The canonical copy of a piece of content is the “truest” copy. This is typically:
- the copy most actively in use, i.e. the copy to which updates are applied
- the copy which you want to lose the least
- the copy from which backups are made
- the copy you want people to link to
Different contexts for “canonical”
“canonical” can be used in a variety of different circumstances, and whilst it’s meaning is approximately the same in all of them it’s worth being aware of the difference.
Canonical copy in software
E.G. within someone’s publishing software there might be several copies of content, the copy which is used to build pages to serve, and which is edited when updates are made; a copy in an index which is used for searching; a copy in a cache for quickly serving pages. In this case it’s the former which is the canonical copy, as the others a) are easily rebuilt from the canonical copy and b) go out of date as soon as an update is made.
Canonical copy on the web
Content on the web can be copied freely both by the publisher (for example POSSE copies) and by others (for example reposts). In each circumstance, the “canonical” copy is the copy on the author’s site.