Introduction to metadata
"Metadata" is simply "data about data". In the case of CloudPublish, metadata refers to data about your eBooks, eJournals, and other electronic documents. This data is used to describe and locate your items, and it is vital that it is both accurate and consistently-presented.
Depending on the type of item in question, CloudPublish accepts either ONIX files (an XML format devised by Editeur and very widely adopted in the ebook industry) or CSV files, with a template created by CloudPublish based on standard industry practice.
There are a few considerations which are common to all types of metadata:
- Character set: To avoid any character encoding issues, please use UTF-8 at all times.
- Authors / names in general: Please use a consistent format, e.g. "Firstname Lastname" or "Initials, A B C". We have algorithms that parse authors into the parts of their names in order to generate correctly-styled citations, but it helps if the original data is well-structured.
- HTML is acceptable in only a few places, most notably in eBook descriptions. It is your responsibility to ensure that this HTML is well-structured.
- Do not use all-capitals, e.g. in titles or subjects, unless this is the convention (such as in Thema or BIC codes). For example, please give "History" not "HISTORY".
We can conduct a dry-run to test your metadata at any time, please let your account manager know if you require this.
It does not matter if you send metadata first, or source files first, or both at the same time. CloudPublish will carry out checks for the existence of corresponding source files or metadata (as appropriate) when processing both. However, it is necessary for metadata to have been received and processed before items added or updated in that metadata can be returned from the CloudPublish API.
For more information about specific types of metadata, please see these pages: