This week’s discussion focuses on visualizing texts. To begin, visit wordle.net and create your own visualization of plain text (you can paste it onto their site) or a webpage (you can paste in a URL).

Although Wordle itself is probably not a suitable research tool, can you imagine using visualization tools like this for your own sources? What kind of features would make such a tool more useful?

12 Responses to “Wordle”

  1. Cathy Hajo Says:

    Hmmm. It makes a neat display, but I’m not sure how to make use of it as it is. I looked at a number of the visualizations on the site and then tried to look at some lengthier texts. I entered whole chapters of my book (http://www.wordle.net/show/wrdl/1296996/Chapter_from_a_book_on_birth_control_clinics) versus the latest volume of the Sanger Papers (http://www.wordle.net/show/wrdl/1296982/Chapter_of_Margaret_Sanger_Papers), which is made up of primary source documents and annotations. I found that the monograph has far more repetition of key words than the primary sources do. What do I make of this? I’m not that sure. Linguistic analysis is not a very important part of my work, though it is interesting and the ease of using this tool might make me think about exploring things that would have been daunting before.

    I think that it might be more useful if you could omit some words from the display, either by parts of speech (remove “the” and other articles, or take out specific words by clicking on them to remove them (abbreviations, etc.)

    I can think of uses for this if the tool were more robust. It would be interesting to compare the word clouds for different decades of Sanger’s articles and speeches, to see graphically how her language changed, and how the terminology of birth control gradually changed to planned parenthood and family planning. To do that you would need to be able to gather texts in meaningful clumps, either by subject, or by date, or purpose. It would be neat to see what the visualization looked like when she spoke before medical or scientific groups and before women’s groups; I would guess that her content would be very different.

  2. Cathy Hajo Says:

    On the second link, remove the ending parenthesis to get it to work:

  3. Jan Kunnas Says:

    It put the whole introductory chapter of my Doctoral thesis through wordle just to see how it works. I gave a very good overall view of the contents, indeed a word count would have done the same. Unfortunately, I am not able to glue it here, it looks quite nice. Soon all the academic book covers will be made this way?! đŸ˜‰

  4. gemunro Says:

    I entered text from an article I am writing. The crossword effect is interesting. I was a bit surprised by the configuration and size of various words, but as I thought about it, I decided it was appropriate.
    I’m not sure anything significant or useful is added to my work. It might make an interesting design to put on my office bulletin board.

  5. tim vermande Says:

    It’s an interesting tool, since many of students are highly visually oriented. To some of them, it probably shouts “replace your mind and job with a computer.” To others, it could be the starting point for further ideas.

    To me (not a designer) it might be a way to at least throw words up on a projection screen when there’s no other form of illustration. A bit of control, such as colors or arrangement (especially as to which word is most prominent) would be helpful.

  6. Melissa Says:

    I think wordle works in a couple of ways. One, it let’s you visually represent your dissertation…it might make for interesting conversation at job talks? And let’s face it…it’s kinda fun! But I actually see it as a teaching tool for writing.

    For students, one of the hardest things to figure out when it comes to historical writing is HOW to shape all of the information they’ve gathered. If they have a rough draft, they can enter it, and see what sort of words stand out — or not! — which may help them tighten up what they are trying to express (for instance, if the word ‘heresy’ doesn’t stand out in a paper they are purportedly writing on heresy, that’s a problem).

    In addition, I’ve used this before to help students realize their overuse of the passive voice. Often, when I have them input a paper into wordle, the word ‘IS’ stands out like a sore thumb. They can’t see the passive voice when reading their drafts, but when it’s visually represented by wordle, they ‘get it!’

  7. wilssearch Says:

    I have never thought of this as a teaching tool or good for anything except fun and graphic design. I have seen Wordle used to design post cards as flyers for events in the Department at DePaul. Until reading the responses here, I never would have considered it a tool for history.

    But now that I have seen how others have used it here – I will be trying it out on a paper I have due next week.

  8. Jeff Tenuth Says:

    The wordle configurations are an interesting exercise in invoking (provoking?) a particular response or in making a point by the artist/author, but I don’t see them as being of any use in my academic or object based research. And if wordles are meant to reflect a position, philosophy, argument, etc., there is too great a chance of subjectivity, such as in art. It seems to me that they better represent opinion rather than research or fact. Perhaps they are an artform rather than an academic tool.

  9. Andrew H. Lee Says:

    I put in the text of a Guardian article about an unidentified African American volunteer from the Spanish Civil War:


    What strikes me is the prominence of certain words and the insignificance of others. (Obama and Franco get equal billing ion Wordle.) I could see use of this for getting people to think about what is important based on Wordle results, with the subjectivity highlighted. One could use to say what is apparently important in this article is the photographer and the war, and how the real subject of the article (the anonymous volunteer) is lost. But is he then the real subject? Or is it the search for his identity? I could see it as a point of entrance to a number of discussion on how to analyze texts, the nature of historical subjects, how words and our own knowledge influence how we read. Oh for similar piece of software that would do images.

    And I think it would make great covers to books, especially those Oxford VSI series. I am tired of the Rothko like color field covers they are current using.

  10. Jan Kunnas Says:

    I want to return to this discussion by adding a link to a page that uses a word cloud similar to the output from wordle as a way to navigate through the site. I have seen the same approach on other pages as well, but this is a good example.


    I would be nice to know how to make such links yourself…

  11. allison Says:

    I have to admit I was completely skeptical about the usefulness of this tool – for one’s own work, surely it would just tell you what you already know (I didn’t try my own dissertation, but I’m sure discourse would be in bold print in the center), but I threw in some text from the journals of the artist Vuillard, on which I’ve written. Again, the results were unsurprising, but if I had done this at the beginning of my research of these texts, it might have inspired me to take my analysis in a different direction. Or one could create and compare two different wordles from different periods of an artist’s life, or different literary works. In other words, it strikes me as essentially a tool for brainstorming.

    The suggestion of one commentator to use it as a teaching tool is brilliant – I also teach a lot of fine arts students – having them type in either artists’ statements or their own statements and then getting them to both elaborate the wordle (as an art work) and analyze it with a subsequent paper would make the basis for a great assignment.

  12. tdmackie Says:

    I missed this earlier. This technology has little use for my applications. It seems a useful for expression lessons but I was at a loss on how to use it in my field.

Enhancing Historical Research With Text-Mining and Analysis Tools