Re: Improvement of www.schema.org/menu

Hi Ralph,

Thanks for the update.  Glad to hear that you intend to support UN/ECE Rec
20 codes. I hope my suggestion about a bidirectional mapping makes sense.

Let me know directly how you might like some help filling in the gaps in
coverage and I'll try to help if I can.

Best wishes

Mark

On 16 Jan 2017 21:39, "Ralph TQ [Gmail]" <rhodgson@topquadrant.com> wrote:

> Hi Mark,
>
> I and my QUDT colleagues welcome collaboration on this work. There are
> some important naming conventions we are following for all units of measure
> and for dimensional analysis. Mappings to UN/ECE and UCUM have been done to
> different levels of completeness.  QUDT Schema R2.0 has the properties that
> are needed (see below).  These mappings were in some cases performed by
> getting data into RDF and then running SPIN rules for transformations.
>
> QUDT work has started this year with more help from other parties. Also
> QUDT.org <http://qudt.org> is going through the membership process with
> W3C.  One motivation for this is the high interest in the W3C SHAPES SHACL
> standard. SHACL is of interest for constraint checking of RDF graphs in IOT
> projects and other industrial and finance application areas.
>
> I will check the status of the mappings with other QUDT editors and give
> an you an update in due course.
>
> Regards,
>
>
> Ralph Hodgson, @ralphtq <http://twitter.com/ralphtq>
>
> CTO, TopQuadrant, Inc., www.topquadrant.com @TopQuadrant
> <http://twitter.com/topquadrant>
> *cell: +1 781-789-1664 <(781)%20789-1664> / fax: 703 299-8330
> <(703)%20299-8330> / main: 919 300-7945 <(919)%20300-7945>*
> *TQ Blog:* *The Semantic Ecosystems Journal*
> <http://topquadrantblog.blogspot.com/>,  and
> <https://www.linkedin.com/company/topquadrant>
> <https://twitter.com/TopQuadrant>
> <https://twitter.com/TopQuadrant>
>
> On Jan 16, 2017, at 7:35 PM, Mark Harrison <mark.harrison@cantab.net>
> wrote:
>
>
>
> On Mon, Jan 16, 2017 at 6:34 PM, Wes Turner <wes.turner@gmail.com> wrote:
>
>>
>>
>> On Monday, January 16, 2017, Mark Harrison <mark.harrison@cantab.net>
>> wrote:
>>
>>> Sorry to be pedantic ;-) , but in fact, the SI base unit for mass is the
>>> kilogram.
>>> See
>>> https://en.wikipedia.org/wiki/International_System_of_Units#Base_units
>>> and http://physics.nist.gov/Pubs/SP330/sp330.pdf (section 2.1.1.2) and
>>> http://www.bipm.org/en/publications/si-brochure/kilogram.html
>>>
>>> The gram is the base unit for mass in the old CGS system - but not in
>>> the modern SI system.
>>>
>>
>> Good catch. I guess that makes standarization on unit specifiers all the
>> more important.
>>
>>
>>>
>>> QUDT will be more useful and more complete when version 2.0 is finally
>>> released - and has very useful machine-readable triples for expressing the
>>> dimension of each unit (is it a mass, a length, etc.) and conversion
>>> factors and offsets between units that belong to the same physical
>>> dimension (e.g. to convert between various units of mass or between various
>>> units of length).
>>>
>>> In GS1 and the GS1 web vocabulary, for the value of
>>> http://gs1.org/voc/unitCode we use a string value indicating a
>>> Measurement Unit from UN/ECE Recommendation 20 code tables, e.g. GRM for
>>> gram, KGM for Kilogram, MGM for milligram and MC for microgram.  A 2005
>>> edition of the code tables is available at http://www.unece.org/filead
>>> min/DAM/cefact/recommendations/rec20/rec20_rev3_Annex3e.pdf
>>>
>>
>> .unit=(unitStr/URI, unit system)
>>
>>
>>>
>>> Personally, I'd much prefer the QUDT approach but industry does
>>> currently use the UN/ECE Rec 20 code tables for expressing units of
>>> measure, even if some of these UN ECE code strings are completely opaque
>>> and non-intuitive, e.g. 28 = kilogram per square metre.
>>>
>>>
>> - Are there mappings from the UN/ECE codes to labels and URIs?
>> - Are there mappings from the UN/ECE system to QUDT?
>>
>
>
> As far as I am aware, the UN/ECE codes are only published in spreadsheets
> or PDF documents.
> I've never seen any RDF datasets from UN/ECE that map from UN/ECE codes to
> anything useful such as labels, conversion factors and offsets or QUDT URIs.
>
> Good Relations do have this useful table on their site:
> http://wiki.goodrelations-vocabulary.org/Documentation/UN/
> CEFACT_Common_Codes
> but it appears to be just an HTML table without any inline markup and
> appears to be for the most frequently used units - only a subset of the
