Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

[Pedro Andres Aranda Gutierrez]

[-- Attachment #1: Type: text/plain, Size: 936 bytes --]

Hi,

@Eli Zaretskii <eliz@gnu.org>
>evaluation cannot load any files, nor calls functions like
>@code{posn-at-point} or @code{window-in-direction}, which themselves
>+evaluate the mode line, as doing so could cause infinite recursion.

Shouldn't that be 'cannot load any files nor call functions'?

Best, /PA

PS: @list I was just listing to the 'allow ...' discussion and you may have
overseen that allow + infinitive is the 1:1 translation of 'permitir +
infinitive' in Spanish ;-)
Anyhow, my question would be if we always need 'allowing' or if we can say
that "variables _control_ how things are done" and functions _do_ things.
Wouldn't that be easier to read (and understand)?

-- 
Fragen sind nicht da, um beantwortet zu werden,
Fragen sind da um gestellt zu werden
Georg Kreisler

Headaches with a Juju log:
unit-basic-16: 09:17:36 WARNING juju.worker.uniter.operation we should run
a leader-deposed hook here, but we can't yet

[-- Attachment #2: Type: text/html, Size: 2091 bytes --]

Re: Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

[Eli Zaretskii]

> From: Pedro Andres Aranda Gutierrez <paaguti@gmail.com>
> Date: Wed, 4 Sep 2024 09:02:18 +0200
> 
> >evaluation cannot load any files, nor calls functions like
> >@code{posn-at-point} or @code{window-in-direction}, which themselves
> >+evaluate the mode line, as doing so could cause infinite recursion.
> 
> Shouldn't that be 'cannot load any files nor call functions'?

No, because "calls" refers to 'evaluation", which is singular, not
plural.

> PS: @list I was just listing to the 'allow ...' discussion and you may have overseen that allow + infinitive is the 1:1
> translation of 'permitir + infinitive' in Spanish ;-) 
> Anyhow, my question would be if we always need 'allowing' or if we can say that "variables _control_ how things
> are done" and functions _do_ things. Wouldn't that be easier to read (and understand)?

Yes, where "controls" is appropriate.  It isn't always.  Here's one
random example:

  Turning on option 'foo' allows you to observe a wonderful vista.

Thanks.

Re: Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

[Stephen Berman]

On Wed, 04 Sep 2024 15:25:20 +0300 Eli Zaretskii <eliz@gnu.org> wrote:

>> From: Pedro Andres Aranda Gutierrez <paaguti@gmail.com>
>> Date: Wed, 4 Sep 2024 09:02:18 +0200
>>
>> >evaluation cannot load any files, nor calls functions like
>> >@code{posn-at-point} or @code{window-in-direction}, which themselves
>> >+evaluate the mode line, as doing so could cause infinite recursion.
>>
>> Shouldn't that be 'cannot load any files nor call functions'?
>
> No, because "calls" refers to 'evaluation", which is singular, not
> plural.

The clause beginning with "nor" is a complement of "can", so the verb
must be a bare infinitive, i.e. "call" and not "to call", but also not
"calls", since that's not an infinitive form.  If you find "nor call
functions" hard to parse, the more explicit "nor can it call functions"
is an alternative.

Steve Berman

Re: Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

[Stephen Berman]

On Wed, 04 Sep 2024 15:55:29 +0200 Stephen Berman <stephen.berman@gmx.net> wrote:

> On Wed, 04 Sep 2024 15:25:20 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
>
>>> From: Pedro Andres Aranda Gutierrez <paaguti@gmail.com>
>>> Date: Wed, 4 Sep 2024 09:02:18 +0200
>>>
>>> >evaluation cannot load any files, nor calls functions like
>>> >@code{posn-at-point} or @code{window-in-direction}, which themselves
>>> >+evaluate the mode line, as doing so could cause infinite recursion.
>>>
>>> Shouldn't that be 'cannot load any files nor call functions'?
>>
>> No, because "calls" refers to 'evaluation", which is singular, not
>> plural.
>
> The clause beginning with "nor" is a complement of "can", so the verb
> must be a bare infinitive, i.e. "call" and not "to call", but also not
> "calls", since that's not an infinitive form.  If you find "nor call
> functions" hard to parse, the more explicit "nor can it call functions"
> is an alternative.

After seeing your new rewording, I see that my "more explicit"
alternative phrasing is not good in the full sentence, which I didn't
see when I responded (a correct alternative is "nor that it can
call...", which sound somewhat stilted, or simply (and better) "and
cannot call...").  (Your new rewording is fine, though I think "neither
loads any files nor calls functions like..." might be understood to
imply nothing else is excluded from evaluation, while with "does not
load any files or call functions like..." that implication seems
weaker.)

Steve Berman

Re: Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

[Pedro Andres Aranda Gutierrez]

[-- Attachment #1: Type: text/plain, Size: 2753 bytes --]

Hi,

I've continued thinking about this and have come up with:

Make sure this
evaluation does not load any files, or call functions like
@code{posn-at-point} or @code{window-in-direction}, which themselves
evaluate the mode line, as doing so could cause infinite recursion.

or

Make sure evaluating @var{form} does not load any files, or call functions
like
@code{posn-at-point} or @code{window-in-direction}, which themselves
evaluate the mode line, as doing so could cause infinite recursion.

