Comments on: Why I Do Not Swivel Data http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/ Strength in Numbers Wed, 17 Mar 2010 23:24:30 -0400 http://wordpress.org/?v=2.9.2 hourly 1 By: 8 Reasons Why I Do Not Like Data360 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-1167 8 Reasons Why I Do Not Like Data360 Fri, 11 Jan 2008 20:56:46 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-1167 [...] page, to the drab graphics and slideshows, I feel bored looking at it. I’ve criticized Swivel enough, but at least they make the data somewhat [...] [...] page, to the drab graphics and slideshows, I feel bored looking at it. I’ve criticized Swivel enough, but at least they make the data somewhat [...]

]]> By: inquisant » Blog Archive » Whats wrong with Swivel? http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-103 inquisant » Blog Archive » Whats wrong with Swivel? Thu, 01 Nov 2007 20:29:46 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-103 [...] for creating and standing behind data exposes Swivel to all kinds of valid criticism like this excellent post at flowing data. Yahoo creates the applications and the content. Google stays away from creating content and [...] [...] for creating and standing behind data exposes Swivel to all kinds of valid criticism like this excellent post at flowing data. Yahoo creates the applications and the content. Google stays away from creating content and [...]

]]> By: FlowingData / Swivel Gets a Data Guy http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-27 FlowingData / Swivel Gets a Data Guy Wed, 03 Oct 2007 08:56:51 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-27 [...] should be good for Swivel, who has seemed to be missing the “data guy” piece of the puzzle. Will Swivel’s visualization tools improve? Will data become more [...] [...] should be good for Swivel, who has seemed to be missing the “data guy” piece of the puzzle. Will Swivel’s visualization tools improve? Will data become more [...]

]]> By: Tal Galili http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-26 Tal Galili Wed, 12 Sep 2007 10:32:17 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-26 I have nothing much to add to the discussion, besides mentioning it is very interesting to read - thank you for your post. I have nothing much to add to the discussion, besides mentioning it is very interesting to read – thank you for your post.

]]> By: Nathan http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-25 Nathan Fri, 07 Sep 2007 23:18:22 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-25 @vozome: You bring up some excellent points and it's GREAT that the OECD is aiming to make their data more available. I agree that Swivel is playing a good role in changing the way people consider publishing data. As strictly a data store, Swivel is on the right track, but is this what Swivel aims to be? I don't think it is. Swivel markets itself as an application to upload <em>and</em> explore data. With that in mind, how many people in Swivel's wide audience will download the data and explore with their desktop tools? That's when all the Swivel graphs start to trouble me, because, as you noted, they're not especially useful and many (or most) are useless. Then the trouble begins when people start "interpreting" the graphs, perhaps seeing something that isn't really there. @vozome: You bring up some excellent points and it’s GREAT that the OECD is aiming to make their data more available. I agree that Swivel is playing a good role in changing the way people consider publishing data.

As strictly a data store, Swivel is on the right track, but is this what Swivel aims to be? I don’t think it is. Swivel markets itself as an application to upload and explore data.

With that in mind, how many people in Swivel’s wide audience will download the data and explore with their desktop tools? That’s when all the Swivel graphs start to trouble me, because, as you noted, they’re not especially useful and many (or most) are useless. Then the trouble begins when people start “interpreting” the graphs, perhaps seeing something that isn’t really there.

]]> By: vozome http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-24 vozome Fri, 07 Sep 2007 16:38:42 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-24 Hi Nathan, I work at the OECD. In 3 words my job there is to publish official data. Early on at the OECD we have been interested in working with Swivel as we like their approach. Traditionally, we publish large & complex datasets complete with detailed source & methods documentation. Obviously we are going to continue doing that but we feel that people outside of statistical offices, central banks and research labs should also get access to numbers from the OECD that they can comprehend. Enter swivel. As you can guess when I first tried it I had a number of gripes. obviously the data visualization tools were not as powerful as what we have inhouse. plus it seemed much easier to build a completely irrelevant graph than something meaningful. Yet there're quite a few things that swivel did very well: allow us to add extra information to our heart's contents, tags, interact with users via comments, etc. So while I was less than thrilled by the content we could post I was quite excited by how it could be published. After a while Swivel and us started talking. We were so concerned that people could produce junk graphs out of our content which would still bear our brand, that together we came up with the concept of "official source", which they implemented in a couple of days. Then, each time we provided feedback on content & how to do stuff better on the purely statistical front they listened to us and we saw the application evolve very rapidly. We still talk and exchange ideas. Furthermore, ever since we started exploring swivel, quite a few other data-oriented IGO's and official data providers have been curious. We talk together, they talk to swivel, who proves to be very receptive. Swivel are working on quite a few ideas which I believe would dramatically improve the site & the service and some of these ideas and suggestion come from veteran statisticians all around. Eventually, swivel caters to a wide audience, so they'll focus on the kind of data that's understandable by many, rather than on expert analysis. You'll still see fun graphs put forward which may hurt the credibility of more relevant data. Yet I believe that it has a role to play. At the very least it challenges the way we "official" data people consider publishing data, which is already quite an accomplishment!!! Hi Nathan,

I work at the OECD. In 3 words my job there is to publish official data.

