bug#49037: flyspell word correction frustrating

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 1407 bytes --]


I have been working with flyspell and have found correcting words with flyspell

spectacularly frustrating.  



"flyspell-correct-word" is annoyingly only bound to a mouse click. 



"flyspell-correct-word-before-point" is also bound to a mouse click or one can

use "M-x flyspell-correct-word-before-point".



Another big problem is that the menu of possible corrections.that is displayed

covers the buffer.  If I have a long list of completions, the menu covers the text.



Although ispell works better, the way one selects the replacement word is also

cumbersome.



emacs has the icomplete (incremental minibuffer compeltion preview) which works

very well without affecting the viewing of the buffer.  One drawback with icomplete

is that the completions are not shown in a vertical way as done in ivy.



On Apr 27, 2020, Omar Antolín Camarena released icomplete-vertical.el, but the 

functionality has not really been introduced in the official emacs and improved upon.



If there was the capability for vertical spelling correction with vertical listing mimicking

ivy, things will be much better for correcting words with flyspell.



Omar Antolín Camarena also wrote orderless.el, a very useful package that provides
orderless completion style in the minibuffer. This would be very good to have in 

official emacs as well. 
 


[-- Attachment #2: Type: text/html, Size: 3018 bytes --]

bug#49037: flyspell word correction frustrating

[Lars Ingebrigtsen]

henri-biard@francemel.fr writes:

> I have been working with flyspell and have found correcting words with flyspell
> spectacularly frustrating.  
>
> "flyspell-correct-word" is annoyingly only bound to a mouse click. 

`C-c $' is the command to use from the keyboard.

I didn't read the rest of the bug report -- you previously said (under a
different name) that you were a "journalist" performing some sort of
"experiment".  If the experiment is to test how many irrelevant bug
reports you can file before getting banned (or something similar) --
please stop the experiment.

Closing.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 1382 bytes --]



One developer full of shit trying to maintain a package whilst closing 

bug reports he doesn't like.  Now the banning bullshit starts.  



You might as well ban people considering you are getting in the habit to close request

anyway.



There have been many criticisms through the years that emacs developers are slow to

react and improve.  Lately there have been discussions on emacs modernisation but it

always gets to deaf ears.




From: Lars Ingebrigtsen <larsi@gnus.org>
To: henri-biard@francemel.fr
Subject: bug#49037: flyspell word correction frustrating
Date: 15/06/2021 17:10:51 Europe/Paris
Cc: 49037@debbugs.gnu.org

henri-biard@francemel.fr writes:

> I have been working with flyspell and have found correcting words with flyspell
> spectacularly frustrating. 
>
> "flyspell-correct-word" is annoyingly only bound to a mouse click. 

`C-c $' is the command to use from the keyboard.

I didn't read the rest of the bug report -- you previously said (under a
different name) that you were a "journalist" performing some sort of
"experiment". If the experiment is to test how many irrelevant bug
reports you can file before getting banned (or something similar) --
please stop the experiment.

Closing.

-- 
(domestic pets only, the antidote for overdose, milk.)
bloggy blog: http://lars.ingebrigtsen.no





[-- Attachment #2: Type: text/html, Size: 1835 bytes --]

bug#49037: flyspell word correction frustrating

[Lars Ingebrigtsen]

henri-biard@francemel.fr writes:

> You might as well ban people considering you are getting in the habit
> to close request anyway.

We have no mechanisms to ban anybody.  I was pointing out that if your
"experiment" was to see how much you can abuse the bug reporting
facilities (with trivial user questions that are already answered in the
manual) before you get banned -- it's not going to happen, so you can
just stop the experiment right away.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#49037: flyspell word correction frustrating

[Robert Pluim]

>>>>> On Tue, 15 Jun 2021 17:23:52 +0200 (CEST), henri-biard@francemel.fr said:

    henri-biard> One developer full of shit trying to maintain a package whilst closing 
    henri-biard> bug reports he doesn't like. Now the banning bullshit starts.  

You reported a bug that was not a bug. Lars closed it, rapidly. And
politely, unlike you.

    henri-biard> You might as well ban people considering you are getting in the habit to close request
    henri-biard> anyway.

You'd prefer we left bugs that are not bugs open?

(hmm, flyspell mode binds "C-." and "C-,", which donʼt work on
tty. Perhaps we should invent alternate bindings, but thatʼs a
separate bugreport)

Robert
-- 

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 1728 bytes --]



> We have no mechanisms to ban anybody. 



Let's hope so.  But you brought it up, I believed you.



I was pointing out that if your "experiment" was to see how much you can abuse the

bug reporting facilities (with trivial user questions that are already answered in the
manual) before you get banned -- it's not going to happen, so you can just stop the

experiment right away.



It's has only been your conclusion about what you call "experiment".



It is you who is inventing all these things.  At least Eli responds well, with no accusations

of sorts.



Can you tell me exactly where the information is in the manual because I could only

find a reference to aspell in "16.4 Checking and Correcting Spelling"



If it was good information I would not complain.



Instead of delving into accusations, it is always better to educate and assume other

are not as bad as you might believe.







