Re: Scientific Measurements

Units of measurement are important for quality data, and IMHO the W3C 
missed the opportunity to include a clean mechanism for units into RDF 
or RDF Schema itself early on. As a result, vocabularies such as QUDT 
were created by 3rd parties (my employer TopQuadrant in this case) and 
RDF developers are still confused what to do.

Maybe can do a better job. Just some random input. With QUDT 
some of the use patterns are to

- use the units as XSD datatype of the RDF literals (this would not work 
with right now, because the datatypes there are a fixed 
enumeration while in RDF this is open-ended).

- use "reified" objects that are pairs of a value plus another property 
for the unit.

- attach the (default) unit to the property itself. This makes 
computations based on those values easiest because no normalization 
needs to be done, but obviously not everyone (e.g. in the US) would be 
happy to convert everything into metric units and prefer to use units 
such as feet.

A 4th option is used by in some places, allowing values to be 
strings consisting of a number plus a unit's abbreviation, e.g. "10 ft".

I do believe that there is no harm (but many benefits) if the 
properties would point to a default unit so that people just need to 
enter a number as the property value. This could be part of a comment, 
but better would be a property attached to the property itself. If 
another unit than the default is needed then the string encoding (option 
4) might be the most pragmatic approach. Engines that use 
data could quite easily translate those values into proper, normalized 
RDF literals before writing them into an RDF triple store. Ontologies 
such as QUDT would be useful for such normalizations because they 
already contain the unit abbreviations and their conversion factors in 
(SPARQL-friendly) machine readable form. I had written up details on how 
to achieve that with SPIN here

In general, I see the role of twofold: for collecting data should make it easy for people to use, and therefore support 
flexibility in how such values are encoded as strings. But for 
processing and storing data, a normalized, strict version of 
can be used.


On 6/6/2013 6:08, William L. Anderson wrote:
> And just to add to the soup there is Catalog QUDT: The QUDT, or 'Quantity, Unit, Dimension and Type' collection of ontologies define base classes, properties, and instances for modeling  physical quantities, units of measure, and their dimensions in various measurement systems.
> -Bill Anderson
> On Jun 5, 2013, at 12:35 PM, Alex Milowski <> wrote:
>> On Wed, Jun 5, 2013 at 10:04 AM, Thad Guidry <> wrote:
>> But that Type stems from eCommerce & Trade ... but it's all primarily based on UNECE codes... and those happen to have Measurements defined even scientific ones if you dig deep enough in it.
>> See the "master list" here:  and scroll down and you will see the UNECE MeasurementUnit sections and others.
>> BTW, poke me in eye with a red hot poker for making me read an XML Schema to find a three letter code.  ;)
>> Too bad they (or doesn't have a flat list of the codes.
>> -- 
>> --Alex Milowski
>> "The excellence of grammar as a guide is proportional to the paucity of the
>> inflexions, i.e. to the degree of analysis effected by the language
>> considered."
>> Bertrand Russell in a footnote of Principles of Mathematics

Received on Wednesday, 5 June 2013 22:30:07 UTC