Principles / The longer argument
How we think.
Digital work crosses language, structure, publishing and engineering. These are the principles we use to make those connections work.
Everything changes.
The underlying system doesn't.
Technology keeps changing. The information problems underneath it don’t.
Search engines, publishing platforms, AI, frameworks, automation and interfaces are successive mechanisms. The fundamental problem remains information.
Most digital problems aren't problems inside disciplines. They're problems between them. A search issue is often a language issue. A publishing issue is often a structure issue. An acquisition issue is often a communication issue.
We understand the whole machine. Not as a metaphor, but as a working reality. Journalism, design, development, search, publishing, automation, AI — these are not separate kingdoms. They are contiguous territories.
Language is infrastructure.
Words and taxonomy are the foundation the whole system runs on.
Content isn't filler between design and technology. Language is something the system operates on.
Writing, journalism, editorial direction, content strategy, linguistic frameworks, taxonomy, terminology. These are not afterthoughts. The words are the interface before the interface. The taxonomy is the architecture before the architecture.
When you understand language as infrastructure, you stop asking "what should the page say?" and start asking "what does the system need to be able to express?" That distinction changes everything.
Structure makes information executable.
Well-structured information can be used by any machine, now or later.
HTML, CSS, information architecture, semantics, schema, entities, data structures, APIs, Node, front-end systems, scalable web architecture.
This is an understanding that how information is structured determines what can be done with it. A well-structured document is machine-readable, human-navigable, search-discoverable, AI-comprehensible and future-adaptable all at once.
The structure is the promise you make to every system that will ever touch your information.
The interface is where the system meets a human.
Interfaces translate between machine logic and human understanding.
Design, UI/UX, journalism, visual hierarchy, editorial judgment, human interpretation.
Communication is not decoration. It is the translation layer between machine precision and human understanding. A technically perfect system that humans cannot navigate is a failed system. A beautiful interface that cannot scale is a liability.
The boundary is where most projects break down. Someone knows what to say but not how to show it. Someone knows how to design but not how to build. Someone knows how to build but not what the system actually needs to express. The difficult part is the handoff.
Search is what happens when information has been structured well enough to be understood.
Being findable is a byproduct of being well-structured, not a bolt-on.
Search is part of the cake. It isn't the bill.
Search is not the hero. It is the consequence. When language is precise, structure is semantic, and publishing is systematic, discovery becomes a property of the publishing system.
You do not optimise for search engines. You build information systems that search engines can interpret more consistently.
Eventually, a website stops being a website.
At scale, a website is a system of rules, not a pile of pages.
Once information starts moving at scale, the page stops being the interesting part.
What matters is the system underneath: taxonomy, hierarchy, templates, entities, editorial rules, automation, syndication, relationships between information, and the logic that determines what gets published, where it appears and how it can be found.
The technology has changed repeatedly. The problem hasn't. Whether in media, institutional publishing, real estate, education or large digital environments, the same underlying challenge appears: how to make information findable, consistent and sustainable at scale.
An acquisition system is bigger than a landing page.
Leads are a pipeline. Most organisations leak after the form.
Information moves through a pipeline.
Acquisition is not a page. It is a flow. From form submission through delivery and syndication, every handoff is an opportunity for signal loss or signal gain.
The landing page is just the door. The system is the building. What happens after someone expresses interest is where most organisations leak information, lose context and waste the effort that brought the person there in the first place.
The difficult part isn't any individual discipline. It's what happens between them.
Every discipline affects every other. The hard problems live between them.
Taxonomy, terminology, editorial logic
HTML, schema, entities, APIs
What the system expresses
AI is another machine.
AI amplifies whatever system it touches — good or bad.
The useful question is whether the underlying information system gives the machine something worthwhile to operate on.
AI is not a revolution that replaces information, language, structure or systems. It is a mechanism that amplifies them — or exposes their absence.
An AI system operating on poor structure, vague language and broken publishing pipelines will produce poor results faster. An AI system operating on precise structure, clear language and robust pipelines will produce outcomes that scale. The machine is only as good as the system it touches.