Re: [HACKERS] Statistics "dependency"

On 04/23/2017 12:44 PM, Simon Riggs wrote:
> On 23 April 2017 at 09:17, Dean Rasheed <dean.a.rasheed@gmail.com> wrote:
>> On 23 April 2017 at 03:37, Bruce Momjian <bruce@momjian.us> wrote:
>>> In looking at the new multi-column statistics "dependency" option in
>>> Postgres 10, I am quite confused by the term "dependency".  Wouldn't
>>> "correlation" be clearer and less confusing as "column dependency"
>>> already means something else.
>>>
>
> I also asked that exactly that question...
>
>> Actually, the terms "dependency" and "correlation" are both quite
>> broad terms that cover a whole range of other different things, and
>> hence could be misleading. The precise term for this is "functional
>> dependency" [1], so if anything, the option name should be
>> "functional_dependencies" or some shortening of that, keeping a part
>> of each of those words.
>
> ...and got that answer also.
>
> For us "functional dependency" would sound like something to do with
> functions (e.g. CREATE FUNCTION), so just "dependency" appears to me
> to be the best term for this.
>

Not really. Functional dependency is a term well-defined in relational 
algebra, particularly in definition of normal forms. It has nothing to 
do with functions, and I'm sure it's not the only possible "ambiguity".

I actually considered functional_dependency when working on this, and I 
think it might be a better choice, but seemed a bit longish.

> There are multiple statistics for dependency stored, hence
> "dependencies". I don't like it, but its the best term I can see at
> present.

That is a good point, actually. It should probably be plural.

regards

-- 
Tomas Vondra                  http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services