The Automotive Ontology Working Group is an informal group of individuals and corporations who want to advance the use of shared conceptual structures in the form of Web ontologies for better data interoperability in the automotive industry, and this at Web scale. In particular, we want to develop
extension proposals for schema.org so that automotive information can be better understood by search engines and
OWL Web ontologies for the automotive industry.
Also, we want to provide a forum for bringing together researchers and practitioners who are working on advancing the field.
Note: Community Groups are proposed and run by the community. Although W3C hosts these
conversations, the groups do not necessarily represent the views of the W3C Membership or staff.
Chairs, when logged in, may publish draft and final reports. Please see report requirements.
Some time ago we received suggestion that it would be easier to comment on the automotive extension to schema.org if we had draft versions of the schema.org application.
It comes a bit later than I expected, nonetheless I’m happy to inform you that as the results of our discussions here we are now proposing the following changes to schema.org types and properties related to our automotive extension:
Description for http://auto.schema.org/emissionsCO2 has been changed, because there is no UN/CEFACT Common Code for g/km.Before: ‘The CO2 emissions in g/km. The property uses a number instead of a QuantitativeValue, since g/km is the dominant unit of measurement, and there is no UNCEFACT Common Code for g/km.’After: ‘The CO2 emissions in g/km. When used in combination with a QuantitativeValue, put “g/km” into the unitText property of that value, since there is no UN/CEFACT Common Code for “g/km”.’
The text “This should be considered a pre-final preview release; final changes may be made after wider community review.” has been removed from the extension home page.
There is also an issue being discussed on the GitHub whether http://schema.org/vehicleSpecialUsage should be in the core or in the extension. As soon as this is resolved and some errors related to this property range and domain, we will lunch the Pull Request that will end this phase of Community review.
We came to the end of the review process. There were few changes proposed, and after some discussion we have one change to make (thanks to Tom Adamich):
Broaden the range of emmisionsCO2 property.
We suggest adding QuantitativeValue to the range of emmisionsCO2 property. One could then use “name” or “valueReference” to relate to test protocols or standards. This change also requires modification of property description.
We will now proceed to get this change accepted and activated !
To help with the review we just launched, please find a simple indented list of all terms in auto.schema.org.
We use red color for types, turquoise for individua and default color for properties.
All terms are hyperlinks to respective auto.schema.org description.
Let’s initiate a process of GAO community review of auto.schema.org extension.
The final shape of the extension can be explored at http://auto.schema.org/
Whenever you find some very important term (type or property) related to cars missing from the extension, or you would name or define existing property differently, please let us know.
Our intention is to close this review process in about 2 weeks (i.e. until ~ October 18th), organize the second teleconference and finally inform schema.org team about the community group position.
Please find some important remarks and a technical guidance about the initial phase of works on GAO. Feel free to comment on that matters – our goal is to create the process that is not too complicated and assumes use of simple tools (at least in the beginning)
1. Some ontological remarks
We start building GAO from collecting the concepts recognised as relevant within the automotive domain.
The concepts will be collected in a conceptual map called „bag-of-concepts”. Adding the new concepts to the conceptual map, we can take advantage of the auto.schema.org or go beyond the set of concepts we have there. The first set of concepts is stored in GAO.mm file.
By the collaborative development we shall create the conceptualisation of the automotive domain shared by all the participants of Automotive Ontology Community Group.
At this stage of GAO development process we shall not impose any semantical structure on the bag-of-concepts. The concepts are to be loosely linked by one relation which covers all the SKOS-like hierarchical and associative relations such as “broader”, “narrower” and “(somehow) related”.
Of course, in the end, GAO is intended to be a formal specification (RDF(S) or OWL) of the conceptualisation of the automotive domain (shared by all the participants of Automotive Ontology Community Group). So just after the community decides the work on the conceptual map of the domain was completed, we shall transform the bag-of-concepts into the first version of GAO
2. A brief technical guidance
At this stage to contribute to GAO it is enough to have the most up-to-date conceptual map of GAO concepts represented by the mind mapping diagram (see GAO.mm file). The GAO.mm file can be easily modified by FreeMind tool (available here: http://freemind.sourceforge.net/wiki/index.php/Download).
Our first community group conference call was convened on Thursday July 30, 12 PM EDT. We had about 10 participants.
After a brief introduction of all participants we have discussed the following matters:
The current status of schema.org automotive vocabularies, both in the core (schema.org) and in the forthcoming extension (auto.schema.org)
The planned deliveries after the first
The potential change of the work plan to include brainstorming as the first step to build the initial “bag-of-concepts” that, together with concepts we already have in schema.org, will form the seed for GAO vocabulary.
The use of GitHub as the collaborative tool for our common development.
The use of FreeMind as the supporting tool for brainstorming.
Referring to matters (1) and (2) Martin Hepp explained that while we can propose some modifications to schema.org automotive vocabularies, particularly to the forthcoming auto.schema.org extensions, we should not make revolutionary changes. Both core and auto.schema.org are reviewed extensions and there is always a risk of long review processes and a danger promoting terms that are not important from global search engines perspectives. Martin also explained that more manufacturer specific terms can us schema.org Property-Value (https://schema.org/PropertyValue) mechanism.
The rest of the call was devoted to the plans to introduce brainstorming process as a first step in going beyond the set of terms we have in schema.org. In that context, the use of “mind mapping” diagrams was recommended as they do not presuppose any specific kind of relation (like is_a relation) in the initial collection process.
The use of FreeMind tool (available here) has been proposed.
The initial diagrams for schema.org vocabularies and first attempt to create a bag of concepts for GAO were presented.
The current content of the two GitHub repositories:
is limited to Freemind native files (‘mm’ extension)
The next call is planned after summer holidays in September. Meanwhile the practical methods of building the set of terms will be proposed by us so that we can start fully productive activity after the next call.
The diagrams presented during the call:
1) Current schema.org core set of terms:
2) Proposed (awaiting publication) auto.schema.org set of terms:
(Note: there are some small difference between the diagram and current proposal)
3) The first trial set of terms for GAO brainstorming:
We had a very nice and informative first Community Group Conference call last week (The summary is coming here soon).
One of the results of the call is that we agreed to use GitHub as our repository of electronic materials, source code, future ontology files etc.
Our profile is https://github.com/W3C-GAO and has been temporarily assigned to the email account gao@makolab.com (until we practically figure out how to use our internal w3.org email facility of this group for such purposes).
Both contain source files for FreeMind tool (available here) that we will use for sketching the structure and content of our vocabularies as the initial tool for brainstorming. I will shortly describe it all in the call summary.
Anyone from this community group who plans to contribute to the development (all are warmly welcomed) please send the respective request to me (sopekmir@makolab.pl).