The general rule seems to be:
If Go's default value has the same semantics we'd use for an unset
value, don't bother with a pointer.
I'm not sure how well that squares with [1]:
We want a consistent way to identify unset settings.
But if the falsy values count as "unset", maybe the "null is a
consistent identifier for unset" approach was never really viable.
Qiang points out that pointers are required to opt-out of boolean
settings where both true and false would require action [2], so I've
worded the exception to only apply when the Go default for the type is
expicitly a no-op in the spec.
I'm also not sure if the new style extends to integers where zero has
the same semantics as unset values. It sounds like Michael was ok
with no pointers for those values [3], but OOMScoreAdj (where zero
clearly means "do nothing") got a pointer in #233 [4]. More clarity
on the threshold would be nice; in this commit I've laid out the logic
and not explicitly listed the types it applies to.
[1]: https://github.com/opencontainers/specs/pull/233#discussion_r47829711
[2]: https://github.com/opencontainers/specs/pull/317/files#r50932706
[3]: https://github.com/opencontainers/specs/pull/233#issuecomment-155250592
[4]: https://github.com/opencontainers/specs/pull/233/files#diff-34c30be66233f08b447fb608ea0e66bbR206
Signed-off-by: W. Trevor King <wking@tremily.us>
pandoc/LaTeX is not happy with this shady character.
```
! Package inputenc Error: Unicode char \u8:↔ not set up for use with
LaTeX.
```
Signed-off-by: Vincent Batts <vbatts@hashbangbash.com>
So we have something to cite to avoid rehashing established decisions.
Provide some motivation and links to the backing discussion so folks
can re-open these if they have new information that wasn't covered in
the original decision.
Like the glossary (18734986, glossary: Provide a quick overview of
important terms, 2015-08-11, #107), I've used subsection titles for
each entry to get link anchors.
Signed-off-by: W. Trevor King <wking@tremily.us>
W. Trevor King
Vincent Batts