cursor_image

[slashell(at)partners.org (Sean La Shell)]

IDL help for Cursor_image keyword says:

Specifies the cursor pattern. The value of this keyword must be a
16-line by 16-column bitmap, contained in a 16-element short integer
vector. The offset from the upper left pixel to the point that is
considered the hot spot can be provided via the CURSOR_XY keyword.

I assumed that each of the elements of the vector could be determined
by writing the decimal equivalent of a binary number where the binary
number specified which of the pixels in that row should be set and
which should not.  I further assumed that the least significant bit
would be the rightmost bit:
32768,16384,8192,4096,2048,1024,512,256,128,64,32,16,8,4,2,1

Thus, to set only the third pixel from the left on a given row, one
would specify the number 8192 for that row.

What I seem to have discovered is that the order of the bits is:
512,1024,2048,4096,8192,16384,32768,1,2,4,8,16,32,64,128,256

Thus, to set only the third pixel from the left on a given row, one
would specify the number 2048 for that row.

Is this correct?  If so, why??????????  This arrangement makes no
sense to me.

Looking for sense in a crazy software environment,
Sean