Tasty Data Goodies

Towards a better upload

We know that it can be challenging to get numbers into Swivel.  In the past few weeks, we've been working on a world-class solution for this problem.

After examining dozens of ideas, we came up with 3 different approaches to evaluate in more detail.

1. Guided graph creation: upload numbers to Swivel in any format. When you graph the numbers, Swivel uses your input to understand the structure of the data.
2. Interactive error correction: when you upload numbers that Swivel doesn't understand, it helps you correct the errors and get your data set into an accepted format.
3. Adapters for common data sources: quickly pull in and graph public numbers that are relevant to you.  For example, stock prices for your company, leads from Salesforce, or trends from Google. 

The video below shows how each solution might work.  What do you think?

User contributed context

We're designing Swivel to let everyone participate and contribute. And it isn't only about the numbers; it's also about your observations and insights into those numbers. Context is what makes numbers meaningful as they relate to a team's project, a business, an industry, or the world.

So we're pretty stoked about a new feature called notes, which Gerad also mentioned last week. And people are already asking for more with notes; rather than annotating a single point in one data column, our early beta users want to add context that spans their entire business. We're already working on new design solutions to these challenges. Here's a sample design we've been kicking around for a shared workspace-wide timeline:

We have a huge backlog of ideas for context we'd like to integrate into Swivel: email conversations, online calendars, RSS feeds, Twitter timelines... the sky is the limit. What contextual time data would you pull into Swivel?

Swivelers at InfoVis 2007

Yesterday Sara and I took a field trip to Sacramento for the InfoVis conference to participate in a panel on The Impact of Social Data Visualization.

It was a great crew of leaders in the field: Martin Wattenberg and Fernanda Viega (Many Eyes) Ola Rosling (Google), all organized and moderated by Robert Kosara (UNCC). I gave a short presentation focusing on our approaches to data sharing, collaboration, and visualization, with some hints at cool emerging topics we're encountering in designing the upcoming version of the site, like privacy and sharing, data ownership, provenance, live data streams, the api, and so on.

Slides from my talk (swivel_vis2007.pdf; 5MB PDF)

I, Robot

1) A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2) A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
3) A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
- Three Laws of Robotics

Robots have come and taken notice of Swivel.  With their help, a swiveler can have a much easier time in extracting data from the world and feeding it into Swivel.  I'm talking about the robots of OpenKapow like the Netflix rental history bot that I created. 

Although the robot did not run into any moral dillemas, crimes of intrigue, or black hole generating end of world doomsday devices, it does snappily retrieve my netflix rental history and output it in csv format.  Then using the Swivel Upload Bot, they faithfully aided me in the pleasure of viewing my rental history in swivel (don't you love it when robots work together?):

You can use this bot too! Just go here and enter the following in the BotURL field (replacing the username and password):

http://service.openkapow.com/tao/netflixrentalhistory?Username=kalendae79%40yahoo.com&Password=XXXXXX

then fill out the form and submit the form and VOILA you will get to see your rental history graphed in its full glory.  Then don't forget to compare it to your work productivity graph =)

As OpenKapow and Swivel both mature more, get more powerful, and become easier to use, I shudder at the wonderful limitless potential!

The era of the robots is upon us and I am glad they found a way to Swivel.

Update: If you are trying the robot out on your netflix account, make sure you look at your uploaded dataset and make sure it's not showing your password. (the title of the dataset, the source, and the source url fields could have it.  all those fields are editable)