home *** CD-ROM | disk | FTP | other *** search
- Subject: Re: Colour.
- Date: Tue, 31 May 1994 10:03:52 +1000
- From: Warwick Allison <warwick@cs.uq.oz.au>
- Precedence: bulk
-
- In message <Pine.3.87.9405300202.A15807-0100000@undergrad>, you wrote:
- >On Mon, 30 May 1994, Warwick Allison wrote:
- >> ... colour palette handling by GEM programs ...
- >>
- >> - The standard 16 colours
- >> - They're different under MultiTOS (right?)
- >> - When should a program desire to change them?
- >When it's window is topped.
-
- ie. when it receives WM_TOPPED? What if it allows clicks on backgrounded
- windows? Should the program then force a palette-set whenever it is clicked
- on? Not realistic for 256 colour modes.
-
- >> - When should a user desire to change them?
- >Expound on this questions please.
-
- How much can we allow a user to set there own 16 colours? For example,
- I'm not satisfied with any of the std 16 colours as a colour for my
- background window, nor for my std window element colours, nor the colours
- available for 3D forms. So I change them all. How should programs handle
- this (eg. remapping colours used in icons to look for closest match?)
-
-
- >> - Sharing colours
- >> - No point each program allocating its own Purple
- >> - When colours must be changed, it would be nice for
- >> the actual palette change to be minimized (eg. purple
- >> changes to dark purple, not green)
- >Are you sure programmers are going to want to put forth that effort?
- >It's much easier to just stick together your palette. How about someone
- >write a library routine that, given a new palette, sorts them with respect
- >to the palette in place?
-
- Exactly. If a library routine is to be provided, then a greater amount
- of effort is justified compared to if just a statement of a standard is
- required. For a std, it should be simple, or else people won't follow it.
- For a lib routine, extra effort can be employed and the code shared.
-
-
- >> - True Colour emulations
- >> - Some programs use a 5x5x5 or 6x6x6 colourspace, and
- >> these should all be the same space.
- >>
- >> - When to change palettes
- >> - When window is topped?
- >Only this one.
- >> - When window is touched in any way?
- >In most cases, this would top the window. Under MultiTOS or others,
- >where you can use the gadgets without topping the window, the palette
- >should NOT be changed to the one for that window... how would you know
- >when to change it back?
-
- You can also touch windows without topping them. And a good thing this
- is, too. Ask any X11 user. Take for example a screen where 2 applications
- are running. The windows are such that they do not overlap. The user is
- swapping back and forth between the two windows quite regularly (eg. they
- are cutting and pasting, with cut occuring using the left button and pasting
- by using the right button). Should the user have to produce an extra click
- every time they change applications? The visual effect of this extra click
- under multitos, with window element colours configured to something useful,
- would be quite minor.
-
- I support clicking on background windows. So does WINX. So does MultiTOS.
-
-
- >> - When mouse enters window?
- >Too difficult... requires that something watch the mouse pointer.
-
- I tend to agree, unless there is some centralized way this mouse-in-window
- rather than window-on-top focus can be implemented.
-
-
- >Your proposal is sound, however, since not all apps will follow this, and
- >the WM_TOPPED signal is broadcast globally, perhaps apps should set the
- >palette BACK to what it was before when it's window is untopped?
-
- This doesn't work. In fact, the more applications that follow the
- scheme, the worse it gets, and the more confusing it gets for the user.
- Only if EVERY application follows the protocol will it work... in which
- case there is no need to attempt to restore some other application's
- palette.
-
-
- >For colors, an app should scan all available colors and see what it can
- >match to. When changing the palette, it should favor using those above 15.
-
- Match exactly? That won't work unless you adopt the X11 approach of naming
- colours. The colour space is too large otherwise (ie. WHICH purple?).
-
-
- >Since GEM is too free about this, this one is going to be tough for us to
- >work out, especially since no-one before will have followed these rules,
- >yet we'll have to tollerate their apps.
-
- Yes. Whatever the solution, older apps must be catered for in some way.
- Fortunately, many older applications where monochromatic.
-
- --
- Warwick
-