Decisions related to the creation of two new SKOS collections for Scientific Event Loggers

[gwemon]

We decided to create 2 new controlled vocabularies (SKOS collections) to be hosted on the NVS to hold the persistent identifiers and definitions of terms needed for "Actions" and "Processes" as part of scientific event loggers on board research vessels. The content is largely drawn from the terms used by the EARS ontology. The terms have been reviewed by an international informal group that started meeting online in 2021 but had been in communication for some time before that, with members representing multiple organisations or existing initiatives or projects including WHOI/R2R, RBINS/EuroFleet/EARS, Oregon State University/CORIOLIX, British Antarctic Survey (BAS) and NOC-BODC/NVS. The main objective of the group was to consolidate existing work on vocabularies to support event loggers and in particular ensure that terms used by EARS and R2R were aligned.

These done, we now need to create the collections on the NVS. What name shall we give to these collections? EARS/R2R Actions and EARS/R2R Processes? Shall we create a new governance group for them or can we use an existing one already set up in C30 (SeaVoX maybe)?

Proposing that the identifiers for the 2 new collections in the NVS be EL1 for the Actions vocab and EL2 for the Processes vocab.

Any preferences? Please feel free to comment below.

[rhudak2020]

@gwemon It is good to see all the progress going on with the eventlogger controlled vocabularies! I like your suggestion for the identifiers to be EL1 and EL2 for the new collections. On the topic of governance, I'm not very knowledgeable about all of the requirements for governance groups. But I am of the mindset that if it can go under a pre-existing governance I think that would be the best possible route. This way we avoid added complexity but also maintain good linkages.

[gwemon]

I am reviewing this proposal in order to deploy new content onto our production server. I am proposing a slight modification of the proposed titles in order to specify that the vocab is "harmonised" (i.e. the result of looking at existing resources from, in our case, 2 main sources, and selecting one harmonised solution), and it also refers to research "platforms" (i.e. not just research vessels). Also I am mentioning EARS and R2R in the description but not in the title to avoid overloading the title. Please could all interested in this activity, have a look below and let me know if anything needs changed: For the Action terms, it is suggested that the collection have the following metadata:

Identifier: EL1 Title: Harmonised Marine Research Platform Event Logger Actions Short_title: eLogger Actions Description: Terms used to identify actions performed on board research platforms operating at sea, and logged in an on-board scientific event log book. It is the result of an harmonisation exercise carried out by an international group of marine data managers keen to implement the vocabularies used by the Eurofleets Automatic Reporting System (EARS) in Europe and by the Rolling Deck to Repository (R2R) project in the US as persistent controlled vocabularies. The following definitions are derived from the EARS ontology: an "Action" defines a step in an ongoing process, and can be deliberate or incidental (http://ontologies.ef-ears.eu/ears2/1#Action); a "Process" defines a general ongoing activity a tool or vessel can do and consists of a sequence of punctuated actions (including malfunctions and deliberate actions) (http://ontologies.ef-ears.eu/ears2/1#Process); an "Event" defines a practical type of event by associating a tool, process, action and possibly an accompanying property (http://ontologies.ef-ears.eu/ears2/1#EventDefinition).

For the Processes: Identifier: EL2 Title: Harmonised Marine Research Platform Event Logger Processes Short_title: eLogger Processes Description: Terms used to identify processes occurring on board research platforms operating at sea, and logged in an on-board scientific event log book. It is the result of an harmonisation exercise carried out by an international group of marine data managers keen to implement the vocabularies used by the Eurofleets Automatic Reporting System (EARS) in Europe and by the Rolling Deck to Repository (R2R) project in the US as persistent controlled vocabularies. The following definitions are derived from the EARS ontology: a "Process" defines a general ongoing activity a tool or vessel can do and consists of a sequence of punctuated actions (including malfunctions and deliberate actions) (http://ontologies.ef-ears.eu/ears2/1#Process); an "Action" defines a step in an ongoing process, and can be deliberate or incidental (http://ontologies.ef-ears.eu/ears2/1#Action); an "Event" defines a practical type of event by associating a tool, process, action and possibly an accompanying property (http://ontologies.ef-ears.eu/ears2/1#EventDefinition).