From: Lars Ingebrigtsen <larsi@gnus.org>
To: henri-biard@francemel.fr
Subject: bug#49037: flyspell word correction frustrating
Date: 15/06/2021 17:48:22 Europe/Paris
Cc: 49037@debbugs.gnu.org

henri-biard@francemel.fr writes:

> You might as well ban people considering you are getting in the habit
> to close request anyway.

We have no mechanisms to ban anybody. I was pointing out that if your
"experiment" was to see how much you can abuse the bug reporting
facilities (with trivial user questions that are already answered in the
manual) before you get banned -- it's not going to happen, so you can
just stop the experiment right away.

-- 
(domestic pets only, the antidote for overdose, milk.)
bloggy blog: http://lars.ingebrigtsen.no





[-- Attachment #2: Type: text/html, Size: 2286 bytes --]

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 2164 bytes --]

Politely ???



He closed and continued with



>I didn't read the rest of the bug report -- you previously said (under a
>different name) that you were a "journalist" performing some sort of
>"experiment". If the experiment is to test how many irrelevant bug
>reports you can file before getting banned (or something similar) --

>please stop the experiment.



Can see how it is extremely polite !  Full of accusations about user's ill will.

Things do not become pleasant because one adds please at the end.



>You'd prefer we left bugs that are not bugs open?



No, but people need time to respond and get some clarifications.  We all understand

that a bug report gets closed.  Have worked in many places where bug reports

just need some clarification.



Where can one report on fixes (things one could technically argue are not fundamentally

bugs) or to discuss improvements ?



You contributed to the conversation 



>(hmm, flyspell mode binds "C-." and "C-,", which donʼt work on
>tty. Perhaps we should invent alternate bindings, but thatʼs a
>separate bugreport)



This has nothing to do with the bug report, but I get the lynching, and you don't.



From: Robert Pluim <rpluim@gmail.com>
To: henri-biard@francemel.fr
Subject: bug#49037: flyspell word correction frustrating
Date: 15/06/2021 17:55:04 Europe/Paris
Cc: Lars Ingebrigtsen <larsi@gnus.org>;
   49037@debbugs.gnu.org

>>>>> On Tue, 15 Jun 2021 17:23:52 +0200 (CEST), henri-biard@francemel.fr said:

henri-biard> One developer full of shit trying to maintain a package whilst closing 
henri-biard> bug reports he doesn't like. Now the banning bullshit starts.  

You reported a bug that was not a bug. Lars closed it, rapidly. And
politely, unlike you.

henri-biard> You might as well ban people considering you are getting in the habit to close request
henri-biard> anyway.

You'd prefer we left bugs that are not bugs open?

(hmm, flyspell mode binds "C-." and "C-,", which donʼt work on
tty. Perhaps we should invent alternate bindings, but thatʼs a
separate bugreport)

Robert
-- 





[-- Attachment #2: Type: text/html, Size: 2795 bytes --]

bug#49037: flyspell word correction frustrating

[Eli Zaretskii]

> From: henri-biard@francemel.fr
> Date: Tue, 15 Jun 2021 18:34:00 +0200 (CEST)
> Cc: 49037@debbugs.gnu.org
> 
> Can you tell me exactly where the information is in the manual because I could only
> find a reference to aspell in "16.4 Checking and Correcting Spelling"

That's the place.

> If it was good information I would not complain.

What is not good about it?

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 1207 bytes --]

>> If it was good information I would not complain.

>What is not good about it?



For instance:



1. It does not tell me that aspell is the default



2. It does not tell me that even when I use ispell-word, it might actually

do something else (aspell-word maybe).



3. It does not describe that it searches PATH.



I could delve into that and add to the documentation.  It has some information

but not comprehensive enough to actually set things up in practice without

any doubt. 



These comments were written to help based on my experience.








From: Eli Zaretskii <eliz@gnu.org>
To: henri-biard@francemel.fr
Subject: bug#49037: flyspell word correction frustrating
Date: 15/06/2021 18:54:21 Europe/Paris
Cc: larsi@gnus.org;
   49037@debbugs.gnu.org

> From: henri-biard@francemel.fr
> Date: Tue, 15 Jun 2021 18:34:00 +0200 (CEST)
> Cc: 49037@debbugs.gnu.org
> 
> Can you tell me exactly where the information is in the manual because I could only
> find a reference to aspell in "16.4 Checking and Correcting Spelling"

That's the place.

> If it was good information I would not complain.

What is not good about it?





[-- Attachment #2: Type: text/html, Size: 1639 bytes --]

bug#49037: flyspell word correction frustrating

[Eli Zaretskii]

> From: henri-biard@francemel.fr
> Cc: larsi@gnus.org,
>  49037@debbugs.gnu.org
> Date: Tue, 15 Jun 2021 19:08:48 +0200 (CEST)
> 
> >> If it was good information I would not complain.
> 
> >What is not good about it?
> 
> For instance:
> 
> 1. It does not tell me that aspell is the default

Why should it?  You set up your system with whatever spell-checker you
want, and Emacs uses that.  It is not Emacs's business to tell you
which spell-checker to install and use.

