I still build websites. I just no longer believe the website is the interesting part.
Semantic architecture · Schema.org-first Drupal · Independent tourism
For almost thirty years, the tools changed. The question did not: how can a small organization keep control of the meaning and value of what it knows?
I went to India for a sabbatical year expecting, if I am honest, perhaps a little enlightenment. I came back in 1996 without the enlightenment but with a damaged career plan, which turned out to be more useful. After five years in communication and advertising, and a Federal marketing qualification, the early web seemed interesting because a small business could publish directly without first needing the budget, media access or permission that conventional communication usually required.
I had continued building websites, worked around tourism and hospitality marketing, studied psychology and gained experience as an incoming tour operator. In 2010 I moved to Kyrgyzstan intending to start a tour operator built partly around the modular building-block idea I had seen in India. A revolution began a few days after I arrived. It was not a particularly clever moment to launch a tour company. So I stayed, learned Drupal and started building websites again: restaurants, pubs and nightclubs in Bishkek, before Instagram absorbed so much of that digital role.
Nomad's Land became the main turning point. What began as the owner's self-built website gradually became something much harder to call a website: destinations, journeys, multilingual content, travelers, partners and operational processes started becoming parts of the same system. Work with WTK / Helvetas and other tourism projects added the same lesson from different angles. The interface was becoming the least interesting part of the work. BlueMelon was where the craft accumulated: Drupal builds, tourism platforms, structured content, SEO, multilingual systems and client projects. Tamaga did not replace that history. It emerged from it when the recurring problem became clearer than the medium used to solve it.
jrockowitz's Schema.org Blueprints work changed something fundamental for me. Instead of asking first which pages a site needed, I began asking which things actually existed and how they related: a hotel, a room, a place, an offer, a person, an image. I became rather enthusiastic about Schema.org, which is perhaps not the most exciting sentence ever written about a professional turning point, but the work genuinely changed the way I understood a website. Once you start thinking that way, it becomes difficult to go back to a website made primarily of page types.
Laurent Bourrelly's work around semantic cocoon, audience and branding pushed me back toward the human side of the problem: what does the organization want to mean, and what does the traveler actually need to decide? Christian Méline's work around semantic language sharpened another part of it: words drift, especially when teams, markets and machines start reusing them. LLMs and AI-assisted discovery made the missing layer impossible for me to ignore. The website may be canonical, but the final answer about a hotel, route or destination can now be assembled somewhere else. Who established this fact? Under which condition? Who can change it? Who still has to say yes?
I thought the building blocks were pages. Then journeys. Then entities. Now I think the important blocks are meaning, decisions, entities, sources, conditions and responsibility.
Tamaga is where I am trying to make those blocks into infrastructure.
Independent knowledge should remain close to the people who can correct it, even when platforms and AI systems make the final answer somewhere else.
A useful entity graph is not separate from the traveler. It should help connect brand meaning, structured data and the decision a person is actually trying to make.
Schema.org can describe a public edge, but markup is not authority. The visible claim, its source and the operational fact behind it still need their own boundaries.
Inspect the authority modelI do not want semantic strategy to survive only as a workshop deck. I am testing how the Travel Platform Blueprint can become entities, registries, workflows, review states and software inside Drupal.
A strategy that only exists in a presentation is not much of a strategy.
A travel platform where destinations, journeys, multilingual publishing, traveler workflows, partners and operational knowledge meet.
What it taught me: It taught me that a travel website can gradually become part of the operating system.
Inspect Nomad's LandWork around destination ecosystems, media, image rights, tourism communication and local organizations.
What it taught me: Infrastructure can also mean giving other people usable material, not only publishing another site.
Inspect the WTK LibraryA controlled environment for House Record, representation integrity, request versus confirmation, Schema.org and Host Desk patterns.
What it taught me: A fictional environment can isolate failure modes safely, but it is not client proof.
Inspect the Hotel DemoA travel project for testing categories, routes, seasons, hosts, omissions, sources and AI-assisted review.
What it taught me: The same infrastructure problem extends beyond hotels into routes and destination knowledge.
Inspect OutseasonOther tourism work includes specialist outdoor and destination projects in Switzerland and Kyrgyzstan, including Nendaz Parapente and related BlueMelon work.
Schema.org Blueprints and Schema.org-first Drupal architecture changed how I think about things, relationships and pages.
Semantic cocoon, audience thinking and branding helped move the question from keyword and page thinking toward decision structures.
Semantic language and concept discipline sharpened my attention to lexical drift and the continuity of meaning.
A Research Briefing on representation integrity: why valid machine-readable data can still lose a condition a traveler or staff member needs to rely on.
Read the briefingPublic-source research into how important hotel facts change across the modern travel web.
Read the researchResearch and method notes as they become ready for public reading.
Visit the NewsroomTechnical provenance and public contribution history.
Inspect the Drupal profileFictional reference implementation for controlled hospitality experiments.
Inspect the demoDan Metille is the Swiss founder of Tamaga, based in Kyrgyzstan. He has worked with Drupal since 2009 and has spent more than fifteen years building tourism platforms, multilingual systems and operational tools. His current work focuses on travel knowledge infrastructure: connecting brand meaning, traveler decisions, structured entities, Schema.org, sources and operational authority for independent tourism. Earlier work through BlueMelon includes long-running DMC, destination and specialist tourism projects in Switzerland and Kyrgyzstan.
Citation: Dan Metille, founder of Tamaga, works on travel knowledge infrastructure, semantic architecture and Schema.org-first Drupal for independent tourism.