My two main questions are: 1) Is it okay to say "platform" instead of "research vessel"? 2) What should the governance group be? Is it better to come under the SeaVox's umbrella (I will email the SeaVox mailing list to have some feedback on this)? or shall we set up a new governance group for it like e.g. Event Logger Vocabulary Harmonisation governance group? which we would then operate entirely via this github repo.

[PetratenH]

Hello Gwen, In my opinion it would be better to use "platform" to be able to use the vocabs also for events from other platforms, e.g. autonomous vehicles. I'm not very familiar with implications of the governance group decision. Are there any limitations if we are under the SeaVox's umbrella that would prevent us from governing this? Many thanks for deploying the content.

[rhudak2020]

Hello all, Correct me if I am wrong, but I believe that "platform-specific" vocabulary already exists in a repository, maybe in BCO-DMO? I think we want to specify this group of terms for a vessel. In my mind this is primarily due to the fact that you would "deploy" other "platforms" from a vessel. I think it would be worthwhile to keep it to a "vessel" and not specify a "research vessel". I think this will let us use the eventlogger terms to be used for other ships that may not be research specific (ie a commercial vessel deploying ARGO floats).

I'm also not sure how using SeaVox might affect the vocabulary. What are L22 terms governed under?

Thank you again for pulling all of this together.

[gwemon]

@rhudak2020 I understand your point about using vessels and not research vessels but could you try and find out more about the "platform-specific" vocabulary? Could we get somebody from BCO-DMO to comment on or provide links to these vocabularies so that we can compare content? L22 is owned by SeaVox but it is effectively managed by BODC. We don't have a consultation process when a request for a new instrument model is submitted. I think this would need to be different for EL1 and EL2. New terms or changes to existing ones would need to go through an approval process. We can give ourselves a name and set ourselves as the governance group, using github for communication or we can broaden the governance and reach out to all those registered in SeaVox. That being said, SeaVox was set up a long time ago and membership would probably gained from being refreshed.

[alex-tate]

1. I prefer the term platform as the event model is useful for many things beyond what people understand by the term 'vessel'.
2. Same as @rhudak2020 , I would ditch the 'research' part of the title.
3. I would also get rid of the 'logger' part of the title. The actions and processes are related to events. An event logger is the (digital or paper-based) thing that records them.
4. I'm ambivalent about the 'harmonised' word but we'd need to set ourselves up as a named group to provide anything different/better.
5. The concepts are generic enough that you could even think about dropping the word 'marine'. I see no reason why we wouldn't use these vocabs for instrumentation events on Antarctic stations or on our planes.

[rhudak2020]

@gwemon I think I might have just been a little confused. BCO-DMO has vocabulary defining types of platforms and I think they were using ELOGs action terms. In my conversations with them they have expressed future plans on connecting to SSN/SOSA vocabulary: https://www.w3.org/TR/vocab-ssn/ ; But this still does not address "actions".

As such, I think we can move forward with using "platform" instead of "vessel" so we can be more inclusive. And if we are going to go this direction I agree with @alex-tate to remove "marine" and open it up to all "platforms".

I also agree with replacing 'elogger' with 'event'. We will most likely use these vocabulary primarily for eventloggers but we should not limit its use.

I think it might be worthwhile to email the SeaVox mailing group to get their input and then decide from there. I think it would be beneficial to be in the SeaVox umbrella. We can always fall back and make our own governing group since we know that we are currently active.

[gwemon]

Thank you everyone for your feedback. There are pros and cons in making the vocabulary specific to "vessels" or open to all platform types. I will circulate to SeaVox too to invite comments.

For reference, this is the proposed content for EL1, the action vocab: EL1_Action_Vocab_Proposed_Content.xlsx

And this is the proposed content for EL2, the process vocab:
EL2_Process_Vocab_Proposed_Content.xlsx

If we go for generic "platform" then we may need to review the definitions to make sure they do apply to any relevant platform types.

[gwemon]

To recap, the preference seems to got for being non-specific about the type of platforms the action and process terms apply. For reference, a platform is defined in SOSA as: IRI: http://www.w3.org/ns/sosa/Platform Platform - A Platform is an entity that hosts other entities, particularly Sensors, Actuators, Samplers, and other Platforms. Example A post, buoy, vehicle, ship, aircraft, satellite, cell-phone, human or animal may act as Platforms for (technical or biological) Sensors or Actuators.

