Re: New year - Out with the old!

Re: New year - Out with the old!

[novim]

[-- Attachment #1: Type: text/plain, Size: 2220 bytes --]

> Everyone new to Emacs finds how different it is. That can be like
> discovering a new continent. For some, it's marvelous. For others,
> it's just primitive, populated by savages, a decaying culture, or is
> just no longer relevant. For some, it's a mix.

The question is: what is the goal?

Is it getting Emacs used by more people, so the ideals of free
software reaches more people? Or is it keeping it as is, a niche tool
which free software enthusiast can discover if they want to?

If the former then the problem is emacs is very alien for new users
compared to popular software, so it takes a great deal of
determination to push through and discover the strengths of emacs.

Users usually complain about having to tinker for emacs quite a while
to get productive. It includes learning the keybindings and
configuring emacs for their usecases. Many users are turned away by
this, because they don't have the time or inclination to tinker to get
basic stuff work, but they may stick with emacs if it's usable for
them right away and they could discover the strengths of emacs if they
keeep using it.

As for the keybindings emacs should have selectable keyboard configs
similar to other software, just like other tools have vim, emacs,
etc. key emulations. These configs should be selectable right from the
startup screen for a new user, so he can switch to familiar keys with
a simple click. Familiar keys include copy/paste and other keys
standard on the platform (e.g. on windows C-c, C-v). Of course, the
documentation uses the default keys when descripting commands, but I
can imagine a mode which replaces the displayed keys in the
documentation automatically with overlays if an other key config is
selected.

As for productivity many users complain they have to tinker to get
completion set up for their programming languages. This should be
built in, so, for example, for java, c++, etc. completion and
documentation lookup should work automatically out of the box, using
LSP.

All of the above is for the case the goal is to get more users for
emacs. If the goal is to have it as a powerful tool which rewards only
those who are willing to spend a significant time on it then things
are good as they are.

[-- Attachment #2: Type: text/html, Size: 2894 bytes --]

Re: New year - Out with the old!

[Jean Louis]

* novim <laszlomail@protonmail.com> [2020-12-23 17:40]:
> As for the keybindings emacs should have selectable keyboard configs
> similar to other software, just like other tools have vim, emacs,
> etc. key emulations. These configs should be selectable right from
> the startup screen for a new user, so he can switch to familiar keys
> with a simple click. Familiar keys include copy/paste and other keys
> standard on the platform (e.g. on windows C-c, C-v).

Emacs already has CUA keys selectable in the Options menu, from start
and all the time.

If you mean vim has it, I do not see anything selectable right from
startup. On GNU/Linux there are many editors and many do not support
thoe CUA keys. Some simpler editors do. Any editor with history may
have rather its own keys. Not even on Windows all editors especially
older ones, do not support it. And I use computers since long time
really back to 1986 and since then I actually used rarel CUA keys on
windows as even on Windows I used different specific editors in DOS
mode as I considered such superior.

In general myself I never even expected that CUA keys must be
everywhere and I use it in browser on GNU/Linux. Nowhere else.

That is also one user experience.

> Of course, the documentation uses the default keys when descripting
> commands, but I can imagine a mode which replaces the displayed keys
> in the documentation automatically with overlays if an other key
> config is selected.

Aha that is good idea. Documentation could be dynamically
self-documenting. But in general it is from one side viewed. As if
user press C-h C-v when CUA mode is not enabled it says scroll up and
if user enables CUA keys C-h C-v says it is (cua-paste) function. From
that view point it is dynamically self documenting the keys at least.

> All of the above is for the case the goal is to get more users for
> emacs. If the goal is to have it as a powerful tool which rewards
> only those who are willing to spend a significant time on it then
> things are good as they are.

Bindings and features are alright as from features' side.

In my opinion, and based on the review of various statistics, there
are probably 5-10 millions of Emacs users already. According to my
latest analysis Emacs usage is growing, this year was a lot of new
popularity growth for Emacs. Many things are attracting people.

What is more important is free software promotion. Users should be
coming in general to GNU/Linux as it is free software. Emacs is free
software editor. That is how I have discovered it, by first finding
free software operating system, then finding Emacs and other editors
and software altogether.

Re: New year - Out with the old!

