pivot. The first one is that it only works on Windows 7 and Vista. For most of the world, that’s not a deal breaker, but for some people it will be a tough pill to swallow.

The other thing is that the data needs to be put together into a “Collection”, which is basically a database of things that will be in your Pivot display. The website for Pivot, http://www.getpivot.com, has descriptions of how you can create your own Collections, different kinds of Collections, etc. It also has a free downloadable plug-in for Excel 2007 (Windows, of course) that can assist you in building these collections. There are also other tools available and probably more to come.

Why don’t I just show you an example of Pivot?

If you’ve got a Windows 7 or Vista machine, you can just download the free Pivot application and just start punching buttons yourself. It’s not very hard at all to get a feel for what it does, and that will probably help you to imagine other ways that Pivot could be deployed. For me, personally, I’d love it to take a crack at our inventory systems we use at work, both our own and at the distributors we use. Or put together a Collection of all the sales transactions we’ve ever done.

But enough of that. Here’s an example using something fairly simple, which is a Collection of the “Top” 500 Movie Actors. To help you imagine what’s going on, picture a big stack of Baseball cards, but instead of players there are Actors on each. Instead of stats on the back, there’s information on what movies they appeared in, a bio, their age, who they co-starred with, etc. Also picture a stack of cards with every movie ever made, with info on who was in the movie, when it was made, what awards it won, etc. Here’s what the Collection would look like in Pivot:

This is a histogram-type view, where the actors are grouped in sets based on how many  movies they appeared in. I know this because above the 1st column is a control that says “Sort: Number of Movies”. On the left hand side of the screen, there are controls that we can use to filter the data. For example, we can click on the term “Number of Movies, and we can grab the handles on the bar underneath the graph to narrow down, slide, or expand the filter set. Since we’re dealing with fewer things, the scale of the images gets a little bigger, and we can now start picking out faces from the deck.

Let’s say that from that group, we want to find all the Actors that have Co-Starred with Steve Buscemi, for example. To do that, I’d just click on the Filter heading that says “Co-starred With”, which will bring up a list of the actors and you can just click on the ones you want to include in the filter. Another way to do it would be to click on “Filter By Keyword” and type in what you’re looking for. But we all love Steve Buscemi, so let’s stick with that. Here’s the result:

We see that Steve has been a busy guy. But now that there are fewer entries, the pictures are bigger, and I just happen to spot Phillip Seymour Hoffman in the Collection. If I want to, I can click on his picture and up will come all kinds of data from the back of his “baseball card” showing on the right hand side of the screen:

It’s hard to see on the last picture, but if you look closely at the top right hand corner of Mr. Hoffman’s picture, you’ll see a green rectangle. On that rectangle is the word “More”, and if you click on that you’ll get a collection of Movies that he’s been in:

From there, you can get the idea. I had forgotten for a moment that he was in “Almost Famous”, but that picture of Kate Hudson reminded me.

Reality Sets In

That may or may not have been a useful thing for anybody to know, but you should be able to see where this would come in handy. Going back to my wish for a unified theory of inventory databases, this could make quick work of trying to find the right part to order for a specific machine made several years ago, or the right software license to order for a particular client, and perhaps even a picture of the widget that I’m looking for. If you’re doing that using a web search engine, then you already know how frustrating that can be. With Pivot and a collection of available parts, I could really speed up that process, and even find alternatives that I may not have been looking for originally.

Another tack would be to sort and filter by other tagged data, such as product movement, compatibility with other parts, or whatever. You could also use this for figuring out trends in customer purchases, identifying key clients in segments (known or unknown), things like that.  If you can sort and filter, and still do goofy things that may or may not help, but it’s definitely more fun than your typical DAM or search engine.

Going Forward

This blog entry doesn’t show some of the more attractive features of Pivot, such as the way things move around when sorts or filters are applied. It also doesn’t show how you can use the history tools to see where you’ve been, undo things, or go back to scratch. To see that, you’ll just have to download it and kick the tires.

Pivot can be embedded in your own websites, and there are different ways in which you can put Collections together. Things can be relatively simple like this Actors Collection, or more complex involving lots of different sets of data interacting in complex ways. Again, the website at www.getpivot.com has more info for developers, and it’s worth taking a look at. As for me, I’m waiting for the definitive Collection on guitars…