Ann: TeXCAD 4.1

Ann: TeXCAD 4.1

[Gautier Write-only]

TeXCAD is a program for drawing or retouching {picture}s in LaTeX.
More infos @ http://www.mysunrise.ch/users/gdm/texcad.htm

I'm still looking for people interested in developing the
GUI part for other than "native" Windows:
 - GtkAda
 - "native" Mac OS X
 - ?

A large part of TC is OS & GUI-independent (all TC.* packages),
so it is "only" a question of windows, menus, mouse etc.
________________________________________________________
Gautier  --  http://www.mysunrise.ch/users/gdm/gsoft.htm

NB: For a direct answer, e-mail address on the Web site!

Re: Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-16, Gautier Write-only <gautier@fakeaddress.nil> wrote:
> TeXCAD is a program for drawing or retouching {picture}s in LaTeX.
> More infos @ http://www.mysunrise.ch/users/gdm/texcad.htm

Great app! Keep up the good work!

> I'm still looking for people interested in developing the
> GUI part for other than "native" Windows:
>  - GtkAda
>  - "native" Mac OS X

Why native Mac OS X if one has GtkAda? I thought GtkAda compiled and
worked on Windows and Mac OS X. Anybody here that can verify this?


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: 
: Why native Mac OS X if one has GtkAda? I thought GtkAda compiled and
: worked on Windows and Mac OS X. Anybody here that can verify this?

Why not improve GtkAda so it arrives at least at the level of usability
that OS X offers? :-)



-- Georg

Re: Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-17, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> Why not improve GtkAda so it arrives at least at the level of usability
> that OS X offers? :-)

You mean remove possibility to right-click?

I don't know how usable Mac OS X is or not and frankly I don't care. Usability
is relative to the user. My point is why make three versions if you can
make and maintain one.


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: On 2004-03-17, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
:> Why not improve GtkAda so it arrives at least at the level of usability
:> that OS X offers? :-)
: 
: You mean remove possibility to right-click?

Not true.
 
: I don't know how usable Mac OS X is or not and frankly I don't care. Usability
: is relative to the user.

Well, I think the user is the all important person if you want to sell
him/her your program. If you sell to a company there will still be
people who have to provide input to the program, and react to the
visual reactions of the program. What if they complain?

: My point is why make three versions if you can
: make and maintain one.

Usually you can sell all three version and the feedback might
be that the respective version fits better with the rest of the
operating system than a Gtk interface (compare the GIMP's file
open dialog to the native; does d&d work outside GNOME...)


-- Georg

Re: Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-17, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
>  
>: I don't know how usable Mac OS X is or not and frankly I don't care.
>Usability : is relative to the user.
>
> Well, I think the user is the all important person if you want to sell
> him/her your program. If you sell to a company there will still be
> people who have to provide input to the program, and react to the
> visual reactions of the program. What if they complain?

Who is selling anything here? Anyway from screenshots I cannot say I can
see what is so much better with the Mac OS X toolkit. No, I'm not
talking about the other setup things.

> Usually you can sell all three version and the feedback might be that
> the respective version fits better with the rest of the operating
> system than a Gtk interface (compare the GIMP's file open dialog to
> the native; does d&d work outside GNOME...)

Well take a look at Gtk 2.4 that was released today.

But I guess you would like to do a native Mac OS X version?

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: On 2004-03-17, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
: 
: Who is selling anything here?

I (also) meant that as a metaphor.

: Anyway from screenshots I cannot say I can
: see what is so much better with the Mac OS X toolkit.

That misses the "feel" compopnent of L&F. I _use_ a computer,
I do not just look at it :-)

: Well take a look at Gtk 2.4 that was released today.

Apart from a new file open dialog window (at last!) the new features
include some technicallities like support for the fact that not everyone
writes from left to right, and many graphical niceties.

However, as can be learned from advanced UIs before OS/2 and even
from OS/2, file open dialogue windows are a great leap, uhm, backwards.
Though to appreciate this one might need the help of a different approach
to UI design, like applications truly integrated with the OS, including
its GUI, literally. Like a document centered approach, or like the integrating
approach of Plan 9 (the *user* defines the user interface using what the
OS provides for this.)

: But I guess you would like to do a native Mac OS X version?

There is GNUStep, as has been mentioned. Mac OS X (relative of NeXTStep)
can get in your way too, why else have they invented Exposee? But as long
as GNU/Linux interface programmers (who by profession are not designers,
or perception psychologist, or ...)
are trying to catch up
with Windows(TM) there cannot be improvements in L&F, i.e. usability, as
they do not seem to see the great number of ideas absorbed by MS that
the producers of windows are carefully hiding behind the curtains.

One absurdity that is even present in KDE and that not even Qt could
prevent is deeply nested menues as a replacement for a whole usable
screen.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote on 18/03/2004 (12:05) :
> However, as can be learned from advanced UIs before OS/2 and even
> from OS/2, file open dialogue windows are a great leap, uhm, backwards.
> Though to appreciate this one might need the help of a different approach
> to UI design, like applications truly integrated with the OS, including
> its GUI, literally. Like a document centered approach, or like the integrating
> approach of Plan 9 (the *user* defines the user interface using what the
> OS provides for this.)

Now you are not talking any more about GtkAda versus the Mac OS X
toolkit so this is completely irrelevant.

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+abuse@pvv.org> wrote:
: Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote on 18/03/2004 (12:05) :
:> However, as can be learned from advanced UIs before OS/2 and even
:> from OS/2, file open dialogue windows are a great leap, uhm, backwards.
:> Though to appreciate this one might need the help of a different approach
:> to UI design, like applications truly integrated with the OS, including
:> its GUI, literally. Like a document centered approach, or like the integrating
:> approach of Plan 9 (the *user* defines the user interface using what the
:> OS provides for this.)
: 
: Now you are not talking any more about GtkAda versus the Mac OS X
: toolkit so this is completely irrelevant.

(Mac OS X is not mentioned in the paragraph above, but I admit that
I'm always carried away by human interface design issues. I think they
are important, and as long as programmers are empowered to make user
interface design decisions, improvements aren't likely.)

Well, irrelevant..., the OP had mentioned or implied, to me at least,
that low level layers (below Gtk) as well as the high level layers 
invite criticism that can be well expressed in typical "Ada" terms.
(Which is what I thought reading pages from the XCB site.)
Thinking about a new interface it is an opportunity to think differently
about interfaces, taking human interface research into consideration.
Interestingly this maps to programming language syntax design decisions,
as presented in a collection of articles entitled "The Psychology of
Programming" (I don't have it here, but I can provide details when
I'm at home.)

(For example, is a dashboard oriented panel design, as is more or
less hardcoded in (the spirit of) Gtk(Ada) really an adequate choice
when in comes to applications that have nothing to do with engineering,
or with engineers?)


In addition, the new Gtk has adopted an XML based mechanism to
describe certain interface components. This is another level of 
abstraction. Dr. Leif might be pleased to hear about this ;-)

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-18, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
>
> Well, irrelevant..., the OP had mentioned or implied, to me at least,
> that low level layers (below Gtk) as well as the high level layers 
> invite criticism that can be well expressed in typical "Ada" terms.

What do you mean? I really don't understand what you are saying.

You said: That one should bring GtkAda up to the level of Mac OS X.

Then you go on comparing apples and oranges.

> Thinking about a new interface it is an opportunity to think differently
> about interfaces, taking human interface research into consideration.
> Interestingly this maps to programming language syntax design decisions,
> as presented in a collection of articles entitled "The Psychology of
> Programming" (I don't have it here, but I can provide details when
> I'm at home.)

Please I don't want yet another debate about Human interface. Rather
read : http://developer.gnome.org/projects/gup/hig/

and give me a link to apples whitepaper on this.

> (For example, is a dashboard oriented panel design, as is more or
> less hardcoded in (the spirit of) Gtk(Ada) really an adequate choice
> when in comes to applications that have nothing to do with engineering,
> or with engineers?)

What do you mean?

> In addition, the new Gtk has adopted an XML based mechanism to
> describe certain interface components. This is another level of 
> abstraction. Dr. Leif might be pleased to hear about this ;-)

Are you talking about libglade?


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-18, Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
> Please I don't want yet another debate about Human interface. Rather
> read : http://developer.gnome.org/projects/gup/hig/
>
> and give me a link to apples whitepaper on this.

OK found it myself.

You might also want to look at www.xfce.org


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: On 2004-03-18, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
:>
:> Well, irrelevant..., the OP had mentioned or implied, to me at least,
:> that low level layers (below Gtk) as well as the high level layers 
:> invite criticism that can be well expressed in typical "Ada" terms.
: 
: What do you mean? I really don't understand what you are saying.

For example, as the GNOME guidelines say, a user using a computer
with a graphical user interface should see a consistent (see "Ada
culture", see also Ada RM :-) user interface accross applications,
whether these applications appear as one application (document
centered approach) or not.
But GNOME /= Windows, thus a GNOME style application and a Windows
style application both running on Windows might have the very same
Ada parts, but they must fail to have the same user interface parts.

This is why I said it might be worthwhile writing applications
that specifically use an operating system's features.

 I know many who prefer PhotoShop over GIMP not because
GIMP lacks functions they need, but because it doesn't fit as well
with the operating system and with their interface expectations.

Dress code violation, so to speak.


: You said: That one should bring GtkAda up to the level of Mac OS X.

That one had a smiley :-)

: Then you go on comparing apples and oranges.

O.K., maybe what I said is confusing or not explicit enough. I'll
try.

: Please I don't want yet another debate about Human interface. Rather
: read : http://developer.gnome.org/projects/gup/hig/

This is much more a "How to Use Gtk/GNOME Elements".
It argues at a very different level, namely it shows how not to
misuse the GNOME interface elements, not at the level of
  "Why We Have Chosen The Controls That You Find in Gtk
But Possibly Nowhere Else".
When things important for "human interface" considerations come up,
they refer you to unnamed books,
"There are a large number of professional interaction designers who
 write books and teach courses on design methods that can help with
 this process, many of which are extremely useful."