[novim]

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Wednesday, December 23, 2020 4:58 PM, Jean Louis <bugs@gnu.support> wrote:
>
> > As for the keybindings emacs should have selectable keyboard configs
> > similar to other software, just like other tools have vim, emacs,
> > etc. key emulations. These configs should be selectable right from
> > the startup screen for a new user, so he can switch to familiar keys
> > with a simple click. Familiar keys include copy/paste and other keys
> > standard on the platform (e.g. on windows C-c, C-v).
>
> Emacs already has CUA keys selectable in the Options menu, from start
> and all the time.

I was not clear enough, because I don't limit these to the copy/paste keys
only.

Popular editors usually have keyboard maps of other popular editors
(like Sublime, Atom, Visual Studio, etc.) so new users coming from them
can use the editor right away with their favorite bindings.

For example, from the VSCode docs:

Keymap extensions

Keyboard shortcuts are vital to productivity and changing keyboarding habits can be tough. To help with this, File > Preferences > Keymap Extensions shows you a list of popular keymap extensions. These extensions modify the VS Code shortcuts to match those of other editors so you don't need to learn new keyboard shortcuts. There is also a Keymaps category of extensions in the Marketplace.

https://code.visualstudio.com/docs/getstarted/keybindings#_keymap-extensions


The same thing could help new Emacs users. When a new user starts Emacs then he could select
a preferred keymap if he wants to, so he didn't have to start learning new keys,
he could use the keys he's familiar with right away for opening files, saving, switching
files, etc.

And the displayed info docs could immediately reflect this, so, for example, a user coming
from VSCode could use C-s for saving file, C-f for for searching in the current file, etc
and the info docs would show:

‘C-s’
     Save the current buffer to its file (‘save-buffer’).

Re: New year - Out with the old!

[Jean Louis]

* novim <laszlomail@protonmail.com> [2020-12-23 19:20]:
> 
> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
> On Wednesday, December 23, 2020 4:58 PM, Jean Louis <bugs@gnu.support> wrote:
> >
> > > As for the keybindings emacs should have selectable keyboard configs
> > > similar to other software, just like other tools have vim, emacs,
> > > etc. key emulations. These configs should be selectable right from
> > > the startup screen for a new user, so he can switch to familiar keys
> > > with a simple click. Familiar keys include copy/paste and other keys
> > > standard on the platform (e.g. on windows C-c, C-v).
> >
> > Emacs already has CUA keys selectable in the Options menu, from start
> > and all the time.
> 
> I was not clear enough, because I don't limit these to the copy/paste keys
> only.
> 
> Popular editors usually have keyboard maps of other popular editors
> (like Sublime, Atom, Visual Studio, etc.) so new users coming from them
> can use the editor right away with their favorite bindings.

I see that as various possible emulations. Emacs has thousands of
packages so I guess if any of Emacs package developers would miss
those emulations they would already make it just as the emulation of
vi editor exist. In my opinion those modes are not so hard to make.

> For example, from the VSCode docs:
> 
> Keymap extensions
> 
> Keyboard shortcuts are vital to productivity and changing keyboarding habits can be tough. To help with this, File > Preferences > Keymap Extensions shows you a list of popular keymap extensions. These extensions modify the VS Code shortcuts to match those of other editors so you don't need to learn new keyboard shortcuts. There is also a Keymaps category of extensions in the Marketplace.
> 
> https://code.visualstudio.com/docs/getstarted/keybindings#_keymap-extensions

I understand that propaganda for editor may go over various emulations
of other editors. That is useful and handy. Well, somebody could make
those emulations.

> The same thing could help new Emacs users. When a new user starts
> Emacs then he could select a preferred keymap if he wants to, so he
> didn't have to start learning new keys, he could use the keys he's
> familiar with right away for opening files, saving, switching files,
> etc.

Definitely, I agree fully on that. The more emulations there would be
by default easier would be for some users to start using the editor.

When Emacs get stuck in someting, then I switch to using Zile editor
which is pretty much like Emacs by all its basic keybindings. Then I
may use `mg' editor which is simpler and has Emacs bindings. The
reason I use those is that I am used to Emacs keybindings, but they do
not become my first priority, rather just replacement. Then also `e3'
which has all common keybindings such as for vi `e3vi', pico like
`e3pi', then nedit, vi, wordstar. I may sponsor some of your mentioned
emulations little later in time. That would make Emacs easier usable
by other people.

