VOUnits RFC

[Rob Seaman]

Hi Norman and the Semantic Faculties,

> The VOUnits recommendation process trundles on...

I wonder if you have seen the proposed new SI system:

	http://nvlpubs.nist.gov/nistpubs/jres/116/6/V116.N06.A01.pdf

There may be some implications for VOUnits from their seven-fold path.

Also, Andrew Main (zefram of perl) gave a presentation (draft attached) at the recent UTC meeting with his vision of how to support multiple timescales.  He addresses the issue of (the equivalent of) unit conversions:

	"It is worth examining the extent to which multiple dispatch is required. In many non-time conversion situations, all conversions can pivot around a single canonical format, thus breaking the general double-dispatch conversion case into two single-dispatch legs. That is not possible in the general case for time scale conversions. For example, consider the UT1 ↔ UT2 conversion, which is exactly defined for any time for which UT1 is defined. If forced to go via an atomic time scale such as TAI, the conversion would be incorrectly rendered imprecise for any future time (for which the requisite EOP measurements don’t yet exist). Conversely, UTC ↔ TAI would be incorrectly rendered imprecise if forced to go via an Earth rotation time scale. There is no pivot time scale that satisfies both cases, so single dispatch does not suffice." (p. 7)

The assertion is not that all non-time conversions benefit from canonical formats, but only that many do.  Some may not.

> The VOUnits Recommendation still intends _not_ to be innovative, in the sense that it is intended to indicate the _intersection_ of the existing syntaxes for unit strings, as much as possible.


I guess innovative might imply the conveyance of semantics within the format / thesaurus?  Whereas the intent of the recommendation is to convey only syntax and leave semantics for applications to sort out?  Whichever it is, the structure of the conversions should be general enough not to assume that all time-like or mass-like or distance-like conversions are commutative / idempotent / reversible / transitive.

Rob
--

-------------- next part --------------
A non-text attachment was scrubbed...
Name: AAS_13-518.pdf
Type: application/pdf
Size: 186839 bytes
Desc: not available
URL: <http://www.ivoa.net/pipermail/semantics/attachments/20130726/9cfdb4ca/attachment-0001.pdf>
-------------- next part --------------