 And the guide is contradictory in many ways. For example,

  "Put the User in Control"
versus
  "what you want to enable them [users] to do"

("learn what your users want" is not present
in the same bulleted list, but only elsewhere.)

 
:> (For example, is a dashboard oriented panel design, as is more or
:> less hardcoded in (the spirit of) Gtk(Ada) really an adequate choice
:> when in comes to applications that have nothing to do with engineering,
:> or with engineers?)
: 
: What do you mean?

As a brute force example, we have switches and push buttons,
but no muffineers or graters.
 
: Are you talking about libglade?

"GtkUIManager allows the application to specify a set of
 "actions", then create menus and toolbars using those actions
 from an XML-like interface description."

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-18, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
>  I know many who prefer PhotoShop over GIMP not because
> GIMP lacks functions they need, but because it doesn't fit as well
> with the operating system and with their interface expectations.

Sure as long as they don't have to pay for PhotoShop or need the extra
functionality.

> but no muffineers or graters.

What are muffineers and graters?

I guess you dont mean:

   http://www.spencermarks.com/html/e269.html


>: Are you talking about libglade?
>
> "GtkUIManager allows the application to specify a set of
>  "actions", then create menus and toolbars using those actions
>  from an XML-like interface description."

Ok something similar to libglade


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Randy Brukardt]

"Georg Bauhaus" <sb463ba@l1-hrz.uni-duisburg.de> wrote in message
news:c3cl75$9mf$1@a1-hrz.uni-duisburg.de...
> For example, as the GNOME guidelines say, a user using a computer
> with a graphical user interface should see a consistent (see "Ada
> culture", see also Ada RM :-) user interface accross applications,
> whether these applications appear as one application (document
> centered approach) or not.
> But GNOME /= Windows, thus a GNOME style application and a Windows
> style application both running on Windows might have the very same
> Ada parts, but they must fail to have the same user interface parts.
>
> This is why I said it might be worthwhile writing applications
> that specifically use an operating system's features.

Right. That's why we built Claw as Windows-specific; that's less interesting
than a GUI that works everywhere, but it makes it possible to build an
application that looks and feels like other Windows applications. (I only
wish more Windows programs actually did that!)

                     Randy.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: 
: Sure as long as they don't have to pay for PhotoShop or need the extra
: functionality.

No, actually some people pay for software.

:> but no muffineers or graters.
: 
: What are muffineers and graters?

See? :-)  muffineers and graters are domain specific tools rarely met
as UI controls though they might be useful. (Their absence is puzzling me
as many known programmers have not only recommended cooking for learing
about computers, but have also published recipes, with tool descriptions.)

: I guess you dont mean:
: 
:   http://www.spencermarks.com/html/e269.html

Yes, I do mean this tool. I'd venture a guess that not many engineers
are familiar with tools of this sort although it is in frequent use.
I can imagine many more or less metaphorical uses of a muffineer control
in GUIs.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-19, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
>: 
>: Sure as long as they don't have to pay for PhotoShop or need the extra
>: functionality.
>
> No, actually some people pay for software.

If you pay 500-600 euro for a software that does not give you any
additional features but only a slightly different GUI, then either 1)
you are not paying or 2) you have too much money or 3) you extemely
picky about unimportant details.
>
> Yes, I do mean this tool. I'd venture a guess that not many engineers
> are familiar with tools of this sort although it is in frequent use.
> I can imagine many more or less metaphorical uses of a muffineer control
> in GUIs.

In the beginning of the sentence you were talking about spesific widgets
so I thought these where some special widgets the Mac OS had.

Anyway I'll stop this discussion for my part and rather do some coding
in Ada :-)

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
:> No, actually some people pay for software.
: 
: If you pay 500-600 euro for a software that does not give you any
: additional features

Have you looked at the software you are talking about?

: but only a slightly different GUI, then either 1)
: you are not paying or 2) you have too much money or 3) you extemely
(The essentials cost just about 80)
: picky about unimportant details.

(featuring normative ontology; slightly arrogant intended or not)
This brings us the software that is not respecting the rule,
"Know what the *users* of the system need/want."
It isn't the *programmer* who decides what is an important detail to a
graphics artist (in the case of a pictures editting program).

If these details aren't part of the requirements specification
what kind of users do you have in mind for your programs?

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-19, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
>:> No, actually some people pay for software.
>: 
>: If you pay 500-600 euro for a software that does not give you any
>: additional features
>
> Have you looked at the software you are talking about?

Please! Try to restain from changing the premissions every time.

You said: I know people who choose Photoshop over Gimp, *NOT* because of
the missing features, but because that it is more intergrated.

So I assume that the missing features is not a premise for buying
Photoshop. Of course if somebody *need* the missing features, the matter
is different.

As to if I have tried Photoshop, yes I have. However not on Mac OS X,
but on the older Mac OSes which I found dumbed-down and frustrating as I
find Windows. But I do not wish to say anything with respects to Mac OS
X (the Unix clone Mac had to change to as their old OS was beyond
repair) as I haven't tried it, I expect it to be better than Windows,
but I do not know. But please don't thing the users are a homogenous
group of people. User friendliness is not the same for everybody. Some
want clicky-micky-click-click interface (and developing problems with
the muscles in the hand) others want a UI where they can use the
keyboard etc...

> If these details aren't part of the requirements specification
> what kind of users do you have in mind for your programs?

Whatever

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Randy Brukardt]

"Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
message
news:slrnc5lp8q.o0.randhol+valid_for_reply_from_news@k-083152.nt.ntnu.no...
> On 2004-03-19, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> > Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
> >:
> >: Sure as long as they don't have to pay for PhotoShop or need the extra
> >: functionality.
> >
> > No, actually some people pay for software.
>
> If you pay 500-600 euro for a software that does not give you any
> additional features but only a slightly different GUI, then either 1)
> you are not paying or 2) you have too much money or 3) you extemely
> picky about unimportant details.

I think you forget the most important reason for using a "standard" tool:
the ability to get training/advice/instructions from your peers. And that
help from your peers usually is very specific to the user-interface of the
particular tool. Another tool that provides the same functionality, but a
different interface, is a lot harder to use -- and for a novice, might be
impossible.

To take a personal example: I've learned a lot about how to format cells,
specify label rows, draw lines, apply page formatting and the like in Excel.
Little of that knowledge would be usable on another spreadsheet program,
even though the other spreadsheet probably would have similar features and
may even be able to read/write the same files.

Another personal example: I'd like to get rid of Quicken. But other than
Microsoft Money (which would just be jumping from the boiling kettle into
the fire), I haven't been able to find anything with sufficient features and
a sane interface to be able to replace the Intuit junk. So it appears I'm
stuck (unless I want to write a replacement in Ada!)

                     Randy.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: On 2004-03-19, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
:> Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
:>:> No, actually some people pay for software.
:>: 
:>: If you pay 500-600 euro for a software that does not give you any
:>: additional features
:>
:> Have you looked at the software you are talking about?
: 
: Please! Try to restain from changing the premissions every time.
: 
: You said: I know people who choose Photoshop over Gimp, *NOT* because of
: the missing features, but because that it is more intergrated.

True. (And I didn't think of prices.)
As Randy has explained, interfaces are quite important when it
comes to a decision for using this or that program. Sometimes there is
also a prestige/status issue which is technically not relevant but
it drives decisions. There are peer groups of program users influencing
decision making etc etc etc. Not always the most desirable factors
in decision making, at least from some perspectives.

Ada Tutorials [was:] OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Randy Brukardt <randy@rrsoftware.com> wrote:
 

:  Another tool that provides the same functionality, but a
: different interface, is a lot harder to use -- and for a novice, might be
: impossible.

By a level of indirection this brings me to Ada tutorials.

(<<may_be_skipped>>The same tool with a different interface is
a lot harder to use not because it is more difficult to use, but because 
(a) it requires learning
(b) it requires that you forget the old ways.
If the interfaces are similar enough, they will introduce another level
of complexity, because you will have to switch between old memories and
fresh memories.)


Short Ada tutorials are (freely) available for different audiences.
I guess some if not all of them are made for programmers with
some experience. Is there one that is both short and suitable for
studying by people with less experience?

When I had an opportunity to introduce people to computers, or to
specific computer programs like text processing programs, it was always
refreshing to see that
(a) people like to learn how things work (e.g. context menues, para styles)
(b) people are capable of using operating system techniques 
   (e.g. folder integration)
But it is somewhat saddening to see that (a) and (b) don't become lasting
skills when there are only the equivalent of "left-mouse-button peers",
and when they are reluctant to RTFM because it *appears* bulky.
Absurd as it may be, when there is a tutorial part in the FM,
the thickness of FMs can hide the good short tutorial part.
It is a BIG mistake from both a pedagogical and economic point of view
not to draw attention to these tutorials because a good tutorial is what
gets you going.  It prevents frustration, it might lead to adoption.
(For example, people get regularly excited when they learn that there
is a mechanism in just about any text processing program that allows for
boxes of text being placed somewhere on the page without having to play
tricks using tables, or worse spaces and tabs... The tutorials introduce
this on one or two pages, conceptually, without listing all the steps
you need not understand in order to see and try out what a frame is.)

Is there a good short hands-on Ada beginners tutorial that does not
talk about Ada's history, safety, the new features, and this and that,
but starts in medias res of programming, just makes you want to write
programs using the nice-simple-powerful facilities of the language that
you have just seen?

So far I have only found Ada presentation slides by R. Dewar (C) 2004 
available from the NYU as "course ware", but I guess they lack spoken
words and exercises.


