Re: [Hydro.dwg] [EXTERNAL] Re: json waterml2 / timeseriesml?

The beauty of an automation is we can set it up to repeat as updates come
in.

Better to test your model with example data during the development process
than try to retrofit it.

So as long as we don't push intermediate solutions we can gain early
visibility of which aspects are easily automated and which may require hand
coded solutions, such as adaptation to covjson or other hand built schemas.

I know UML makes this hard, but having an effective pathway to testing json
implementation should allow us to go faster, better.

My experience with a far more complex application schema for cadastre
survey gives me a lot of confidence this will save time and produce a far
better result.




On Sun, 16 Jun 2024, 10:03 Kathi Schleidt, <kathi@datacove.eu> wrote:

> Hi Rob,
>
> maybe slow down just a tad here, we're currently in the process of
> updating TSML to OMS. If you pull the current 1.3 model, you'll run into
> conflicts as depends on both O&M as well as the old coverage models. Paul
> and I have started work on the update, but we're not there yet. I'm worried
> that if you now start work on the old model, we'll end up in another epic
> muddle like we already have on Observation Types.
>
> In addition, based on experiences with the OMS updates, we'd really like
> to do a bit of prototyping before we roll this out, like we've been doing
> via the GeoTech and WQ IEs for the STA 2.0 update
>
> :)
>
> Kathi
> On 15.06.2024 23:27, Rob Atkinson wrote:
>
> Great.  The link I sent is already based on Features, so it should be
> straightforward to set up timeseries and waterml specialisation hierarchy.
>
> Do we have a draft updated tsml UML Model yet? Otherwise I can start with
> the old one and update as required.
>
> On Sat, 15 Jun 2024, 16:48 Kyle Onda, <konda@lincolninst.edu> wrote:
>
>> Looping in Tom Karalidis and Ben Webb here. Due to ongoing work with WMO
>> on WIS2, we will Be embarking on an experimental waterml2 json encoding,
>> delivered over OGC API Features.
>>
>> Get Outlook for iOS <https://aka.ms/o0ukef>
>> ------------------------------
>> *From:* Blodgett, David L <dblodgett@usgs.gov>
>> *Sent:* Thursday, June 13, 2024 4:33 PM
>> *To:* Rob Atkinson <ratkinson@ogc.org>; Kathi Schleidt <kathi@datacove.eu
>> >
>> *Cc:* Paul Hershberg - NOAA Federal <paul.hershberg@noaa.gov>;
>> Hydro.dwg@lists.ogc.org <hydro.dwg@lists.ogc.org>; Schaaf, Hylke van der
>> <hylke.vanderschaaf@iosb.fraunhofer.de>; SDW WG (public-sdw-wg@w3.org) <
>> public-sdw-wg@w3.org>; Simon Cox <scox@ogc.org>; Kyle Onda <
>> konda@lincolninst.edu>; Grellet Sylvain <S.Grellet@brgm.fr>
>> *Subject:* RE: [Hydro.dwg] [EXTERNAL] Re: json waterml2 / timeseriesml?
>>
>>
>> Yeah, Sylvain, an hdwg best practice or hdwg contributed annex is fine by
>> me either way. I just know we’ve seen this need and had no practical
>> blockers for a long long time. How do we make sure it gets done?
>>
>>
>>
>> I’m glad the OMS/TSML work is moving ahead. Is there a plan for a json
>> encoding as part of that? I don’t see any activity on github.
>>
>>
>>
>> *From:* Rob Atkinson <ratkinson@ogc.org>
>> *Sent:* Thursday, June 13, 2024 1:45 PM
>> *To:* Kathi Schleidt <kathi@datacove.eu>
>> *Cc:* Paul Hershberg - NOAA Federal <paul.hershberg@noaa.gov>;
>> Hydro.dwg@lists.ogc.org; Schaaf, Hylke van der <
>> hylke.vanderschaaf@iosb.fraunhofer.de>; SDW WG (public-sdw-wg@w3.org) <
>> public-sdw-wg@w3.org>; Simon Cox <scox@ogc.org>; Kyle Onda <
>> konda@lincolninst.edu>; Blodgett, David L <dblodgett@usgs.gov>; Grellet
>> Sylvain <S.Grellet@brgm.fr>
>> *Subject:* Re: [Hydro.dwg] [EXTERNAL] Re: json waterml2 / timeseriesml?
