We’re doing more and more with content these days. There was a time when you wrote a page for the web and that was that. Hyperlinking created the webbyness of the net. But basically, a page was on the web, with a URL to locate it, and that was that.
Since search engines came along and started breaking pages up and detecting meaning, the experience of the web has started to be a feel more like 50% content and 50% aggregations. Aggregations allow us to make the best use of our content – they give us more bang for our buck. But we don’t get them for free. Aggregations can be automatic, but this is a long way from saying that they are un-designed experiences.
Using any piece of content in more than one place increases the complexity of the content (the fact that this increased complexity might require simplification is the sort of paradox that I like to point out within a sentence in parentheses). Even search engines give us an aggregation to be worried about, search engine results pages (SERPs). Millions of hours have already been invested in optimising content both for the ‘page’ you’re creating and the potential SERP that will lead an audience to the page.
Then there are content management systems. Now templates and teasers can make everyone’s life easier. Content can be shared across the system and appear in multiple places. It’s easy to think of the pages created by automatic aggregations as “free”. The category pages on this blog feel pretty free, now. And they are automatically generated, but they’re not un-designed. They had a cost, but it was in the past, and the benefit vastly outweighs it.
In the case of this blog I’ve chosen the ‘tags’ that I use to categorise posts and which populate the menu. I’ve thought carefully about how these collections of posts, drawn together by common themes or subject matter should be displayed. I’ve tweaked the design of the template and I even think about the ‘teaser’ text that appears on the aggregation. This is the key – aggregations are all very well, but they place design requirements on the content you’re creating. These requirements can’t be ignored if your product relies on automatic or dynamic aggregations. In fact an aggregation will often be the place when you’re content needs to be clearest, users are choosing between similar links, so clarity is key.
I’ve used the examples of SEO and CMS to talk about content optimisation for reuse, but it’s not all initialisms <pedantry> an acronym should create a word of its own </pedantry>. When you’re creating apps and content formats you need to exercise the same (actually probably more) care and attention. Dynamically generated pages can be difficult to think about – you switch your brain to a different mode. Recognising that this as a different skill might just help us all do it justice.
Automatic is a dirty word
Aggregations require design, whether they’re created by a search engine or an integral part of your content format. Reusing content requires effort. The benefit is that this effort is often worth it. Using content in more than one place gives you value for money, but you still have to spend money in the first place. You need to design and create the content with each representation in mind.
Designing aggregations and the content that will be aggregated are two skills. An aggregation is created by a machine – it transforms content. But you need the content in the first place and you need to design the machine. There’s a danger of becoming hypnotised by this complexity. Designing the content and the machine that will render it in different representations might feel like two many variables to juggle. But being aware that you’re in control of the design of all the variables is actually empowering.
Imagine you’re designing an engine to create timelines. You’re aware that you’ll be creating content for timelines for various people, but that often these people will share events. You’re building the timeline engine and creating content for the timelines. You can therefore ask yourself, “what can we re-use?” This is probably one of the most complex forms of aggregation. It’s not just a list, it assumes some coherence between the content nodes that might be being drawn from diverse sources.
In this example ‘time’ provides an overarching paradigm that will help to ensure coherence. It orders the aggregation using a familiar and therefore easily accessible model – time. But often, especially where the aggregation is a content format, coherence isn’t enough. It’s also desirable to have consistency within our dynamically generated elements. Editorial standards can ensure this – just in the way that in a CMS we can establish rules about the length of snippets or the size of images, content standards should be established for tone of voice, perspective (first person etc) and anything else that could affect the experience of the content.
As we come to leverage semantic technologies and linked data, computers will be creating more and more pages for us. But it’s unlikely that they will be designing them too. Dynamic publishing isn’t magic, it obeys rules and requires design. Content publishing can be automatic, new forms of content can be derived from relationships between content formats and templates. And reusing content might get you 3 pages for every bit of content you create. But don’t be surprised if takes you twice as long to design and create the content in the first place. The benefits are still substantial. The thing to remember if you’re going to rely on automation and aggregations is that you don’t get anything for free.