home *** CD-ROM | disk | FTP | other *** search
- Subject: The Undigestable Digested
- Date: Sun, 12 Jun 1994 17:30:22 +0200 (MDT)
- From: Annius.Groenink@cwi.nl (Annius Groenink)
- X-Face: "E3Hm]k]&:,OEP<{D2ixJf>-9[qOGLebNa0&cQyFL-a~)kTM3&&I"gFw=fJ]K%1IduGjOE`
- ZGu]&~G]QNGa7i/L!+#Xng<|+}HKYHj~5?fTInUEUh0$I1gBI7jrA!&_|e/pR1[cX:^xgJTPsrjA_9
- m8Zli[|.-u{]+c1(6C7mL*m`/_J\>.{4!:g
- Mime-Version: 1.0
- Precedence: bulk
-
-
-
- Neil:
-
- > I think the ST Book had a blue 'Atari' key, which meant certain keys with blue
- > figures on could be used as a numeric keypad... I don't know wether they work
- > the same way or not, but it is laid out exactly the same...
-
- Ah. A nice way to represent the num pad key combinations (in blue) in menus.
-
- -------
-
- Warwick:
-
- > A MUCH safer technique is to use Shift-FULLER. It also trains users to
- > look on the right-hand side for the SMALLER. It's also very meaningful,
- > since the FULLER already means `Change Size', which is what iconification
- > is all about.
-
- Makes sense. I'll change my code. Should this be part of the standard?
-
- -------
-
- Proposal v5:
-
- >CTRL Home - Move to top of page
- >Shift+CTRL Home - Move to bottom of page
- >ClrHome - Move to top of document
- >Shift+ClrHome - Move to bottom of document
-
- More standard is
-
- Control Uparrow Move to top of page
- Control Dnarrow Move to btm of page
- Shift Uparrow Page up
- Shift Dnarrow Page dn
-
- ------
-
- Michael Nolte proposes to add Ctrl to home/shift home for top/bottom of document:
-
- > - Safety. If you accidentally press ClrHome, you'll find back to where you
- > were easier, if the cursor has only jumped to the top of the frame/page
- > instead of jumping to the top of the document. CTRL-ClrHome is harder
- > to hit by accident.
-
- This is much the same story as Control A being dangerous---an application should
- provide a way to jump back to the original position when a user hits HOME by
- accident. (for example by placing a marker). It is not the code that is dangerous,
- but the limitated safety measures (no airbag) of most applications.
-
- ------
-
- Stefan:
-
- > What about standard key for fulling a window?
- > I use: *
- > And for zooming the window contents?
- > I use:
- > + for zoom in, more details
- > - for zoom out, fewer details
- > 0 for original zoom factor
-
- Yes. Pretty standard! I proposed this earlier. It should at least be implemented
- optionally, as I've done in Edith. It applies to practically any GEM application.
- (In Edith +/- means smaller/larger font).
-
- -----
-
- Warwick on Redo/Repeat
-
- > VI users would propose CTRL-.
-
- (and so would the few who have already got used to Edith Pro...) Seriously, Undo
- and Redo have 'do' in common, so I really think having Shift Undo or Control Undo
- as redo is nice. But we argued earlier that redo and do again is not the same!
- Redo is un-undo, whereas for 'do it again', no undo operation needs to have occurred.
- 'Do it again' after Undo could mean 'Undo the operation BEFORE the one just undone.'
- (if an application has multi-stage undo).
-
- -----
-
- Michel in reply to my HELP philosophy
-
- > Why strictly ASCII files? The main advantage of using a program like
- > ST-Guide is that you can have images, diagrams, boxes, circles, text
- > effects, and other features to make your online help presentable. Since
- > it is an external program, it also does not add to the size of your
- > application.
-
- The reason I want plain ASCII files is that it is then very easy to drag
- all the documents on your hard disk into a directory. Write a short description
- of each file, call the file containing the descriptions WHATIS (cf. Unix man)
- and the thing works!
-
- > Er, what? I'm not sure I follow this point. Is there any particular
- > reason why you would not want your data files in the same directory
- > as your program?
-
- Well... I recently divided my HD into applications, resources and user-specific
- data. Of course, on a multi-user system, this would always be the case.
- And yes, I feel happier if all help files are in the same hierarchy, apart from
- the applications.
-
- -----
-
- Michel:
-
- > Ofir, I just noticed; there is no "abandon" key in the standard. The
- > standard (used by all programs I have seen) is Control-H. Since that
- > is not used in the proposal, could we add this key?
