All metadata is text-based; in that, it is composed of characters from a well-defined character set or sets. To ensure consistency when copying or moving data, the same international standard as used for XML, which is UTF-8, should be used minimally. There is UTF-16 and other supersets of UTF-8 that can also be used.
As was covered in Chapter 1, What is Unstructured Data?, data that is stored as raw text is in effect, unstructured. This is due to the fact that there are no rules or controls that govern it. Though text-only metadata is flexible in its entry, it is easy to introduce errors. A good example that most museums encounter is representing the date in a text field.
In the following list, determine the actual dates:
12-Dec-01
10/11/12
19 June
30th February 2010
Februry 10th 1870
50-60a.d.
In addition, how easy would it be to do a date search range on these values?
The following are the issues identified with the previous dates:
It is not clear what the year is. Is it 2001, 1901, 1801...