> entire UN/ECE Rec 20 common code table.
>
> The v1.1 QUDT units resource at http://qudt.org/vocab/unit# appears to be
> offline and I can't find a complete corresponding v2.0 QUDT resource for
> the units - it looks like they have published their schema so far [ see
> http://www.qudt.org/release2/qudt-catalog.html and
> http://qudt.org/doc/2016/DOC_SCHEMA-QUDT-v2.0.html ], and some datasets
> for some units - but certainly not for all units - nor did they release the
> dataset for SI units first.  Various units datasets are still in progress
> or in quality control, according to http://www.qudt.org/
> release2/qudt-catalog.html .
>
> One example of a QUDT 2.0 units resource is http://qudt.org/vocab/unit/
> KG-PER-MOL
> Looking at that, I don't see a triple that links that resource to even the
> string UN/ECE Rec 20 code string, which is D74 in the case of kilogram per
> mole.
>
> I think QUDT 2.0 will be a very useful resource for everyone when it is
> complete and online.
> Not sure whether they're under-resourced and could appreciate some help to
> complete and test the units dataset.
>
> What could be a very useful extension of QUDT 2.0 is if they can provide a
> mapping to/from the UN/ECE Rec 20 common code string in both directions so
> that we could have triples such as
>
> http://qudt.org/vocab/unit/KG-PER-MOL   owl:sameAs
> http://qudt.org/vocab/unece/D74 .
>
> appearing within the triples for each resource within the
> http://qudt.org/vocab/unit/ namespace
>
> and also a dataset at a new namespace, for example:
> http://qudt.org/vocab/unece/
>
> providing triples such as
>
> http://qudt.org/vocab/unece/B15  owl:sameAs  http://qudt.org/vocab/unit/J-
> PER-MOL .
> http://qudt.org/vocab/unece/D74  owl:sameAs  http://qudt.org/vocab/unit/KG
> -PER-MOL .
>
>
> When QUDT 2.0 is complete and online, it could then be appropriate for
> schema.org, GoodRelations, the GS1 web vocabulary to introduce a new
> dedicated property schema:qudtUnit  etc. that expects a URI from within the
> http://qudt.org/vocab/unit/ <http://qudt.org/vocab/unit/J-PER-MOL>
>  namespace.
>
> I've copied Ralph Hodgson on this e-mail thread, since he is one of the
> original developers or QUDT and is still on their board of directors.  He
> and I exchanged a couple of emails along these lines around 18 months ago.
> Now that there are some examples of what QUDT 2.0 units resources look
> like, some of us might be able to help QUDT fill in some of the gaps.
>
>
>
>
>
>>
>>>
>>>
>>>
>>> On Mon, Jan 16, 2017 at 4:27 PM, Wes Turner <wes.turner@gmail.com>
>>> wrote:
>>>
>>>> http://schema.org/NutritionInformation
>>>>
>>>> http://schema.org/servingSize r: Text
>>>> "The serving size, in terms of the number of volume or mass"
>>>>
>>>> Other NutritionInformation attributes have a r:ange of Mass.
>>>>
>>>> - Does this suggest a need for a Volume class?
>>>> - Could/should the servingSize range be Quantity?
>>>>
>>>> - Should Quantity have a 'unit' property with r: URL?
>>>>   http://schema.org/Quantity
>>>>
>>>>   - QUDT defines URLs for many (powers of) physical units
>>>>     - Unfortunately, there are a number of vocabularies for physical
>>>> units
>>>>   - The SI unit for Mass is always g(ram)
>>>>
>>>> ...
>>>>
>>>> https://wrdrd.com/docs/consulting/units#rdf-and-units
>>>>
>>>> https://wrdrd.com/docs/consulting/linkedreproducibility#csv-
>>>> csvw-and-metadata-rows ... "Table with 7 metadata header rows"
>>>>
>>>> On Monday, January 16, 2017, Dan Brickley <danbri@google.com> wrote:
>>>>
>>>>> On 15 January 2017 at 07:42, Xavier Gonsalves <axv4444@gmail.com>
>>>>> wrote:
>>>>> > Many have talked and requested about this but w3 seems to avoid it.
>>>>>
>>>>>
>>>>>
>>>>> > Schema should add more properties under restaurant menus like dish
>>>>> price,
>>>>> > cuisine, spiciness, dish name, ingredients, veg, nonveg, vegan
>>>>> category,
>>>>> > description .etc.. so that search engines can implement the
>>>>> following in the
>>>>> > future:
>>>>> >
>>>>> > https://searchenginewatch.com/sew/news/2328869/google-tests-
>>>>> restaurant-menus-in-card-results/
>>>>> >
>>>>> > It can be ordered such that these properties can be put on the
>>>>> webpage of
>>>>> > the URL of the menu.
>>>>> >
>>>>> > Please look into it ASAP.
>>>>>
>>>>> Please comment on the draft at http://webschemas.org/MenuItem in
>>>>> Github, https://github.com/schemaorg/schemaorg/issues/1288
>>>>>
>>>>> Dan
>>>>>
>>>>>
>>>
>
>