-
- Control H was originally in a note saying: Atari suggest control H for
- Replace (as in swap text block and selection fragmentwise). This is a very
- good way of looking at search and replace. It should remain available
- optionally. And since we're talking about UNDO anyway, what about:
-
- Undo Undo
- Sh Undo Redo (shift being a modifier that sometimes
- means `in the opposite direction')
- Ctl Undo Undo the whole lot (i.e. Abandon or Restore)
-
- Ofir replies:
-
- > There is, although I put it in the wrong place.
- >
- > CTRL D - Abandon Window (iconify or place in menu)
- >
- > as suggested by Wilfried Behne.
-
- That's not the same kind of 'abandon'...
-
-
-
- Michel [I was over-expounding on active/top windows...]
-
- > Annius, I feel that this is too specific. Making a distiction between
- > the top window and the active window only applies to Edith. No other
- > program I have seen (ever) have made this distiction. Making it a part
- > of the standard would be too complex, when we are trying to keep things
- > simple.
-
- You're probably right. I should find my own way of doing this within the
- standard as it is proposed. However, there are a few programs that support
- key input into background windows (ASCREEN is one, and I've got two other
- German goodies on my disk but I forgot which ones).
-
- -----
-
- Timothy ``my little finger'' Miller writes:
-
- > The block=big-cursor method seems the most natural to me, and one of the
- > reasons I like it is that the cursor commands are heterogenious, where
- > the same commands that work on text work on blocks just the same, and
- > another reason I like it is that it takes a lot fewer total keypresses to
- > deal with
-
- ...and a lot fewer total keypresses to unintentionally get rid of your documents!
-
- -----
-
- Warwick in the ``what should preceed what'' discussion:
-
- > action:key:
- >
- > *Dialog*.Confirm.Key: Return
- > *Dialog*.Confirm.Colour: Green
- >
- >
- > key:action:
- > Dialog*Key.Return: Confirm
- > Dialog*Colour.Green: Confirm
- > # huh?
-
- I'd say there are sections
-
- 'key' maps a key code to a value (an action)
- 'colour' maps a user interface component to a value (a colour)
-
- hence
-
- Dialog*Key.Return: Confirm
- Dialog*Colour.Confirm: Green
- # hm.
-
- I think I'll stay low level for a while. Things are getting ever more complicated...
- I think at this point, reading the rest of Warwick's story, I agree with
- 'action-key'. My original thought was 'key-action' would be easier to implement
- (an array 'indexed' by the keys for quick access to their corresponding actions)
-
- -----
-
- Gerd:
-
- > When we talk about text blocks, we should have in mind, that text blocks
- > can be discontinuous!
-
- > In the case of block operations: Have discontinuous blocks in mind.
-
- I am! Actually, I've got one in my window right now!
-
- -----
-
- Timothy 'Control-A' Miller replies to someone arguing about ShCtl S <-> Ctl M
- (it takes 10 minutes to get used to ShCtl S)
-
- > Ah, but you use the opposite logic when saying why we should keep
- > Ctrl-A. If they'll take 10 minutes to get used to the new standard, then
- > they'll take 10 minutes to learn something other than ctrl-a, and they
- > won't be bothered by it since ctrl-a will do absolutely nothing.
-
- I *actually* agree here! Small sacrifice to swap ^A and Shift^A!
-
- -----
-
- Tim:
-
- > But I've never seen a text editor (I haven't seen the one you mention)
- > that did discontinuous blocks. Sounds like it would be a pain to deal
- > with... I'd need to keep track of ANOTHER linked-list in memory.
- > <grumble> Memory management....
-
- Papyrus is a word processor, not an editor. Edith (which I wrote and
- ZFC distributes) is an Editor which does discontinuous blocks. BTW, I
- don't use a linked list at all, I simply use an array and a smart
- re-allocation scheme. But I agree, discontinuous blocks is not
- something you can properly do in a few weeks... It took me years to
- implement a proper mechanism.
-
- > >You're right. How about CTRL+<, CTRL+>, CTRL+SHIFT+< and CTRL+SHIFT+>?
- >
- > I disagree... Ctrl-< is the same thing as ctrl-, This gets icky. I
-
- And besides, on a German keyboard Control + < is the same as Shift +
- Control + > as '<' and '>' are on the same key!!
-
- > Don't ever use any software that uses Ctrl-A for Select-All
- > Don't ever use any software that uses Shift-Backspace to delete a line.
- > Unusual tagline, isn't it?
-
- EXCEPT software where these combinations are harmless. I have to thank
- Timothy sincerely though, as I dicovered and fixed a problem concerning
- Shift Backspace in Edith. You can now safely hit shift + backspace 5
- times (by accident) without losing the original contents of the line.
- (this used to be only once.) Simply hit UNDO and on you go. I'll add
- Timothy to the 'acknowledgements'.
