Re: Question on roadway weather event standards

NTCIP 1204 was developed largely based on WMO BUFR and I believe we still have many references to the BUFR codes within the standard. I believe this standard is used for virtually all modern deployments of roadside weather stations in the US and in many other countries. While the data is not presented as an “ontology”, it is defined as an SNMP MIB within the standard and inferring the ontology is a fairly straight forward process from there.

I am happy to convert this into a formal ontology, but this raises two important questions:
1. What conventions should we use to document the ontology?
2. How do we manage the copyright on the generated material?

I propose the following answers:
1. I believe we need both a formal documentation language (e.g., OWL) as well as a graphical notation to facilitate human consumption. Assuming we use OWL as the formal language, I would suggest the Ontology Definition MetaModel (ODM) provides a reasonable graphical equivalence, but I find a few of its notations rather cumbersome and unnecessarily space consuming. I am happy to consider alternatives, but I think one of our first priorities should be to decide on how we will convey the model and we should strive to be as consistent as possible.

2. The nature of a large scale ontology is that there will be many, many experts that need to contribute their information to the ontology. IMHO, it is unreasonable to assume that any one standards development organization (SDO) will really own this ontology, it should be considered a public resource. From that perspective, I propose that it be permanently hosted on a public Github site and we establish rules on how we (i.e., multiple SDOs) formalize and approve content  into the “approved” branch. If the effort is to become successful, we have to have buy-in from a large community and the only way that will happen is if they all believe they have a sense of ownership and input into the final product.


Regards,
Ken Vaughn

Trevilon LLC
6606 FM 1488 RD #148-503
Magnolia, TX 77354
+1-936-647-1910
+1-571-331-5670 cell
www.trevilon.com

