HerrnhutUNESCO World Heritage
DE Plan a visit

Inside the demonstrator

Why this site looks the way it does

This website tells Herrnhut's story and organises the knowledge behind it, so that guests and search systems find the same places and relationships. This guide explains in plain language what matters and which assumptions are evidenced.

See the evidenceData & downloadsTourism home page

The model's five layers

They work together and describe the same places from different perspectives.

Visible explanationunambiguous entitiesrelationships & geographystructured datasources & maintenance

Research status, 3 November 2026

Evidenced, plausible, unproven: 69 claims about AI search checked

We compared 69 common claims about AI search with operators' documentation and published measurements, then rebuilt this site accordingly. Almost all figures in this field come from companies that sell software or consulting on the subject, so each figure names its origin.

Evidenced

Google treats AI search as search

  • AI Overviews and AI Mode run on Google's normal search and quality systems.
  • Structured data is not required for them. If used, it must match the visible text.
  • Google ignores the llms.txt file.
  • For local businesses, Google names the Business Profile as a route into AI answers, without figures.
  • 17 per cent of holiday travellers in Germany have already used AI for excursions, sights or events (Bitkom, August 2026).
  • 71 to 80 per cent of businesses appeared in at most every second run (BrightLocal, vendor; USA, UK, Australia). A single test question proves nothing.
Plausible

Visible text and third-party mentions probably count

  • In a small test, AI systems read the visible text when fetching a page and ignored the code (searchVIU, vendor, one test product).
  • Brands mentioned often online appear more often in AI Overviews (Ahrefs, vendor; a correlation, not a cause).
  • In the lab, citing sources raised a text's share of AI answers: by 28 per cent in simulation, by about 11 per cent on Perplexity (Aggarwal et al., 2024). Nobody has tested this for World Heritage places.
  • Where little material exists online, a single page can shape answers (BBC experiment, February 2026). That this applies to Upper Lusatia is our assumption.
Unproven

Four common promises without a measurement

  • “JSON-LD brings more AI mentions.” Among 1,885 already heavily cited pages that added JSON-LD, citations in AI Overviews fell by 4.6 per cent; AI Mode and ChatGPT showed no measurable difference (Ahrefs, vendor). Small businesses were not measured.
  • “Name your neighbours and you will be found more easily.” The evidence runs the other way: Google counts how many websites link to a business.
  • “AI systems trust statements that confirm one another.” Consistent details lower the risk of wrong answers. Nobody has measured more mentions as a result.
  • “Number one in AI.” Google and Bing promise no inclusion, and rankings rarely repeat.
Consequences for this model site

Address, hours, arrival and access are in the visible text of every place page and readable without scripts. We justify the code by data quality, presentation in Google and data exchange with portals. That is why the events at Zinzendorf Manor are marked up: Google has no special presentation for places and towns, but it does for events. Links to neighbours appear where they help guests. Business Profiles, reviews and portal listings lie with the operators and cannot be demonstrated here.

Sources, retrieved November 2026

Putting it to the test

Seven questions for any website, answered here

The Upper Lusatia GEO assistant asks every website the same seven questions. An answer counts only if it stands word for word on the page.

Guests ask these questions without any AI. The list shows where this model site answers them. For God's Acre the operators state no access times, so the page says “to verify” and the code says nothing.

What is it?
The German component 1468bis-002 of a UNESCO World Heritage property. Read the answer
Where is it?
In Upper Lusatia, Görlitz district, with addresses per place. Location and arrival
When is it open?
Hours for the information point, manor and tourist information. No times are stated for God's Acre. Visit data
Who is it for?
World Heritage travellers, families and guided groups, distinguished per place. Audiences
How do you get there?
Löbau station, PlusBus 10, A4 Weißenberg, three mapped parking areas. Arrival
Who are the neighbours?
Moravian Stars and KOMENSKÝ, marked as separate facilities. Related places
Who is behind it?
Sources per statement, data maintained by polisMEDIA, no official operator. Sources

The visitor story layer

Storytelling and the knowledge graph belong together

A person remembers the fitting door. A machine additionally needs to know which places it connects and how strongly the claim is evidenced.

The page “Seven stories connecting Herrnhut with the world” therefore combines memorable people, spaces and objects with place, time, entity relationship, source and evidence note. Stories have stable anchors, matching German and English versions and a dedicated story dataset.

The door demonstrates the method: the familiar Bethlehem association remains understandable, while a fact-check explains that the official source more broadly names Moravian buildings in the USA or Gracehill. Narrative strength and factual reliability support one another.

See the story layerOpen structured story data

The idea in one sentence

A shared truth layer

GEO stands for Generative Engine Optimization. For Google, being easy to find in normal search lays the groundwork for AI answers. So the work starts with readable, evidenced and current details.

