I wrote last week about using country detection to allow or deny access to information. And earlier this week, I posted about using referral URLs to get your researchers to publishers. This will be the last post about tools to manage a visitor’s experience. I’ve been playing around with a WordPress plugin that allows me to look at other browser attributes to try to customize a visitor’s experience, including language.

The plugin Redirection has been around for a decade and has a large install base. It’s got a number of uses that you might want to implement in a law library. Like other WordPress plugins, it allows you to get some technical benefits without a deep technical skillset.

Simple Page Redirection

I started using Redirection in order to fix broken links. For example, when I started blogging about 15 years ago, I was on a different platform. When I migrated to WordPress in 2012, my landing page went from:

http://ofaolain.com/FrontPage

to

https://ofaolain.com/about/

Like any link created by someone else, you don’t necessarily have the ability to get it updated. And you don’t need to. If you have URLs that are not persistent, it’s useful to redirect them for the user. Search engines will eventually crawl and update their index but if your page has been linked elsewhere, you can avoid breaking those links.

You can manage this at the server level. When I migrated a corporate web site, and all of the URLs changed, we ran a redirect list for a couple of years. The Redirection plugin shifts that functionality into the web app – WordPress in this case – so you don’t need any special privileges.

More Custom Redirects

One constant goal for web content is to try to get people to their information as quickly as possible. While the term personalization is often used, I think it’s wrong. We can’t really gather enough information from visitors without a login to personalize access.

Short of that, we’re talking about a more customized experience. This is nothing new to libraries. We put out book displays that are tailored to our walk-in researchers. We use topic-based email announcement lists to let lawyers know about the latest cases or events in their practice area. It’s not tailored to the individual but to a group. That custom focus balances resources with outcomes.

Check Referers

For reasons unknown to me, while the normal word is referrer, the spelling in the HTTP space is referer. As I explained in my last post, we can transmit a specific, secured URL to legal publishers. They can read that URL – the referer – and accept those visitors as authenticated users.

Similarly, you can intercept visitors from external sites and funnel them to the content that may be of the most interest to them. For example, if you use a persona-based navigation and differentiate public and professional researchers, you could drop people on the more-likely-to-be-useful content.

If the link is coming from a public library or legal aid site, you might expect that person to be looking at self-represented legal resources. A link from a bar association might be more likely to be a lawyer. “But…,” you say! Right. That’s the issue. You won’t get it right every time because you can’t know in advance.

One referer I’m playing around with involves Facebook. I think Facebook is a destructive influence and source of negative information. With Redirection, I can intercept visitors and advocate that they stop using Facebook.

Facebook is one of a number of media platforms – LinkedIn is another – that remove information about where their visitor came from. So, instead of responding to specific Facebook links, you can only identify the incoming visitor as having come from somewhere from the Facebook platform.

Check Language

Canadian web content can sometimes come in two languages, French and English. European web sites often have multiple languages. U.S. public law library web sites may have both English and Spanish legal resources. You can’t always know the visitor’s spoken language ability but you can look at their browser.

There is a standard list of languages hosted by the Library of Congress (although, ironically, you get a 404 if you try to get to it from the IETF; maybe LOC needs a redirect?). You can also look at your visitor log. If I go into my Matomo analytics, I can click on a visitor profile and view an XML file. It will show me the information the browser passes to my server, as you can see below:

Screenshot of an XML file showing browser information for a Russian visitor from a Matomo analytics report.

As you can imagine, I attempt to give visitors from Russia a customized experience on my blog. I’ve used the language tag as well as attempting to geo-fence these visitors. Matching visitors are redirected to a custom page.

The Redirection plugin has a simple form for configuring rules, as shown in this screenshot. Language is one of the default options so you really just need to know your language codes. You can include multiple languages in a single rule.

The question then becomes whether you make an attempt to guess. A Spanish speaker may be using a browser that has a language identifier of en-US (US English). And someone using a language identifier for a Spanish language group (es-CO or es-MX for Colombia or Mexico) might also be bilingual. You may end up overriding their or misinterpreting their language preference.

It occurs to me that a sweet spot might be two items of information. A link from a Spanish-language-oriented legal services site and a language identifier from a Spanish-language group would be a stronger indicator than any one browser element. I’m not sure there’s a simple way to manage that sort of rule-base though.

But a plugin like Redirection would support your testing this. For example, you could redirect based on the language identifier. Or you could wait until visitors reached content in a particular language and ask them to set a cookie to reflect their language preference. Redirection supports cookie validation as well and you could create a rule to intercept and redirect users based on that choice.

It’s out of scope for this post but there are content management systems that support multi-lingual content (one content ID, languages iterate). Those, plus using techniques like third level domains (es.ofaolain.com, en.ofaolain.com, fr.ofaolain.com), can make language-specific content easier to wrangle from the creation side.

Other Checks

You could catch spam or other negative visits, like bots, this way as well. But I think you are better off pushing these sorts of problem visitors out away from your web site. If you’re using WordPress, you should have Akismet for spam control. I use Cloudflare’s free service to firewall my site from a lot of that unwanted traffic. You can block specific types of visitors like bots or place challenges or blocks based on country.

There may be other identifiers that would allow you to customize access for your visitors, but language and source seem to be the low hanging fruit. You would also want to decide how the rest of that visitor’s experience goes. Your rule should not limit their ability to choose a different page, in case your custom guesswork is mistaken.

Otherwise they will continue to be forwarded the same starting point. And instead of a positive customized experience, you’ll just be giving them a negative one. That’s the trick of any attempt at customization, let alone personalization. You are, to a certain extent, trying to anticipate need. When it works, it’s like magic. But it should be flexible enough to take into account reality.