Qualified usage of `private` vs. shared caches

[mpdude]

Section 3 of draft-ietf-httpbis-cache says:

A cache MUST NOT store a response to any request, unless: [...]
– the "private" response directive (see Section 5.2.2.6) does not appear in the response, if the cache is shared.

Section 5.2.2.6 then reads:

The "private" response directive indicates that the response message
is intended for a single user and MUST NOT be stored by a shared
cache. [...]

If the private response directive specifies one or more field-names,
this requirement is limited to the field-values associated with the
listed response header fields. That is, a shared cache MUST NOT
store the specified field-names(s), whereas it MAY store the
remainder of the response message.

To me, this last paragraph was a surprising read after Section 3 made a rather clear "if it is private and the cache is shared, don't cache it" statement.

Maybe it would help to amend the wording in Section 3 to read (emphasis mine)

the "private" response directive (see Section 5.2.2.6) does not appear in unqualified form in the response, if the cache is shared

[royfielding]

Right, though I expect "qualified" or "unqualified" is a little too vague to explain it. Maybe the "private" response directive without a list of field names?

[mnot]

Currently, we have:

if the cache is shared: the "private" response directive is not present in the response (ref to section)

Proposal:

if the cache is shared: the "private" response directive, if present, allows the response to be stored (possibly with modifications; see ref to section)

[mpdude]

@mnot I am not sure I get you right :-/. Which section exactly are you referring to with your suggestion?

[mnot]

Sorry - here

[royfielding]

I think that would be sufficient to close. Maybe specifically say "allows the response to be stored by a shared cache (with modifications)".