Nobody can promise a mention by ChatGPT, Gemini or Copilot. This site improves the conditions and gives guests the details they need anyway.

How the site was made

From a human prompt to a tested GEO model project

It began with a detailed brief from the project initiator to Codex: research the German World Heritage component at Herrnhut far beyond the obvious results, examine its digital representation and build an idealised GEO website from scratch.

Starting pointA human briefwith a purpose, stakeholder perspective and quality ambition
Research base115 sources & data servicesidentified, assessed and classified by source role
ImplementationVibe coding with Codexfrom the data model to site, planner and editor
Result18 pages · DE & ENplus JSON-LD, story data, entity register, GeoJSON and documentation
1

Prompt and target

The human set the problem, audience and ambition: intensive research, a German Herrnhut focus, bilingual content, ALT text, real visitor conditions and a demonstrator that stakeholders could explore.

2

Research and source critique

Codex reviewed the project knowledge base, researched official and tourism sources, and distinguished primary evidence, operator information, contextual sources and changeable visitor data.

3

Knowledge model before layout

Entities, identifiers, part-whole relationships, spatial sectors, operator roles, sources and freshness rules were modelled before designing the pages.

4

Vibe coding and testing

Natural-language instructions became HTML, CSS, JavaScript, PHP and data files. Browser, mobile, link, JSON and content checks then drove further revisions.

Collaboration, not one-click magic

The division of labour between human and AI

H

Human direction

The project initiator set purpose and limits, contributed the stakeholder perspective, made taste and priority decisions, tested deployment and supplied concrete feedback - including the comment that the criteria box was hard to understand.

AI

AI contribution

Codex provided research breadth, comparison and synthesis, developed the entity and data model, drafted the content, generated working code, documented decisions and ran reproducible technical checks.

↔

Shared iteration

The result did not emerge from one uncontrolled prompt. Drafts were inspected, questioned, tested on the server and improved. Feedback flowed back into content, design, data and project memory.

What “vibe coding” means here

A human describes in everyday language what an application should do and how it should feel; the AI translates that intent into working code. This project went beyond rapid prototyping: research, source critique, a data model, visual checks, validators and human approval gave the process its professional frame.

Why deliberately idealised?

The demonstrator shows in one coherent model how guests' frequent questions can be answered: with direct answers, entities, relationships, geography, multilingual content, structured data, sources, image context, planning conditions, freshness and governance. A real operator website need not copy the design or every file; it can adopt the components that match its role and connect them to a shared knowledge network.

The site is therefore a reference architecture and learning object - not an official World Heritage website and not proof of a particular AI ranking. Factual approval, real photography, operator maintenance and institutional responsibility remain human tasks.

What to notice as you click

Six visible traces of the GEO logic

The demonstrator deliberately makes its method visible. These details help people and machines identify, relate and plan.

1

Direct answers

Headings and answer boxes address real questions: What is part of the property? What is merely nearby? What must be checked before visiting?

2

Clear roles

The site distinguishes the town, the World Heritage component, the Moravian Church, individual places, operators and related visitor services.

3

Explicit relationships

“Component of”, “contained in”, “operated by” and “thematically related” communicate more than a vague association with Herrnhut.

4

Planning conditions

Indoor or outdoor, slopes, duration, weather, mobility and seasonal access make places usable in context-sensitive itineraries.

5

Sources and review date

Stable World Heritage facts point to primary sources. Opening and access information should be checked with the responsible operator.

6

Images with context

ALT text describes the motif. Current images are clearly marked as AI placeholders so they cannot be mistaken for documentary photographs.

Officially precise, plainly explained

Why are there strange-looking codes?

Codes remain where they make an entity unambiguous. Their plain-language meaning appears alongside them.

1468bis

The serial property

This is the official identifier used for the whole Moravian Church Settlements property. “bis” is part of the identifier, not a value judgement.

1468bis-002

The Herrnhut component

The suffix identifies the German component within the serial property - not the whole municipality and not all four settlements together.

(iii) & (iv)

Cultural testimony and settlement model

In simple terms, criterion (iii) recognises exceptional testimony to Moravian culture; criterion (iv) an outstanding type of planned Protestant settlement.

45.56 ha + 185.08 ha

Property and buffer zone

The first number is the property area and the second its buffer zone. A representative coordinate is not an official boundary.

3 sectors

One spatially distributed component

The historic centre, God's Acre/Hutberg and Berthelsdorf are connected parts of the Herrnhut component, not an arbitrary attraction list.

reviewed …

Knowledge needs maintenance

A review date records editorial checking. Timetables, openings and access still require current operator information.

The design principle:

The official code creates precision in the data space. Its plain-language translation keeps people from being stranded in internal terminology.