This means EL1 and EL2 can contain terms for actions and processes that apply to any instances of sosa:Platform: we are not restricting the terms to neither ships nor just marine platforms. This is fine by me but we have to consider 2 things: governance and content diversity. For the governance: We are all marine experts and would have a good knowledge of the terms needed for marine observing platforms including probably planes and satellites. But I would be nervous if we start receiving requests for terms related to non-marine platforms other than our familiar ones. Or are there so few platform types that we feel confident we can cover the requirements for them all? Another thing to bear in mind I think is that if we want to use the list of terms for platform-specific forms (such as e.g. a ship event logger) and we allow the vocabulary to accept terms that do not apply to that platform then we will need to build reasoning or filtering on the top of the EL1 and EL2 vocabulary. Which is completely possible of course but something I wanted to bring to the group's attention.

[gwemon]

To respond specifically to the suggestions made by @alex-tate

1. I prefer the term platform as the event model is useful for many things beyond what people understand by the term 'vessel'. Agree about non-vessel specific but nervous about what I don't know about other kind of platforms. (see comment above)
2. Same as @rhudak2020 , I would ditch the 'research' part of the title. Agree
3. I would also get rid of the 'logger' part of the title. The actions and processes are related to events. An event logger is the (digital or paper-based) thing that records them. Agree
4. I'm ambivalent about the 'harmonised' word but we'd need to set ourselves up as a named group to provide anything different/better. Yes that's correct. Let's see what other suggestions we get.
5. The concepts are generic enough that you could even think about dropping the word 'marine'. I see no reason why we wouldn't use these vocabs for instrumentation events on Antarctic stations or on our planes. I can see the advantage of doing this - just consider comments above too.

[PetratenH]

If updates of the terminologies are done upon an open discussion of the governing group, that can reach out to other experts, I would not be too worried about going beyond purely marine use. If the terminologies are used for various platforms there will always be terms that don’t apply to that specific platform. For example the term EL1:Act28 OnBoard is ship-specific, so we either have to change that or assume that it will be used only when appropriate. Should we not leave it up to the user to decide which term applies to them?

[rhudak2020]

Is there a way to have a subcategory for 'marine' or 'ship-specific'? In the L22 vocabularies we have 'Broader' and 'Narrow' linkages and terminology. So in a similar logic can we narrow down an action term on what platform it can be used on?

[dr-shorthair]

I'm a big fan of being able to recognise individual actions or processes during scientific investigations. This is at the heart of O&M and SOSA (Samplings, Observations and Actuations are the high-level classes) and also PROV-O (Activities vs Entities is the high-level division) and BFO (Occurrents vs Continuants). So having a useful taxonomy of these would support them being routinely recognised in data activities.

[gwemon]

Is there a way to have a subcategory for 'marine' or 'ship-specific'? In the L22 vocabularies we have 'Broader' and 'Narrow' linkages and terminology. So in a similar logic can we narrow down an action term on what platform it can be used on?

@rhudak2020 yes there are various ways of doing this, some more costly in maintenance than others. I was just mentioning this to make sure that we were all aware of the consequence of making these vocabularies applicable to any platforms. I can sense that the consensus is there and I completely understand the reasons/advantages. So yes, providing we keep in mind that if we want a platform-specific list of terms we will need to facture in additional development. Alternatively, the default, will be the list of terms as published on the NVS.

@PetratenH yes, if we have a responsive governance group, and access to experts then this might not be such an issue. Also I imagine that demand for new terms won't be that frequent.

[gwemon]

A couple of feedback comments related to the Process vocabulary came from my meeting with @neil-ices-dk and his team at ICES:

• Processes terms should be more grammatically consistent so that if we have a term for pro1="sampling" then pro11 should really be "tracking" and not "track"
• Also Pro19 "AtHarbour" is defined as "Arrival at harbour" which seems to indicate that the process is only about the ship arriving in port and not being in port. Should we either change the preflabel to ArrivalAtHarbour to make the term unambiguous?
• Also, if these vocabularies are to be applicable to multiple platforms then we may need to review some of our definitions.
• Finally, when going through the list after meeting Neil and his team I also realised that we may need to add examples of usage for some of these terms.