I personally would highlight this. Putting that in a separate paragraph may
help
or adding a bold *CAVEAT* before. My feeling is that being in the paragraph
dilutes the importance of this kind of advices, not only here, but also in
other
places in the manuals. I know this can may take long, but IMHO this would
improve
the manuals a lot.

Best, /PA

On Wed, 4 Sept 2024 at 17:11, Stephen Berman <stephen.berman@gmx.net> wrote:

> On Wed, 04 Sep 2024 15:55:29 +0200 Stephen Berman <stephen.berman@gmx.net>
> wrote:
>
> > On Wed, 04 Sep 2024 15:25:20 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
> >
> >>> From: Pedro Andres Aranda Gutierrez <paaguti@gmail.com>
> >>> Date: Wed, 4 Sep 2024 09:02:18 +0200
> >>>
> >>> >evaluation cannot load any files, nor calls functions like
> >>> >@code{posn-at-point} or @code{window-in-direction}, which themselves
> >>> >+evaluate the mode line, as doing so could cause infinite recursion.
> >>>
> >>> Shouldn't that be 'cannot load any files nor call functions'?
> >>
> >> No, because "calls" refers to 'evaluation", which is singular, not
> >> plural.
> >
> > The clause beginning with "nor" is a complement of "can", so the verb
> > must be a bare infinitive, i.e. "call" and not "to call", but also not
> > "calls", since that's not an infinitive form.  If you find "nor call
> > functions" hard to parse, the more explicit "nor can it call functions"
> > is an alternative.
>
> After seeing your new rewording, I see that my "more explicit"
> alternative phrasing is not good in the full sentence, which I didn't
> see when I responded (a correct alternative is "nor that it can
> call...", which sound somewhat stilted, or simply (and better) "and
> cannot call...").  (Your new rewording is fine, though I think "neither
> loads any files nor calls functions like..." might be understood to
> imply nothing else is excluded from evaluation, while with "does not
> load any files or call functions like..." that implication seems
> weaker.)
>
> Steve Berman
>


-- 
Fragen sind nicht da, um beantwortet zu werden,
Fragen sind da um gestellt zu werden
Georg Kreisler

Headaches with a Juju log:
unit-basic-16: 09:17:36 WARNING juju.worker.uniter.operation we should run
a leader-deposed hook here, but we can't yet

[-- Attachment #2: Type: text/html, Size: 3954 bytes --]

Re: Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

[Eli Zaretskii]

> From: Pedro Andres Aranda Gutierrez <paaguti@gmail.com>
> Date: Fri, 6 Sep 2024 10:13:20 +0200
> Cc: Eli Zaretskii <eliz@gnu.org>, emacs-devel@gnu.org
> 
> I've continued thinking about this and have come up with:
> 
> Make sure this
> evaluation does not load any files, or call functions like
> @code{posn-at-point} or @code{window-in-direction}, which themselves
> evaluate the mode line, as doing so could cause infinite recursion.
> 
> or 
> 
> Make sure evaluating @var{form} does not load any files, or call functions like
> @code{posn-at-point} or @code{window-in-direction}, which themselves
> evaluate the mode line, as doing so could cause infinite recursion.

The current text says

  Make sure this evaluation neither loads any files nor calls functions
  like @code{posn-at-point} or @code{window-in-direction}, which
  themselves evaluate the mode line, as doing so could cause infinite
  recursion.

which I think is equivalent to what you suggest, but slightly less
ambiguous, grammar-wise.

> I personally would highlight this. Putting that in a separate paragraph may help
> or adding a bold *CAVEAT* before. My feeling is that being in the paragraph 
> dilutes the importance of this kind of advices, not only here, but also in other
> places in the manuals. I know this can may take long, but IMHO this would improve
> the manuals a lot.

I don't see a reason to make each caveat in the manual stand out.
There are too many of them.

Re: Small correction to commit ae2463796f236b8ee2cef3b5e38bffa13abd2233

[Pedro Andres Aranda Gutierrez]

[-- Attachment #1: Type: text/plain, Size: 757 bytes --]

 I know and that may be a long task. But highlighting tricky points may
help in more than one direction:
1.- reduce frustration in people customising their Emacs (at least because
you understand what you did)
2.- encourage people to investigate more and potentially come up with a way
of getting rid of these caveats

My .2c, /PA

On Fri, 6 Sept 2024 at 13:04, Eli Zaretskii <eliz@gnu.org> wrote:

>
> I don't see a reason to make each caveat in the manual stand out.
> There are too many of them.
>

We

-- 
Fragen sind nicht da, um beantwortet zu werden,
Fragen sind da um gestellt zu werden
Georg Kreisler

Headaches with a Juju log:
unit-basic-16: 09:17:36 WARNING juju.worker.uniter.operation we should run
a leader-deposed hook here, but we can't yet

[-- Attachment #2: Type: text/html, Size: 1371 bytes --]