Re: Partial Mode in Aggregate Functions - Mailing list pgsql-hackers

From David G. Johnston
Subject Re: Partial Mode in Aggregate Functions
Date
Msg-id CAKFQuwbnkdMxagNvFTr8R7=_q+1PzrakBw3yyMRcrSnz+n4SvQ@mail.gmail.com
Whole thread
In response to Re: Partial Mode in Aggregate Functions  (Marcos Pegoraro <marcos@f10.com.br>)
Responses Re: Partial Mode in Aggregate Functions
List pgsql-hackers
On Sunday, March 1, 2026, Marcos Pegoraro <marcos@f10.com.br> wrote:
Em sex., 27 de fev. de 2026 às 22:58, David G. Johnston <david.g.johnston@gmail.com> escreveu:
I commented about the phrasing for this being not ideal for the target audience of the functions reference page.  Here's some other wording to consider:

I sent this link to some colleagues and asked, "Tell me what 'Partial Mode' means." Some are developers, some are DBAs, and some are just starting out. Those who responded in less than 20 seconds obviously didn't read the headers; they just deduced it by looking at the table title. And yes, some said it related to aggregating data on fewer records. So, putting an explanation in the header probably won't completely solve the problem, since many people only look at the table, and if they can draw any conclusions from it, they ignore the rest of the text.


I can’t come up with a better header than Partial Mode.  Given the fairly benign outcome for such a misunderstanding - this isn’t critical to understand knowledge for writing queries - I can live with people who don’t read the documentation being misinformed. Though it does make my idea of just ripping it out, and putting it elsewhere where optimizations are talked about, more appealing.

David J.

pgsql-hackers by date:

Previous
From: Yura Sokolov
Date:
Subject: Re: Fix bug in multixact Oldest*MXactId initialization and access
Next
From: Daniil Davydov
Date:
Subject: Re: POC: Parallel processing of indexes in autovacuum