If the belief is true, would it not then be knowledge?
Isn't the difference between belief and knowledge the fact of truth and justification (reasonable and necessary plausible assertions/evidence/guidance [italics taken from wiki]) a for the idea held?
But where do you draw the line?
Using Moegreche's example, does the man who
believes his directions to be accurate truthfully possess knowledge even if they were given to him by a friend--whom he deems trustworthy--but never personally verified? To take it a step further, what if the directions were given by a friends' friend, and tested only by the friends' friend. Does that still count as knowledge?
We could circumvent this question by defining knowledge as
empirical data that can be proven true by the individual. However, this creates another problem because by this definition a man could claim knowledge that, when tested by the individual, proves false. But if knowledge must also be
true, then we now know the man never possessed knowledge in the first place. Thus, the definition of knowledge must change to
empirical data that has been proven true by the individual.
This brings us to the crux of the question:
where are we measuring the value of knowledge? We're asking whether
someone else's knowledge is more valuable than
someone else's belief, but if my definition of knowledge is correct--and I'm not necessarily saying it is--then until the answer-seeking individual proves the answer true, it's not knowledge. Thus, to the answer-seeking individual, directions based on another's knowledge equate to directions based on another belief. To the asker, they are
both a form of belief until proven true, although some beliefs (as defined in this sense) might have a much, much higher chance of being truthful.
Taking this into account we could argue that, to an individual, knowledge is more valuable than belief because it will already have been proven true by that individual. Otherwise, it's not knowledge. However, I think I may have taken a rather divergent path from the original question, so I'll attempt to bring it back to the comparison of
external knowledge and true belief.
In a sense, we already know that knowledge is more valuable than belief. If belief were always true it wouldn't
be belief. And once the belief proves true, it becomes knowledge rather than belief, even if it's not labeled as such by the believer.
Even so, the value of one or the other is, as always, dependent on the circumstance. If we're seeking directions or specific factual information then knowledge is always more valuable than belief because, even if the belief is true (which isn't always the case), the knowledge had already been proven true by the information giver, thus removing any uncertainty.
Conversely, if we're lost in the desert with two men, one of whom (because this is philosophy and we're not very good at character development) only answers in what he knows, belief might be more valuable than knowledge. This, I think, is the point Salvidian was trying to make. We could ask the factual man to tell us where to search for an oasis, but he'll only be able to tell us he doesn't know where it is (and that it's currently in our line of sight). However, the other man might believe he knows where an oasis is and lead us in that direction. Even if he was wrong, we'd still be doing
something. In this situation, belief is more useful and thus, more valuable.
It's kind of like comparing abstract apples and oranges. Sometimes we might prefer one, other times we might prefer another, and the overall value of each is determined by the value placed on each by the individual, as well as the value placed on the situations in which one or the other might prove more useful.