-- Georg

Neat (Pascal) and Scruffy (C) Languages [was:] OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
: Interestingly this maps to programming language syntax design decisions,
: as presented in a collection of articles entitled "The Psychology of
: Programming" (I don't have it here, but I can provide details when
: I'm at home.)

The article "The Nature of Programming" by T R G Green in the book
("Psychology of Programming", Hoc, Green, Samurc,ay, and Gilmore, Ed.s
it is not the one by Weinberg)
adds some interdisciplinary substance to the C/Ada arguments.
It covers syntax, style, culture, and more from a prespective of
cognitive psychology.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-19, Randy Brukardt <randy@rrsoftware.com> wrote:
> To take a personal example: I've learned a lot about how to format cells,
> specify label rows, draw lines, apply page formatting and the like in Excel.
> Little of that knowledge would be usable on another spreadsheet program,
> even though the other spreadsheet probably would have similar features and
> may even be able to read/write the same files.

Look at gnumeric.


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[chris]

Georg Bauhaus wrote:

> But GNOME /= Windows, thus a GNOME style application and a Windows
> style application both running on Windows might have the very same
> Ada parts, but they must fail to have the same user interface parts.
> 
> This is why I said it might be worthwhile writing applications
> that specifically use an operating system's features.

I'm developing an application now and hadn't considered using different 
UI toolkits on Windows and Linux targets.  It increases development 
time, but I suppose targetting windows with GWindows and Linux with 
GtkAda (or something else?) is a good idea.


>  I know many who prefer PhotoShop over GIMP not because
> GIMP lacks functions they need, but because it doesn't fit as well
> with the operating system and with their interface expectations.
> 
> Dress code violation, so to speak.

One of the reasons for this application was because people disliked the 
GIMP interface on Windows (I was going to mod Gtk+ to be better behaved 
on Windows).  There are other reasons so it's not the only one, but it 
certainly topped the list of reasons people chose not to use GIMP on 
Windows - an other was because it crashed too much.


> As a brute force example, we have switches and push buttons,
> but no muffineers or graters.

What do you mean by a muffineer or grater in UI terms?


> : Are you talking about libglade?
> 
> "GtkUIManager allows the application to specify a set of
>  "actions", then create menus and toolbars using those actions
>  from an XML-like interface description."

Interesting.  Has anyone used this in practise?


Regards,
Chris

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-24, chris <spamoff.danx@ntlworld.com> wrote:
> I'm developing an application now and hadn't considered using different 
> UI toolkits on Windows and Linux targets.  It increases development 
> time, but I suppose targetting windows with GWindows and Linux with 
> GtkAda (or something else?) is a good idea.

I haven't used Gwindows but if the argument is that one should use the
underlaying GUI in the OS then why don't one use the Windows API?

I would say: Develope your application in GtkAda and compile it on
Linux, Windows and Mac OS (if there are anybody out there using Mac OS
and Ada). Then you can see how it works on Windows. I think it should be
just fine and you can concentrate on developing more features in your
application and let purist port it for you to the GUI of their choice.

> One of the reasons for this application was because people disliked
> the GIMP interface on Windows (I was going to mod Gtk+ to be better
> behaved on Windows).  There are other reasons so it's not the only
> one, but it certainly topped the list of reasons people chose not to
> use GIMP on Windows - an other was because it crashed too much.

I haven't tested Gtk 2.x much on Windows yet, but I believe it should
work much better than the old version which wasn't good. GPS works fine
at least. 

   http://gtk-wimp.sourceforge.net/screenshots/

> What do you mean by a muffineer or grater in UI terms?

Nothing.
>
> Interesting.  Has anyone used this in practise?

No because Gtk 2.4 was released some days ago and GtkAda isn't bound to
it yet. But please help contribute to the GtkAda binding.


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

chris <spamoff.danx@ntlworld.com> wrote:
: 
: What do you mean by a muffineer or grater in UI terms?
 
Apart from use in cooking related programs?
For example there was once a way to dry signatures (written in
real ink). That could be used in illustration programs.

Grind an eMail using a GnuPG/PGP mill. Or a mincer?

Add a dash of randomness to mathematically created MIDI music.

Season source code ingredients of an executalbe with
debugging or optimization powder :-)

add a watermark to the videostream and stir using a ...


Georg

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Martin Dowie]

"Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
message > I would say: Develope your application in GtkAda and compile it on
> Linux, Windows and Mac OS (if there are anybody out there using Mac OS
> and Ada). Then you can see how it works on Windows. I think it should be
> just fine and you can concentrate on developing more features in your
> application and let purist port it for you to the GUI of their choice.

The only GtkAda app I've tried porting to Windows came to a crunching halt
when tasking was introduced.

For Windows, use Claw www.rrsoftware.com.

I once started a 'toy' signature package (based on Claw more than GWindows
or GtkAda) that was a front-end to both Claw and GtkAda. As a assumption
prover it worked (i.e. it displayed a simple application window with its own
icon, etc) but the effort in producing something that 'complete' would be
enormous. Also, with Ada0Y specifying such an interface would hopefully be
easier using, er, interfaces!

Be interesting to hear if anyone has ever tried porting Claw to a
Linux-based environment.

-- Martin

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-24, Martin Dowie <martin.dowie@btopenworld.com> wrote:
> "Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
> message > I would say: Develope your application in GtkAda and compile it on
>> Linux, Windows and Mac OS (if there are anybody out there using Mac OS
>> and Ada). Then you can see how it works on Windows. I think it should be
>> just fine and you can concentrate on developing more features in your
>> application and let purist port it for you to the GUI of their choice.
>
> The only GtkAda app I've tried porting to Windows came to a crunching halt
> when tasking was introduced.

Why?


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[chris]

Preben Randhol wrote:
> I haven't used Gwindows but if the argument is that one should use the
> underlaying GUI in the OS then why don't one use the Windows API?

Because it is, only a thicker than the Win32 binding?  My experience of 
GWindows and Win32 suggests GWindows is a very convienant extensible 
powerful binding to the Windows API.


> I would say: Develope your application in GtkAda and compile it on
> Linux, Windows and Mac OS (if there are anybody out there using Mac OS
> and Ada). Then you can see how it works on Windows. I think it should be
> just fine and you can concentrate on developing more features in your
> application and let purist port it for you to the GUI of their choice.

That's possible, I've not decided yet what to do with the UI layer. 
Right now I'm working on supporting (dib bitmap) images and loading them 
on Linux.  After that I will start prototyping the UI.


> I haven't tested Gtk 2.x much on Windows yet, but I believe it should
> work much better than the old version which wasn't good. GPS works fine
> at least. 
> 
>    http://gtk-wimp.sourceforge.net/screenshots/

I'm considering that.  How close is it to Windows L&F?


Chris

p.s. anyone know of a pie menu implementation for Windows or Gtk?

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: I haven't used Gwindows but if the argument is that one should use the
: underlaying GUI in the OS then why don't one use the Windows API?

Because the Windows API (assuming Win32), versatile as it may be,
is at a comparatively very low level.

Some toolkits provide reusable components well suited to both
the operating system and the programming language used.

BTW, does GtkAda have d&d of data icons (like files in windows folders)
or of other documents' objects into Gtk views?

:  Then you can see how it works on Windows. I think it should be
: just fine and you can concentrate on developing more features in your
: application and let purist port it for you to the GUI of their choice.

It's not a question of purism, but of usability and features that just
cannot be provided because they aren't supported in the portable GUI
toolkit.
 
:> What do you mean by a muffineer or grater in UI terms?
: 
: Nothing.

I is not true that I had meant nothing (unless in your world
something = nothing).
However, I have been observing now for some years that engineers
tend to be condescending when it comes to creativity not involving
the technical toys they are used to.  Old habits die hard.

In fact, you might have observed that some household items have
a *long* tradition in GUIs, e.g. the trash bin. IBM had the opposite,
very visibly, namely templates. This is another feature that is barely
persent in Windows, but I wouldn't know how to do this in GtkAda,
systemwide.

Things could become more portable across modern
operating systems if Microsoft could agree to standards.
CORBA is there. The "new piece of paper" button is there in just
about every application, each time signifying a *typed* data item.
Does GtkAda support adding a Windows component object to an application's
data objects? Does CLAW? Can we have XParts on Windows in a GtkAda
Windows application?

I should add that I do not want to demean GtkAda.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-24, chris <spamoff.danx@ntlworld.com> wrote:
> I'm considering that.  How close is it to Windows L&F?

It is better.

> p.s. anyone know of a pie menu implementation for Windows or Gtk?

What do you want that for?

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-24, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> BTW, does GtkAda have d&d of data icons (like files in windows folders)
> or of other documents' objects into Gtk views?

I don't know what you mean by Gtk views, but yes GtkAda has d&d.

Perhaps you should try it out?
-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Martin Dowie]

"Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
message
news:slrnc63gqr.b1m.randhol+valid_for_reply_from_news@k-083152.nt.ntnu.no...
> On 2004-03-24, Martin Dowie <martin.dowie@btopenworld.com> wrote:
> > "Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
> > message > I would say: Develope your application in GtkAda and compile
it on
> >> Linux, Windows and Mac OS (if there are anybody out there using Mac OS
> >> and Ada). Then you can see how it works on Windows. I think it should
be
> >> just fine and you can concentrate on developing more features in your
> >> application and let purist port it for you to the GUI of their choice.
> >
> > The only GtkAda app I've tried porting to Windows came to a crunching
halt
> > when tasking was introduced.
>
> Why?

Can't remember off hand (I'm at home) - I'll post the info when I can get
back to my work PC.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[chris]

Preben Randhol wrote:
> On 2004-03-24, chris <spamoff.danx@ntlworld.com> wrote:
> 
>>I'm considering that.  How close is it to Windows L&F?
> 
> It is better.

Better is good, but is it Windows?  Does it behave like Windows people 
will expect?


