Reference
Words used around the lab
Crawl
A process that requests resources and follows discoverable addresses. A crawler sees only the material it successfully retrieves within its rules and limits. A stopped or blocked crawl is not a complete inventory.
Index
A search system's stored representation of information it may use to answer queries. A successful request or an index-permitting directive does not prove that a particular search engine has included a page.
Canonical URL
An address declared or selected as the preferred representation of a page or equivalent group of pages. The publisher's declaration and a search engine's selection are related but distinct observations.
Internal link
A link from one page to another address on the same website. Its usefulness depends on the relationship between the source, label, destination, and reader's task, not simply on the number of links.
Metadata
Information that describes a resource, such as its title or a short description. Metadata should accurately represent the resource rather than promise unsupported benefits or substitute for useful content.
Rollback
A bounded operation that restores an earlier state after a change. A dependable rollback requires a known previous version and verification that the restoration actually occurred. Restoring a file does not automatically reverse every external effect.