Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[John Yates]

On Tue, Mar 6, 2018 at 3:51 PM, Richard Stallman <rms@gnu.org> wrote:
> Once we can do proper formatting of paragraphs
> with variable-width text, with a few kinds of alignment, and we can
> save these in files, we will be able to use it for writing letters and
> handouts, instead of LibreOffice.

I agree that filling with proportional fonts is a crucial piece of
building a WYSIWYG editor.  But I suspect that the number of folk for
whom that is the threshold impediment to using Emacs for "writing
letters and handouts" is exceedingly small.  That user experience
would be akin to the earliest versions of MS Word or WordPerfect: all
styling applied manually.

Today Emacs is able to understand the structure of the text in a
buffer based on its mode (either because the contents is created and
maintained entirely by elisp code or because the mode knows how to
parse the buffer's contents).  Given such understanding the mode can
then do an intelligent job refilling, reformatting and rendering.

Until Emacs has some way of modeling the structure of a text document
I doubt that we will see any significant uptake in its use a
LibreOffice replacement.

/john

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[Eli Zaretskii]

> From: John Yates <john@yates-sheets.org>
> Date: Tue, 6 Mar 2018 16:45:56 -0500
> Cc: John Wiegley <johnw@gnu.org>, daniele@grinta.net,
> 	Emacs developers <emacs-devel@gnu.org>
> 
> Today Emacs is able to understand the structure of the text in a
> buffer based on its mode (either because the contents is created and
> maintained entirely by elisp code or because the mode knows how to
> parse the buffer's contents).  Given such understanding the mode can
> then do an intelligent job refilling, reformatting and rendering.
> 
> Until Emacs has some way of modeling the structure of a text document
> I doubt that we will see any significant uptake in its use a
> LibreOffice replacement.

I think you have in mind much more sophisticated word processor than
it would take to make significant progress in this area relative to
what we have now.  Extending enriched.el to do a few things it
currently can't will already be a huge step forward, in that it will
allow users who want that to produce reasonably pretty documents in
Emacs, send enriched/text emails, etc.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[John Yates]

On Wed, Mar 7, 2018 at 12:18 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> From: John Yates <john@yates-sheets.org>
>>
>> Until Emacs has some way of modeling the structure of a text document
>> I doubt that we will see any significant uptake in its use a
>> LibreOffice replacement.
>
> I think you have in mind much more sophisticated word processor than
> it would take to make significant progress in this area relative to
> what we have now.  Extending enriched.el to do a few things it
> currently can't will already be a huge step forward, in that it will
> allow users who want that to produce reasonably pretty documents in
> Emacs, send enriched/text emails, etc.

I am quite curious what "few things" you have in mind.

I have no doubt that with filling in the presence of variable width fonts
and with some extensions to enriched.el it will be possible to compose
documents which, when viewed by a third party, look workman-like.

My skepticism is around the experience of authoring those  documents.
Until we have something akin to  paragraph styles and templates that
experience is going to feel primitive and tedious.  Hence my contention
that if we are going after the WYSIWYG use case then we need more
or uptake simply will not occur.

(My understanding is that Richard's vision includes WYSIWYG editing,
though I could be wrong.)

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[Eli Zaretskii]

> From: John Yates <john@yates-sheets.org>
> Date: Wed, 7 Mar 2018 18:29:45 -0500
> Cc: Richard Stallman <rms@gnu.org>, John Wiegley <johnw@gnu.org>, daniele@grinta.net, 
> 	Emacs developers <emacs-devel@gnu.org>
> 
> My skepticism is around the experience of authoring those  documents.
> Until we have something akin to  paragraph styles and templates that
> experience is going to feel primitive and tedious.  Hence my contention
> that if we are going after the WYSIWYG use case then we need more
> or uptake simply will not occur.

We have very different experiences with styles and templates, it
seems.  I hate them and rarely if ever use them in the documents I
need to write on my daytime job (and I write quite a few of them).
IME, styles in Office like applications get in the way too much, in
that you change something at some place, and suddenly many unrelated
places change as well.