Entities and relationships

Every place stands in named relationships

For guests it matters whether a building belongs to the World Heritage property or merely stands next to it. So this site states for each place what it is and what it connects to.

serial World Heritage property · ID 1468bisMoravian Church Settlements
contains
German component · ID 1468bis-002Herrnhut
connects
Historic settlement centre
God's Acre & Hutberg
Berthelsdorf
not identical with · but related to
MunicipalityMoravian ChurchOperatorsMoravian StarsAccommodationFood
The relationship is information: “part of the property” is different from “located in Herrnhut”, “operated by” or “a useful onward destination”.

One body of knowledge, several representations

Five layers, one content

Core statements appear on several layers. Every detail in the code also appears in the text, and a check script compares the two before each publication.

01Visible textanswers, advice, descriptions
02Semantic HTMLheadings, lists, navigation
03JSON-LDtypes, IDs, relationships
04Entity register & GeoJSONshared reference data
05Primary sourcesevidence and correction

Inspect the machine layer

The demonstrator makes central files public. They describe the same places and relationships as the text and serve data quality, presentation in Google and exchange with portals. An advantage in AI citations is not evidenced.

From data to a travel answer

Why weather, time and mobility belong here

Generative travel advice is often expected to say not only what exists, but what makes sense under particular conditions.

A place therefore needs more than a name and coordinate. Indoor or outdoor setting, slope, duration, access, interests and operator responsibility support a reasoned route.

Try the itinerary planner

2 daysrainfew slopeshistory

A different route from two sunny hours

Indoor places receive more weight. Hutberg is not silently added despite the mobility constraint. Opening and access still need a live check.

Planning relationships with evidence status

Accessibility, parking and paths are knowledge-graph data

For many guests these details decide whether a visit is possible at all. They therefore appear individually and with their evidence status.

A

Accessibility as features

Step-free entrance, lift, toilet, door width, surface, gradient and sensory/cognitive support must be kept separate. “Accessible” is not a universal yes/no field.

P

Parking as an entity

A parking facility has its own coordinate, operator, capacity, fees, opening and designated spaces. Nearby does not mean owned or currently available.

→

Path as a directed edge

A connection has a source, destination, mode, distance, duration, surface, gradient, steps, source and review status. Return travel may differ.

01Operator statementfor a clearly named scope
02Map signalexistence, not a promise
03Route modeldistance/duration, not accessibility
04Field checkfeatures, date, responsibility
05Maintenancechange and live status
Herrnhut example

The exhibition fit-out is described as accessible, but that does not establish the complete building route. A car park is mapped near Zinzendorf Square, but that does not establish capacity or free spaces. The path to God's Acre is modelled, but that does not establish surface, segment gradient or wheelchair suitability.

Inspect evidence and data gapsOpen the machine dataset

Several organisations, one region

Jointly maintained details help guests first

One good site is a start. When municipalities, attractions, hosts, restaurants, transport operators and regional portals describe the same places alike, guests meet fewer contradictions.

1

Organisations maintain data

Official name, type, location, operator, access, booking, review date and reliable sources.

2

Entities are connected

Shared identifiers and named relationships show that several pages mean the same place.

3

Regional data becomes more useful

Portals, maps, entity registers and Open Data structures can reuse checked details without retyping them.

4

Wrong details become rarer

When details agree everywhere, there is less reason for wrong or contradictory answers, whether from a person at the counter or from an AI.

We link where it helps guests.

For local prominence, Google counts how many websites link to a business, so the advantage lies with the one being named. Naming your neighbours helps guests and gives the neighbours a genuine pointer. Nobody has measured that it makes you easier to find yourself, and Google explicitly calls link exchanges spam.

A mini guide for operators

Ten steps for your own website

The order matters: readable first, then answered visibly, machine-readable last. A large knowledge graph is not a precondition.

  1. Make it readable: no noindex, search crawlers allowed in robots.txt, content in the HTML without scripts.
  2. Name: official name and clear short description.
  3. Locate: full address, region in the text, verified coordinate.
  4. State conditions: opening with closing time, duration, individual accessibility features, booking and audiences.
  5. Describe arrival: station, motorway and parking, with a source.
  6. Relate: part of, operated by, near, booked through or thematically related.
  7. Evidence: primary sources, responsible contact and last review date. Change the date only when the content has changed.
  8. Structure: JSON-LD matching the visible text, stable URLs and identifiers. It serves data quality and presentation in Google.
  9. Compare: visible text, structured data and regional feeds must not disagree.
  10. Maintain beyond the website: Google Business Profile, Bing and Apple, answering reviews, genuine coverage in press and portals.
This guide follows the principles it describes.

It has a stable URL, German/English language relationship, semantic headings, visible statements, structured TechArticle markup, sources, a review date and internal links to the data. Reviewed 9 November 2026.