CCCT Court Web Site Guidelines – Recommendations 3 and 4: Content, Search, Navigation & Taxonomies – Keep It Simple
Last Saturday, I shared on Slaw the first two recommendations of the CCCT IntellAction Working Group on Court Web Site Guidelines. In this post, I’m sharing our draft recommendations 3 and 4.
(note: in the next few days, I will share our two remaining recommendations, 5 and 6)
As usual – your comments and suggestions are welcome! Please let us know if you think we are in the right direction…
Recommendation #3: Recommended Content
Although each court may end up with more or less information published on its web site, there are categories of information deserving of publication by all courts, because courts are in a better (if not unique) position to ensure the timeliness and accuracy of published information, and because the information relates to the fundamental mission of the court.
Such categories include:
- court decisions (integral)
- rules of the court
- notices to the Profession
- guidance to litigants (lawyers and self-represented litigants), including contact information and alternative dispute resolution information
In addition to the above information, the following categories of information should also be made available:
- docket information
- current list of judges (with links to biographical information, pictures, contact details)
- role of the court
- annual reports of the court, if published
- guidelines for serving as a juror, attending court, accessing court records, etc.
- frequently asked questions
Finally, the following information would be a welcome addition to court web sites:
- historical and general information about the court (creation, important historical moments, judges over time, pictures of the court buildings and courtrooms)
- court fees
- court forms
- educational materials
Recommendation #4: Search, Navigation & Taxonomies – Keep It Simple
When several categories of information are published on a web site, “search, navigation & categories (or taxonomies)” issues always arise. The question becomes: how can the right information be made available to each audience intuitively, taking into account the various backgrounds of court web site users (lawyers, self-represented litigants, researchers, etc.)?
The answer to that question has three facets. The first facet is navigation. When navigation is considered, “usability” is key. Usability is a disciplined approach to web site development. It can ensure that complex sites are easy to use. The CCCT recommends that the top navigation be structured along top-level navigation portals for at least the following audiences:
- members of the Public (general)
- self-represented Litigants
- practitioners: Lawyers, Paralegals, Stenographers, Translators, etc.
- researchers: Law Professors, Law Librarians, Law Students
Resources permitting, the following top-level portals are also recommended:
- law Publishers
Below the top-level portals, there should be second-level content categories. Some of these content categories can be common to more than one top-level navigation portal. For example, “court decisions” is a second-level content element that is likely to be found on all top-level navigation portals. Most commercial and open source Web Content Management Systems can handle sites with such a flexible content architecture.
A note of caution on annotated content. Court resources are limited. Courts, both historically and traditionally, are not in the business of annotating and/or organizing court decisions the way law publishers have been doing for decades. The CCCT recommends the publication of court decisions, subject to publication bans and privacy rules, but does not recommend that courts start annotating such decisions (for example augmenting decisions with headnotes or cross-references to other court decisions).
The second facet of the “navigation, search and categories” issue is searching. Searching on a web site is an experience that can be frustrating or simple, with all shades of effectiveness in-between, and with corresponding implementation costs. A simple Boolean search feature, which returns search results from the entire content database without sorting, categorization or pertinence ranking, can be implemented quite easily. However, this would probably result in more complaints than if no search were available on the site.
At the other extreme, a sophisticated natural language search engine with highly refined usage and lexicological parameters can provide unprecedented effectiveness, but at very high startup and ongoing implementation costs. How can the cost effectiveness balance be found between Boolean searching and sophisticated natural language searching engines?
The CCCT recommends finding this balance using a two-step approach. First, select a Web Content Management System (WCMS) most capable of adhering to the 11 principles laid out in this report. Next, assess the built-in WCMS search module against third party search modules that can be integrated with that WCMS. At a minimum, users should be able to execute multi-faceted search and refinements. In other words, they should be able to selectively search within categories of content.
Finally, the third facet is categorization of content, also called “taxonomies”. When content is published on a site that contains many categories of content, it is customary to annotate the content with metadata, or “terms” taken from a “controlled vocabulary”. This is another way of saying that web site administrators have set aside a number of categories (also called “controlled vocabularies”), each containing a list of terms that can be associated with content. Examples of controlled vocabularies: “Location”, “Topic”, “Jurisdiction”, “Judge”.
Taxonomies are driven and controlled by web site administrators. These administrators decide what terms and vocabularies are to be used, and which ones are made available to what categories of content. When authors publish new content on the site, they can pick terms and thus “categorize” their content. When content is categorized in this fashion, it can be made to appear on relevant pages or in specific search queries.
There is another kind of metadata built on “folksonomies”. Folksonomies are user-driven. As opposed to web site administrators deciding what are the controlled vocabularies and terms within each vocabulary, folksonomies simply allow “free tagging”. Content authors can decide on-the-fly when they publish content what terms (or “tags”) they want associated with their content.
There are pros and cons to using each approach (taxonomies and folksonomies). The CCCT recommends using both approaches, which complement each other. Taxonomies should be used in a limited fashion for the most obvious categories of content, leaving room for folksonomies to evolve over time with the nature of the published content. The ability to add free tagging to content should be reserved for court staff, to ensure a minimum consistency and coherence in developing the categories and terms associated with content.