So I prefer applying styles by hand, locally wherever I need that.  It
works wonders on my productivity.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[Yuri Khan]

On Thu, Mar 8, 2018 at 8:36 PM, Eli Zaretskii <eliz@gnu.org> wrote:

> We have very different experiences with styles and templates, it
> seems.  I hate them and rarely if ever use them in the documents I
> need to write on my daytime job (and I write quite a few of them).
> IME, styles in Office like applications get in the way too much, in
> that you change something at some place, and suddenly many unrelated
> places change as well.

That is a consequence of two possible reasons:

* Office-like applications do not make you very aware of styles. To
see what style is applied, you have to position the point and look at
the style selection box. That is in contrast with markup-based editing
workflows, where you have opening and closing tags in your face all
the time.

* Your lexicon of styles is poorly designed, that you use one style
where you need two or more.

> So I prefer applying styles by hand, locally wherever I need that.  It
> works wonders on my productivity.

How big are your documents, how long-lived are they, and are they part
of a bigger collection that needs to be consistently formatted?

I might forgo styles in case of a self-sufficient document under two
pages that I need to print out once and not even bother to save to a
file.

Otherwise I’d have the opposite problem of having to do the same
change in too many places and never being sure I got them all.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[John Yates]

On Thu, Mar 8, 2018 at 8:36 AM, Eli Zaretskii <eliz@gnu.org> wrote:
> We have very different experiences with styles and templates, it
> seems.  I hate them and rarely if ever use them in the documents I
> need to write on my daytime job (and I write quite a few of them).
> IME, styles in Office like applications get in the way too much, in
> that you change something at some place, and suddenly many unrelated
> places change as well.

I agree that styles as exposed in MSOffice and vaguely copied in
LibreWord are not great.  Quite the contrary.  And I am sure that
many users of today's crop of word processor would concur.

I was not arguing that Emacs should mimic that model.  What I am
arguing is that those style models attempt to answer an actual need.
That they mostly do a poor job is actually an opportunity.  Failure
to acknowledge that need and to provide any answer will limit severely
adoption of Emacs as a WYSIWYG document preparation tool. OTOH, were
Emacs to offer a more attractive style model and user experience
that could become an important selling point.

My preference would be the model used in Interleaf TPS and Samna
Ami Pro (later Lotus Word Pro).  In both products the structure tree
of the paragraph styles was displayed in a narrow left-hand pane
with each node aligned to the first line in the right-hand pane of the
paragraph it described.  Clearly the two panes scrolled in lockstep.

I cannot tell you how many times I demoed Ami Pro to a user of
MSWord or WordPerfect and made an instant convert.

It has been too many decades to recall much more.  For instance I do
not recall either product offering style inheritance (styles based
on other styles).  Clearly such inheritance has power but is also the
source of much user frustration.  A style tree browser clarifying the
source of properties (inherited versus locally defined) might help.
An ability to restrict the display to a single style property (font,
indentation, etc) could further help.

My bottom line: The style models in MSWord and LibreWord represent
years of incremental tweaking to ill-conceived initial  models.  Emacs
has an opportunity to start afresh.  If it really is going to pursue
WYSIWYG document preparation it should not squander that opportunity.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[Eli Zaretskii]

> From: Yuri Khan <yuri.v.khan@gmail.com>
> Date: Thu, 8 Mar 2018 21:35:49 +0700
> Cc: John Yates <john@yates-sheets.org>, John Wiegley <johnw@gnu.org>, 
> 	Daniele Nicolodi <daniele@grinta.net>, rms@gnu.org, Emacs developers <emacs-devel@gnu.org>
> 
> * Office-like applications do not make you very aware of styles.

Trust me, I'm aware of my styles.

> * Your lexicon of styles is poorly designed, that you use one style
> where you need two or more.

That could be, because Someone Else (TM) designed them.  But I don't
think the number of styles is the problem, more likely their
relations.

> > So I prefer applying styles by hand, locally wherever I need that.  It
> > works wonders on my productivity.
> 
> How big are your documents, how long-lived are they, and are they part
> of a bigger collection that needs to be consistently formatted?

Quite large (up to hundreds of pages), quite long-lived (years), and
usually maintained all the time.

> Otherwise I’d have the opposite problem of having to do the same
> change in too many places and never being sure I got them all.

