External IIIF Images

Many museums, libraries, and archives make their collections available through a technology called IIIF (International Image Interoperability Framework). This means you can build Telar stories around high-resolution images from institutions worldwide — without downloading or hosting the images yourself.

Finding IIIF images

Look for IIIF resources at:

IIIF Guide to Finding Resources
Major museums (British Museum, Getty, Smithsonian, Rijksmuseum)
Digital libraries (Internet Archive, Europeana, Gallica)
University collections and special archives

When an institution supports IIIF, you’ll typically find a manifest URL — a link that describes the image and its metadata. It usually ends in info.json or manifest.json.

Adding an external image

In your objects spreadsheet (Google Sheet or objects.csv):

Create a row with a unique object_id (e.g., museum-textile-001)

Add the IIIF manifest URL in the source_url column:

https://example.org/iiif/image/abc123/info.json

That’s it. Telar will fetch the image directly from the institution’s server when viewers visit your site.

Manifest URL formats

IIIF URLs typically look like:

Image API: https://example.org/iiif/2/abc123/info.json
Presentation API: https://example.org/iiif/2/abc123/manifest.json

Telar supports both formats, and both IIIF versions (2.0 and 3.0).

Mixing self-hosted and external images

You can use both your own images and external IIIF images in the same project. Leave source_url blank for objects that use self-hosted images:

object_id,title,...,source_url
my-textile,My Textile,,,
museum-map,Museum Map,...,https://example.org/iiif/manifest.json
my-ceramic,My Ceramic,,,

Automatic metadata extraction

When you provide a source_url, Telar can automatically fill in metadata from the IIIF manifest — title, description, creator, period, location, and credit. This saves you from typing information that the institution has already recorded.

How to use it

Add the IIIF manifest URL to your spreadsheet and leave the metadata fields blank:

object_id,title,description,source_url,creator,period,location,credit
map-001,,,https://example.org/iiif/manifest.json,,,,

When your site builds, Telar will:

Fetch the IIIF manifest
Extract available metadata
Fill in any fields you left blank
Keep any values you entered yourself

Your data always takes priority

You have full control over what gets extracted:

Leave fields blank and Telar fills them from the manifest
Fill in some fields and Telar only fills the rest
Fill in everything and the manifest metadata is ignored

Example — partial override:

object_id,title,description,source_url,creator,period,location,credit
map-001,My Custom Title,,https://example.org/manifest.json,,,,

Telar will use “My Custom Title” (your value) and extract the description, creator, period, location, and credit from the manifest.

Language detection

Metadata extraction respects your site’s language setting (telar_language in _config.yml):

English sites (en) — Prioritizes English metadata, falls back to other languages
Spanish sites (es) — Prioritizes Spanish metadata, then English, then others

Smart credit detection

For the credit field, Telar filters out legal boilerplate (Creative Commons URLs, generic rights statements) and looks for meaningful attribution lines — the institution name, the rights holder, or the credit line.

What Telar looks for

Different institutions label their metadata differently. Telar searches for common variations:

Field	Looks for
title	Title, Label, Name
description	Description, Summary, Note
creator	Creator, Artist, Maker, Author
period	Date, Period, Creation Date, Date Created
location	Repository, Holding Institution, Current Location
credit	Attribution, Rights Holder, Credit Line, Provider

When to override

You might want to enter your own values when:

The manifest is in a language your audience won’t understand
The institution uses abbreviations or codes
The description is too technical for your audience
You want a shorter or simpler version of a field

Build-time processing

Metadata extraction happens automatically during the site build:

GitHub Pages: Runs during deployment — no action needed
Local development: Run python3 scripts/csv_to_json.py when you add or update manifest URLs

Validation

During the build, Telar checks each manifest:

Valid manifest — Metadata extracted successfully
Manifest unavailable — The URL didn’t respond; will retry on the next build
No metadata found — The manifest is valid but doesn’t contain extractable fields

Check your build logs for extraction status.

Limitations

Only works with external IIIF manifests (self-hosted images don’t have manifests to extract from)
Manifests must be publicly accessible (no login required)
HTML tags are removed from extracted text
Some manifests may not include metadata fields

Troubleshooting

Image not loading

Verify the manifest URL is correct — try opening it directly in your browser
Make sure the resource is publicly accessible (no authentication required)
Check that you pasted the full URL including info.json or manifest.json

Metadata not appearing

Check your build logs for extraction warnings
The manifest may not include the fields you expect
Try entering the values manually in your spreadsheet

Slow loading

External images load from the institution’s server — speed depends on their infrastructure
Some institutions may throttle requests from unfamiliar sites