On Thu, 22 Feb 2001, Seth Burgess <[EMAIL PROTECTED]> wrote:
> I know you're trying to make a point, but I'd actually
> agree with that assessment too, if for slightly
> different reasons.  The user has no way of knowing
> what shade of gray they are about to place on the
> image.  I don't think this is used anywhere near as
> commonly as layers/transparency, so isn't perceived as
> such a big problem.  But it should get some attention
> as well.  

It is even worse when you are working with indexed images (yes, there
are people who are working with GIF images) because you can use only a
limited set of colors and many tools do not work as expected.  In
fact, the support for indexed images in the Gimp is not very good, but
in some cases it is better to live with the limitations of the indexed
mode than converting to RGB and then back to indexed.  For example,
some games require a fixed palette, or sometimes you need to assign a
specific index value to some pixels.

I don't think that converting everything to RGB automatically is a
good solution.  It is not the solution for indexed images and not for
grayscale images either.

Now, about the alpha channel...  When I posted my first message in
this thread, I vaguely remembered what we discussed last year and I
though that always adding an alpha channel would be the best solution.
However, I re-read the discussion from last year and I saw that at
that time I proposed to add an alpha channel as soon as a second layer
is added, but not before.  You can read the old thread with the
subject ``Is "Add alpha channel" really necessary?'' on this page:

I think that it makes more sense to add an alpha channel when a second
layer is created, not before (but I could change my mind again :-).

Basically, we have to choose the lesser of two evils:

1) Adding an alpha channel to all images (even with a single layer)
   will confuse some users who think of their images as "flat" things
   (no layers, no transparency).  They would see that the new version
   of the Gimp adds transparent areas when they use the eraser or when
   they cut a selection.  And then they would be told to flatten their
   image before saving it, because the file format they are using did
   not support transparency.

2) Adding an alpha channel as soon as a second layer is added will
   confuse some users when they see that adding a layer causes the
   eraser or the cut operation to behave differently on the background
   layer (actually, they would behave in the same way as in all other
   layers, but not as they did before when the image was flat).

I don't think that the second case is a real problem.  These users
already know about about layers and transparency, and they will expect
all layers to behave in the same way.  I think that the current
situation is worse, because the background layer behaves differently
from all other layers until you explicitely use "Add Alpha Channel".
If you read the discussion from last year, you will see that I was
surprised to discover that some users used the following solution when
they did not understand why the background layer was behaving
differently: they duplicated the background layer, then deleted the
original one.  For them, it was faster (two mouse clicks) and
(surprisingly) more intuitive than finding "Add Alpha Channel" in some

I think that confusing the first category of users (those who do not
think in terms of layers) would be worse.  Many users are still using
the Gimp as they would use xv, Windows Paint or other simple tools:
they do not care about layers and they do not even know what they
mean.  I have witnessed this when I observed several of my colleagues
using the Gimp.  Although the concept of layers is powerful and simple
to understand, this is simply not necessary for many users (not
professionals, just casual users -- and this is the majority).


Gimp-developer mailing list

Reply via email to