ROR Identifier Pattern

The preferred form of a ROR identifier (a "ROR ID", the id element in a ROR record) is the entire URL: https://ror.org/02mhbdp94. However, the ROR API will also recognize forms such as ror.org/02mhbdp94 or 02mhbdp94, so these forms of the ROR ID may also be used.

The entire URL is the preferred form of a ROR identifier because URL changes can easily be made backward compatible via 301 redirects, whereas IDs stripped of URL components can make it difficult for humans and machines to determine how to resolve that ID. See https://github.com/ORCID/ORCID-Source/issues/6520 for a fuller discussion of this issue.

The unique strings in ROR identifiers are assigned randomly, not sequentially, and contain no organizational information; therefore, the ROR ID of one organization cannot be predicted from the ROR ID of a related organization.

The unique portions of ROR identifiers have a consistent pattern and can be validated with regular expressions. The unique string consists of 9 characters: a zero, 6 characters that can be either lower-case letters or integers, and 2 concluding integers. The unique string always begins with a zero so that ROR ID landing pages such as https://ror.org/02mhbdp94 can be easily differentiated from ROR web pages such as https://ror.org/about.

The ROR ID pattern uses base 32 Crockford encoding, which excludes letters "I", "L", "O", and "U". The last 2 digits of the ROR identifier are a checksum that follows the ISO/IEC 7064:2003 standard. See the code that generates the ROR ID and the checksum calculation for further details.

Regular expressions to validate the unique portion of the ROR ID pattern include the following:

^0[a-z|0-9]{6}[0-9]{2}$

^0[a-hj-km-np-tv-z|0-9]{6}[0-9]{2}$