HL7 uses tooling to aid both the standards development process and the conversion of artifacts into published form for balloting and for use by implementers
|Requirement||Information necessary to allow "rendering" of artifacts must be maintained as part of the artifact in circumstances where that information may need to be determined by humans (i.e. conscious decision required) rather than auto-generated by algorithmic processes|
|Rationale||In order to produce rendered versions of the artifacts in an automated way from the source of truth we need to avoid human intervention in the publication process. That means all human intervention must occur in the authoring process, even when that intervention is solely to define how the information is rendered rather than defining aspects of the artifact that actually affect implementation|
|Requirement||Information necessary to provide guidance to implementers, reviewers and various other participants must be captured in a way that allows the content to rendered differently (both in terms of whether it is exposed and how it is exposed) depending on the target audience and intended use of a particular published view of the artifact.|
|Rationale||Not every audience needs the same information. Because HL7's artifacts tend to be large it's important to filter the information exposed so that a given reader can easily consume the information that's relevant to them without having to wade through or skim past a large amount of information that isn't.
Specifically categorizing information into different groups with specific objectives and targets also leads to more rigor in the creation of the artifact documentation by forcing authors to be more conscious of a given piece of documentation's intended use and flagging other areas in which documentation may be appropriate.