Early on at the OECD we have been interested in working with Swivel as we like their approach.

Traditionally, we publish large & complex datasets complete with detailed source & methods documentation. Obviously we are going to continue doing that but we feel that people outside of statistical offices, central banks and research labs should also get access to numbers from the OECD that they can comprehend.

Enter swivel. As you can guess when I first tried it I had a number of gripes. obviously the data visualization tools were not as powerful as what we have inhouse. plus it seemed much easier to build a completely irrelevant graph than something meaningful. Yet there’re quite a few things that swivel did very well: allow us to add extra information to our heart’s contents, tags, interact with users via comments, etc. So while I was less than thrilled by the content we could post I was quite excited by how it could be published.
After a while Swivel and us started talking. We were so concerned that people could produce junk graphs out of our content which would still bear our brand, that together we came up with the concept of “official source”, which they implemented in a couple of days. Then, each time we provided feedback on content & how to do stuff better on the purely statistical front they listened to us and we saw the application evolve very rapidly. We still talk and exchange ideas. Furthermore, ever since we started exploring swivel, quite a few other data-oriented IGO’s and official data providers have been curious. We talk together, they talk to swivel, who proves to be very receptive. Swivel are working on quite a few ideas which I believe would dramatically improve the site & the service and some of these ideas and suggestion come from veteran statisticians all around.

Eventually, swivel caters to a wide audience, so they’ll focus on the kind of data that’s understandable by many, rather than on expert analysis. You’ll still see fun graphs put forward which may hurt the credibility of more relevant data. Yet I believe that it has a role to play. At the very least it challenges the way we “official” data people consider publishing data, which is already quite an accomplishment!!!

]]> By: Nathan http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-23 Nathan Sat, 01 Sep 2007 21:10:19 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-23 @Brian: Thanks for hearing me out. I know it might seem like I'm completely against you and Swivel, but you should know that I'm rooting for Swivel's success. I just hope that it's success for all the right reasons. @Tom: Data360 (among others) have been in my field of vision for a while now, so a review of some sort is definitely on my todo list. @Brian: Thanks for hearing me out. I know it might seem like I’m completely against you and Swivel, but you should know that I’m rooting for Swivel’s success. I just hope that it’s success for all the right reasons.

@Tom: Data360 (among others) have been in my field of vision for a while now, so a review of some sort is definitely on my todo list.

]]> By: Tom Paper http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-22 Tom Paper Sat, 01 Sep 2007 06:17:34 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-22 Nathan, Swivel has a lot going for it and, as you pointed out, it also has challenges. All the data visualization sites are in the early stages, so each one has ability to respond to feedback. Data360 is website where people can find, analyze and present data. We see ourselves as more serious than Swivel and we think that our analytical, presentation and reporting features are more robust, as well as more dependent upon the judgment of the person posting the data. My background is business finance and strategy. I would be grateful for your review of our site, which can be found at www.data360.org. Best regards, Tom Paper Nathan, Swivel has a lot going for it and, as you pointed out, it also has challenges. All the data visualization sites are in the early stages, so each one has ability to respond to feedback. Data360 is website where people can find, analyze and present data. We see ourselves as more serious than Swivel and we think that our analytical, presentation and reporting features are more robust, as well as more dependent upon the judgment of the person posting the data. My background is business finance and strategy. I would be grateful for your review of our site, which can be found at http://www.data360.org. Best regards, Tom Paper

]]> By: Brian Mulloy http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-21 Brian Mulloy Sat, 01 Sep 2007 00:27:32 +0000 http://flowingdata.com/2007/08/31/why-i-do-not-swivel-data/#comment-21 Thanks for your very thoughtful post about Swivel. We whole-heartedly agree with your concerns about visualization, data quality and statistical validity. Swivel has been alive for less than 9 months and we really see ourselves as approaching the start line not the finish line. We hope to address many of your concerns over the coming months. We are a business and it is our deeply held belief that we will do well by doing good. Creating a Web site that allows people to engage with data in a meaningful way is the key to enabling great decisions and improving people's lives both in the private sector and the public sector. Brian Mulloy CEO & Co-founder Swivel brian@swivel.com 415.680.3641 Thanks for your very thoughtful post about Swivel. We whole-heartedly agree with your concerns about visualization, data quality and statistical validity.

Swivel has been alive for less than 9 months and we really see ourselves as approaching the start line not the finish line. We hope to address many of your concerns over the coming months.

We are a business and it is our deeply held belief that we will do well by doing good. Creating a Web site that allows people to engage with data in a meaningful way is the key to enabling great decisions and improving people’s lives both in the private sector and the public sector.

Brian Mulloy
CEO & Co-founder
Swivel
brian@swivel.com
415.680.3641

]]>