>>p.s. anyone know of a pie menu implementation for Windows or Gtk?
> 
> What do you want that for?

A(n?) usability experiment.  I want to see if people find Pie Menus more 
productive in practice.


Chris

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

chris <spamoff.danx@ntlworld.com> wrote on 25/03/2004 (14:50) :
> Preben Randhol wrote:
> >On 2004-03-24, chris <spamoff.danx@ntlworld.com> wrote:
> >
> >>I'm considering that.  How close is it to Windows L&F?
> >
> >It is better.
> 
> Better is good, but is it Windows?  Does it behave like Windows people 
> will expect?

What does windows people expect except that the application should crash?

> A(n?) usability experiment.  I want to see if people find Pie Menus more 
> productive in practice.

Thought you wanted Window L&F

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: On 2004-03-24, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
:> BTW, does GtkAda have d&d of data icons (like files in windows folders)
:> or of other documents' objects into Gtk views?
: 
: I don't know what you mean by Gtk views, but yes GtkAda has d&d.

What I mean is can I drag a drawing object from Corel Draw into a GIMP
image? Can I embed a GIMP image in a Word document and have the
(Word) document window's menu adapt itself to the object in focus?

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-25, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> What I mean is can I drag a drawing object from Corel Draw into a GIMP
> image? Can I embed a GIMP image in a Word document and have the
> (Word) document window's menu adapt itself to the object in focus?

I don't know. I don't use Windows. You can try. I doubt very much that
M$ wants to implement anything coming from GIMP though.

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
: On 2004-03-25, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
:> What I mean is can I drag a drawing object from Corel Draw into a GIMP
:> image? Can I embed a GIMP image in a Word document and have the
:> (Word) document window's menu adapt itself to the object in focus?
: 
: I don't know. I don't use Windows. You can try. I doubt very much that
: M$ wants to implement anything coming from GIMP though.

In this particular case MS has nothing to do with it.  If GIMP
chooses to conform to a COM, for example, other Windows programs
could embed GIMP objects in their data structures and provide
embedded views.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[chris]

Preben Randhol wrote:
> chris <spamoff.danx@ntlworld.com> wrote on 25/03/2004 (14:50) :
> 
>>Better is good, but is it Windows?  Does it behave like Windows people 
>>will expect?
> 
> What does windows people expect except that the application should crash?

I use Linux every day for work and use Free Software everday and X 
crashes or hangs more than Windows ever did IME... that said it isn't 
*that* often.  Windows applications do not crash so much users expect it 
to happen.  In fact on my system no application has crashed in a long 
while, and windows 2k never has - 98 did but things have come along way 
since then and my box is well administered IMHO.  Anyway you know fine 
well I was talking about the look and feel behaviour of Windows in UI 
terms... it's very clear from the context of the message.  The fact that 
windows crashes or doesn't crash is irrelevant to UI behaviour.


>>A(n?) usability experiment.  I want to see if people find Pie Menus more 
>>productive in practice.
> 
> Thought you wanted Window L&F

Yes, but the Pie Menu is an experiment to determine if it increases 
productivity - as they are publicised to.  Linear menus will be the 
default, but Pie Menus are a possible option.


Chris

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[chris]

Georg Bauhaus wrote:

> What I mean is can I drag a drawing object from Corel Draw into a GIMP
> image? Can I embed a GIMP image in a Word document and have the
> (Word) document window's menu adapt itself to the object in focus?

I like that sort of thing, but not many windows apps do it :|  M$ apps 
tend to, but some apps have problems.  How hard is it to do in practise?


Cheers,
Chris

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-25, chris <spamoff.danx@ntlworld.com> wrote:
> I use Linux every day for work and use Free Software everday and X 
> crashes or hangs more than Windows ever did IME... that said it isn't 
> *that* often. 

If X hands look at your installation. Have no problem with this.

> Anyway you know fine well I was talking about the look and feel
> behaviour of Windows in UI terms... it's very clear from the context
> of the message. The fact that windows crashes or doesn't crash is
> irrelevant to UI behaviour.

Sure, but there is no L&F in Windows.

> Yes, but the Pie Menu is an experiment to determine if it increases
> productivity - as they are publicised to.  Linear menus will be the
> default, but Pie Menus are a possible option.

I don't know of any pie menus in GTK. There is however some patch here.
http://atrey.karlin.mff.cuni.cz/~0rfelyus/GtkPieMenu/

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote on 25/03/2004 (15:30) :
> In this particular case MS has nothing to do with it.  If GIMP
> chooses to conform to a COM, for example, other Windows programs
> could embed GIMP objects in their data structures and provide
> embedded views.

Well, this doesn't have to do with d&d.

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

chris <spamoff.danx@ntlworld.com> wrote:
: Georg Bauhaus wrote:
: 
:> What I mean is can I drag a drawing object from Corel Draw into a GIMP
:> image? Can I embed a GIMP image in a Word document and have the
:> (Word) document window's menu adapt itself to the object in focus?
: 
: I like that sort of thing, but not many windows apps do it :|  M$ apps 
: tend to, but some apps have problems.  How hard is it to do in practise?

see here:
http://www.adapower.com/gnatcom/

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+abuse@pvv.org> wrote:
:> could embed GIMP objects in their data structures and provide
:> embedded views.
: 
: Well, this doesn't have to do with d&d.

It does from a usability perspective, and this is what user interfaces
are about. For example, if you want to fix a photograph on a sheet of
paper, do you move the photograph with your hand or do you have buttons
and menus built into your desk that you can press and that, in effect,
will arrange for a robot to fetch the photo and move it to some place
on the sheet of paper lying in front of you? 

Direct manipulation is the obvious way.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Pascal Obry]

"Martin Dowie" <martin.dowie@btopenworld.com> writes:

> The only GtkAda app I've tried porting to Windows came to a crunching halt
> when tasking was introduced.

This is not Windows. Tasking and GtkAda must be handled with care but there is
nothing impossible.

> For Windows, use Claw www.rrsoftware.com.

But it is only Windows.

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|         http://perso.wanadoo.fr/pascal.obry
--| "The best way to travel is by means of imagination"
--|
--| gpg --keyserver wwwkeys.pgp.net --recv-key C1082595

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-25, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> Preben Randhol <randhol+abuse@pvv.org> wrote:
>:> could embed GIMP objects in their data structures and provide
>:> embedded views.
>: 
>: Well, this doesn't have to do with d&d.
>
> It does from a usability perspective, and this is what user interfaces
> are about.


No, I mean a technical view as you asked if GtkAda did this. It is not
part of d&d.

Preben

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-25, Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote:
> chris <spamoff.danx@ntlworld.com> wrote:
>: Georg Bauhaus wrote:
>: 
>:> What I mean is can I drag a drawing object from Corel Draw into a GIMP
>:> image? Can I embed a GIMP image in a Word document and have the
>:> (Word) document window's menu adapt itself to the object in focus?
>: 
>: I like that sort of thing, but not many windows apps do it :|  M$ apps 
>: tend to, but some apps have problems.  How hard is it to do in practise?
>
> see here:
> http://www.adapower.com/gnatcom/

What prevents you from using this with GtkAda?

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote:
:> see here:
:> http://www.adapower.com/gnatcom/
: 
: What prevents you from using this with GtkAda?

(a) With GtkAda as opposed to provided by GtkAda.

(b) Does using GNATCOM mean that magically GIMP's fly-out menu
    disappears and that a GtkAda component embedded into another
    document magically replaces Word's menu bar with it's own? Is
    this portable? No.

(c) I would be using an OS-depended toolkit anyway, and I'm
    not sure that mixing toolkits is the way to go.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Stephen Leake]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> writes:

> Sure, but there is no L&F in Windows.

This is absurd.

Perhaps you were trying to say "Windows applications do not have a
consistent look and feel". This is simply not true. Sure, there are
applications that choose to present a different look and feel; the
same is true for any other GUI.

Preben, please refrain from commenting about Windows; you clearly know
nothing about it. 

Just as we don't like it when people who know nothing about Ada try to
trash it anyway, it is not ok for you to trash Windows when you know
nothing about it.

-- 
-- Stephe

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-26, Stephen Leake <stephen_leake@acm.org> wrote:
> Perhaps you were trying to say "Windows applications do not have a
> consistent look and feel". This is simply not true. Sure, there are
> applications that choose to present a different look and feel; the
> same is true for any other GUI.

? So it is true then. Shift+Insert, Ctrl-C...

What I would like to know is which particular Window L&F that is missing
in GtkAda 2.x? OK the file-selector is different, but other than that?

If the argument is that one cannot use GtkAda for Window bcs you cannot
d&d an object into Word I find this very odd. This means you cannot use
Java either in Windows.

> Preben, please refrain from commenting about Windows; you clearly know
> nothing about it. 

I wish I didn't because that would mean I didn't have to use it, but
unfortunately I do.

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Marius Amado Alves]

> What I would like to know is which particular Window L&F that is missing
> in GtkAda 2.x?....

I find it funny that a constant look and feel is so assumed to be a good 
thing. It is not. It only makes beginners confuse all applications. Vive la 
différence.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Marius Amado Alves <maa@liacc.up.pt> wrote:
: I find it funny that a constant look and feel is so assumed to be a good 
: thing. It is not. It only makes beginners confuse all applications. Vive la 
: diff?rence.

How does consistency imply indistinguishable?

I don't think that Mac users think or have thought that controlling
an application using familiar dialogues is confusing.

Most importantly, users shouldn't even have to know about applications.
In general, they should be manipulating data and the OS should offer
tools to do so

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Marius Amado Alves]

On Friday 26 March 2004 12:54, Georg Bauhaus wrote:
> Marius Amado Alves <maa@liacc.up.pt> wrote:
> : I find it funny that a constant look and feel is so assumed to be a good
> : thing. It is not. It only makes beginners confuse all applications. Vive
> : la diff?rence.
>
> How does consistency imply indistinguishable[ness]?