>>
>>
>>
>> it should be trivial to define waterml as a profile of timeseriesml
>> schema - this is why we have been developing the building blocks tooling to
>> support transparent extensibility.:-)
>>
>>
>>
>> we just define the waterML property set to require waterml extensions and
>> the core timeseries
>>
>>
>>
>> allOf:
>>
>>  - $ref: {timeseriesML schema}
>>
>> - properties:
>>
>>    waterml_prop1:
>>
>>      .{scheme def}
>>
>>
>>
>> anf then all examples will be tested against both, and the JSON-LD
>> context will be transitively inherited from timeseries (and sosa
>> underlying that!)
>>
>>
>>
>> also SHACL rules for logical consistency get inherited
>>
>>
>>
>>
>> *Rob Atkinson*
>>
>> *Senior Research Engineer ** | Open Geospatial Consortium (OGC)*
>>
>> Mobile: +61 419 202973
>>
>> ratkinson@ogc.org | ogc.org | @opengeospatial
>>
>>
>>
>>
>> *Sign up for OGC News*
>> <https://ogc.us4.list-manage.com/subscribe?u=704e02f81107a6caab1568067&id=4e4528fd9d>
>>
>>
>>
>>
>>
>> On Fri, Jun 14, 2024 at 4:27 AM Kathi Schleidt <kathi@datacove.eu> wrote:
>>
>> Hi all,
>>
>> adding Paul Hershberg as Mr. TSML as a lot of this discussion is circling
>> around TSML.
>>
>> There is an activity underway to update TSML to 2.0, taking the recent
>> updates on both OMS and Coverage models into account, should be linked to
>> whatever goes down here
>>
>> :)
>>
>> Kathi
>>
>> On 13.06.2024 20:10, Grellet Sylvain via Hydro.dwg wrote:
>>
>> Ø  *that includes key waterml2 domain details that are not included in
>> TimeseriesML as an informative annex. *
>>
>> I’d rather keep those extra WaterML2.0 domain details handled in the
>> HydroDWG (thus WaterML2.0 line) unless
>>
>> -          they are very generic (not so waterish afte rall)
>>
>> -          we just have 1 or 2 of them (to avoir proliferation of specs
>> as well)
>>
>>
>>
>> If we end-up with having no need for creating water specific Class & al.
>> what we could do would rather be turn WaterML2.0 Part 1 – TimeSeries into a
>> Best Practice document explaining how to use TSML for quantitative water
>> data.
>>
>> That’s exactly the road the Water Quality IE will suggest : everything
>> covered in OMS, no need for extra-formal specs rather document how to use
>> it for Water Quality Data to ease the uptake.
>>
>>
>>
>> Sylvain
>>
>>
>>
>> *De :* Blodgett, David L <dblodgett@usgs.gov> <dblodgett@usgs.gov>
>> *Envoyé :* jeudi 13 juin 2024 19:34
>> *À :* Rob Atkinson <ratkinson@ogc.org> <ratkinson@ogc.org>; Kyle Onda
>> <konda@lincolninst.edu> <konda@lincolninst.edu>
>> *Cc :* Grellet Sylvain <S.Grellet@brgm.fr> <S.Grellet@brgm.fr>;
>> Hydro.dwg@lists.ogc.org; Schaaf, Hylke van der
>> <hylke.vanderschaaf@iosb.fraunhofer.de>
>> <hylke.vanderschaaf@iosb.fraunhofer.de>; SDW WG (public-sdw-wg@w3.org)
>> <public-sdw-wg@w3.org> <public-sdw-wg@w3.org>; Simon Cox <scox@ogc.org>
>> <scox@ogc.org>
>> *Objet :* RE: [EXTERNAL] Re: [Hydro.dwg] json waterml2 / timeseriesml?
>>
>>
>>
>> Thanks for the dialogue on this. Turns out similar questions are floating
>> around in a few places.
>>
>>
>>
>> Whatever we come up with should be portable between custom APIs, OGC-API
>> EDR (Position and Locations), STA, and the forthcoming OGC-API CONSYS part
>> 2.
