Interesting question, although my answer might be more historical than you'd like ;-)
One answer is that it gives you all the notes of the diatonic scale on the white keys, so by transposing to C major you can play any major-key melody that doesn't modulate using only the white keys.
Another way of saying this: assume that you are working in our musical system, which has twelve-tone equal temperament as the background "system", but within that the diatonic major scale is the most commonly used set of pitches. Then assume that you want to have one particular diatonic scale easy to play, and that you'll put the other pitches on harder-to-reach keys. Subtract the diatonic pitches C-...-B from the set of all twelve pitches and you are left with C#/Db, D#/Eb, F#/Gb, G#/Ab and A#/Bb. Put these "between" the diatonic keys, in the right order in the chromatic scale, and you have something very close to the standard piano keyboard. (You can't add any more "half-steps" between E and F, or between B and C, without expanding your tuning beyond 12-tone equal temperament).
Wikipedia and Grove Music online (subscribers only, unfortunately) note that the original organ keyboards (13th century) had only the pitches of the C major scale, plus B flat, because that made up more or less the entire pitch resources of the religious music sung at that time (and instruments would have been used only for accompanying sung music -- at least in church). On those keyboards B and B flat were both "white keys", with no "black keys" at all. The first surviving organ with a fully chromatic keyboard, from the late 14th century, still has B flat as a "diatonic"/"white" key.
I would guess that as keyboard music developed as its own genre, it became much more useful to be able to play fast runs in the major scale -- lots of early keyboard music is based on existing pieces of vocal music with the addition of fast, "improvisational"-sounding ornaments. At least at first, the chromatic notes would have largely been used in chords, not scales, so it would be an acceptable trade-off to have those keys harder to get to in exchange for being able to play the main scale quickly. Even quite a bit later, around 1600, there are pieces which are written in G major or F major, but where all the fast bits ignore the key signature sharps or flats and just use the diatonic keys -- it was easier to play fast scales on the "white" keys with the technique they used.
Finally, it's worth noting that people have often built keyboards with more than twelve tones to the octave. In the 16th century it was common to have the Eb/D# key "split", with the front half playing one of the two pitches (Eb) and the back half the other (D#). This was done by people who valued having perfectly-tuned chromatic notes over being able to easily navigate all scales at high speed. The extreme of this way of thinking, pre-20th century, is probably the 1555 Archicembalo, which has 36 keys to the octave! And 20th/21st century microtonal musicians have done lots of similar things. There is a nice introduction to different tuning systems both historical and modern at Kyle Gann's page.