Re: [Texmacs-dev] Keybindings(2)

[Álvaro Tejero Cantero]

On Mon, 2002-08-12 at 21:05, Joris van der Hoeven wrote:
> 
> > The logical/physical markup confusion was a silly mistake of mine. I see
> > now the rationale, but I'd prefer the logical markup access keys to be
> > more accesible than the physical ones. I think all logical markup should
> > be fit into A-* keys, like the sectioning commands. Anyway, F5/F6 are
> > really far from my fingers for everyday use.
> 
> But that is already the case: you also have A-s, A-m, etc.
> for strong, emphasize, etc.

Fine!.

> My only regret with the new conventions is that the Greek letters
> are harder to obtain with F5. How do you feel about that?

Bad. I have to read the keybinding files more thoroughly to propose any
alternative... It's really far. I'll think over it.

> On the other hand, we do have fast keystrokes for bold,
> caligraphic and fraktur using F6, F7 and F8.
> These are not yet available via A-* keys, but that *is* a plan.

I'm not sure I understand... would you _mix_ within the same modifier (A
for example) the structural formatting *and* the visual formatting, with
different keys?. I don't think it's a good idea to share the
'namespace'. 

For me physical formatting should be about as accesible as the preamble
commands, because they should be used only in style files (at least I
see no use of them out of that). I would give them a second-level
keybinding (that is modifier -- physical selector -- bold). All I'm
saying is with regard to the text mode, I am not sure about math, since
it's so much more difficult to do hard-typing of mathematical
expressions and since math is sometimes as much a drawing as a text.

My concern with this is that new texmacs users should use the editor in
a structured manner. Other than that, if there are keys free and easily
bindable to user-defined macros, the biggest problem is educational,
since if one wants each first appearance of a technical term to appear
in sans-serif one should do a \fa macro and assign, say, A-f to it.

What did you decide about evaluating the \ commands with space? this
would encourage the use of macros, because it's a little faster to type.
The problem arises only if whitespace is allowed in macro names.

> > Regarding the math issue: I would like to have keybindings for
> > \equation, \equation*, \eqnarray* and the forthcoming \eqnarray*. 
> 
> You may use "\ [ return" for \equation*.
> You are right that we should have bindings for the others.
> Which bindings would you suggest?

Still not sure:: A-[, A-], A-$... BTW is there any reason for choosing
\equation over an one-line \eqnarray? The spacing seems more sensible in
the second... Would it be possible(interesting to make \equation an
special case of \eqnarray?


> > Perhaps this could be done by establishing one cycle key. This cycle key
> > would cycle amongst table types and figure types as well. It's a "kind
> > cycle key". Others could be thought of, for example the "style cycle
> > key" for cycling amongst numbering schemes for enumerations, bullet
> > characters for bulleted lists, citation styles and more (perhaps
> > numbered/unnumbered equations would fit here). And there should be a
> > "hierarchical cycle key" to cycle between depth levels in section
> > titles, depth levels in enumerations, etc. etc. The point is to keep
> > cycles short, at most with 4/5 possible states, else back/forth cycle
> > keys would be necessary (or returning to the original state would be
> > slow). This is why I suggest to search for orthogonal cycles
> > (kind/style/depth, for example). 
> 
> I agree that we need cycling, but I did not yet have time
> to implement this :^(
> 
> > Anyway, these are my dreams and I understand this poses a big problem
> > for the typesetter, because heavy reprocessing would be needed to show
> > the results while cycling. 
> 
> No, that is OK, since the enclosed structures (an equation or the text
> of a theorem) are usually small. Possible exceptions are big itemize
> or enumerate lists, but one or two pages are retypeset reasonably fast
> too anyway (about 0.25 sec on a 600MHz machine).

Great, what about the renumbering of sections/figures/equations of the
document?. You can delay it I guess, till the affected part is visible.

> > It would be nice to have TeXmacs typeset it's own keybindigns with some 
> > macros
> > and thus form a "reference card". I don't know how to do this.
> 
> That is also a project; it presupposes some other stuff related
> to buffer handling.
> 
> > I'm still discovering the new keybinding set, so give me some time...
> 
> Yep.
> 
> > have fun,
> 
> That aplies to *you*

:). á.

 
-- 
álvaro.tejero.cantero
alqua.com, la red en estudio
"La perfection est atteinte non quand il ne reste rien à
ajouter, mais quand il ne reste rien à enlever"
Saint-Exupéry.