Many Eyes

This week’s resource up for discussion is a tool that transforms plain data into eye-catching visuals.

On the website, users upload a standardized set of data (like from an Excel spreadsheet) and can then inspect it from a variety of perspectives. You can browse the gallery of visuals created by users, but the real fun is uploading your own data (it takes about 15 minutes to go through the process) and seeing it in new ways.

Let us know what you did during your visit to Many Eyes. Is their website easy to understand and use? What did you think of existing visualizations? Did you upload your own data? Did you get any new ideas? What is lacking from this kind of visualization tool?

12 Responses to “Many Eyes”

  1. Jan Kunnas Says:

    I just took a look at the Many Eyes -page, or to be more correct tried to. At least with my connection it seemed to be very slow. There seemed to be a lot of different nice ways to visualize your data.
    I would though think twice using it to visualize work in progress, as you first have to upload your preliminary datasets. You could of course upload your data, create your visualizations and the take your dataset out again. Before that: “If you have made a visualization of the data set, you’ll need to delete the visualization before the data set.”
    Another thing would be to use it for already published data, if you still have the copyright in it.

  2. Cathy Hajo Says:

    Well, that was an exercise in frustration. I got very excited about this application, because I have a database that I want to attach to a map to play with. I read over the instructions, selected and uploaded my data, and tried to get the visualization to work. First of all, the fact that you need to upload your data and “everyone can see it” but you can’t remove it if you want to correct it is annoying. Even though the instructions said that the application could read abbreviations for states, I had to hand match each state and county combination. The application offered suggestions in some cases that made sense, in others it was wildly offbase, and in some cases, it did not allow me to select the correct combination at all. I edited my data and tried it again, correcting all the errors I could, and loading about 2950 records. My records were simple–a year, a city, county, state, and type of birth control clinic (4 choices). I identifed the state and county fields, and told it to treat years as text, not numbers.

    When the map came up, http://manyeyes.alphaworks.ibm.com/manyeyes/visualizations/frustrating-map-of-clinics the years were treated as numbers, and the resulting map is pretty much useless. When I look at “type” it seems to select one of the clinics in a county, maybe the most numerous, I don’t know, and display it. So New York County is colored blue for independent clinics, but there were also hospital clinics and settlement clinics there. When I look at year, it has treated it like a number, not text, so that the circles are all essentially meaningless–they are all in the top range, because they are all years between 1916-1940.

    I didn’t try the other visualizations, because I spent about 3 hours on this one, cleaning data, trying again, etc. It seemed like a way to do some GIS-light work on my data, but I don’t think that it is something I’ll want to mess around with again any time soon.

  3. tim vermande Says:

    It also seems slow to me, and not suited for preliminary work. But for my visual students, it would be a great tool for materials when I had the time to get all of the data together and wait for it to churn out something (because even as slow as it is, it would be faster than other methods available here at the present).

  4. Kelly Kennington Says:

    My internet connection is also slow, so I was not able to upload my data or create my own visualization. That being said, I like the variety of types of visualizations one can make. They would benefit visual learners and would make interesting illustrations in publications, helping the reader visualize the author’s data.

  5. Jeff Tenuth Says:

    As with the other examples we’ve looked at, this also is an interesting visualization of data that can be useful if one is looking at distribution, percentage, range, or other statistical data, but beyond that I don’t see much of a use for historical analysis. This is the type of approach that I would begin an analysis with, but I would never end it there because it doesn’t tell me anything more that what’s already there. Many Eyes is simply a variety of ways of looking at data. They help present data as part of an initial anaylsis, but as for understanding events, trends or patterns in history, I don’t see it. Visualizations cannot yet show context, origins or consequences; at least not that I’ve seen so far.

  6. Andrew H. Lee Says:

    First thoughts. Okay, I need to get my data into it to really evaluate it but reading Cathy’s comments maybe I need to wait until I am snowed in. In order to look at what others had as data sets that i may be interested in I searched for Spain and got a fair number of results. But also “Common Sense” by Thomas Paine. Is the site using some fuzzy logic searching? And the colors on the site are too pastel — I would look better with other colors. Again, these are initial impressions but unlike Stanford’s site which made me want to really explore it the Many Eyes site does not. And that they have no maps of Spain… judging a book by its cover but I expect IBM could do better.

