Because automating tedium is the purpose of the machine. Most of what we do is in pursuit of allowing the user to do more with less of their capacity.
The URL itself is a naming abstraction to let users identify things. And it's not even a great one; it has some sharp edges from the end-user's point of view (why does the end user care about protocol? They just want the thing. Why do they care about domain? They just want the thing. Why is the path nine slashes long? The thing they saw yesterday was just "history of mesopotamia;" if they type that in, they should just get the thing). Allowing the bar to accept and intelligently identify multiple different ways to identify things is an increase in capability, not a decrease.
However, what I take issue is with "intelligently identify multiple ways to identify "the thing"" sure, in you're example if someone wants to get to the top search result for "the history of mesopotamia" it might be doing alright...but, who is the one who decided what that search result is, if I type in "history of mesopotamia" and I get shunted directly to the wikipedia entry, maybe that's all ever get, I won't see the thing I really might have wanted which was a specific book about stone tablets of mesopotamia.
The point is, offloading our intelligence to "the machine" only makes as smart or as agenda driven as the machine and its maker, this overall must have some limiting effect. even if doing more with less capacity is good, if the capacity itself is limited to the machine, then it's like saying, I can eat more pie, but the pie is also shrinking.
> Because automating tedium is the purpose of the machine.
Getting people to think of knowing where they are and what they're doing as "tedium" is exactly what The Man wants. Congratulations on your succesful assimilation.
"As we approach the limit of efficiency, movement because becomes so little that it is practically inert, nothing has to move because it's already there, where it is supposed to be" - me