Re: New year - Out with the old!

[Barry Fishman]

On 2020-12-23 08:38:22 GMT, novim wrote:
>> Everyone new to Emacs finds how different it is. That can be like
>> discovering a new continent. For some, it's marvelous. For others,
>> it's just primitive, populated by savages, a decaying culture, or is
>> just no longer relevant. For some, it's a mix.
>
> The question is: what is the goal?
>
> Is it getting Emacs used by more people, so the ideals of free
> software reaches more people? Or is it keeping it as is, a niche tool
> which free software enthusiast can discover if they want to?

If you ask a loaded question, you get a useless answer.

This seems to be the usual commercial goal of giving people fewer
choices, and have every product go after the same (biggest) market,
Maybe each with a slight twist that makes it seem better but
fundamentally is the same as all the others.

Is the goal to get more people to use Emacs and then somehow (?) have
that lead to them using more Free Software, or is it getting people to
prefer and use Free Software because it is better being designed for and
by its users, and not distorted by corporate attempts at controlling,
and extracting as much money as possible from its users?

There are plenty of editors out there, free and non-free; but in many
ways Emacs doesn't have alternatives.  Getting Emacs to work like other
editors is not providing something that does not already exist, but in a
real way taking something away from people who don't have other good
choices.  I know I have spent a long time with a section of my Emacs
init file labeled, "Forward into the Past", undoing new changes to
the Emacs API that made things harder for me.  But as a *user* developed
application, usually, these changes were eventually rolled back or
modified to be less of a problem.

If one really wants to provide freedom for computer users, one should
not try to build the ultimate application that the plurality of people
like, but build a set of applications that meet the needs of each user
community.  Maybe sharing code (where free software has a clear
advantage over the proprietary alternatives).

<Off topic>

Personally, I would think it would be useful to fork Emacs (probably
starting with Gemacs) and produce an editor that integrates well with
Guile-3, Gtk-4, and the Gnome desktop.  It would focus on the common set
of goals: produce applications that anyone can use, and share a common
user interface, GTK, (and hopefully programming interface, Guile), at the
expense (initially) of a good chunk of Emacs functionality.

It could then act as a framework to port useful Emacs functionality to a
new Guile based editor in a way that worked well in a Graphical
environment and even show up in other applications.  We could learn more
about whether this made sense.

We would have more reason to invest in things like a Jit compiler that
was automatically available to a wide set of applications.  The Guile
framework is already looking at supporting JavaScript which is used by
much of Gnome.

The main Emacs project could then focus on programmers who want to
preserve the (programmable) easily modified editor/framework it has
always been, which will run in a wide variety of situations.  And then
its loyal users would not be antagonized to make it into a less familiar
(ever more complex) editor chasing the Microsoft's or Apple's attempts
at keeping their software (proprietorially) distinct, but familiar to
the plurality of users.

--
Barry Fishman

Re: New year - Out with the old!

[novim]

[-- Attachment #1: Type: text/plain, Size: 568 bytes --]

> Getting Emacs to work like other editors is not providing
> something that does not already exist, but in a real way taking
> something away from people who don't have other good choices.

Apparently, you misunderstood my suggestion. I didn't talk about
changing emacs defaults.

I suggested creating keybinding emulations which a new user can
turn on the first time he starts emacs, so he doesn't have to
start learning new keybindings right away when trying out emacs.

So it's an optional thing which can make emacs less alien for
users coming from other editors.

[-- Attachment #2: Type: text/html, Size: 790 bytes --]

Re: New year - Out with the old!

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > If you ask a loaded question, you get a useless answer.

I love that!  I will try to remember to use it sometimes.

  >   I know I have spent a long time with a section of my Emacs
  > init file labeled, "Forward into the Past", undoing new changes to
  > the Emacs API that made things harder for me.

Would you like to tell me what they were?  Perhaps off the list, as it would
be a tangent.

-- 
Dr Richard Stallman
Chief GNUisance of the GNU Project (https://gnu.org)
Founder, Free Software Foundation (https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)