I never need to change the style, because I consistently use the same
styles in all the documents.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[Eli Zaretskii]

> From: John Yates <john@yates-sheets.org>
> Date: Thu, 8 Mar 2018 09:54:43 -0500
> Cc: Richard Stallman <rms@gnu.org>, John Wiegley <johnw@gnu.org>, daniele@grinta.net, 
> 	Emacs developers <emacs-devel@gnu.org>
> 
> I agree that styles as exposed in MSOffice and vaguely copied in
> LibreWord are not great.  Quite the contrary.  And I am sure that
> many users of today's crop of word processor would concur.
> 
> I was not arguing that Emacs should mimic that model.  What I am
> arguing is that those style models attempt to answer an actual need.

And my point was that I can manage very well without using any styles.
The reasons why I started doing that are mostly irrelevant to this
discussion.  What's important is that according to my experience, one
can write reasonably well formatted documents without ever touching
any styles.  Which means that Emacs with advanced text-formatting
features can be very useful even if it doesn't support style sheets of
any kind.  (Of course having styles would be even better.)

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[John Yates]

On Thu, Mar 8, 2018 at 10:09 AM, Eli Zaretskii <eliz@gnu.org> wrote:
> What's important is that according to my experience, one
> can write reasonably well formatted documents without ever touching
> any styles.  Which means that Emacs with advanced text-formatting
> features can be very useful even if it doesn't support style sheets of
> any kind.

In a sense that goes to the heart of the undertaking.  Does it hope
merely to make it possible for users who have already committed to
Emacs to carry out an incrementally larger portion of their authoring
activities within that familiar environment?  Or does it hope to attract
users for whom WYSIWYG editing represents a more dominant portion
of their authoring activities?

The former is merely some subset of the Emacs community scratching
an itch.  The latter would represent a strategic initiative.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[Eli Zaretskii]

> From: John Yates <john@yates-sheets.org>
> Date: Thu, 8 Mar 2018 11:02:38 -0500
> Cc: John Wiegley <johnw@gnu.org>, daniele@grinta.net, Richard Stallman <rms@gnu.org>, 
> 	Emacs developers <emacs-devel@gnu.org>
> 
> In a sense that goes to the heart of the undertaking.  Does it hope
> merely to make it possible for users who have already committed to
> Emacs to carry out an incrementally larger portion of their authoring
> activities within that familiar environment?  Or does it hope to attract
> users for whom WYSIWYG editing represents a more dominant portion
> of their authoring activities?
> 
> The former is merely some subset of the Emacs community scratching
> an itch.  The latter would represent a strategic initiative.

Time will tell.  For now, neither seems to be lifting off the ground.
IME, nothing serious in Emacs gets done without a motivated and
capable individual who just starts doing that.  And it doesn't look
like we have one for this job yet.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[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. ]]]

  > My bottom line: The style models in MSWord and LibreWord represent
  > years of incremental tweaking to ill-conceived initial  models.  Emacs
  > has an opportunity to start afresh.  If it really is going to pursue
  > WYSIWYG document preparation it should not squander that opportunity.

I use Libre Office much in the way of themes or styles.  It's fine for
the simple things I want to do.  I can't be the only one who uses it
in a basic way.  That is why I am sure that WYSIWYG word processing in
Emacs, if implemented at the same low level of sophistiation,
will be useful.

That doesn't mean I'm against those customization features.  Quite the
contrary: given a WYSIWYG facility, I'm sure Emacs developers will
work on various kinds of style features and make them good.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[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. ]]]

  > In a sense that goes to the heart of the undertaking.  Does it hope
  > merely to make it possible for users who have already committed to
  > Emacs to carry out an incrementally larger portion of their authoring
  > activities within that familiar environment?  Or does it hope to attract
  > users for whom WYSIWYG editing represents a more dominant portion
  > of their authoring activities?

I hope it will eventually do the latter.  But since our changes are
implemented one feature at a time, I think it is best to start with
the former.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.

Re: Towards a WYSIWYG word processing (was: What improvements would be truly useful?)

[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. ]]]

I wrote

  > I use Libre Office much in the way of themes or styles.

but that was an editing error.  I meant to write

  > I use Libre Office without much in the way of themes or styles.

-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)
Skype: No way! See https://stallman.org/skype.html.