[jats-list] Why is archiving JATS with a DOI not common?

["Castedo Ellerman castedo@xxxxxxxxxxx" <jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

I'm new to JATS and a bit confused why archiving JATS with a DOI is not 
common. I have yet to see an example actually. The closest example I 
have seen is eLife publicly storing JATS XML files in 
github.com/elifesciences/elife-article-xml

Are there any examples of JATS being archived with a DOI?

I believe many in this forum would agree that a DOI should refer NOT to 
a manifestation of an article (in FRBR Group 1 terms) but rather the 
work or expression. https://jats.nlm.nih.gov/archiving/rationale.html reads:
B  "The intent of the Archiving Tag Set is to provide a standardized 
format in which to preserve the intellectual content ..."

So shouldn't it be common that DOIs reference the intellectual content 
in the form of JATS XML?

As it stand now, it seems most DOIs to scientific articles are actually 
IDs to manifestations at journal websites which are optimized for the 
business interests of the publisher.

Thank you in advance if someone can help me understand why JATS XML DOIs 
are not common.

-Castedo