  7. Shawn Barron Says:

    In my opinion many of these presentations are visual aids not really scholarly sources. I think that the programs uploaded to this site provide both fascinating springboards for further exploration as well as making bold visual examples for presentations. Hearing about how much more severe US industrial emissions is one thing, but having those emissions given dimensions is something else.
    It seems that anyone can submit to this site which is both a curse and a blessing. More ideas are always positive, however many of the visualizations were either of limit interaction or else didn’t explain how they worked very well. In the end i think this is a pretty fascinating site which maybe will inspire em to make a visualization of my own.

  8. ajopp Says:

    Many Eyes is a tool that allows people to explore different visual images of large amounts of data and share it with others to help them collectively make better sense of the information.
    Their site says “Many Eyes is a bet on the power of human visual intelligence to find patterns. Our goal is to “democratize” visualization and to enable a new social kind of data analysis. Jump right to our visualizations now, take a tour, or read on for a leisurely explanation of the project.”

    I liked its interactivity; I could move the cursor over a spot on the visualization and see the principal data. However, I am not sure if the data people put in to build visualization is correct. In addition, there seem to be a lot of data sets that were uploaded as tests and never removed. As a user I found it difficult to navigate among the datasets. Also, I was not able to figure out how Many Eyes can mix and match parts of different datasets.

    I definitely liked the variety of topics -from bird sightings to the distribution of wealthy people by state. This set of interactive visualizations provides insight into different topics. Besides, presenting the data in visual form helps to understand any material better.
    The tool supports different types of visualization including network diagrams, bar charts, bubble charts, block histograms and world maps.

    This tool is one more example of how knowledge can be distributed in new media.

  9. AUrbanski Says:

    I am most impressed about this tool because it is a great example of how far the digital age has come. In class we saw how (relatively) simple it is to place your data on a spreadsheet and then upload it onto this site. What also makes this site attractive to users, I’m sure, are the varieties of visualizations you can selet to display your data. I have no data to upload, unfortunately, but if I did I would be interested to see what I can come up with on this site. I’m not sure why I thought about this, but it would be really cool to create a visualization showing which U.S. states have produced the most military leaders over time, or which states have sent the most men to war over time. (Yes, I know the north and south would be favored, but it still might be interesting to see and show, especially if you were doing a Veteran’s Day project).

  10. Nadine Q Says:

    I was really excited about using this in my final project, but it took forever uploading the data and I wasn’t sure if it worked correctly. However, I did find the other examples to be very interactive and visually appealing. The array of choices to use for the visual presentation is what attracted me to use this tool. With the different ways of presenting your data, you can compare which one is more effective than the other. Using this tool, it would be really great to see which state holds the largest private collection of works of art versus which state has the largest public collection in museums.

  11. Nabeel Siddiqui Says:

    I actually want to discuss how these tools may help a digital historian instead of just history in general. I think other people have already discussed how we can use these programs for historical research, but one of the visualizations got me thinking about how we could use this to expand digital history. Namely, it was a visualization of some code and what functions, variables, etc. were used in a program.

    In my opinion, this can be extremely helpful in creating standards that can help digital history expand, and I am very surprised that I did not ever think about mining source code. I think this will be extremely useful in determining standards for the internet. For example, we can take all the historical sites and then we can examine the source code to examine how the standards of programming can be used for mining. I think this can be really helpful for how we can help expand digital history.

  12. Ashley Anttila Says:

    I have to agree with Shawn. Many Eyes provides great visualization options for teaching or demonstrating but I don’t think it’s a good option for research. It seems fairly user friendly and there are plenty of options for displaying data in unique ways. But it doesn’t allow for the kinds of data manipulation available in SEASR, for example. In my opinion, the value of Many Eyes is in the ability to make a visualization quickly, not to manipulate data for the purpose of research and analysis.

Enhancing Historical Research With Text-Mining and Analysis Tools