I’ve long been a fan of deep linking. If you can provide a lawyer or researcher a direct link to a resource within a larger database, you do two things. First, you bypass what is often very poor user interface design. Second, you eliminate the need to understand the terms of art that layer legal information finding. Deep links to a resource can be pretty easy. Deep links using search need some real care to do right.
There was a period where law library portal pages seemed very common. I created some in Ohio and re-created them here in Canada. The basic idea is:
- you know the major practice area focuses of your audience
- you tailor a set of resources just for them
- your consumers are happy because they don’t have to constantly navigate a wide system when their only goal is to go deep
Deep Links and Tools
Practice group-specific intranet pages are probably the norm in most law firms that have law librarians. Legal publishers have done practice-oriented collections (practice pages, practice centers, etc.) as well. It’s one of those things that seems to cycle in and out of fashion.
This sort of effort doesn’t scale, though. If the publisher’s aren’t helping, it can be difficult to create on your own. It’s not the initial effort, it’s the maintenance.
It’s one reason I was particularly intrigued by efforts like the AALL 2019 innovation tournament candidate Andre Davison. It’s like a research provider’s holy grail to get researchers to the information as fast as possible and with the least amount of friction.
Search is Tricky
A direct link to a database source or sub-source should, in theory, be pretty easy. A direct link that builds on search can be a lot harder. I’ll start with a simple obvious failure exemplified by that second rate web search engine, Microsoft Bing.
Microsoft Rewards is an online quiz and attention capture project. It’s a bit like the points programs the legal publishers create to capture law student attention. Another cyclical phenomena. In any event, Rewards creates small tiles that allow you to click and Bing will auto-magickally return your desired search result.
Some of these work pretty well. Type weather and [city name] and get a weather tile.
Some searches that have very few results may bring back slightly enhanced results. This is a page on my site that Bing shows as a left-to-right scrollable result, rather than just a search snippet.
But for each hit, there seem to be those that only barely land. And some are complete misses. Here’s an example. The Reward tile contains the custom search link that is supposed to take you to resources that Bing returns about the city of Lisbon.
The #YouHadOneJob issue exists. If you use a deep link or a custom search link, it has to work.
But let’s assume your search works and results are returned. The whole purpose is to get the researcher closer, with less friction, to the result than something they could have typed themselves. Once again, Bing doesn’t quite get a tile – intended to encourage vacationing in the country of Cuba by listing the best hotels – quite right.
The tool will drive the result and your ability to create custom deep links. We had a recent example where one of the courthouse law libraries in our union catalog wanted a more custom experience for their researchers. The default search is to retrieve results from all nearly 50 law libraries in Ontario. We use the Ex Libris Primo discovery search to incorporate other digital repositories as well as catalog information.
In this case, we could just mine the URL for a solution. It’s surprising what you can find in a search URL:
- base url: http://www.infolocate.ca/primo_library/libweb/action/search.do?
- [lots of other stuff]
- scope: %3A([library location])
- [more stuff]
If the goal is to just create a deep link that sends people to a custom search page, all I need is the SCOPE parameter. If I return to our search page, add this element to it, I can create a link that returns a search form with the law library pre-selected.
In the event, the courthouse law library wanted catalog plus digital resources so our technical team made a custom Primo view. This way, they get the custom search over their catalog materials but return digital repository materials that are managed by our library for the whole province.
Deep links using search are harder the further you are from controlling the result. As new content is indexed, you may find that a well-crafted tailored search returns less relevant results. Like catalog records to digital resources or web links on your site, it’s another resource that requires some regular maintenance and verification. Even when it’s your system, if the search technology isn’t helpful – for example, if the details are limited to the session – you may have to hunt around.
Another example. We have a digital repository of free CLE papers. If I search for content, the URL I get is: http://lx07.lsuc.on.ca/R/BISLGAXU7YLJHBB16EGSS9TLH3UAPAMRNU52A74TTPXQENX33Q-01731. That doesn’t work a second time – it’s strictly kept for that visitor.
But if I return to our discovery search that keeps this information in the URL, I can rerun the same search, narrow it to the CLE papers, and get the same results. That link can then be chopped down to a re-useable deep link.
I’ve used this approach to create a deep search link to a Web site. It means relying on Google or another web search engine, but you can point it and delimit the search with site:[site domain] to get around obfuscated search URLs.
The Bing results sort of make me glad, even if they’re such failures. Even folks with deep pockets and lots of resources don’t necessarily get this right. Deep links – whether direct to a resource or built on a search – are hard. It can be great to create them but then you need to consider how stable they and the underlying content or tools are.