> On Feb 18, 2020, at 10:51 AM, George Percivall <gpercivall@ogc.org> wrote:
> 
> Ted,
> 
> You are correct that OGC, TC211 and W3C have observation and measurement ontologies.  Development of roadway weather observations would do well to use those mature standards as they have been deployed around the globe in sensor webs.   Unfortunately, I am not aware of an existing ontology for roadway weather observations.
> 
> An ontology for roadway weather observations could review the aviation weather information models, e.g.., the WXXM standards for aviation weather developed by WMO and ICAO/FAA/EUROCONTROL using ISO/OGC standards.  The aviation Weather Information Exchange Models standards could be an input to a road surface moving vehicle weather event that included Observations and Measurements.
> 
> George
> 
> 
> 
> Weather Information Exchange Models
> The Weather Information Exchange Model specifications support the data-centric environment. It supports MET information collection, dissemination and transformation throughout the data chain. The 3-tiered model (WXCM-WXXM-WXXS) is referred to as a single entity, the term used is 'WXXM'.
> http://wxxm.aero/ <http://wxxm.aero/>
> 
> WXXM implementation status  https://www.ofcm.gov/groups/OD/meetings/workshop/10a-iwxxm_faa.pdf <https://www.ofcm.gov/groups/OD/meetings/workshop/10a-iwxxm_faa.pdf>
> Schema  http://schemas.wmo..int/iwxxm/3.0/ <http://schemas.wmo.int/iwxxm/3.0/>
> 
> OGC Testbed-14: Semantically Enabled Aviation Data Models Engineering Report
> http://docs.opengeospatial.org/per/18-035.html <http://docs.opengeospatial.org/per/18-035.html>
> 
> 
> WXXM https://en.wikipedia.org/wiki/WXXM_(data_model) <https://en.wikipedia.org/wiki/WXXM_(data_model)>
> iWXXM https://en.wikipedia.org/wiki/IWXXM <https://en.wikipedia.org/wiki/IWXXM>
> 
> 
>> On Feb 18, 2020, at 10:51 AM, Ted Guild <ted@w3.org <mailto:ted@w3.org>> wrote:
>> 
>> Moving discussion to public-transportation-data mailing list that I
>> created and announced last week for this sort of topic.
>> 
>> https://lists.w3.org/Archives/Public/public-automotive/2020Feb/0060.html <https://lists.w3.org/Archives/Public/public-automotive/2020Feb/0060.html>
>> 
>> On Tue, 2020-02-18 at 15:22 +0000, Rittmuller, Robert (VOLPE) wrote:
>>> No objections! I agree, this would be well served by having the
>>> conversation moved over to the mailing list. Thanks again for
>>> elevating this up to the right level. 
>>> 
>>> Thanks,
>>> RJ
>>> 
>>> On 2/18/20, 10:21 AM, "Ted Guild" <ted@w3.org> wrote:
>>> 
>>>    RJ,
>>> 
>>>    Any objection to me moving this over to the new mailing list?
>>> 
>>>    George or Clemens,
>>> 
>>>    OGC and TC211 both have observations ontologies. Roadway weather
>>>    event/conditions may be served with combination of the route
>>> ontology
>>>    Clemens is working on, observations over time (TimeseriesML) and
>>> NTCIP
>>>    1204. That would follow a core/modular approach over monolithic.
>>> 
>>>    https://www.opengeospatial.org/pressroom/pressreleases/3158
>>> 
>>> 
>>> https://www.ntcip.org/wp-content/uploads/2018/11/NTCIP1204v0308r3withErrata.pdf
>>> 
>>>    On Mon, 2020-02-17 at 09:28 -0600, Kenneth Vaughn wrote:
>>>> While I am unaware of a complete data model for road-weather
>>> per se
>>>> (e.g., a timeline analysis), NTCIP 1204 does provide data
>>> definitions
>>>> for reporting current road-weather conditions. If this is of
>>>> interest, I could easily turn this into a UML class diagram and
>>> post..
>>>> 
>>>> Regards,
>>>> Ken Vaughn
>>>> 
>>>> Trevilon LLC
>>>> 6606 FM 1488 RD #148-503
>>>> Magnolia, TX 77354
>>>> +1-936-647-1910
>>>> +1-571-331-5670 cell
>>>> www.trevilon.com
>>>> 
>>>>> On Feb 14, 2020, at 1:20 PM, Megan Katsumi <
>>> katsumi@mie.utoronto.ca
>>>>>> wrote:
>>>>> 
>>>>> Thanks, Ted!
>>>>> Hi RJ,
>>>>> To comment from the ISO JTC1 side of things - we don't
>>> currently
>>>>> have weather conditions in our ontology, but that was
>>> definitely an
>>>>> item that we had marked for future work. Would definitely be
>>>>> interested in seeing  the model that Volpe is working on, as
>>> well
>>>>> as if other groups have addressed this. As Ted mentioned,
>>> maybe
>>>>> this might be a good question to post to the new mailing
>>> list.
>>>>> 
>>>>> Best,
>>>>> Megan
>>>>> 
>>>>> On Fri, Feb 14, 2020 at 11:40 AM Rittmuller, Robert (VOLPE) <
>>>>> robert.rittmuller@dot.gov> wrote:
>>>>>> Thanks Ted, here is some additional background on what the
>>> Volpe
>>>>>> team is looking for.
>>>>>> 
>>>>>> The machine learning / AI team here at Volpe is developing
>>> a
>>>>>> weather detection model specifically for the SHRP2
>>> Naturalistic
>>>>>> Driving Study data set. The purpose of this feature
>>> detection
>>>>>> model will be to assist transportation safety researchers
>>> in
>>>>>> being able to identify adverse weather conditions for trips
>>> as
>>>>>> recorded in the SHRP2 Roadway Information Database (RID).
>>> We are
>>>>>> in the formative stages of developing the data
>>>>>> annotation/labeling strategy and are seeking input on what
>>>>>> standards might be informative to the development of those
>>>>>> labels. The core concept is to be as standards-based as
>>> possible
