Bruce Momjian <bruce@momjian.us> writes:
> On Thu, Apr 6, 2017 at 06:13:44PM -0400, Tom Lane wrote:
>> Not really convinced by the s/regular/normal/g changes. That doesn't seem
>> to clarify much either. Maybe it would help to define "normal" near here.
>> There's a definition in func.sgml (that they are not ordered-set or
>> hypothetical-set aggregates) but that's too far away.
> This change is not on my patch yes but what do you think about changing
> "regular" and "normal" aggregate designations to "generic"? That seems
> clearer because the aggregates work in non-window and window contexts.
I don't think "generic" is le mot juste here. What's more generic about
avg() than about mode(), for instance?
My dictionary says "generic" means "characteristic of a class or group
of things", which doesn't seem to have anything to do with this.
I actually think "regular" might be a better choice than "normal" for this
purpose, mainly because somebody looking for statistical aggregates might
think that "normal" has something to do with normal distributions. So
maybe we should leave that one group of uses alone and s/normal/regular/
in section 9.20. But in any case it would help to define the term
explicitly in both sections.
regards, tom lane