[plt-scheme] alpha-colors in universe/world

PS: be sure to explore test cases carefully with him if you do the
below. He'll find that some of the pixels in a circle already have
fractional alpha values (ie, not 0 and not 255). The pixels around the
edge of the circle.

Robby

On Wed, Sep 30, 2009 at 7:41 PM, Robby Findler
<robby at eecs.northwestern.edu> wrote:
> On Wed, Sep 30, 2009 at 5:35 PM, Jordan Johnson <jmj at fellowhuman.com> wrote:
>>
>> Hi all,
>>
>> One of my particularly enthusiastic students discovered the
>> documentation for alpha-colors in the image.ss teachpack, and
>> subsequently discovered that the image primitives (triangle, circle,
>> ...) don't accept alpha-colors.
>>
>>  * Is there anything I can do to get his hands onto a set of primitives
>>   that DO accept the alpha-colors and that are still compatible with
>>   the HtDP approach?
>
> He can write them! You'd compose a regular function, say the circle
> function, with a conversion to alpha colors and the you'd multiply in
> the alpha value you wanted and convert it back to a image.
>
> Not exactly efficient, but perhaps more enlightening?
>
>>  * Is this something that will likely change in the spiffy new version
>>   that's coming out sometime in the next few months?
>
> Yeah, sure. I can try to build that in this time (making it more
> efficient). I've also considered a function that, given an image,
> returns another image with a different transparency value (applied to
> the whole image). I think that's possible in the low-level toolkit.
> Thanks for the suggestion!
>
>>  * If not either of those, can you suggest any answer I can give that's
>>   better than "sorry, not possible"? :)
>>
>> This is a youngster I *really* would like to inspire to do neat things
>> with DrScheme, since he's bright, highly motivated, and already pushing
>> the limits of what we can do (or at least what limits I'm aware of) with
>> the image libraries -- and politically, this would be a nice thing to
>> get right, since a parent is a programmer/professor who is both highly
>> involved with the school and highly skeptical of my use of DrScheme
>> here.
>
> Good luck with that. But, I have to say if it comes down to the
> particular set of image functions that drscheme supports, then I'm not
> sure I'd have a lot of respect for this parent/programmer/professor's
> opinion. :(
>
> Robby
>