An article posted on Techcrunch (see also the interesting comment stream following the article) introduces Swivel, a site and online service of obvious interest to the more nerdy among us, and less obviously of much deeper significance.
Swivel allows users to upload data, which can then be viewed and analysed in various ways — it’s frequently described in marketing-speak as ‘YouTube for data’.
However, it’s not so much the core functionality that is interesting (you can do far more extensive analysis with the open source package ‘”R”:http://www.r-project.org/’, or commercial tools like Matlab and Mathematica), as what it represents in terms of the relationship between where data originates and where it ends up. Swivel represents a kind of ‘freedom of data movement’ where data generated and perhaps pre-processed in one location is able to migrate to another part of the world drawn by the pull of a specialised service offering that is not available elsewhere.
In a sense this is ‘Service Orientated Architecture’ (SOA) writ-large, in the spirit of Amazon’s recent grid computing and storage offerings. The concept itself isn’t new, but this application of it has a certain eye-catching X-factor about it akin to concepts from the world of ubiquitous computing whereby information and functionality are able to freely coalesce and co-locate for greatest utility.
An obvious question on everyone’s lips is “so, will Google buy them or blow them away?” Personally I would fear the latter, as Google is already vastly endowed with number crunching ability, and services like Google analytics and Google trends show considerable charting capabilities. Time will tell; it probably makes sense for Google and others to wait and see what direction Swivel goes and grows in based on the demands both of its user-base and commercial necessity.
December 10th, 2006 at 3:14 am
Tom,
Thank you for the thoughtful post. Neither Brian nor I can really take credit for the “Youtube for Data” line, although it seems to be catching on. The model we aspire to is probably closer to that of Wikipedia. Hope you’ll get a chance to play with Swivel and give us your feedback.
Best regards,
Dmitry Dimov
Product Chief & Cofounder
http://www.swivel.com
December 10th, 2006 at 6:35 am
Thanks for the post about Swivel.
You’re insight about the core functionality is right on. The software package that I most admire across any category is Mathematica. It’s untouchable, in my opinion, for digging into and visualizing data. And it’s beautiful looking and beautifully architected and consistent.
Where we think Swivel provides an interesting take on core functionality is in comparison with a spreadsheet. A spreadsheet is, of course, a grid of rows and columns. In a spreadsheet about public schools the column C could be for ‘Program’ info and then row 9 could have a value of, say, ‘Japanese’.
In Swivel we move from that cartesian geometry to the geometry of the Web. That same cell, C9, in Swivel becomes a full blown Web page (http://www.swivel.com/data_columns/show/1032871) with a myriad of links, graphs, pictures and colors. With Swivel the cell in a grid becomes a node on a web. The value to the user takes us back to the original vision of Tim Berners-Lee: using hyperlinks to link any unit of information to any other unit of information. When a user approaches the data, they ought to experience a velocity of learning (through surfing) that we all get every day surfing Web content.
So, we think Swivel is valuable even with just with one user analyzing one data set.
Time will tell. Thanks again for the post.
Brian Mulloy
CEO & Cofounder
http://www.swivel.com