I've seen users confusing the web browser with the word processor with the 
spreadsheet processor...

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Marius Amado Alves]

> Most importantly, users shouldn't even have to know about applications.
> In general, they should be manipulating data and the OS should offer
> tools to do so

That would be IEE (Integrated Everything Environment), to be out in 2023.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Marius Amado Alves]

> How does consistency imply indistinguishable?

I've seen users confusing the web browser with the word processor with the 
spreadsheet processor...

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Björn Persson]

Marius Amado Alves wrote:

> I find it funny that a constant look and feel is so assumed to be a good 
> thing. It is not. It only makes beginners confuse all applications. Vive la 
> différence.

I'll tell you why I like neither Emacs nor VI: Because they are so 
different from all the other programs I use. I want to be able to press 
the same key combinations for common commands like save, exit, copy and 
paste, I want close, minimize and maximize buttons in the same places in 
all windows, and I want similar menu entries in similar places. I *can* 
learn a lot of different user interfaces, but I don't like it.

And I have a theory about why Emacs has everything but a kitchen sink. I 
think Emacs users want to do everything in Emacs because they find all 
other programs so different from Emacs.

People want consistent user interfaces. It lets them concentrate on what 
they're doing instead of unimportant details like which command saves 
the file in this particular program.

-- 
Björn Persson

jor ers @sv ge.
b n_p son eri nu

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Marius Amado Alves]

On Friday 26 March 2004 17:35, Björn Persson wrote:
> Marius Amado Alves wrote:
> > I find it funny that a constant look and feel is so assumed to be a good
> > thing. It is not. It only makes beginners confuse all applications. Vive
> > la différence.
>
> I'll tell you why I like neither Emacs nor VI: Because they are so
> different from all the other programs I use. I want to be able to press
> the same key combinations for common commands like save, exit, copy and
> paste, I want close, minimize and maximize buttons in the same places in
> all windows, and I want similar menu entries in similar places. I *can*
> learn a lot of different user interfaces, but I don't like it.
>
> And I have a theory about why Emacs has everything but a kitchen sink. I
> think Emacs users want to do everything in Emacs because they find all
> other programs so different from Emacs.
>
> People want consistent user interfaces. It lets them concentrate on what
> they're doing instead of unimportant details like which command saves
> the file in this particular program.

Ok, but that's not "look and feel", it's shortcut keys. I use the X version of 
emacs, with point and click menus, so I don't have to memorize keys (which I 
can't).

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-26, Bj�rn Persson <spam-away@nowhere.nil> wrote:
> I'll tell you why I like neither Emacs nor VI: Because they are so 
> different from all the other programs I use. I want to be able to press 
> the same key combinations for common commands like save, exit, copy and 
> paste, I want close, minimize and maximize buttons in the same places in 
> all windows, and I want similar menu entries in similar places. I *can* 
> learn a lot of different user interfaces, but I don't like it.

It tells more that you haven't really tried neither. If you had you
would see how much more efficient they are (especially vim).

It is folly to say that every program should be equal when it clearly
hinders efficiency. It is like saying one should only have programs with
a Graphical UI. Well try do a repetitive task 200 times in a GUI
application by clicking with the mouse and compare it to a command line
script.

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Randy Brukardt]

"Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
message
news:slrnc68641.29f.randhol+valid_for_reply_from_news@k-083152.nt.ntnu.no...
...
> If the argument is that one cannot use GtkAda for Window bcs you cannot
> d&d an object into Word I find this very odd. This means you cannot use
> Java either in Windows.

That's true, really. For Quick and Dirty apps, it really doesn't matter, but
if you are planning to distribute the apps to a large number of users, you
really need a tailored interface (Windows, Linux, whatever). Java apps on
Windows are clunky, ugly, and tend to sink below other windows. Pretty much
the same list of things leveled at GtkAda.

             Randy.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Dale Stanbrough]

Randy Brukardt wrote:

> "Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
> message
> news:slrnc68641.29f.randhol+valid_for_reply_from_news@k-083152.nt.ntnu.no...
> ...
> > If the argument is that one cannot use GtkAda for Window bcs you cannot
> > d&d an object into Word I find this very odd. This means you cannot use
> > Java either in Windows.
> 
> That's true, really. For Quick and Dirty apps, it really doesn't matter, but
> if you are planning to distribute the apps to a large number of users, you
> really need a tailored interface (Windows, Linux, whatever). Java apps on
> Windows are clunky, ugly, and tend to sink below other windows. Pretty much
> the same list of things leveled at GtkAda.
> 
>              Randy.
> 

Posiedon (the UML editor in Java) seems to be doing ok. I presonally find
the interface to be woefully inadequate, but obviously many others don't.


Dale

-- 
dstanbro@spam.o.matic.bigpond.net.au

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Florian Weimer]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:

> It does from a usability perspective, and this is what user interfaces
> are about.

Users typically don't create COM objects using drag and drop.

-- 
Current mail filters: many dial-up/DSL/cable modem hosts, and the
following domains: postino.it, tiscali.co.uk, tiscali.cz, tiscali.it,
voila.fr.

Re: OT: GUI

[Jacob Sparre Andersen]

Martin Dowie wrote:

> Be interesting to hear if anyone has ever tried porting Claw to a
> Linux-based environment.

But would that be any better than GtkAda?  Just another non-native GUI
library.

I must agree that GtkAda isn't exactly perfect.  But I had to use a
Microsoft GUI a few days ago, and that was even further from perfect.

Also.  For people like me are GUI's more of a nuisance than something
important.  I don't mind GUI's written with Tcl/Tk.  And GtkAda hides
most of the difficulties with coding a GUI, when I absolutely have to
do it.  Something more complicated from the programmer POV is unlikely
to be used for "hobby" projects [1].

Greetings,

Jacob

[1] I get paid for writing those programs I make GUI's for, but in a
    sense I still consider them hobby projects.
-- 
�But you have to be a bit wary of a ship that collects
 snowflakes.�                                  -- Diziet Sma

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[chris]

Preben Randhol wrote:
> On 2004-03-26, Bj�rn Persson <spam-away@nowhere.nil> wrote:
> 
> It tells more that you haven't really tried neither. If you had you
> would see how much more efficient they are (especially vim).

They are efficient for you.  Just because something is efficient for you 
does not mean it is efficient for everyone else.  C is efficient for 
some people, but not others.  Ocaml is efficient for some, not for 
others.  Ada for some not for others.  Windows for some, not for others. 
  Linux for some, not for others.  Too many people think they know it 
all and that all is their little bit of the world!  Closed minds are A 
Bad Thing.

> It is folly to say that every program should be equal when it clearly
> hinders efficiency. It is like saying one should only have programs with
> a Graphical UI. Well try do a repetitive task 200 times in a GUI
> application by clicking with the mouse and compare it to a command line
> script.

No one says every app should be equal, just consistent in user 
expectations of it's interface.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

Randy Brukardt <randy@rrsoftware.com> wrote on 26/03/2004 (22:50) :
> That's true, really. For Quick and Dirty apps, it really doesn't matter, but
> if you are planning to distribute the apps to a large number of users, you
> really need a tailored interface (Windows, Linux, whatever).


Ah, yes, Gimp, GPS and GVD (for example) are really quick and dirty apps.
http://gtk-wimp.sourceforge.net/screenshots/gfx/gimp-winxp.png
http://libre.act-europe.fr/gps/img/gps_800x600.jpg
http://libre.act-europe.fr/gps/img/debug_800x600.jpg

http://gtk-wimp.sourceforge.net/screenshots/

I have have seen dozens of ugly and clunky apps using the native Windows
API. So it has more to do with programmer/designer. Apart from that I
don't like Java either.

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-26, chris <spamoff.danx@ntlworld.com> wrote:

> They are efficient for you.  Just because something is efficient for
> you does not mean it is efficient for everyone else.  

No, but saying that the strength of two programs are their weakness
isn't exactly what one would expect.

> Closed minds are A Bad Thing.

My point exactly.

> No one says every app should be equal, just consistent in user
> expectations of it's interface.

Please, try to get my point.


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Ludovic Brenta]

Marius Amado Alves <maa@liacc.up.pt> writes:

> > Most importantly, users shouldn't even have to know about applications.
> > In general, they should be manipulating data and the OS should offer
> > tools to do so
> 
> That would be IEE (Integrated Everything Environment), to be out in 2023.

Or the OS/2 Workplace Shell, out in 1994.

OS/2 was a great, truly object-oriented environment.  Too bad it
wasn't free software.  IBM dropped it and now it's dead.

-- 
Ludovic Brenta.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Ludovic Brenta]

Björn Persson <spam-away@nowhere.nil> writes:

> I'll tell you why I like neither Emacs nor VI: Because they are so
> different from all the other programs I use.

[...]

> And I have a theory about why Emacs has everything but a kitchen
> sink. I think Emacs users want to do everything in Emacs because they
> find all other programs so different from Emacs.

Yes! Emacs rules!

> People want consistent user interfaces. It lets them concentrate on
> what they're doing instead of unimportant details like which command
> saves the file in this particular program.

C-x C-s

:)

-- 
Ludovic Brenta.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Ludovic Brenta]

Marius Amado Alves <maa@liacc.up.pt> writes:

[Emacs and VI]

> Ok, but that's not "look and feel", it's shortcut keys. I use the X
> version of emacs, with point and click menus, so I don't have to
> memorize keys (which I can't).

Funny.  Whenever I use a "standard" text editor, I find myself typing
M-q to fill, C-a and C-e to move, M-d to delete words, etc.  And I
FEEL frustrated when they don't work.  Not a look-and-feel, eh?  I say
key bindings are an essential part of look-and-feel.

BTW, yes I read news from within Emacs, and I have not touched the
mouse once in more than an hour.