-
- -----
-
- Evan K. Langlois:
-
- > the traffic, but I recently saw some sort of suggestion about replacing
- > the keyboard table! This table is system wide. Replacing it is suicide
- > for MultiTOS.
-
- MultiTOS and suicide are synonyms anyway...
-
- > There is a little trick you can do to get all 16 colors in any dialog and
- > have it perfectly readable in ALL modes. And it looks fine in mono too!
- > What you do is set the dialog background color to the opposite color
- > that you want it, now set the SELECTED bit
-
- Ah! The same as the '!!' trick in C... Smart, but a bit cumbersome...
-
- -----
-
- Michel:
-
- > Ofir, I am not sure that you understand what I meant. As written, the
- > above keyboard shortcut implies that the window should be iconified if
- > the key is pressed. I mean we need a key that causes any changes you
- > have made to your document/whatever to be thrown away. Everyone,
- > even Atari, use Control+H for this...
-
- Not true. The latest Atari guidelines specify ^H for Replace (not
- search AND replace, but just replace. Control + Undo would be a better
- choice for this (undo all).
-
- -----
-
- Julian Reschke:
-
- > I thought that the idea of his proposal is more POSIX-like. Not to *change*
- > existing practice, but to document it. I doubt that any major developer
- > will change his existing code just to make the readers of this list happy.
-
- I think you're right. But the proposal doesn't change all that much,
- except perhaps for ShiftControl S. That's why I suggested to keep
- Control + M as well as ShiftControl S. (I saw that Ofir has said the
- same somewhere else). You'll see that in the end, Sh^S is the code that
- will survive, as it is much easier to remember. It is also the
- standard on other platforms than Atari.
-
- Julian: do you know a 'POSIX' like *explanation* for the fact that
- Control M is used as a shortcut for Save As...?
-
- ------
-
- -- Barry S. Munro -----------------------------------:-(
-
- > If the .SYS idea is to be implemented then there are no problems at all
- > with CTRL+A as it could be changed using one line in this file!
-
- And any program that bothers about our standard will probably also
- bother to implement the DEFAULTS.INF specification! So whatever we do
- with control + A, any program whose author has read our proposal will
- provide a way out for Timothy Miller.
-
- -----
-
- Michael Nolte:
-
- > I want Sh-CTRL-V as "Paste from...", because you can do it easily with one
- > hand. For "Shift-CTRL-I" you need two hands (or very long fingers).
-
- Weird argument... Is "Paste from..." something you do *so* regularly
- that you want to be able to do it with one hand?
-
- -----
-
- Ofir:
-
- > Even the so called German standard is not that well defined - CTRL+I in
- > Everest selects a word, in Kobold it shows Object Info, in Papyrus it does
- > something else (I forget what) and the list goes on.
-
- Good point. as to ^I: in a standard ^I for show info would be the most
- general we can think of. If you make a table counting the number of
- programs that use ^I for each different purpose, you'd find that 'Show
- Info' gets most points. A good extra note in the guidelines would be
- that if ever a programmer decides to use a keycode from the proposal for
- something DIFFERENT, s/he should watch out that it is not dangerous
- when a user assumes the standard meaning (as set out in the
- proposal) and simply hits the key without RTMF.
-
- -----
-
- Rick Flashman (Gribnif)
-
- > (because it makes more sense), we would have a substancial revolt among
- > our users (you should hear how many complaints our beta testers gave us
- > as we tried to change our close window from the original NeoDesk
- > CONTROL-W to the more accepted Alt-Esc).
-
- Well that sure is funny. We were just discussing changing CYCLE WINDOWS
- (Control W) to the more accepted Alt-Esc! :-) Does this mean that the
- confusion between cycling and closing windows is something universal,
- that goes beyond just the Atari world? :-?
-
- -----
-
- Ofir says control A is not open for discussion; Tim Miller replies:
-
- > I think is most certainly IS open for discussion. The fact that is has
- > been brought up and demonstrated to be potentially hazzardous is enough
- > to make it important.
-
- Well it HAS been discussed for more than a week now and the results are
- clear. Control A is there to stay (although I'd settle for adding shift
- and having control A for deselect all). Let's close the discussion Now.
-
-
-
- ------
- Is anyone actually reading upto this point?
-
- regards,
- --
- Annius V. Groenink | E-mail: avg@cwi.nl | Private & ZFC:
- CWI, Kruislaan 413 | Room: M233 | P.O. Box 12079
- 1098 SJ Amsterdam | Ext: 4077 | NL 1100 AB Amsterdam
- Netherland | Phone: +31 20 592 4077 | Phone: +31 20 695 9901
-