Actually, if you look at the definition of "currency" from various sources, you'll come across stuff like this:
"1. Money in any form when in actual use as a medium of exchange, especially circulating paper money."
(http://www.thefreedictionary.com/currency)
... and for "money":
"1. A medium that can be exchanged for goods and services and is used as a measure of their values on the market, including among its forms a commodity such as gold, an officially issued coin or note, or a deposit in a checking account or other readily liquefiable account."
(http://www.thefreedictionary.com/Money)
The problem doesn't seem to be that the definition is incorrect; the problem seems to be that people assume their own definition based on what they're used to, rather than adhering to something you'd find in a dictionary - commonly the motivation appears to be using "it's not a currency" as a (fallacious) reason to express their dislike of something like Bitcoin.
This idea that a currency needs all features of historical currencies to be considered a currency is ridiculous, and doesn't accommodate a "start-up" currency which is something we've never seen before - at least not globally. At the very least the definition of currency ought to change to something like: "A currency possesses at least 3 out of 5 of these qualities."
None of these qualities are permanent, unchanging qualities. That seems to be the other critical mistake economists are making - failing to see the forest for the trees. That comes from an institutional mindset where only governments (or government-connected organizations) can issue currency and so currency has the aura of permanence.