-- 
Ludovic Brenta.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Ludovic Brenta]

Florian Weimer <fw@deneb.enyo.de> writes:

> Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:
> 
> > It does from a usability perspective, and this is what user interfaces
> > are about.
> 
> Users typically don't create COM objects using drag and drop.

But programmers make it possible to drag and drop by creating COM
objects.  COM is the preferred integration mechanism in the Windows
world (the ActiveX thing is really an extension to COM).

-- 
Ludovic Brenta.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Björn Persson]

Preben Randhol wrote:

> No, but saying that the strength of two programs are their weakness
> isn't exactly what one would expect.

Every time I want to go to the end of the line I get sent all the way 
down to the end of the file and totally lose my place because I pressed 
End without thinking. Is that the strength of Emacs?

Yes I know I could learn Emacs-Lisp and rewire the entire user 
interface. I prefer to use an editor that I don't have to rewire, and 
program something more interesting in Ada instead.

-- 
Björn Persson

jor ers @sv ge.
b n_p son eri nu

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Bj�rn Persson <spam-away@nowhere.nil> wrote:
 
: Every time I want to go to the end of the line I get sent all the way 
: down to the end of the file and totally lose my place because I pressed 
: End without thinking. Is that the strength of Emacs?

This has been done. Emacs has emulations of keyboard and mouse
handling of vi, CUA (OS/2 and Windows style), and others.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Preben Randhol <randhol+abuse@pvv.org> wrote:
: Randy Brukardt <randy@rrsoftware.com> wrote on 26/03/2004 (22:50) :
:> That's true, really. For Quick and Dirty apps, it really doesn't matter, but
:> if you are planning to distribute the apps to a large number of users, you
                                                 ^^^^^^^^^^^^^^^^^^^^^^^
:> really need a tailored interface (Windows, Linux, whatever).
: 
: 
: Ah, yes, Gimp, GPS and GVD (for example) are really quick and dirty apps.

Again, is there a "large number of users" of these programs when you
compare the number of users of PhotoShop, Visual Studio,
and <whatever standalone debugger>? Why not?
(I think noone wants to demean GtkAda per se.)

And I think that you can feed (young?) programmers or machine operators
a lot that you cannot feed "ordinary" computer users.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Florian Weimer <fw@deneb.enyo.de> wrote:
: Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:
: 
:> It does from a usability perspective, and this is what user interfaces
:> are about.
: 
: Users typically don't create COM objects using drag and drop.

Let MS usability labs have that turned into a button in the
xyz bar. Is there a button to place a GIMP object in some document?
Can I manipulate the object in place?

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Simon Wright]

Bj�rn Persson <spam-away@nowhere.nil> writes:

> I'll tell you why I like neither Emacs nor VI: Because they are so
> different from all the other programs I use. I want to be able to
> press the same key combinations for common commands like save, exit,
> copy and paste, I want close, minimize and maximize buttons in the
> same places in all windows, and I want similar menu entries in
> similar places. I *can* learn a lot of different user interfaces,
> but I don't like it.

I tend to stay in Emacs and get annoyed when M-q doesn't fill a
paragraph in Outloook. I would be less annoyed if Outlook had a 'fill
paragraph' feature .. (and if there is one, please tell me, I will be
eternally grateful) ..

> And I have a theory about why Emacs has everything but a kitchen
> sink. I think Emacs users want to do everything in Emacs because
> they find all other programs so different from Emacs.

.. but I have to admit that even in Emacs there is no real consistency
for extensions; the Ada mode uses C-c C-c to mean 'build', whereas in
CVS it means 'I've finished updating the log entry, commit' -- which
_is_ close to the semantics in gnus & rmail, 'send this message'.

-- 
Simon Wright                               100% Ada, no bugs.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Simon Wright]

Bj�rn Persson <spam-away@nowhere.nil> writes:

> Preben Randhol wrote:
> 
> Every time I want to go to the end of the line I get sent all the
> way down to the end of the file and totally lose my place because I
> pressed End without thinking. Is that the strength of Emacs?

This Emacs is GNU Emacs 21.1.1 (i386-mandrake-linux-gnu, X toolkit,
Xaw3d scroll bars) of 2002-03-03 on montreal.mandrakesoft.com and
(with emacs -q) end is bound to end-of-line .. C-end is bound to
end-of-buffer.

-- 
Simon Wright                               100% Ada, no bugs.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Stephen Leake]

Björn Persson <spam-away@nowhere.nil> writes:

> And I have a theory about why Emacs has everything but a kitchen sink.
> I think Emacs users want to do everything in Emacs because they find
> all other programs so different from Emacs.

That's part of it; Emacs runs on every host OS I need to use, so I
don't need to learn another user interface.

And I will spend time learning how to do something new in LaTeX (in
Emacs), rather than use Word for a "quick memo". The new knowledge
will be useful later, and I don't risk Word trashing the memo as it
lives beyond the initial use.

But more importantly, keyboard interfaces are simply more productive
than WIMP interfaces; it takes less time to get the job done. Keyboard
interfaces do take longer to learn, which is why I want to use only
one everywhere.

> People want consistent user interfaces. It lets them concentrate on
> what they're doing instead of unimportant details like which command
> saves the file in this particular program.

Yes.

-- 
-- Stephe

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Stephen Leake]

Björn Persson <spam-away@nowhere.nil> writes:

> Yes I know I could learn Emacs-Lisp and rewire the entire user
> interface. I prefer to use an editor that I don't have to rewire,

What editor is that?

Does it work on every host OS you have to use?

> and program something more interesting in Ada instead.

I learned how to rewire Emacs 20 years ago, and have been benefiting
from that knowledge ever since. Time spent learning good tools is
amortized over many projects. 

Emacs gives me more time to work on interesting Ada projects!

-- 
-- Stephe

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Stephen Leake <stephen_leake@acm.org> wrote:
: But more importantly, keyboard interfaces are simply more productive
: than WIMP interfaces; it takes less time to get the job done.

There are interfaces with a strong emphasis on the mouse (3 buttons
and  no nervousness wheel) that are still surprisingly productive.
Plan 9 for example.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Stephen Leake]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:

> Stephen Leake <stephen_leake@acm.org> wrote:
> : But more importantly, keyboard interfaces are simply more productive
> : than WIMP interfaces; it takes less time to get the job done.
> 
> There are interfaces with a strong emphasis on the mouse (3 buttons
> and  no nervousness wheel) that are still surprisingly productive.
> Plan 9 for example.

Productive at what sort of task?

When writing Ada code, the result is text. Unless you can enter
characters into a file more efficiently using the mouse, text is
generated by typing on the keyboard. Therefore, anything that takes my
hands off the keyboard is wasting time.

For editing video or sound, a mouse interface does make more sense.

So I need to qualify: "Emacs is the most productive environment I know
of, for tasks that are mainly text-based". That includes writing code
and documents, which is the overwhelming part of what I use computers
for.

-- 
-- Stephe

Re: OT: GUI

[Martin Dowie]

"Jacob Sparre Andersen" <sparre@nbi.dk> wrote in message
news:plptaz2ntn.fsf_-_@sparre.crs4.it...
> But would that be any better than GtkAda?  Just another non-native GUI
> library.

Well, Claw is atleast native on Win32 platforms - is Gtk(Ada) native on
anything? I don't mean to be rude, I really just don't know!

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Georg Bauhaus]

Stephen Leake <stephen_leake@acm.org> wrote:
:> There are interfaces with a strong emphasis on the mouse (3 buttons
:> and  no nervousness wheel) that are still surprisingly productive.
:> Plan 9 for example.
: 
: Productive at what sort of task?

For example at editing and manipulating text. The whole OS
interface is drawn on a graphics display, and every piece of
text, not matter what kind, can be used both as text data, and as
a command. The distinction is made by "activating" the relevant
portion of text with the middle mouse button for commands.
Have you ever succeded in copying the text of a command button
off that user interface control?

E.g. if I write | fmt in this paragraph and I want to reshape
say the above paragraph, I mark the above paragraph and then drag
the mouse along ^^^^^ above with the middle mouse button pressed.
Effect: The marked paragraph is piped through fmt.
The paragraph can be anywhere in any window. Another level of
OS integration.

It is really very handy. If you want to do cut&paste in Emacs
from the keyboard, you mark a region, kill, move (e.g. by searching
or switching buffers), stop the search or adjust point in the
buffer, and yank.

In ACME (a multitrack programmer's editor, but may be used for
news reading/writing, file system browsing, or whatever, sounds
familiar? :-) this is all done very quickly using just the mouse
without interruption. That is, the mouse can cut, no menu or keyboard
involved, can be used for navigation and searching, and it can then
paste.
(Please note the intermediary navigational step, this is neither
the same as dragging, nor is it the same as copy&paste.)

You can thus build your own toolboxes (a.k.a menues) by just
displaying a file of commands or boilerplate text for TeX, Ada etc.
No per application knowledge of how to find/get the template text is
required, it's part of the OS, and completely "scriptable" (again,
recursively, using the same facilities) :-)

Actually you can write Help files containing the text of commands
and this text can actually be used to produce the effects that the
help file describes, like this.

  +-------------------------------+
  | To produce an executable,     |
  | cd /u/you/foo and then        |
  | run gnatmake -Pfoo.gpr        |
  |  ...                          |

Just drag the the pointer along "cd ....foo" with the middle
mouse button pressed and then along "gnat....gpr". Done.  Is this
integrated or not?  And if you compiler happens to be gcc34-gnatmake
or adabuild, you as a user can adjust the help file.
(Again this is not the same as commands built into Windows or Mac
OS X etc help files. As a user, you cannot start editing Windows help
files, and you cannot add arbitrary commands to an existing help
file, can you?) ;-)

Re: OT: GUI

[Ludovic Brenta]

"Martin Dowie" <martin.dowie@btopenworld.com> writes:

> "Jacob Sparre Andersen" <sparre@nbi.dk> wrote in message
> news:plptaz2ntn.fsf_-_@sparre.crs4.it...
> > But would that be any better than GtkAda?  Just another non-native GUI
> > library.
> 
> Well, Claw is atleast native on Win32 platforms - is Gtk(Ada) native on
> anything? I don't mean to be rude, I really just don't know!

