The Walt Whitman Archive – Warwick 2
2What are its weaknesses? What do you wish it would let you do? What changes would you suggest?
Reader in Digital Humanities, Department of Information Studies; Director, Centre for Digital Humanities – University College London
¶ 1 Leave a comment on paragraph 1 0 It is genuinely difficult to identify serious weaknesses of the Whitman archive. I might suggest a few relatively minor changes to site design. Some of the content is nested quite deeply within the site and requires several clicks to access it. For example in the Disciples section the user is three clicks away from accessing individual poems. Web log analysis suggests that the more clicks a users is required to make, the less likely they are to use that part of a site.[ref]P. Huntington, D. Nicholas, P. Williams and B. Gunter “Characterising the Health Information Consumer: an Examination of the Health Information Sources used by Digital Television Users,” Libri 52:1 (2002): 16-27[/ref] Thus I would suggest that if at all possible some of the intermediate levels of nesting are removed to allow better access to material.
¶ 2 Leave a comment on paragraph 2 0 I also found it difficult to access the newspaper content, since the page images took a very long time to load, on an average home broadband connection. This will not cause problems for academics using the site on fast campus networks. However it might limit its usefulness and usability of these newspapers for members of the public to high school students who might be accessing them on a slower connection.
¶ 3 Leave a comment on paragraph 3 0 I also had some problems using the search function, or rather understanding the results returned. I typed the words ‘metaphysics’ and ‘books’ into the search box as someone searching for the exact origin of a quotation might do. I would have hoped that the first hit in any general search of the archive might have referred me to published works, but instead it took me to instances of the words in criticism, but it was not clear why this happened, and would, I think be counter intuitive for most users. I then clicked on the top hit of published works, and when I clicked on it, this did indeed take me to Leaves of Grass, but only to the title page. Again this would not be intuitive for users accustomed for example to the functionality of commercial resources such as EEBO who would expect to be taken to the exact hit without a need to use the search function in Firefox to locate it.
¶ 4 Leave a comment on paragraph 4 0 These are technical problems which should be relatively easily corrected. However, I would also like to point out something which while not exactly being a weakness may more correctly be thought of as an opportunity that has not yet been exploited: that is the possibility of involving users more extensively in the Archive’s activities. At present, apart from leaving comments, it is difficult to see how academic users or members of the public interested in Whitman might be able to get involved in the project.
¶ 5 Leave a comment on paragraph 5 0 The current model for the Archive seems very much that of conventional publishing whereby the team produces content for its users, in a one-to-many fashion. It may be that this is a deliberate decision on the part of the editors: involving users is not appropriate for every digital project. However, if further content is to be created in future, the Whitman archive may consider taking part in some kind of crowd sourcing project. This is still a somewhat unusual method, but is being used by projects such as Transcribe Bentham at UCL to access crowd sourced transcriptions of page images of manuscripts. The use of crowd sourcing and social networking by memory institutions such as museums, has demonstrated that those outside academia are interested in participating in projects about their heritage and history and it may be that there is a potential new way for the Archive to engage with its user community, and to add further content while doing so.
You raise a good point about how the design of the site speaks to what I perceive to be a philosophical dilemma over what to do with the Disciples section of the site. The Archive directors wisely note that Whitman’s relationship with his disciples (as well as his detractors, influences, etc.) provides an important context for understanding his poetry. Nevertheless, the Disciples section is underdeveloped and, as you note, disconnected from the poetry itself. The trouble with site design here speaks to an uncertainty over what, exactly, to do with Whitman’s disciples: do they really belong on the site? and if they do, who else belongs on the site as well? and if we’re including other folks along with Whitman, does that dilute the site’s focus on Whitman? and would that necessarily be a bad thing to show that Whitman’s life and career was inextricably connected to a network of other people?
I couldn’t agree more. I make some similar comments in one of my short essays as well.
Crowd sourcing is where I get a bit more uncomfortable, though. Maybe it’s just the old-fogey in me, but my inclination is for the Archive to find a way to involve users more without necessarily going to a crowd sourcing model. It seems like there should be a compromise position between crowd sourcing on the one hand, and merely having users email the site editors with suggestions on the other.
Well yes but crowd sourcing does not have to mean low quality content. Our experience with Bentham suggests that users produce very high quality transcription. It also goes through an editorial process before posting, so it’s not unmoderated. Perhaps I should have stressed that there needs to be a balance between user input and expert comment or editing to ensure that the content is of good quality.
In another response to Ed Whitley, I explained that the Disciples section, as now developed, can be understood from a traditional author-centered viewpoint as featuring what we might call the collaborative Whitman, an aspect of his work all but completely ignored in the New York University Press multi-volume print edition of his works. If we later extend the Disciples section to other people who had a close relationship with Whitman but didn’t co-author with him in any sense (Anne Gilchrist to take one example), then we will face questions regarding who to include and why. If the section expands a great deal, does the religious or semi-religious implication of the term still work as well?
An equally important issue has been raised about the internal integration of the site. More can be done as the site develops more fully and we have time to implement what we have envisaged. The Traubel volumes, for example, could be linked to various Whitman texts mentioned in them. The correspondence often mentions journalistic pieces, and we can link to them once we have edited them. The list goes on…..
Claire Warwick suggests that the site be flattened to some degree so that users can get to the content they want with less clicking. We understand this point. We have improved this aspect of the site somewhat over the years. One potential cost of flattening might be to lessen the clarity of the site’s organization, an aspect of the Whitman Archive that everyone seems to like.
Although there is no evidence of it yet on the site, we have been discussing user-generated content, with an appropriate level of editorial vetting, for a few years now. We have been slow to move down this avenue not because of any philosophical antagonism or because of technical hurdles but merely because we have been busy meeting grant deadlines.
It seems likely that our first experiment with user-generated content may involve a database devoted to Whitman’s reading. There are book-length studies of Mark Twain and Emily Dickinson’s reading. What we imagine is the creation of a simple web form with boxes users could fill in. They would supply information about what Whitman read, when he read it (if known), the evidence establishing that he read it, and so on. I imagine that we would set this up so that information wouldn’t immediately go “live” but would be first checked and then ordinarily approved. Similar databases could potentially be developed to advance knowledge of poems speaking back to Whitman, television programs invoking him, advertisements using his image or words, and so on. If we pursued the Bentham model, we could perhaps accomplish tasks we couldn’t imagine completing with our current staff. Still, setting up the framework to do crowdsourcing properly itself takes time.
I’d just like to echo Ken that “setting up the framework to do crowdsourcing properly itself takes time.” At my institution we have been discussing the use of crowdsourcing to create transcripts of oral histories conducted in the 50s and 60s, but everyone is already overworked and unable to set up the templates, figure out to whom we should issue a call, and then review and edit the products. Digital sites have the tendency, I think, to make it too easy to envision how much more/else could be done, without accounting for how many more resources might be required to do it. Without additional resources, crowdsourcing for the Whitman Archive might not be possible without ending work on some other facet of the project.
That’s very true. Also once you start working with a user community you have to think about how to resource this interaction. Users need support and like to interact with resource creators, but that comes at a price in terms of person time. It really depends on the philosophy of the project and its priorities. We could have transcribed more material by Bentham if we hadn’t crowd sourced, but we thought it was important to engage the community outside academia with our work. Public engagement is key to all our work at UCL, thanks partly to Bentham’s own philosophy, and I am a passionate believer in it, whether digital or physical. But that doesn’t have to be the case everywhere of course.