>>
>>
>>
>> I _*think*_ that we should probably pursue a basic (focus on ease of use
>> and adoptability in preference to edge case support) json-schema for
>> timeseriesML that includes key waterml2 domain details that are not
>> included in TimeseriesML as an informative annex.
>>
>>
>>
>> Dave
>>
>>
>>
>> *From:* Rob Atkinson <ratkinson@ogc.org>
>> *Sent:* Thursday, June 13, 2024 11:20 AM
>> *To:* Kyle Onda <konda@lincolninst.edu>
>> *Cc:* Grellet Sylvain <S.Grellet@brgm.fr>; Blodgett, David L <
>> dblodgett@usgs.gov>; Hydro.dwg@lists.ogc.org; Schaaf, Hylke van der <
>> hylke.vanderschaaf@iosb.fraunhofer.de>; SDW WG (public-sdw-wg@w3.org) <
>> public-sdw-wg@w3.org>; Simon Cox <scox@ogc.org>
>> *Subject:* [EXTERNAL] Re: [Hydro.dwg] json waterml2 / timeseriesml?
>>
>>
>>
>>
>>
>> * This email has been received from outside of DOI - Use caution before
>> clicking on links, opening attachments, or responding.  *
>>
>>
>>
>> +1 for not proliferating.
>>
>>
>>
>> The way we could address this is add TimeseriesML/WaterML examples to the
>> existing APIs and test using the Building Blocks.
>>
>>
>>
>> Note Building Blocks can also perform and test transformations (of
>> examples - its about testing specifications).
>>
>>
>>
>> They also support cross-compiling across OpenAPI versions - noting
>> ConnectedSystems has decided to aim for OAS 3.1 but other OGC APIs are at
>> OAS 3.0 - we are solving this by a transpilation function to allow common
>> elements to be shared on different versions - its far better to aim at
>> current versions of JSON-schema than OAS 3.0 which has hacked a very old
>> version...
>>
>>
>>
>> so not proliferating requires a degree of sophistication in how
>> components are designed for reusability.  This is where the OGC COSI team
>> is able to support, but we need the domain experts identifying requirements
>> and developing realistic test cases, implementations and ultimately testing
>> at scale with deployments.
>>
>>
>> *Rob Atkinson*
>>
>> *Senior Research Engineer ** | Open Geospatial Consortium (OGC)*
>>
>> Mobile: +61 419 202973
>>
>> ratkinson@ogc.org | ogc.org
>> <https://urldefense.com/v3/__http:/ogc.org/__;!!KbSiYrE!n_chQWkG0m5feRnrxoXQRdM9EoZUKCQVeFw_yGqLl2ZRoYBQ8jX8VScy0BP3hdcuRbh4jP3lzCtN4zPGKoCz$>
>> | @opengeospatial
>>
>>
>>
>>
>> *Sign up for OGC News*
>> <https://urldefense.com/v3/__https:/ogc.us4.list-manage.com/subscribe?u=704e02f81107a6caab1568067&id=4e4528fd9d__;!!KbSiYrE!n_chQWkG0m5feRnrxoXQRdM9EoZUKCQVeFw_yGqLl2ZRoYBQ8jX8VScy0BP3hdcuRbh4jP3lzCtN44JNtwgN$>
>>
>>
>>
>>
>>
>> On Fri, Jun 14, 2024 at 1:44 AM Kyle Onda <konda@lincolninst.edu> wrote:
>>
>> I think a waterml2 JSON encoding, which could be an output format for
>> hydro-relevant STA or EDR (or any other extant and future API specs)
>> endpoints makes more sense than a new API.