Yes, it is native on GNOME on various operating systems: GNU/Linux,
{Free|Open|Net}BSD, Solaris, perhaps more.

-- 
Ludovic Brenta.

Re: OT: GUI

[Frank J. Lhota]

"Ludovic Brenta" <ludovic.brenta@insalien.org> wrote in message
news:87d66yw033.fsf@insalien.org...
> Yes, it is native on GNOME on various operating systems: GNU/Linux,
> {Free|Open|Net}BSD, Solaris, perhaps more.

I don't mean to be rude, but is there an MS Windows version of GNOME? (Am I
the only one who wants an alternative to Explorer?)

Re: OT: GUI

[Ludovic Brenta]

"Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net> writes:

> "Ludovic Brenta" <ludovic.brenta@insalien.org> wrote in message
> news:87d66yw033.fsf@insalien.org...
> > Yes, it is native on GNOME on various operating systems: GNU/Linux,
> > {Free|Open|Net}BSD, Solaris, perhaps more.
> 
> I don't mean to be rude, but is there an MS Windows version of GNOME? (Am I
> the only one who wants an alternative to Explorer?)

What is Explorer? I don't use Windows.

AFAICT, GNOME, the GNU Object Model Environment, has not been ported
to Windows, and won't be.  One subcomponent of GNOME, the GIMP Tookit,
has been ported and you can run GTK applications on Windows, such as
the GIMP itself, or GtkAda applications.

-- 
Ludovic Brenta.

Re: OT: GUI

[Georg Bauhaus]

Ludovic Brenta <ludovic.brenta@insalien.org> wrote:
: 
: AFAICT, GNOME, the GNU Object Model Environment, has not been ported
: to Windows, and won't be.

However, as GNOME uses CORBA, and as there is a standardised IDL
binding for Ada, there are ways to write software for any system
providing an ORB. That is the theory...

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Robert I. Eachus]

Bj�rn Persson wrote:

> Every time I want to go to the end of the line I get sent all the way 
> down to the end of the file and totally lose my place because I pressed 
> End without thinking. Is that the strength of Emacs?
> 
> Yes I know I could learn Emacs-Lisp and rewire the entire user 
> interface. I prefer to use an editor that I don't have to rewire, and 
> program something more interesting in Ada instead.

You definitely don't have to do all that to rebind one key.  In fact if 
I remember correctly, the escape sequence generated by the "End" key is 
not standard emacs.  That is an "additional" binding someone has added 
for you.  Hmmm.  In XEmacs which I use, the End key is not bound to 
anything by default. It only takes a few seconds to bind it to 
"end-of-line", which seems to me to be reasonable binding for it.  Of 
course, I won't bother to add it to my .emacs file, since I will 
continue to type C-e having learned that before I had a keyboard with an 
End key. ;-)


-- 

                                           Robert I. Eachus

"The terrorist enemy holds no territory, defends no population, is 
unconstrained by rules of warfare, and respects no law of morality. Such 
an enemy cannot be deterred, contained, appeased or negotiated with. It 
can only be destroyed--and that, ladies and gentlemen, is the business 
at hand."  -- Dick Cheney

Re: OT: GUI

[Robert I. Eachus]

Frank J. Lhota wrote:


> I don't mean to be rude, but is there an MS Windows version of GNOME? (Am I
> the only one who wants an alternative to Explorer?)

No, but it is a good example of what this discussion is all about.  When 
I am developing code, I open emacs, and design, code, document, test, 
and debug from inside emacs.  I may have to be aware of Windows quirks 
and quivers if I am targeting the Windows GUI, but other than that the 
only GUI I need for software development--for any target--is emacs.


-- 

                                           Robert I. Eachus

"The terrorist enemy holds no territory, defends no population, is 
unconstrained by rules of warfare, and respects no law of morality. Such 
an enemy cannot be deterred, contained, appeased or negotiated with. It 
can only be destroyed--and that, ladies and gentlemen, is the business 
at hand."  -- Dick Cheney

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Stephen Leake]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> writes:

> Stephen Leake <stephen_leake@acm.org> wrote:
> :> There are interfaces with a strong emphasis on the mouse (3 buttons
> :> and  no nervousness wheel) that are still surprisingly productive.
> :> Plan 9 for example.
> : 
> : Productive at what sort of task?
> 
> For example at editing and manipulating text. The whole OS
> interface is drawn on a graphics display, and every piece of
> text, not matter what kind, can be used both as text data, and as
> a command. The distinction is made by "activating" the relevant
> portion of text with the middle mouse button for commands.
> Have you ever succeded in copying the text of a command button
> off that user interface control?
> 
> E.g. if I write | fmt in this paragraph and I want to reshape
> say the above paragraph, I mark the above paragraph and then drag
> the mouse along ^^^^^ above with the middle mouse button pressed.
> Effect: The marked paragraph is piped through fmt.
> The paragraph can be anywhere in any window. Another level of
> OS integration.

Hmm. I assume the output of fmt is piped back to the same selected
region?

Emacs (on X Windows or MS Windows) doesn't have that same level of
dynamic binding of commands to selection regions; that is interesting.

But, if I want to format a paragraph, I don't want to hunt down a
piece of screen that happens to have "| fmt" in it; I want to hit one
key that formats the paragraph surrounding the insertion point,
according to the language mode of the buffer. Emacs does that out of
the box for many languages, and allows me to define new modes and
specify the formatting function.

> It is really very handy. If you want to do cut&paste in Emacs from
> the keyboard, you mark a region, kill, move (e.g. by searching or
> switching buffers), stop the search or adjust point in the buffer,
> and yank.

Yes. All done with keystrokes; very quick and easy.

> In ACME (a multitrack programmer's editor, but may be used for news
> reading/writing, file system browsing, or whatever, sounds familiar?
> :-) this is all done very quickly using just the mouse without
> interruption. That is, the mouse can cut, no menu or keyboard
> involved, can be used for navigation and searching, and it can then
> paste. (Please note the intermediary navigational step, this is
> neither the same as dragging, nor is it the same as copy&paste.)

The mouse can do that in Emacs as well. And in MS Word, or any other
GUI editor I've used.

But the point is, what are you doing before and after the cut and
paste? Normally, I would be typing along on the keyboard, and realize
I want to copy some text from "over there" to "here". Do I move my
hand over to the mouse, cut and paste, then move back and continue
typing? No; that's a waste of time.

If you are claiming the mouse cut&paste is enough _faster_ than the
keyboard cut&paste that it is a net savings, please say so. I doubt it.

> You can thus build your own toolboxes (a.k.a menues) by just
> displaying a file of commands or boilerplate text for TeX, Ada etc.
> No per application knowledge of how to find/get the template text is
> required, it's part of the OS, and completely "scriptable" (again,
> recursively, using the same facilities) :-)

If you can make the bindings of commands to keys persistent, that
could be a nice way for beginners to start building their own
keybindings. But if I'm going to use those bindings for the next
twenty years, it's worth my time to learn how to write them in a more
structured way.

> Actually you can write Help files containing the text of commands
> and this text can actually be used to produce the effects that the
> help file describes, like this.
> 
>   +-------------------------------+
>   | To produce an executable,     |
>   | cd /u/you/foo and then        |
>   | run gnatmake -Pfoo.gpr        |
>   |  ...                          |
> 
> Just drag the the pointer along "cd ....foo" with the middle
> mouse button pressed and then along "gnat....gpr". Done.  Is this
> integrated or not?  And if you compiler happens to be gcc34-gnatmake
> or adabuild, you as a user can adjust the help file.
> (Again this is not the same as commands built into Windows or Mac
> OS X etc help files. As a user, you cannot start editing Windows help
> files, and you cannot add arbitrary commands to an existing help
> file, can you?) ;-)

I do the same thing in Emacs all the time; I have text files that
contain the same sort of instructions; explanatory text mixed with
commands. Simple keystrokes execute the commands; no mouse needed.

And again, Emacs runs on every host OS I need. So I get this same
level of integration on Windows, Lynx, Linux, Solaris, AIX, etc.

What other operating systems does ACME run on?

-- 
-- Stephe

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Björn Persson]

Since some people seem to feel that they need to defend Emacs, let me 
make one thing clear: The post above was not about which editor is best. 
It was not about mice versus keyboards. It was most definitely *not* 
about the End key. It was about _consistency_ in user interfaces. I did 
not say that Emacs and VI are bad. I said that _big_differences_ between 
programs' user interfaces are annoying.

-- 
Björn Persson

jor ers @sv ge.
b n_p son eri nu

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Martin Dowie]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> wrote in message > > The only GtkAda app I've tried porting to Windows came to a crunching halt
> > when tasking was introduced.
> 
> Why?

I think the problem was the same this one:

http://lists.act-europe.fr/pipermail/gtkada/2002-December/001834.html

Cheers,

-- Martin

Re: OT: GUI

[Preben Randhol]

On 2004-03-27, Martin Dowie <martin.dowie@btopenworld.com> wrote:
> "Jacob Sparre Andersen" <sparre@nbi.dk> wrote in message
> news:plptaz2ntn.fsf_-_@sparre.crs4.it...
>> But would that be any better than GtkAda?  Just another non-native GUI
>> library.
>
> Well, Claw is atleast native on Win32 platforms - is Gtk(Ada) native on
> anything? I don't mean to be rude, I really just don't know!

What is your definition of native? Is Ada native on anything or should
one use assembly?

