Re: Installation instructions vs binaries

2020-07-16 Thread David G. Johnston
On Wed, Jul 15, 2020 at 4:13 AM Magnus Hagander  wrote:

> It's kind of strange that if you start your PostgreSQL journey by reading
> our instructions, you get nothing useful about installing PostgreSQL from
> binary packages other than "go ask somebody else about it". Yet we have
> pretty good step by step instructions on our *website* for it.
>
> Attached patch adds a chapter to the docs about installing from binaries,
> and refers those users to the website download instructions (and updates
> the Windows instructions to include an actual link to the website).
>
> It also adds mention of it in a few other places - -there are probably
> more that could use some help with that in the future. But I've seen a lot
> of people get confused by our documentation assuming everything is from
> source when it comes to initdb and pg_ctl that I think it's worth specially
> mentioning it there.
>

Thanks.  Adding tips calling out common package-specific/handled pieces
seems ok.

One typo for the patch as-is:

+   When PostgreSQL is installed using binary packages, starting and
stopping
+of the system is normally integrated with the service management on the
+platform. Refer to the documentation for the documentation for these
+packages and the platform for more information.

Remove "for the documentation"

However, maybe we should avoid repeated use of the passive "When PostgreSQL
is installed using binary packages".  Consider just: "PostgreSQL binary
packages". e.g.,

PostgreSQL binary packages normally include platform-appropriate service
management (starting and stopping).  Consult the package documentation for
more information.

(the other two can be rewording similarly if this is deemed better - all
three should be consistent).

David J.


Re: strange syntax

2020-07-16 Thread Laurenz Albe
On Tue, 2020-07-14 at 21:33 +, PG Doc comments form wrote:
> this sentence looks strange:
> "Schemas are a purely logical structure and who can access what is managed
> by the privilege system"
> https://www.postgresql.org/docs/12/manage-ag-overview.html
> 
> especially this part: "who can access what is managed by"

It is correct English, but maybe it could be worded better:

"Schemas are a purely logical structure and impose no access restrictions,
apart from those defined by the privilege system."

Yours,
Laurenz Albe





Re: strange syntax

2020-07-16 Thread David G. Johnston
On Thu, Jul 16, 2020 at 11:58 AM Laurenz Albe 
wrote:

> On Tue, 2020-07-14 at 21:33 +, PG Doc comments form wrote:
> > this sentence looks strange:
> > "Schemas are a purely logical structure and who can access what is
> managed
> > by the privilege system"
> > https://www.postgresql.org/docs/12/manage-ag-overview.html
> >
> > especially this part: "who can access what is managed by"
>
> It is correct English, but maybe it could be worded better:
>

Agreed

> "Schemas are a purely logical structure and impose no access restrictions,
> apart from those defined by the privilege system."


But I don't think this is an improvement.

I think "modular" is a better term than "logical" here.

Given that we've already said "multiple databases, which are isolated from
each other" trying to say the same thing with "impose no access
restrictions" doesn't seem to help; and one of the two major benefits of
schemas is to provide privilege system hook less granular than object -
which the original puts forward directly but clumsily and the proposal sets
into an aside ("apart from").  The other benefit, namespace isolation,
isn't even mentioned but seems like a useful addition.

Thus:

Schemas provide modular structure, with namespace isolation and
authorization control.
or
... possibly into separate schemas; which provide modular structure with
namespace isolation and authorization control.
David J.


Re: strange syntax

2020-07-16 Thread David G. Johnston
On Thu, Jul 16, 2020 at 1:38 PM David G. Johnston <
david.g.johns...@gmail.com> wrote:

> On Thu, Jul 16, 2020 at 11:58 AM Laurenz Albe 
> wrote:
>
>> On Tue, 2020-07-14 at 21:33 +, PG Doc comments form wrote:
>> > this sentence looks strange:
>> > "Schemas are a purely logical structure and who can access what is
>> managed
>> > by the privilege system"
>> > https://www.postgresql.org/docs/12/manage-ag-overview.html
>> >
>> > especially this part: "who can access what is managed by"
>>
>> It is correct English, but maybe it could be worded better:
>>
>
> Agreed
>
>> "Schemas are a purely logical structure and impose no access restrictions,
>> apart from those defined by the privilege system."
>
>
> But I don't think this is an improvement.
>

I went ahead and reviewed the whole page and came up with a minor rework
patch instead of just a targeted fix.

https://www.postgresql.org/message-id/CAKFQuwacZW-gNLyScFKFxCg_g7QtT7_HQMOyexuV2zUj%3DeQFAw%40mail.gmail.com

David J.