Aside from anomalies like Tigere Chiriga's Floating Mug, I don't see a lot of design focus on the shape of glassware as it pertains to function. But here are a couple of ways in which the actual form influences the usage in some non-obvious ways.

During my stint as a bartender I was taught a quick trick for chilling a pilsner glass: You stuck it inside the ice bin, so that ice was both inside and outside the glass, and spun it with your fingers until it was chilled. *(If you're going to try this at home, read the warning at the bottom of this entry.)

However, the pilsner glass was only served with bottled beer; guys ordering drafts got beer steins. And there's no way to apply the spin trick to a stein, as the handle precludes rotation. So you had to pre-chill steins in the 'fridge behind the bar, which only had room for five of the bulky things amidst the other crap you had to keep in there, so only the favored regulars got chilled steins.

I doubt any glassware designer will interview a bartender or server—but I wish they would. The martini glass in particular is badly in need of a re-design. I feel they got knocked over the most and they were the worst things for a server to carry on a tray.

Recently, a University of Bristol study revealed the shape of a glass actually influences how fast people drink what's in it—if there's alcohol involved. As reported in The Economist,

[Study participants were asked] to do one of four things: drink beer out of a straight glass; drink beer out of a flute (a glass whose sides curve outward towards the rim); or drink lemonade from one of these two sorts of glass. To complicate matters further, some of the glasses were full whereas others were half-full....

...[Researchers found that] a full straight glass of beer was polished off in 11 minutes, on average. A full flute, by contrast, was down the hatch in seven, which was also the amount of time it took to drink a full glass of lemonade, regardless of the type of vessel. If a glass started half-full, however, neither its shape nor its contents mattered. It was drunk in an average of five minutes.

Researchers surmised that the reason lemonade was drank at the same pace irrespective of glass shape, is because people are simply drinking it naturally. But the beer times differed because, according to the researchers, people are aware they're drinking booze and are trying to modulate their intake. The fluted glass throws off people's perception of volume; in other words, they cannot accurately estimate when they've hit the halfway mark.

Should the study see widespread exposure, bar owners might be interested in selecting glass shapes that will more quickly move product.

*(Warning about the ice-submersion glass-chilling method: Managers routinely frowned on this practice, because if you were not careful, you might chip the rim of the glass while inserting it in the ice. Meaning now you've got a shard of broken glass hidden in the ice that's potentially going to be served to someone in a drink. Whenever a glass broke anywhere near the ice bin, at my bar it was standard practice to then dump hot water down the ice bin, melt all the ice, search for a broken shard with a flashlight, and get the barback to refill the ice bin—a seriously time-consuming pain in the neck.)