>>>>>> with our data labeling strategy for this model so it has
>>> high
>>>>>> utility for transportation researchers working with SHRP2
>>> data. 
>>>>>> 
>>>>>> Thanks,
>>>>>> RJ
>>>>>> 
>>>>>> On 2/14/20, 10:50 AM, "Ted Guild" <ted@w3.org> wrote:
>>>>>> 
>>>>>>    RJ,
>>>>>> 
>>>>>>    As you'll see in separate email, your inquiry and
>>> Megan's
>>>>>> recent
>>>>>>    prompting is encouraging me to hasten a soft launch of
>>> this
>>>>>> group
>>>>>>    instead of waiting for all the formalities.
>>>>>> 
>>>>>>    Adding some specific people to this message to see if
>>> we can
>>>>>> get any
>>>>>>    guidance, perhaps moving the conversation over to the
>>> new
>>>>>> mailing list.
>>>>>> 
>>>>>>    What sort of data models exist, under development or
>>>>>> consideration for
>>>>>>    roadway weather conditions/events? 
>>>>>> 
>>>>>>    On Fri, 2020-02-14 at 13:22 +0000, Rittmuller, Robert
>>> (VOLPE)
>>>>>> wrote:
>>>>>>> Thanks! I did look at some of the SmartCities work
>>> but ran
>>>>>> into
>>>>>>> mostly the same information/situation I was finding
>>>>>> elsewhere. For
>>>>>>> what we are doing, conflating some of the common
>>>>>> designations into a
>>>>>>> common core label structure will work as the model we
>>> are
>>>>>> developing
>>>>>>> is going to be multi-label. Our target data set is
>>>>>> currently the
>>>>>>> SHRP2 NDS data so that limits the scope to where I
>>> don't
>>>>>> think we
>>>>>>> will do much damage if there is no accepted common
>>>>>> standard. I'll
>>>>>>> keep an eye on what's developing in this space so at
>>> some
>>>>>> point in
>>>>>>> the future we can incorporate common standards-based
>>>>>> designations. 
>>>>>>> 
>>>>>>> -RJ
>>>>>>> 
>>>>>>> On 2/13/20, 7:23 PM, "Ted Guild" <ted@w3.org> wrote:
>>>>>>> 
>>>>>>>    Hi RJ,
>>>>>>> 
>>>>>>>    I don't and still relatively new at geospatial
>>> ontology
>>>>>> area.
>>>>>>> There is
>>>>>>>    a mix of things going on in ISO's Intelligent
>>>>>> Transportation
>>>>>>> Systems
>>>>>>>    (ITS), SmartCities and Geospatial activities.
>>> Some
>>>>>> actually gets
>>>>>>>    implemented. Open Geospatial Consortium actually
>>> gets
>>>>>> things out
>>>>>>> there
>>>>>>>    and implemented and I would suggest poking around
>>> there
>>>>>> first,
>>>>>>> ITS
>>>>>>>    second. I have had some access inside OGC plus
>>> they
>>>>>> have a fair
>>>>>>> amount
>>>>>>>    public and recently got ITS.
>>>>>>> 
>>>>>>>    I'm starting to form a cross standards body
>>>>>> coordination
>>>>>>> committee but
>>>>>>>    it isn't up and functional yet as this would be a
>>>>>> perfect
>>>>>>> question
>>>>>>>    there. 
>>>>>>> 
>>>>>>>    On Thu, 2020-02-13 at 21:27 +0000, Rittmuller,
>>> Robert
>>>>>> (VOLPE)
>>>>>>> wrote:
>>>>>>>> Ted, a completely random non-cybersecurity
>>> question
>>>>>> for you! I
>>>>>>> have a
>>>>>>>> project that’s currently looking to train a
>>> roadway
>>>>>> weather
>>>>>>> event
>>>>>>>> classifier for forward-facing video data sets.
>>> I’ve
>>>>>> done some
>>>>>>>> research into what standards might exist for
>>>>>> developing a
>>>>>>> roadway
>>>>>>>> weather label set but keep coming up with the
>>> usual
>>>>>>> meteorological
>>>>>>>> resources. My question is this; are you aware
>>> of any
>>>>>> standard
>>>>>>>> classifications of roadway weather that might
>>> serve
>>>>>> as a good
>>>>>>> base
>>>>>>>> label set for something like this? I’m really
>>> keen on
>>>>>> something
>>>>>>> that
>>>>>>>> is directly automotive and has wide
>>> applicability for
>>>>>>> researchers who
>>>>>>>> might be performing searches on these types of
>>> video
>>>>>> data sets.
>>>>>>>> 
>>>>>>>> Thanks,
>>>>>>>> Robert Rittmuller, PMP CISSP VCP
>>>>>>>> Safety Information Systems Division | V-311
>>>>>>>> Volpe Center | U.S. Department of
>>> Transportation
>>>>>>>> ( O: 617.494.3634 | * robert.rittmuller@dot.gov
>>> |
>>>>>>> 8   www.volpe.dot.g
>>>>>>>> ov
>>>>>>>> ( M: 978.801.1376
>>>>>>>> Advancing transportation innovation for the
>>> public
>>>>>> good
>>>>>>>> “The future always wins. Know that the “way
>>> we’ve
>>>>>> always done
>>>>>>> it”
>>>>>>>> is almost certain to become obsolete—and soon.”
>>>>>>>> 
>>>>>>>    -- 
>>>>>>>    Ted Guild <ted@w3.org>
>>>>>>>    W3C Automotive Lead
>>>>>>>    https://www.w3.org/auto
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>    -- 
>>>>>>    Ted Guild <ted@w3.org>
>>>>>>    W3C Automotive Lead
>>>>>>    https://www.w3.org/auto
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>> 
>>>>> 
>>>>> -- 
>>>>> Megan Katsumi, PhD, MASc
>>>>> Postdoctoral Fellow
>>>>> Enterprise Integration Laboratory
>>>>> Transportation Research Institute
>>>>> University of Toronto
>>>> 
>>>> 
>>>    -- 
>>>    Ted Guild <ted@w3.org>
>>>    W3C Automotive Lead
>>>    https://www.w3.org/auto
>>> 
>>> 
>>> 
>> -- 
>> Ted Guild <ted@w3.org>
>> W3C Automotive Lead
>> https://www.w3.org/auto
>> 
> 
> 
> Keep up with all the OGC news by signing up to our quarterly newsletter at http://newsletter.opengeospatial.org <http://newsletter.opengeospatial.org/>
> 
> Interested in attending the next OGC Technical and Planning Committee Meeting? Find out more at http://www.ogcmeet.org <http://www.ogcmeet.org/>