|

Who are directories useful for?

1 Leave a comment on paragraph 1 0 Some things to consider:

Quinn Dombrowski

3 Leave a comment on paragraph 3 0 On one hand, directories can serve as a sort of practical tour of applied DH for new practitioners. “Here’s a set of resources that you can pick up and use to start doing this kind of work!” People talk about how much they like directories, and how important they are for newcomers to the field.

4 Leave a comment on paragraph 4 0 And certainly, there’s a role for things that look like tool directories in pedagogical contexts, though I think of these more like syllabi: they’re curated collections of things for a particular audience, on a particular occasion. They might be updated the next time the course is run, but there’s no expectation that new tools will be added to syllabi for previous courses.

5 Leave a comment on paragraph 5 0 But here’s a different take on directories as a form of scholarly research infrastructure: they’re a workaround for bad scholarly communication practices, that ultimately undermine the goal of elevating the creation and documentation of tools (along with writing about tools and methods) to be treated as legitimate scholarly output.

6 Leave a comment on paragraph 6 0 Why should the discovery of tools be relegated to “directories” that are a challenge to maintain? Why treat “tool reviews” as so different from book reviews? Recently, Reviews in DH has made progress on creating a venue that so far has been primarily for project reviews, but their scope is inclusive of tools. Why not work on normalizing that — along with the appropriate credit and citation of tools in traditional scholarly outputs like articles and books?

7 Leave a comment on paragraph 7 0 What if the answer to “what tools should I be looking at” was more like other humanities reference questions: “Go check out the literature about projects similar to what you’re interested in. You’ll find the information about tools in the bibliography / footnotes / whatever the conventions* work out to be.” Wouldn’t this contextualized, embedded information be at least as valuable — I’d argue even more so — than a directory, and would come much closer to the goal of valuing that work similarly to written scholarship… all without all the ongoing expense and work of maintaining a directory. Perhaps not everyone involved with directories agree that this is a goal to work towards, but I think it’d have significant positive impacts, particularly for tool developers and other alt-ac folks who are often involved in tool reviews, directories, and the like.

8 Leave a comment on paragraph 8 0 * One “convention” that has worked well for the sciences this way is having an explicit “methods” section in papers. While it’s an attractive approach for some kinds of DH scholarship, I can imagine it causing more angst and divisions within the DH community than it’s worth in terms of which kinds of papers have or don’t have it, are papers with it treated as more “serious”, etc. For that reason, advocating for that particular convention is probably not the best tactic.

Geoffrey Rockwell

9 Leave a comment on paragraph 9 0 It is not clear that directories are useful to anyone. They are too dense and have too little explanation to serve people new to the digital humanities. They are hard to maintain which then makes them less useful to those in the field who might want to know about new tools in their area. And, they are rarely useful to their creators after the initial credit for creating them is exhausted. In fact, we might ask if Google searches aren’t good enough.

10 Leave a comment on paragraph 10 0 For directories to survive I suspect they really only need to be useful to their creators/maintainers. Otherwise there is no reason to maintain them over the long term, they atrophy and eventually are embarrassing enough to be removed. What reasons might there be for their creators and maintainers? Here is a tentative list of what is or could be useful:

  • 11 Leave a comment on paragraph 11 0
  • Communication – a properly configured directory can be used to try to communicate ideas, though not in narrative, but in the selection of what is included, in the interface, and in the organization of information.
  • Credit – a well maintained site can provide ongoing academic credit especially if it becomes the “go to” place for others to start looking for information.
  • Funding Magnet – a directory can be useful to other projects and therefore one can partner with new projects to support them and thereby get funding to continue maintaining the project.

12 Leave a comment on paragraph 12 0 Obviously a directory should be useful to new and experienced users. Here are some thoughts on how we could be useful.

  • 13 Leave a comment on paragraph 13 0
  • It’s not the database that matters, but the way into the information. We often focus on the technical features of online resources rather than their usability. For people new to digital humanities we should probably find ways to hide a lot of the entries and suggest good places to start. We should consider developing guides from the entries that explain why one would try one tool over another for a certain purpose.
  • Stable information that can be quoted for experienced users. For experienced users directories might serve as a sort of DOI or metalink to a tool. This would only work if people could trust a directory to stick around and to be updated. The idea would be that people could reference a tool by referencing the tool entry in a directory rather than an ephemeral web site.

