metaformats

From IndieWeb


metaformats started as an April Fools joke concept to describe how to both publish using microformats class names and openly parse meta tags as a fallback for what should be in-the-body visible data, including backcompat with OGP, Twitter Cards, and meta author, description, and anything else real sites (like search engines) appear to consume.

Why

By using metaformats as a methodology and implementation strategy, developers can minimize their implementation and dependency on proprietary or site-specific meta tag markup, while providing a forward compatible path for publishers to use open microformats2 instead.

Why not

Publishers should avoid invisible metadata (e.g. metacrap problems) and thus avoid publishing data in meta tags in general and instead markup their visible content using microformats2.

How to

How to publish

Don’t. Use IndieWeb link preview guidance instead, and note the explanations therein why you don’t need to publish any.

How to implement

Follow the consuming code implementation requirements in the metaformats living specification as they are developed and iterated.

consuming code use-cases

Variants in markup being very plumbing-specific, the consuming code use-cases for metaformats are likely to also be plumbing-specific, that is, addressing existing markup publishing behaviors, more than any new user-level functionality.

Link Previews

As a link preview display fallback. Metaformats can be used to show a summary of a link on pages missing mf2. This is similar to how social media sites use the tags.

Profiles

Bridgy Fed uses meta tag values for websites that don't have a mf2 h-card.

Repurpose existing meta

Many sites have meta OGP markup, because the result of doing so is immediately visible when sharing links to such pages in Facebook. It's immediately testable, and provides immediate visible benefit. (in contrast, Twitter Cards require your site to be approved first, and it is unknown when any Google suggested metadata will cause any actual specific impact in their search results)

Repurposing existing meta OGP tags is also a positive reframing of avoid more duplication of information already present in meta OGP tags, in particular situations where:

IndieWeb Examples

Brainstorming

There are a many ways to incrementally approach defining metaformats. Here are some that help address the use-case(s).

Which approach first

I'm in favor of first attempting to "define mf2 parsing for meta", as that provides more direct author / publisher control, and involves less special-casing in mf2 parsing code. It also allows migration to other non-OGP meta if/when implementations switch away from OGP to some other meta by some other vendor. - Tantek Γ‡elik 15:24, 13 September 2017 (PDT)

  • Hmm, like the idea of not adding anything to mf2 parsing code. My concern comes when we assume that mf2 will override OGP tags. In other words, I would not want to assume '.h-*:first .p-name' to override a OGP title, as that might not be the intention. OGP titles are, when set, known to be page-scoped, and the first h-* is not assumed to be that way. Perhaps if there is only on h-* ? Ben Roberts 08:49, 14 September 2017 (PDT)

Define mf2 parsing for meta

If the microformats2 parsing were updated to define how to parse microformats class names on meta elements, it may be possible to just add 'class' attributes to existing meta OGP tags as a way of making their information visible to mf2 parsers.

Advantages:

  • Helps avoid DRY violation. By re-using content already being published in a meta OGP tag, even if invisible, the publisher avoids any need to duplicate that content into visible elements for mf2 consumers.
  • Publisher control over what meta OGP they want consumed in what way as which microformats and properties
  • Allows publishers to re-use non-OGP meta for certain properties if they wish, e.g. from Twitter Cards, Dublin Core, or other random meta that they are already publishing / keeping up to date for whatever reason.
  • Likely very easy to upgrade existing implementations.

Disadvantages:

  • Publisher is burdened with understanding the mf2 class equivalents of existing OGP properties, thus using them
    • Possible mitigation: validators could check/compare the OGP property and the mf2 property (in the class attribute) and provide warnings if/when there are mismatches.
  • Only valid in HTML5, older strict validators could have an issue with this.

Map meta OGP to mf2

Another approach is to treat meta OGP as a form of backcompat parsing for various mf2 vocabularies, to fallback to when there is no mf2 or mf1 version of that vocabulary on the page.

Advantages:

  • Helps avoid DRY violation same as noted above.
  • Less work (burden) for publishers who "just use OGP".

Disadvantages:

  • Only applies to meta OGP (absent additional work, other meta like Twitter Cards and Dublin Core would need separately defined mappings, as well as a prioritization to apply, which meta to use first for which property)
  • Inadvertently encourages meta OGP as a common format visible to FB and IndieWeb.
  • No publisher control over how OGP properties are re-interpreted as mf2

This approach is not mutually exclusive with the previous approach, and maybe something to consider also doing, rather than versus.

See current practical work implementing this:

See Also