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

On 2004-03-27, Simon Wright <simon@pushface.org> wrote:
>> Preben Randhol wrote:
>> 
>> Every time I want to go to the end of the line I get sent all the
>> way down to the end of the file and totally lose my place because I
>> pressed End without thinking. Is that the strength of Emacs?

No I didn't

-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: OT: GUI

[Martin Dowie]

"Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
message
news:slrnc7sr4a.15k.randhol+valid_for_reply_from_news@rong.nt.ntnu.no...
> On 2004-03-27, Martin Dowie <martin.dowie@btopenworld.com> wrote:
> > "Jacob Sparre Andersen" <sparre@nbi.dk> wrote in message
> > news:plptaz2ntn.fsf_-_@sparre.crs4.it...
> >> But would that be any better than GtkAda?  Just another non-native GUI
> >> library.
> >
> > Well, Claw is atleast native on Win32 platforms - is Gtk(Ada) native on
> > anything? I don't mean to be rude, I really just don't know!
>
> What is your definition of native? Is Ada native on anything or should
> one use assembly?

Native as in not having to go through a large binding to access the
underlying opertion. Claw is "native" as it sits (almost) directly on top
of the Windows calls. But GtkAda on Windows isn't native as it
doesn't use the underlying OS GUI routines (or at least not in any
way that could be described as close).

Re: OT: GUI

[Jacob Sparre Andersen]

Martin Dowie wrote:
> Preben Randhol wrote:

> > What is your definition of native? Is Ada native on anything or
> > should one use assembly?
> 
> Native as in not having to go through a large binding to access the
> underlying opertion. Claw is "native" as it sits (almost) directly
> on top of the Windows calls. But GtkAda on Windows isn't native as
> it doesn't use the underlying OS GUI routines (or at least not in
> any way that could be described as close).

Since GtkAda is made to be a general, high-level graphics system
independent API for GUI programming, I would be very surprised if it
was native (according to your definition) to any graphics system.

A related note: Hardly any Unix programs are written using a native
(according to your definition) GUI API.  Most Unix'es use X as their
graphics system.  And most programs for X are written using a thick
binding to the X protocol.

Jacob
-- 
�What fun is it being "cool" if you can't wear a sombrero?�

Re: OT: GUI

[Stephen Leake]

"Martin Dowie" <martin.dowie@btopenworld.com> writes:

> "Preben Randhol" <randhol+valid_for_reply_from_news@pvv.org> wrote in
> message
> news:slrnc7sr4a.15k.randhol+valid_for_reply_from_news@rong.nt.ntnu.no...
> > On 2004-03-27, Martin Dowie <martin.dowie@btopenworld.com> wrote:
> > > "Jacob Sparre Andersen" <sparre@nbi.dk> wrote in message
> > > news:plptaz2ntn.fsf_-_@sparre.crs4.it...
> > >> But would that be any better than GtkAda?  Just another non-native GUI
> > >> library.
> > >
> > > Well, Claw is atleast native on Win32 platforms - is Gtk(Ada) native on
> > > anything? I don't mean to be rude, I really just don't know!
> >
> > What is your definition of native? Is Ada native on anything or should
> > one use assembly?
> 
> Native as in not having to go through a large binding to access the
> underlying opertion. Claw is "native" as it sits (almost) directly on top
> of the Windows calls. 

Yes. But Claw is itself a "large binding". So are applications written
using Claw "native windows"?

I think this is a moot point. The issue is what is gained by a native
binding. Usually, the desired gains are execution speed, and ease of
porting other "native" applications.

Any thick binding will be slower than a thin binding, but some are
slower than others. Gtk is noticably slower than Windex, and I suspect
than Claw.

Ease of porting will always be low, between any two thick bindings;
porting an app from Windex to Claw is probably as hard as porting from
Windex to GtkAda.

> But GtkAda on Windows isn't native as it doesn't use the underlying
> OS GUI routines 

Of course it does!

> (or at least not in any way that could be described as close).

Well, that's the whole issue, isn't it? What, exactly, do you mean by
"close".

Let's see, we could measure the number function calls. Suppose I do
'Set_Text (widget, "foo");' in Windex, Claw, and GtkAda. How many
function calls does that go thru before it gets to Win32 API Settext?

Well, Windex.Windows.Edit_Text.Put_Text is a direct binding to Win32
API Send_Message (WM_SETTEXT). I guess that's pretty "close".

I'm not familiar enough with the Gtk source to actually trace this. I
bet it is more calls, though.

But then, if I resize a window, and want all the child windows to
automatically resize to fit, GtkAda is _way_ more efficient than
Windex, because it is already done, and it "just works"; any code I
write to do the same thing will not be as good. No binding that
provides that feature will be "close" to the Win32 API.

So it depends heavily on what features you measure, to get a
definition of "close".

-- 
-- Stephe

Re: OT: GUI

[Preben Randhol]

Martin Dowie <martin.dowie@btopenworld.com> wrote on 15/04/2004 (18:40) :
> Native as in not having to go through a large binding to access the
> underlying opertion. Claw is "native" as it sits (almost) directly on top
> of the Windows calls. But GtkAda on Windows isn't native as it
> doesn't use the underlying OS GUI routines (or at least not in any
> way that could be described as close).

OK, then yes GtkAda is native on Linux, but it is not the only GUI you
have as there isn't *one* company owning the OS.

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI

[Preben Randhol]

Stephen Leake <stephen_leake@acm.org> wrote on 16/04/2004 (08:15) :
> I think this is a moot point. The issue is what is gained by a native
> binding. Usually, the desired gains are execution speed, and ease of
> porting other "native" applications.

Porting where? I would have thought "native" and "portability" to be
mutually exclusive?

> Any thick binding will be slower than a thin binding, but some are
> slower than others. Gtk is noticably slower than Windex, and I suspect
> than Claw.

Which version?

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

Bj�rn Persson <spam-away@nowhere.nil> wrote on 27/03/2004 (07:15) :
> Every time I want to go to the end of the line I get sent all the way 
> down to the end of the file and totally lose my place because I pressed 
> End without thinking. Is that the strength of Emacs?

What does this have to do with anything. Just bind the key to end of
line.

Manuals are ment to be read.

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI

[Martin Dowie]

"Preben Randhol" <randhol+abuse@pvv.org> wrote in message
news:mailman.8.1082099708.314.comp.lang.ada@ada-france.org...
> OK, then yes GtkAda is native on Linux, but it is not the only GUI you
> have as there isn't *one* company owning the OS.

sure it isn't native to 'X'?

-- Martin

Re: OT: GUI

[Preben Randhol]

Martin Dowie <martin.dowie@btopenworld.com> wrote on 16/04/2004 (15:20) :
> "Preben Randhol" <randhol+abuse@pvv.org> wrote in message
> news:mailman.8.1082099708.314.comp.lang.ada@ada-france.org...
> > OK, then yes GtkAda is native on Linux, but it is not the only GUI you
> > have as there isn't *one* company owning the OS.
> 
> sure it isn't native to 'X'?

What do you mean? 

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

Bj�rn Persson <spam-away@nowhere.nil> wrote on 28/03/2004 (15:40) :
> Since some people seem to feel that they need to defend Emacs, let me 
> make one thing clear: The post above was not about which editor is best. 
> It was not about mice versus keyboards. It was most definitely *not* 
> about the End key. 

You brought that up.

> It was about _consistency_ in user interfaces. I did not say that
> Emacs and VI are bad. I said that _big_differences_ between programs'
> user interfaces are annoying.

And consistency is A Bad Thing [tm] when it hinders efficiency. Doing
something in a certain way in one type of application can be efficient
yet at the same time pain-stakingly unefficient in another...

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Preben Randhol]

Georg Bauhaus <sb463ba@l1-hrz.uni-duisburg.de> wrote on 27/03/2004 (07:15) :
> : Ah, yes, Gimp, GPS and GVD (for example) are really quick and dirty apps.
> 
> Again, is there a "large number of users" of these programs when you
> compare the number of users of PhotoShop, Visual Studio,
> and <whatever standalone debugger>? Why not?

What do you think? Comparing an application developed probably in 1980s
with massive press coverage and ads with a newly created program which
has no financial backing. So you are saying Gimp is a quick and dirty
application. Well, please enlighten me what you find wrong with it?

As for quick and dirty I find Windows a quick and very dirty OS which so
many use. Why?

-- 
Preben Randhol -------------- http://www.pvv.org/~randhol/Ada95 --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Simon Wright]

Preben Randhol <randhol+valid_for_reply_from_news@pvv.org> writes:

> On 2004-03-27, Simon Wright <simon@pushface.org> wrote:
> >> Preben Randhol wrote:
> >> 
> >> Every time I want to go to the end of the line I get sent all the
> >> way down to the end of the file and totally lose my place because I
> >> pressed End without thinking. Is that the strength of Emacs?
> 
> No I didn't

sorry

-- 
Simon Wright                               100% Ada, no bugs.

Re: OT: GUI [was:]Ann: TeXCAD 4.1

[Stephen Leake]

Preben Randhol <randhol+abuse@pvv.org> writes:

> Björn Persson <spam-away@nowhere.nil> wrote on 27/03/2004 (07:15) :
> > Every time I want to go to the end of the line I get sent all the way 
> > down to the end of the file and totally lose my place because I pressed 
> > End without thinking. Is that the strength of Emacs?

Why did you think "End" meant "End of line" instead of "End of file"?
Probably because some other IDE did it that way. Well, Emacs probably
defined End before that IDE was born, so it's the one that's wrong :).

> What does this have to do with anything. Just bind the key to end of
> line.

Which is what I do. I often want to go to the end of the line, but
very rarely to the end of the file.
 
> Manuals are ment to be read.

Which is a major strength of Emacs. In GPS, there is very little
capability to bind keys or customize the interface (it is getting
better). Other GUI IDEs are similarly limited.

In Emacs, _everything_ can be customized.

-- 
-- Stephe