Are you absolutely certain? I'm currently detecting a hash:
http://www.beautyoftheweb.com/#/highlights/all-around-fast
^-- this spot
But let's assume there are URLs out there that don't contain them: What kind of issues are you encountering elsewhere? In my private network application, I utilize Really Simple History for managing history, and one can effortlessly navigate back and forth between URLs regardless of whether they have the hash, are app URLs without the hash, or completely different page/app URLs. (Even if you bookmark pages with hashes, the app will return to the correct location when revisiting.) This works in my app by checking the initial hash on load and reloading the state as necessary. (I do not depend on — or even utilize — RSH's data storage feature, just the hash manipulations.)
Edit I just tested out the live demo of the first jQuery history plugin I stumbled upon quickly. It managed going back and forth between "hashed" and "unhashed" URLs seamlessly. I visited the "#2" link, then modified the URL in the address bar to remove the hash entirely and hit Enter. This brought me to the correct page. Then clicking Back took me back to the "#2" page successfully. In fact, if you go here, I am directly linking to the "#2" link, and it loads its state appropriately. I believe you may be experiencing some inexperienced use of history libraries where the developer hasn't handled the initial loading process adequately.