>>
>>
>>
>>
>>
>> [image: A blue and green logo Description automatically generated]
>>
>> *Kyle Onda, PhD*
>>
>> Director, Internet of Water
>>
>> Center for Geospatial Solutions
>> <https://urldefense.com/v3/__http:/www.cgsearth.org/__;!!KbSiYrE!n_chQWkG0m5feRnrxoXQRdM9EoZUKCQVeFw_yGqLl2ZRoYBQ8jX8VScy0BP3hdcuRbh4jP3lzCtN4yNr8Y9I$>
>>
>> Lincoln Institute of Land Policy
>>
>> 303 550 4432
>>
>> Location: Chapel Hill, NC
>>
>> Pronouns: he/him
>>
>>
>>
>> *Please note our business hours are Monday through Thursday, 8 am to 6 pm
>> ET. With some exceptions, emails sent on Fridays will be addressed the
>> following week.*
>>
>>
>>
>>
>>
>> *From: *Hydro.dwg <hydro.dwg-bounces+konda=lincolninst.edu@lists.ogc.org>
>> on behalf of Grellet Sylvain via Hydro.dwg <hydro.dwg@lists.ogc.org>
>> *Date: *Thursday, June 13, 2024 at 11:41 AM
>> *To: *Rob Atkinson <ratkinson@ogc.org>, Blodgett, David L <
>> dblodgett@usgs.gov>
>> *Cc: *Hydro.dwg@lists.ogc.org <hydro.dwg@lists.ogc.org>, Schaaf, Hylke
>> van der <hylke.vanderschaaf@iosb.fraunhofer.de>, SDW WG (
>> public-sdw-wg@w3.org) <public-sdw-wg@w3.org>, Simon Cox <scox@ogc.org>
>> *Subject: *Re: [Hydro.dwg] json waterml2 / timeseriesml?
>>
>> Does a Timeseries API also make sense - or does STA or EDR cover all
>> requirements?
>>
>> I’ll let Hylke & Kathi complement but the way I see it is the following
>>
>> ST API 2.0 is OMS compliant
>>
>> My guts feeling is that TSML will end up 90% (or more) OMS compliant
>>
>> There is a high chance ST API covers (dunno for EDR)
>>
>> If the above is correct : I would suggest trying to reduce the entropy
>> for the external world and not add yet another API. Otherwise for those not
>> following all our internal OGC groups discussion, what the API to implement
>> will be more and more complex.
>>
>>
>>
>> My 2 €
>>
>> Sylvain
>>
>>
>>
>>
>>
>> *De :* Rob Atkinson <ratkinson@ogc.org>
>> *Envoyé :* jeudi 13 juin 2024 17:30
>> *À :* Blodgett, David L <dblodgett@usgs.gov>
>> *Cc :* Hydro.dwg@lists.ogc.org; Simon Cox <scox@ogc.org>; SDW WG (
>> public-sdw-wg@w3.org) <public-sdw-wg@w3.org>
>> *Objet :* Re: [Hydro.dwg] json waterml2 / timeseriesml?
>>
>>
>>
>> There is a functional OGC Building Block for the Observations and
>> Measurements model as Features here:
>>
>>
>>
>> https://github.com/opengeospatial/ogcapi-sosa
>> <https://urldefense.com/v3/__https:/github.com/opengeospatial/ogcapi-sosa__;!!KbSiYrE!m10Fc0eL5fJQIQ6s8LmFxL6rrLEQCDf3UEaCq3k510Xw6QJcJI4QtQ8XqLBN9QrTU66zxOeJMW4zK_GorA$>
>>
>>
>>
>> This is being co-developed with the SSN/SOSA update to match the new OMS
>> specification.
>>
>>
>>
>> can spin up customised profiles of these easily - e.g.
>>
>>
>>
>>
>> https://ogcincubator.github.io/bblocks-examples/bblock/ogc.bbr.examples.observation.vectorObservationFeature
>> <https://urldefense.com/v3/__https:/ogcincubator.github.io/bblocks-examples/bblock/ogc.bbr.examples.observation.vectorObservationFeature__;!!KbSiYrE!m10Fc0eL5fJQIQ6s8LmFxL6rrLEQCDf3UEaCq3k510Xw6QJcJI4QtQ8XqLBN9QrTU66zxOeJMW5FeQQS9w$>
>>
>>
>>
>>
>>
>> there are also experiments and activities to explore how common we can
>> make domain models with different APIs for different packaging of
>> observations, including OGC API EDR, SensorThingsAPI and
>> ConnectedSystemsAPI.
>>
>>
>>
>> So I think there may be a number of alternative encodings required,
>> depending on which API to serve them, but some commonalities we can factor
>> out as BuildingBlocks.
