The headline says it all: File Not Found. “Kids these days” have been cocooned so long in search that hierarchical tools like folders are not intuitive. Color me unsurprised. But it’s something I’m interested in, because law libraries are in an information access role. If browsing isn’t logical or within a researcher’s scope, should we be rethinking delivery that relies heavily on hierarchy and browsing?

Let me start out by saying that:

  • search is a natural starting point for more than Gen Z, which is the focus of that Verge piece. All the common legal research tools present search at the top of their interface and I wouldn’t be surprised if that’s where most lawyers of any age start their legal research. This includes the people that put a URL in the Google search box (or a site, like “Google“).
  • we have a lot of browsable content (blog posts, Libguides, digital repositories) that are valuable but which are subordinate to their own site search or unfindable by web search without some help. This content may be invisible to a researcher who relies on search.

While I know that going from a folder-focused organization to using things like metadata, it had not occurred to me that there was a hierarchy or information relationship literacy issue. A shift off folders can be a change management issue. But perhaps it’s not a change for everyone.

Search and browse are finding methods that we can use at the same time. There is no reason to eliminate organization just because there is a preference for search. A lot of organization can support search but we have to be deliberate in making that happen.

Directory structure isn’t just unintuitive to students — it’s so intuitive to professors that they have difficulty figuring out how to explain it.

File Not Found: a Generation That Grew Up With Google is Forcing Professors to Rethink Their Lesson Plans, Monica Chin, The Verge, 9/22/2021

Search can be between a rock and a hard place. Just because people use it doesn’t mean they are good at it and it doesn’t mean what they are looking for is readily accessible by search. For a long time, we have known that lawyers spend a lot of time searching. In 2007, more than half of lawyers in a LexisNexis survey reported spending between 1 and 4 hours a day using web and commercial search.

Responses to the question “During the average workday, how many hours would you estimate that you spend performing each of the following tasks? Conducting online research using Yahoo, Google, or other research tools such as LexisNexis or Factiva”, with the center column showing lawyers.

Invisible Resources

Fast forward to 2021 and a search provider, Elastic, finds that 60% of respondents to their survey report spend more time searching than answering email. It’s hard to know why they spend that time except that we probably all have experiences where the search and the content are separated. This separation may be in space (search isn’t indexing content on separate system) or in design (search is indexing content but isn’t weighted properly to return relevant results).

Here’s an example. Our organization uses SharePoint. Our intranet is on a separate CMS. HR documents are on both. A search on SharePoint brings up documents within its hierarchy, but omits any that are on the CMS unless your organization has indexed them.

If you are logged into Microsoft 365 in the Microsoft Edge browser and search with Bing, there is a Work tab that allows you to see SharePoint 365 results from your organization.

I was talking with someone recently about their organization, which has also adopted Microsoft 365. But they have retained network file servers despite the availability of SharePoint as a way to create a communal source of what are entirely Microsoft file formats (mostly Word and PowerPoints). This can be mean, ironically, when their less reliable on-premises file servers are unavailable, all content is missing.

They actually have a really easy lift and shift since (a) they already have a group organizational system and (b) it’s not a deep hierarchy of folders. Microsoft has guidance to help prepare for this sort of change. There is still a lot of discussion about folders v. metadata in SharePoint. Since using folders creates URLs that include folder names, and you can only have a 400 character URL, you can have issues if you use descriptive file names and multi-level folders.

On SharePoint 2013, on premises, the limitation was 255. I saw someone create a deep folder structure that left only about 10 characters for the files in the lowest folder. I’m firmly in the use metadata on SharePoint camp.

All of this means a law library needs to know two things. One is what people are searching for and the other is how search engines view your web content. The first is harder but you can at least work on the second aspect.

For example, you can use site search limiters to use a common keyword and focus it just on your site (law Google | Bing). You will also be looking at your web site analytics to see where people are currently arriving from. Can someone find an item in your catalog (or digital repository or whatever) with a web search? Can you, knowing it’s there?

It will require looking at your robots.txt file to see if you are inhibiting search engines from reaching your content, including any public facing library catalog or discovery layer. There can be a mindset that if you build it, they’ll come. But if it’s not accessible via a search, it can be as good as not existing. And this inaccessibility isn’t a question of marketing.

Case in point: this blog. I occasionally tweak my theme, the presentation layer of the blog. And sometimes Google changes what it values, so that your presentation in their search results can drop. If your PageSpeed is low, especially on mobile, you may need to investigate how to raise it (caching, retheming, etc.)

If, like my blog, most of your law library traffic is from search, these things can matter. Practically no-one who comes to my site browses or comes via a link. How about you?

January – September views and visits for this site, from JetPack analytics. June is when I made a theme change; August is when I changed back.

It may mean that, blog-like, a law library can’t set and forget content (like Libguides) which will be considered out of date based on when they were published, even if the content is current. And if you make structural changes or see a drop in your analytics, someone in your law library will need to try to understand why.

A Tangent Into SharePoint

I’ve been thinking a lot about the role of SharePoint 365 in our law library’s delivery. We have internal reference librarian content on SharePoint. We have some content that we incorporate into document delivery, like CD-ROM forms that would otherwise require someone to have the disk and the software installed.

I could definitely see a point where we could expand SharePoint as a delivery point for things like ebooks or other documents. It is already possible to set a time limit on a shared link to a resource and you can use security within the SharePoint library to inhibit downloading.

This is a no-brainer in a law firm. It’s a jump for public-facing law libraries, I think. SharePoint obviously supports external-facing sites. We also may have access to user lists (subscribers, members, licensees, circulation users) to help facilitate the access.

We get discovery search layers to fix some of our findability issues. We get specialized tools (catalogs, digital repositories) because they’re better at their functions than a generic system. But those systems may not findable, even if they’re really rich and well-organized.

Which leads me to this: would we be better off dumping our digital repository software and shifting our collections into a metadata-oriented SharePoint site? In other words, rather than having multiple sites we’re trying to drive researchers too, collapse it back into one that can be web-facing or limited to researchers? A SharePoint site may make more sense to law libraries – which tend to be at the smaller end of catalog implementations, of discovery implementations, etc. – than trying to maintain oversized organizational tools.

Back to the Topic at Hand

The whole idea of researchers struggling with folder or hierarchical organization is fascinating to me. It’s not foreign, though. I use a music player called MusicBee to manage my digital music. And I have had a lot of conversations with one of our kids about the folders I’ve used (which tend to be artist-based, as is common when you rip music through Windows Media or download from iTunes).

There are obviously shortcomings to an artist-based storage hierarchy if you come at the music by composer or conductor or genre. But that assumes that you are only finding through navigation. If, in fact, you’re mostly finding through search, then no hierarchy is even needed. So long as the objects have metadata and a tool to present similarly marked objects, you don’t need folders. See also: lawyers that keep all email in an inbox.

So if our researchers are struggling with hierarchy, we will need to find ways to bridge that gap. I don’t think that means “teach them about folders.” It does mean thinking about how to make findable by search every piece of content or resource we make available to researchers.