Introduction to Signposting

Stian Soiland-Reyes
The University of Manchester
University of Amsterdam

https://orcid.org/0000-0001-9842-9718

FDO Summit 2024, Berlin, DE
2024-03-19

Creative Commons Attributions 4.0 International

Overview

  1. How do PIDs currently resolve to repositories?
  2. Why are the Linked Data practices not enough?
  3. Signposting in HTTP
  4. FDO and Signposting
  5. More "advanced" Signposting

A typical PID resolution

GET https://doi.org/10.abcd/j.mystery.14

HTTP/1.1 302 Found
Location: http://www.example.com/repo/item/14

Browsers follow the redirection to the repository's landing page

GET http://www.example.com/repo/item/14

HTTP/1.1 200 OK 
Content-Type: text/html

<html>
  <head> ... </head>
  <body>

A typical landing page

HTML for human readability and human navigation

A preview of the resource

Metadata (e.g. title, abstract, author) listed as text

Link to download (e.g. a .pdf)
... and many unrelated links

What about the machines?

  • Site-specific JSON -- schema must be already known
  • "Secret" API -- do some URL rewriting
  • API documented as URL patterns
    http://api.example.com/items/{item}.json
    • -- but what is the {item}?
GET http://api.example.com/items/14.json

HTTP/1.1 200 OK
Content-Type: application/json

{ "title": "Data 14",  "download": "http://cdn.example.net/item14.zip" }

Some webby attempts

  • Content negotiation can redirect to alternative content representation -- if format is known
GET http://www.example.com/repo/item/14
Accept: application/json

HTTP/1.1 303 See Other
Location: http://api.example.com/items/14.json
  • ... but which JSON schema is used?
  • What about those download links?
  • What was the persistent identifier? Type?

FAIR Signposting

HTTP/1.1 200 OK 
Content-Type: text/html
Link: <http://cdn.example.net/item14.zip>; rel="item"
Link: <http://api.example.com/items/14.json>; rel="describedby"
Link: <https://doi.org/10.abcd/j.mystery.14>; rel="cite-as"
Link: <https://schema.org/TrainingMaterial>; rel="item"

<html>...

Additional Link headers, just like rel=stylesheet

No need to retrieve the HTML, use HTTP HEAD to only get headers.

https://doi.org/10.17487/RFC8288

rel= FDO concept
cite-as PID
type FDO type
describedby metadata
item bytestream
author (attribute)
license (attribute)

https://signposting.org/

Alternative Signposting methods

<html>
<head>

<link rel="stylesheet"  href="https://cdn.jsdelivr.net/npm/bootstrap..." />
<link rel="cite-as"     href="https://doi.org/10.abcd/j.mystery.14"  />
<link rel="item"        href="http://cdn.example.net/item14.zip"  />
<link rel="describedby" href="http://api.example.com/items/14.json" />
<link rel="type"        href="https://schema.org/TrainingMaterial"   />

</head>
<body>
...

HTML signposting with <link> can be used if you don't have control over the HTTP server, e.g. content delivery network (CDN).

Linkset JSON (rel="linkset")

{ "linkset": [
  { "anchor": "http://www.example.com/repo/item/14",
    "cite-as": [
      { "href": "https://doi.org/10.abcd/j.mystery.14" }
    ],
    "item": [
      { "href": "http://cdn.example.net/item14.zip" }
    ],
    "describedby": [
      { "href": "http://api.example.com/items/14.json" }
    ]
  },
  { "anchor": "..." }
] }

<https://doi.org/10.17487/RFC9264

Specify the content type:

Link: <http://cdn.example.net/item14.zip>; rel="item";
        type="application/zip"

Useful for alternative metadata formats.

For generic types, also specify a profile:

Link: <http://api.example.com/marc/14>; rel="describedby"; type="application/xml";
        profile="http://www.loc.gov/MARC21/slim"

Useful for JSONs with different schemas

Next steps:

Acknowledgements

Herbert Van de Sompel https://orcid.org/0000-0002-0715-6126

FAIR-IMPACT https://doi.org/10.3030/101057344
BY-COVID https://doi.org/10.3030/101046203