LISA SPIRO

14 Leave a comment on paragraph 14 0 When I launched the Digital Research Tools (DiRT) wiki along with a few colleagues back in 2008, I think I had in mind several potential audiences:

  • 15 Leave a comment on paragraph 15 0
  • Digital scholarship specialists who needed to use and recommend tools
  • Novices who wanted to get a sense of different technical approaches to digital scholarship
  • Researchers who needed to find a tool to accomplish a particular task
  • Developers who wanted to examine software similar to their own
  • Students who needed to identify the right tool for a project

16 Leave a comment on paragraph 16 0 Frankly, I’m not sure that DiRT 1.0 (the wiki version, which is distinguished from the more robust Drupal version that Quinn developed) served these audiences in any significant way, given how quickly it became out of date and how terse its descriptions were. But I also built DiRT for me– as a way to learn, to keep track of all of the software I was coming across, and to feel like I was contributing to the DH community. I enjoyed my work on DiRT, but my professional responsibilities changed and left me with little time for it, so I was relieved when Quinn offered to shift it to another platform.

17 Leave a comment on paragraph 17 0 Back in the early 2000s, I was a somewhat naive techno-optimist, hopeful that technology could support innovative approaches to research and teaching. Now, as a more of a techno-skeptic, I wonder if a tool directory might be most useful to a computer historian inventorying past software.

18 Leave a comment on paragraph 18 0 As Geoffrey suggests, making tool directories more useful would require a way to highlight which tool works best for particular purposes. Further, the curators of the directory would need to figure out how to maintain it in the face of frequent changes in software. Like Quinn, I think that tool reviews would offer richer information than a directory and not face the same maintenance problems. Ideally these reviews would be easy to find and access, not locked behind a paywall.

Frank Fischer

19 Leave a comment on paragraph 19 0 Gathering overviews of scientific resources, tools, methods is not a new kind of activity. This was already done in »undigital« times. So creating overviews of digital tools in digital tool directories was and is an obvious idea. However, the life of digital tools is dynamic and diverse and can be hard to keep track of. An appropriate tool directory, if it wants to be »comprehensive«, requires an oftentimes unexpected maintenance effort, which is the main reason behind the failure of most attempts so far. I agree that smaller, more focused projects have a better chance of surviving. It is certainly no coincidence that the most long-lived project, TAPoR, had a focus on text analysis tools and only changed its character to be an all-encompassing directory when it merged with the discontinued DiRT directory.

20 Leave a comment on paragraph 20 0 Today, running a useful tool directory is even more difficult. This is because we now have a much better understanding of the environment of tools, of the different contexts in they are used. This includes datasets, workflows and tutorials (for which there are corresponding platforms, dozens of data repositories suitable for Humanities data, workflow platforms like the SSK and learning material sites like Programming Historian or DARIAH Campus). Tools only really make sense in their contexts, and making these contexts clear is the idea behind the SSH Open Marketplace, which DARIAH is building together with its partners within the SSHOC project.

21 Leave a comment on paragraph 21 0 In this respect I would also disagree with Geoffrey. It might be the case that, to survive, a tool directory must only be useful for its creators, but the idea behind the maintenance of a public tool directory is, arguably, that the larger community can and should benefit from it. For this noble goal, of course, someone has to provide (ongoing) funding and have the expertise for managing larger community projects. The only entities that I see that can deliver this are international research infrastructures like CLARIN or DARIAH. Their backing of a platform is probably the only way out of the »directory paradox« in the long run.

22 Leave a comment on paragraph 22 0 One last point I would like to make. Tool Directories, if they really are technically up to date, should not only be there for the researchers, but also for the machines (for example within the Linked Open Data cloud). If resources have permanent identifiers that lead to open machine-readable data, it’s easier to keep content up-to-date and enhance it. Tools often have Wikidata IDs, papers have DOIs and keywords, and we also have established taxonomies like TADIRAH available. So my argument is that we include machines as addressees of tool directories in our considerations.

Page 4

Source: https://~^(?[\\w-]+\\.)?(?[\\w-]+)\\.hcommons-staging.org$/who-are-directories-useful-for/