[PetratenH]

The feedback has some fair points. Would you like to have a Zoom call to go through the definitions?

[rhudak2020]

I think a zoom call would be beneficial to iron out the definitions and provide examples where needed.

I think we should either add ArrivalAtHarbour and change the definition of AtHarbour or only change the definition of AtHarbour. I think we should also think about adding a term for entering EEZs as that could be a major reason to stop data collection.

[gwemon]

Thanks both @rhudak2020 @PetratenH . I will organise a zoom call.

[gwemon]

Thanks both @rhudak2020 @PetratenH . I will organise a zoom call.

This is still waiting for me to take action! I am going to circulate a group poll to try and find a suitable date and time.

[gwemon]

Meeting today 18/11/2022 at 15UTC

[gwemon]

Reviewed title and definitions of EL1 and EL2 following meeting on 18/11/2022 - with additional post-meeting edits by @gwemon

Identifier: EL1 Title: Sampling and Observation Platform Event Actions Short_title: Platform Event Actions Description: Terms used to identify actions performed on sampling and observation platforms. "Platform" is as defined in sosa:Platform (http://www.w3.org/ns/sosa/Platform) i.e. "an entity that hosts other entities, particularly Sensors, Actuators, Samplers, and other Platforms". This vocabulary is the result of an harmonisation exercise between terms used by the Eurofleets Automatic Reporting System (EARS) ontology and terms used by the Rolling Deck to Repository (R2R) project, for research vessel event logs. The following definitions are derived from the EARS ontology: an "Action" defines a step in an ongoing process, and can be deliberate or incidental (http://ontologies.ef-ears.eu/ears2/1#Action); a "Process" defines a general ongoing activity a tool or platform can do, and consists of a sequence of punctuated actions (including malfunctions and deliberate actions) (http://ontologies.ef-ears.eu/ears2/1#Process); an "Event" defines a practical type of event by associating a tool, process, action and possibly an accompanying property (http://ontologies.ef-ears.eu/ears2/1#EventDefinition).

Identifier: EL2 Title: Sampling and Observation Platform Event Processes Short_title: Platform Event Processes Description: Terms used to identify processes occurring on sampling and observation platforms. "Platform" is as defined in sosa:Platform (http://www.w3.org/ns/sosa/Platform) i.e. "an entity that hosts other entities, particularly Sensors, Actuators, Samplers, and other Platforms". This vocabulary is the result of an harmonisation exercise between terms used by the Eurofleets Automatic Reporting System (EARS) ontology and terms used by the Rolling Deck to Repository (R2R) project, for research vessel event logs. The following definitions are derived from the EARS ontology: a "Process" defines a general ongoing activity a tool or platform can do and consists of a sequence of punctuated actions (including malfunctions and deliberate actions) (http://ontologies.ef-ears.eu/ears2/1#Process); an "Action" defines a step in an ongoing process, and can be deliberate or incidental (http://ontologies.ef-ears.eu/ears2/1#Action); an "Event" defines a practical type of event by associating a tool, process, action and possibly an accompanying property (http://ontologies.ef-ears.eu/ears2/1#EventDefinition).

[gwemon]

EL1 and EL2 collections created and queued for publication. Initial population of EL1 also started. The description of the collections had to be reduced to fit the size of the field in the database (500 characters). The definition of the terms "Platforms", "Actions", "Processes", "Events" are now located in the "usage_notes" of the SKOS collection definitions. These are not currently visible on the NVS. They will be made visible in the future.

[gwemon]

https://vocab.nerc.ac.uk/collection/EL1/current/ and https://vocab.nerc.ac.uk/collection/EL2/current/ are live on the NVS. Population of EL1 initiated. A meeting is scheduled tomorrow to review terms to be added to EL2 (Processes).

[danibodc]

The EL2 (Processes) terms highlighted during the scheduled review have now been loaded, and will be live on the NVS from 0730 UTC tomorrow morning.

[danibodc]

The EL1 (Actions) terms highlighted during the scheduled review have now been loaded, and will be live on the NVS from 0730 UTC tomorrow morning.