>>
>>
>>
>> Does a Timeseries API also make sense - or does STA or EDR cover all
>> requirements?
>>
>>
>> *Rob Atkinson*
>>
>> *Senior Research Engineer ** | Open Geospatial Consortium (OGC)*
>>
>> Mobile: +61 419 202973
>>
>> ratkinson@ogc.org | ogc.org
>> <https://urldefense.com/v3/__http:/ogc.org/__;!!KbSiYrE!m10Fc0eL5fJQIQ6s8LmFxL6rrLEQCDf3UEaCq3k510Xw6QJcJI4QtQ8XqLBN9QrTU66zxOeJMW5K_TD6Ew$>
>> | @opengeospatial
>>
>>
>>
>>
>>
>> *Sign up for OGC News*
>> <https://urldefense.com/v3/__https:/ogc.us4.list-manage.com/subscribe?u=704e02f81107a6caab1568067&id=4e4528fd9d__;!!KbSiYrE!m10Fc0eL5fJQIQ6s8LmFxL6rrLEQCDf3UEaCq3k510Xw6QJcJI4QtQ8XqLBN9QrTU66zxOeJMW6eGMtr4Q$>
>>
>>
>>
>>
>>
>> On Fri, Jun 14, 2024 at 12:36 AM Blodgett, David L via Hydro.dwg <
>> hydro.dwg@lists.ogc.org> wrote:
>>
>> Hi All –
>>
>>
>>
>> In conversation today, the topic of a json encoding of WaterML2 or
>> TimeseriesML came up. Is anyone aware of progress on such a thing? Seems
>> like someone must have done something like that even as an ad-hoc
>> experiment. Is there a world where we could establish a json encoding for
>> hydrologic timeseries data?
>>
>>
>>
>> Thanks.
>>
>>
>>
>> Dave
>>
>> _______________________________________________
>> You may visit our Privacy Policy at http://www.opengeospatial.org/privacy
>> <https://urldefense.com/v3/__http:/www.opengeospatial.org/privacy__;!!KbSiYrE!m10Fc0eL5fJQIQ6s8LmFxL6rrLEQCDf3UEaCq3k510Xw6QJcJI4QtQ8XqLBN9QrTU66zxOeJMW6TU6Ow1g$>
>> .
>> Hydro.dwg mailing list
>> Hydro.dwg@lists.ogc.org
>> You may also unsubscribe or manage your OGC public list subscriptions at
>>  https://lists.ogc.org/mailman/listinfo/hydro.dwg
>> <https://urldefense.com/v3/__https:/lists.ogc.org/mailman/listinfo/hydro.dwg__;!!KbSiYrE!m10Fc0eL5fJQIQ6s8LmFxL6rrLEQCDf3UEaCq3k510Xw6QJcJI4QtQ8XqLBN9QrTU66zxOeJMW6Vdj2VfQ$>
>>
>>
>>
>> _______________________________________________
>>
>> You may visit our Privacy Policy at http://www.opengeospatial.org/privacy.
>>
>> Hydro.dwg mailing list
>>
>> Hydro.dwg@lists.ogc.org
>>
>> You may also unsubscribe or manage your OGC public list subscriptions at
>>
>>  https://lists.ogc.org/mailman/listinfo/hydro.dwg
>>
>> --
>>
>> *Katharina Schleidt*
>>
>> CEO, Data Modeler, Data Networking Expert
>>
>>
>>
>> *DataCove e.U.*,
>> Robert Hamerlingg. 1/14,
>> 1150 Vienna, Austria
>>
>> Tel:
>>
>> Mobile:
>>
>> Skype:
>>
>> E-Mail:
>>
>> Web:
>>
>> +43 (1) 89 234 26
>>
>> +43 (650) 89 234 26
>>
>> Kathi Schleidt
>>
>> Kathi@DataCove.eu
>>
>> www.DataCove.eu <http://www.datacove.eu/>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> *FAIR information cube*,
>> Horizon EU project,
>> https://doi.org/10.3030/101059238
>>
>> E-mail:
>>
>> Web:
>>
>> fairicube@nilu.no
>>
>> www.fairicube.eu
>>
>>
>>
>>