Screens, Research and Hypertext

Powered by 🌱Roam Garden

Assembly in Practice

Assembly on the web is harder than it should be.

Building a better vocabulary for content is a great starting point—after all, naming a problem is the first step to solving it. But separating out authoring from assembling poses both technical and organizational challenges.

Most of our existing CMSs are not particularly well-suited for the world in which we find ourselves.

Most organizations are not well-positioned for staffing the assembly side of the authoring-assembling divide.

Authoring inside a CMS is generally an awful experience. For some types of authoring, GatherContent (which I adore) fills that void. But it’s a tool that works best with highly-structured content—product descriptions or service offerings or the like. It lacks the flexibility needed for the kinds of longform, narrative products that think tanks and research organizations produce.

A CMS like Drupal is much better at assembly tasks (e.g., creating metadata or managing taxonomies). But more complex bits of assembly are difficult, as anyone who has ever worked with Views can attest. And you can forget using it to power a multichannel, modular content publishing system unless you’ve a whole team of good developers on hand.

In the think tank/research organization space, we (rightly) recognize the need for expertise in our authors. Those of us working in communications put a lot of thought into helping shift our authors into digital-first ways of working. But too often, we’re saddling that (legitimate) request with too much additional burden. Often, we expect those same authors to handle assembly (by, for example, asking authors to produce tags and categories) or we outsource the job to a web team that typically hires for technical skills, not for curation skills.

Assembling content—like curating a museum—requires a particular set of skills. Those skills aren’t the same as the ones required for writing content, nor are they the same as the ones required for writing good CSS or applying security updates.

Getting where we need to be requires tackling a bunch of interrelated questions:

What kinds of tools do we need if we’re to move authoring out of Word?

How can we make our CMS a friendlier place for assembling content, not just for the web but for all our digital outputs?

What kinds of staffing do we need if we’re to divide authoring from assembling?

How do we govern a process that divides authoring from assembling?

We’re going to be tackling some of these questions at Soapbox in the coming year. If that sounds like a conversation you’d like to be part of, let me know in the comments.

For more context

We need better terms for distinguishing stuff we author from stuff we assemble.

What to read next

Better tools for curation are just one part of a better web.

Other items of interest

What would it look like to write content meant for assembly?

What sorts of CMS tools do we need for better assembly?

Failure to distinguish between authoring and assembling isn't the only place our CMSs have failed us.