Enigmania has been posting on the sky's being blue, including a post titled The sky's blue, therefore it's an object. Oddly, that title may be an important insight in and of itself. "Colored things are objects" seems like a paradigm case of a necessary truth, knowable a priori, yet the sky seems not to fit many of our common-sense ideas of what objects are. It certainly doesn't fit the common-sense assumptions that used to be made of it, namely that it's a solid dome.
This catches my interest, partly because I've been reading Laurence Bonjour's In Defense of Pure Reason, a modern defense of the idea that there are substantiative a priori claims. That in and of itself seems reasonable enough, as I don't see any alternative to that except Humean skepticism, and I'm pretty clearly convinced that skepticism is self-defeating. However, one of Bonjour's main examples bothers me: the idea that no object can be both red and green all over. Scientifically-minded person that I am, I immediately reflected on the fact that the property of the object is to emit light in a certain way, and what we think of as the redness or greenness of the object is a product of how our minds interact with the object. It is possible for the light-waves to be overlaid, though our minds wouldn't give us simultaneously red and green experiences in that case--I think it would be more of a yellowish orange (I'm trying to remember here what it's like to see green and red diodes placed very close together). Now, might we have an experience of something that seems to be simultaneously red and green? I honestly don't know. It's inconceivable in so far as I cannot imagine it, but that doesn't mean its impossible.
This isn't a huge blow to Bonjour's specific ideas--he emphasizes that apparently a priori knowledge can be fallible. Still, it's interesting to think about.