Difference between revisions of "Talk:EN 15907 Identifier"
From filmstandards.org
(Created page with "about the "numeric" attribute: By definition, two identifiers would only need to be compared for equality. Since each identifier is a string of symbols, equality can be assumed ...") |
|||
| Line 3: | Line 3: | ||
By definition, two identifiers would only need to be compared for equality. Since each identifier is a string of symbols, equality can be assumed if each symbol matches that of another identifier in the same place. | By definition, two identifiers would only need to be compared for equality. Since each identifier is a string of symbols, equality can be assumed if each symbol matches that of another identifier in the same place. | ||
| − | Some identifier systems, however, use values that can be represesented in different encodings. Numbers retain their value if they are converted between different number systems (e.g. Roman or Arabic numerals) | + | Some identifier systems, however, use values that can be represesented in different encodings. Numbers retain their value if they are converted between different number systems (e.g. Roman or Arabic numerals) or between number bases (e.g. decimal, octal, binary, hexadecimal, etc.). Moreover, numbers can be represented with groups of digits using different separators where the meaning of the separator can differ between cultures (e.g. 1,000 means one thousand in the Anglo-American world and exactly one in Germany). |
| + | |||
| + | A notorious example for separators in a sequence of digits is the ISBN which can be written with or without groups of digits separated by a dash ("-"). The same is true for GUIDs. Expressing these identifiers without separators in a known number base allows comparison for numeric equality. | ||
| + | |||
| + | N.B.: Several XML-based data exchange specifications use similar attributes for representing "normalised" values along with the native expression of the value. For example, the EAD (Encoded Archival Description) XML schema has a "normal" attribute for the "unitdate" element. | ||
| + | |||
| + | --[[User:Dbalzer|Dbalzer]] 21:17, 9 April 2011 (CEST) | ||
Latest revision as of 19:17, 9 April 2011
about the "numeric" attribute:
By definition, two identifiers would only need to be compared for equality. Since each identifier is a string of symbols, equality can be assumed if each symbol matches that of another identifier in the same place.
Some identifier systems, however, use values that can be represesented in different encodings. Numbers retain their value if they are converted between different number systems (e.g. Roman or Arabic numerals) or between number bases (e.g. decimal, octal, binary, hexadecimal, etc.). Moreover, numbers can be represented with groups of digits using different separators where the meaning of the separator can differ between cultures (e.g. 1,000 means one thousand in the Anglo-American world and exactly one in Germany).
A notorious example for separators in a sequence of digits is the ISBN which can be written with or without groups of digits separated by a dash ("-"). The same is true for GUIDs. Expressing these identifiers without separators in a known number base allows comparison for numeric equality.
N.B.: Several XML-based data exchange specifications use similar attributes for representing "normalised" values along with the native expression of the value. For example, the EAD (Encoded Archival Description) XML schema has a "normal" attribute for the "unitdate" element.
--Dbalzer 21:17, 9 April 2011 (CEST)