The value of a currency at any time is measured by a figure called the Consumer Price Index: this takes into account the price of lots of different things and comes up with a single figure for that currency. This can then be compared between eg the dollar now and the pound now, or the pound now and the pound 5 years ago, or, a defunct currency.
One way to think of this is to imagine a supermarket (in the country and time in question) that sells everything at a standard, average sort of price (perhaps the RRP), and you go to that supermarket with a shopping list. How many of the currency in question would it cost to get the full shopping list? You can sort of go as complicated as you want when calculating the CPI, bringing in more complicated factors such as weighting for different categories of item. But that's the basic idea. If you went to the supermarket now it would cost more than it would 10 years ago to get the full list, and that is how we say there has been "inflation".
Accurately calculating the CPI for a historic currency requires good records of the prices of common goods at that time. Obviously the list of goods will differ (hardly anyone buys wagon wheels any more) but historians can (probably) agree on how it would be calculated for that period, ie which prices are examined and how they contribute to the final figure.