RE: LBD ontologies - concept of aggregation and hosting

Hi all,

In line with the last plenary call, it would be good to reserve a timeslot in one of the coming calls to check issues mentioned in the Github repo of BOT and other LBD ontologies. Some of them have already been discussed at previous LDAC events, but of course need to be validated by members of the group. Other issues from the reviewers of the BOT journal paper w.r.t. definitions probably also have to be added and discussed. It might be a good idea to add issues in one place, e.g. the Github repo to keep an overview?

Best regards,

Mathias

From: Mathias Bonduel <mathias.bonduel@kuleuven.be>
Sent: woensdag 5 juni 2019 9:34
To: Mads Holten Rasmussen <mhoras@byg.dtu.dk>; public-lbd@w3.org
Subject: RE: LBD ontologies - concept of aggregation and hosting

Hi Mads and other LBD members,

I've been thinking about this issue and I think a differentiation between aggregation and hosting is needed. Imagine the case of a quantity survey over a building model. In an early design phase, you want to count every distinct building element separately (e.g. the wall and the window hosted by it, but not the parts of the window), while in the tendering and construction phase, you probably want to count the distinct building elements (e.g. number of walls and windows) and the aggregated parts of some building elements (e.g. the parts of each window). I would be in favor of splitting hosting and aggregation for the above reasons.

As a compromise, we can keep bot:hasSubElement and define two subproperties: bot:hostsElement and bot:aggregates (or better: bot:aggregatesElement)? People who don't need the distinction can just keep on using bot:hasSubElement and people who do need it can use the subproperties? Additionally, the aggregation property can also be made transitive, while the hosting property cannot be made transitive?

Anyhow, the text definition of bot:hasSubElement should be extended to include the case of aggregation and product:aggregates should be removed from PRODUCT (or better, be marked "depreciated").

Depending on the reactions here, I'll prepare a pull request for BOT and PRODUCT to update them accordingly.

Best regards,

Mathias

From: Mads Holten Rasmussen <mhoras@byg.dtu.dk<mailto:mhoras@byg.dtu.dk>>
Sent: dinsdag 28 mei 2019 9:15
To: Mathias Bonduel <mathias.bonduel@kuleuven.be<mailto:mathias.bonduel@kuleuven.be>>; public-lbd@w3.org<mailto:public-lbd@w3.org>
Subject: Sv: LBD ontologies - concept of aggregation and hosting


Hi Mathias (and co)



I believe bot:hasSubElement can be used to describe aggregations. Actually, I think this was the reason why we decided to change the name after last LDAC. Therefore, both of these examples are valid:



Aggregation:

<AirHandlingUnitA> bot:hasSubElement <FanA> , <FanB> , <FilterA> , <FilterB> .



Hosting:

<WallA> bot:hasSubElement <WindowA> , <WindowB> .



Aggregation:

<WindowA> bot:hasSubElement <MullionA> , <FrameA> , <MullionA> .



Obviously (as it is often the case in BOT) the concepts are very generic and you will need to extend the terminology to for example describe where the fans are located inside the Air Handling Unit from a flow system perspective.



In any case, we should probably change the description of bot:hasSubElement and provide some examples.



Best Mads.

________________________________
Fra: Mathias Bonduel <mathias.bonduel@kuleuven.be<mailto:mathias.bonduel@kuleuven.be>>
Sendt: 27. maj 2019 15:59:13
Til: public-lbd@w3.org<mailto:public-lbd@w3.org>
Emne: LBD ontologies - concept of aggregation and hosting

Dear members of the LBD group,

Based on discussions in the Github repositories of the LBD ontologies BOT and PRODUCT, I want to propose the following:

*        the concept of aggregation: this concerns one construction element that consists of multiple smaller parts, e.g. a column that exists from a base, a shaft and a capital. The terminology is currently located in the PRODUCT ontology, but it probably fits better in the BOT module (topology) as it used to:

o   remove the concept product:aggregates from the PRODUCT ontology (https://github.com/w3c-lbd-cg/product/blob/master/prod.ttl).

o   reintroduce bot:aggregates in BOT v0.4.0

o   in the DUL alignment, bot:aggregates can also be made a subproperty of dul:hasPart: https://github.com/w3c-lbd-cg/bot/blob/AlignmentRevision/DULAlignment.ttl

o   see discussion on Github: https://github.com/w3c-lbd-cg/bot/issues/9#issuecomment-490822351

*        the concept of hosting: this concerns one construction element that is hosted by another, e.g. a door hosted by a wall. The terminology is already in BOT, but the name is confusing w.r.t. the concept of aggregation. It might also be confusing w.r.t. another (super)property bot:hasElement between a bot:Zone and a bot:Element.

o   remove bot:hasSubElement from BOT

o   reintroduce bot:hostsElement in BOT v0.4.0

Instead of removing classes/properties of the ontologies, it is also possible to add rdf:type owl:DeprecatedClass or rdf:type owl:DeprecatedProperty to a resp. a depreciated class or property. I would also like to have some kind of easier access to older versions of ontologies, e.g. via Github releases with the TTL directly available: https://help.github.com/en/articles/creating-releases

Please give your opinion on the matter discussed above and I hope we can take some decisions afterwards.

@Pieter/Kris: can this matter be discussed during the next LBD Telco?

Best regards,


Mathias Bonduel
Doctoral researcher
Faculty of Engineering Technology
Technology Campus Ghent
Gebroeders De Smetstraat 1
9000 GENT
tel. + 32 92 65 86 36
gsm + 32 474 93 52 51
iiw.kuleuven.be/onderzoek/sustainable-buildings<https://iiw.kuleuven.be/onderzoek/sustainable-buildings>
[http://stijl.kuleuven.be/mail/2013-kuleuven-rgb.png]