> 2. It does not tell me that even when I use ispell-word, it might actually
> do something else (aspell-word maybe).

ispell-word is a command, it is unrelated to the program being
invoked.  The manual clearly says that those commands can work with
any one of the 4 supported spell-checkers.

> 3. It does not describe that it searches PATH.

That's an implementation detail.  If you want to know those details,
you have the source to read and study.  The user manual explains how
to use the commands, not how the commands work.

bug#49037: flyspell word correction frustrating

[Glenn Morris]

Lars Ingebrigtsen wrote:

> We have no mechanisms to ban anybody. 

FTR, postings to debbugs are moderated via a standard mailman interface.
Of course, that is trivial to circumvent for a determined bad actor.

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 1992 bytes --]

>From: Eli Zaretskii <eliz@gnu.org>
>To: henri-biard@francemel.fr
>Subject: Re: bug#49037: flyspell word correction frustrating
>Date: 15/06/2021 19:33:50 Europe/Paris
>Cc: larsi@gnus.org;
>   49037@debbugs.gnu.org

>> From: henri-biard@francemel.fr
>> Cc: larsi@gnus.org,
>> 49037@debbugs.gnu.org
>> Date: Tue, 15 Jun 2021 19:08:48 +0200 (CEST)
>> 
>> >> If it was good information I would not complain.
>> 
>> >What is not good about it?
>> 
>> For instance:
>> 
>> 1. It does not tell me that aspell is the default

>Why should it? You set up your system with whatever spell-checker you
>want, and Emacs uses that. It is not Emacs's business to tell you
>which spell-checker to install and use.



I have multiple ones and make comparisons between them.  So I really want to know

what it uses.  After all the information is there and that is what emacs actually does.

Should I not be able to easily know what it is doing under the hood.  Emacs Self

Documentation should help me with that.



>> 2. It does not tell me that even when I use ispell-word, it might actually
>> do something else (aspell-word maybe).

>ispell-word is a command, it is unrelated to the program being
>invoked. The manual clearly says that those commands can work with
>any one of the 4 supported spell-checkers.

>> 3. It does not describe that it searches PATH.

>That's an implementation detail. If you want to know those details,
>you have the source to read and study. The user manual explains how

>to use the commands, not how the commands work.



Explains with no examples or much description of ispell-program-name, which

for me is very important.



Because Gnu Aspell was designed to replace Ispell, how about deprecating Ispell

and simply have emacs-ispell be the general framework for orthographic study.



That would certainly solve the problem.



Would appreciate if you do not consider me a "Bad Actor".  





[-- Attachment #2: Type: text/html, Size: 2447 bytes --]

bug#49037: flyspell word correction frustrating

[Eli Zaretskii]

> From: henri-biard@francemel.fr
> Cc: larsi@gnus.org,
>  49037@debbugs.gnu.org
> Date: Tue, 15 Jun 2021 20:24:27 +0200 (CEST)
> 
> >> 1. It does not tell me that aspell is the default
> 
> >Why should it? You set up your system with whatever spell-checker you
> >want, and Emacs uses that. It is not Emacs's business to tell you
> >which spell-checker to install and use.
> 
> I have multiple ones and make comparisons between them.  So I really want to know
> what it uses.  After all the information is there and that is what emacs actually does.
> Should I not be able to easily know what it is doing under the hood.  Emacs Self
> Documentation should help me with that.

Documentation cannot tell everything; if it tried, it'd become too
long and hard to read.  The details you are after are in the source
code; the documentation of each variable has a link to the source,
where you can see the gory details.

> >> 3. It does not describe that it searches PATH.
> 
> >That's an implementation detail. If you want to know those details,
> >you have the source to read and study. The user manual explains how
> >to use the commands, not how the commands work.
> 
> Explains with no examples or much description of ispell-program-name, which
> for me is very important.

Yours is a rare situation, so you shouldn't expect it to be covered in
the general documentation.

> Because Gnu Aspell was designed to replace Ispell, how about deprecating Ispell
> and simply have emacs-ispell be the general framework for orthographic study.

We already did that.

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 821 bytes --]

>> >That's an implementation detail. If you want to know those details,
>> >you have the source to read and study. The user manual explains how
>> >to use the commands, not how the commands work.
>> 
>> Explains with no examples or much description of ispell-program-name, which
>> for me is very important.

>Yours is a rare situation, so you shouldn't expect it to be covered in
>the general documentation.

Then I suggest maintainers stop making accusations on users simply because their use

case is a rare occurrence.  One cannot expect politeness once you denigrate others

about what is impartant to them or what they are trying to do. 



It occurs to me that you do understand, and would be a big improvement if others 

follow your example.



Many Gratitudes Eli.











[-- Attachment #2: Type: text/html, Size: 1031 bytes --]

bug#49037: flyspell word correction frustrating

[henri-biard]

[-- Attachment #1: Type: text/plain, Size: 281 bytes --]


Could you make ispell use the minibuffer for correction instead of making a separate

temporary buffer?  With the possibility of using the arrow keys for selection, taking ideas

from ivy.  Is there an icomplete vertical in official emacs?  It would be neat to have.



[-- Attachment #2: Type: text/html, Size: 308 bytes --]