Constrain <si:distributionXMLList> und <si:distribution> to Enum Values to ensure correct interpretation
According to the discussion in the DKD Querschnits-Ausschuss https://gitlab1.ptb.de/d-ptb/dcc/TCs/dcc-ua-querschnittsthemen/-/issues/54 I sugest constraining si:distributionXMLList und si:distribution to a redefined Datatype:
<xs:simpleType name="distributionType">
<xs:annotation>
<xs:documentation xml:lang="en">
String providing a distribution (statistical distribution of measurement values).
Only predefined distributions are allowed.
</xs:documentation>
</xs:annotation>
<xs:restriction base="xs:string">
<xs:enumeration value="normal"/>
<xs:enumeration value="uniform"/>
<xs:enumeration value="exponential"/>
<xs:enumeration value="gamma"/>
<xs:enumeration value="beta"/>
<xs:enumeration value="log-normal"/>
<xs:enumeration value="weibull"/>
<xs:enumeration value="chi-squared"/>
<xs:enumeration value="student-t"/>
<xs:enumeration value="f-distribution"/>
<xs:enumeration value="pareto"/>
<xs:enumeration value="cauchy"/>
<xs:enumeration value="rayleigh"/>
<xs:enumeration value="other"/>
<!-- Add additional distribution names as needed -->
</xs:restriction>
</xs:simpleType>
Or similar the actual included distributions needs to be discussed.
if the distribution have been discussed we could develop python /c++ what ever libraries the allow to get the PDFs and draw from this distributions in an standardized way. To make actual use of the knowledge of the distribution. When using str like in the momentary implementation
<xs:simpleType name="distributionType">
<xs:annotation>
<xs:documentation xml:lang="en">
String providing a distribution (statistical distribution of measurement values).
</xs:documentation>
</xs:annotation>
<xs:restriction base="xs:string"/>
</xs:simpleType>
We end with endless possible ways to express the same distribution :
normal,Normal,gauss,Gauss,gausian,Gausian,Нормальное,
resulting in incompatibility of any software using this Information.