Future-Proof, AI-Friendly Product Literature

This article was originally published as part of a series of articles on our LinkedIn page:

If you’re interested in both battle-tested, reduce-to-the-max workflows (I am) and bleeding-edge technology (hello again!), you’ll likely stumble into two extremes that actually make for an excellent pair:

plain text; one of humanity’s oldest cultural tools and
large language models, commonly referred to as “AI”.

I will assume that you haven’t been living under a rock for the last three years, so I’ll spare you the usual “What Large Language Models are, and how they’ll turn your life and business upside down” blurb.

Whether you love ’em or hate ’em, LLMs are the new plumbing / electricity: You (probably) have it, and after accepting that it probably won’t kill you, you take it for granted. Both you and your clients will use ChatGPT, Claude, Geminini & Co. to research complex topics, create and improve documents, and engage in conversations about topics both banal and profound. Most people do this through their favorite LLM’s web interface, which often provides well-structured answers with headings, lists, and tables. Important information is highlighted in bold type or italics.

Now, when people want to use this LLM-generated content elsewhere, they will usually click / tap the “Copy” icon and then paste the content from the clipboard, only to see markers around words – *asterisks*, _underscores_ and [square brackets].

What often happens next is heartbreaking for a single-source publishing guy like me. Users will manually delete “those funny characters”, one by one, only to “add formatting” in their software of choice – a process that can easily take 15 minutes for a long document. This is like a food lover watching another guest rinse the divine Spaghetti Arrabiata that the chef put so much work and soul into under warm water, only to add their own, bottled sauce (“I like it better that way, maestro!”). It may not be a crime or a sacrilege, but it’s … inefficient, to put it mildly.

That content you pasted already has formatting, structure, and even metadata – so why strip it away?

The “funny characters” are Markdown, a popular lightweight markup language (LML) used in thousands of publishing tools and workflows everywhere – just Google it; I’ll wait.

The beauty of Markdown is that it’s not something that you have to buy or license, but something you learn. And once you have done that (it usually takes less than 15 minutes), no one can take it from you. As a publishing technique, it’s as powerful as letters are for writing.

This is a pendulum that swings both ways. Markdown-formatted content can be copied from and to the web interfaces of many text editors, content management system frontends, and large language models, resulting in a well-structured, formatted document with headings, lists, tables, hyperlinks, and even images (which can be easily referenced in Markdown).

The ability to work with formatted text across system borders (large language models, text editors, content management systems) is a huge timesaver. And it gets better.

By now, many large language models have web access, and new search engines such as Perplexity can process web content and return it to users in digestible form. Google Search has also added AI summaries to its results pages.

This means that if your brand publishes product literature, such as manuals and quick guides, in well-formed, semantic HTML, these engines can process and present the relevant information more effectively, increasing product understanding and, accordingly, user satisfaction.

This isn’t the stuff of Science Fiction. To see it in action, give ChatGPT the URL of a web manual and ask it to return a digest. After a few seconds, you should see a summary of the product’s features and basic operation instructions. You should also be able to extract information about a specific product feature or ways to troubleshoot a problem.

Now imagine hundreds or thousands of people doing this without having to contact support (or angrily shake their fists at the sky). If you don’t see the savings potential in this, not even the smartest LLM may be able to show you the light. 😉

A growing ecosystem

The ecosystem of applications supporting both Markdown and large language models is constantly expanding. Notion, the “application for everything”, is another popular productivity and project management tool with AI support, allowing users to create, structure, and process formatted content using an LLM, and it has supported Markdown from day one.

So there’s a lingua franca, and it’s spoken both by humans and machines. What’s not to like?

Are you using lightweight markup languages? Are they part of your daily work or your product literature publishing workflow? And if the answer is “No”, can I interest you in a single-source publishing tool that was literally built on Markdown?

Questions, questions, questions. And more on this and other topics next week.

Next week: Open Source Tools and Digital Sovereignty

↻ 2025-10-10