Re: Historical events

I think that it is problematic to have a subtype "HistoricEvent", because it will mean that the type for a object will depend on the temporal context. E.g., a day after a conference is over, it wil become a HistoricEvent.

My feeling is that a lot of confusion stems from the fact that schema.org tries to avoid subclasses as long as they do not change the processing of respective entities by search engines. For instance, we do not need a subtype of places as long as all places are rendered the same way by Google et al.

As a consequence, we attach properties that make sense only for some instances directly to the more abstract classes.

This sounds unintuitive if you come from a knowledge engineering context, but it is actually so by design. A type in schema.org serves as a hookpoint for triggering / grouping computational operations over the data. If you have no algorithmic steps that require a certain distinction, then we do not have respective specializations of a type. 

Best

Martin

 
-----------------------------------
martin hepp  http://www.heppnetz.de
mhepp@computer.org          @mfhepp




> On 12 Jun 2018, at 09:08, Muri, Allison <allison.muri@usask.ca> wrote:
> 
> Many, many thanks to those who have contributed to this interesting discussion. I am working on examples for two types: first, HistoricalEvent as a subtype of Event. The more I work through this as a possibility the more I wonder how usable it would be.  The second is Occurrent as a new type. This would have certain advantages and disadvantages, just as HistoricalEvent as would described above.
> 
> I will add examples tomorrow and then hopefully generate a discussion about the value and applicability of http://schema.org/Event/HistoricalEvent as subclass of Event versus something ttp://schema.org/Event/Occurrent to distinguish between an event as is now used for mostly future events, and the following for anthying that is an occurrencw the past.
> 
> More to come tomorrow.
> 
> - Alliaon
> 
> 
> On Jun 4, 2018, at 5:11 AM, Phil Barker <phil.barker@pjjk.co.uk> wrote:
> 
>> 
>> On 03/06/18 00:07, Muri, Allison wrote:
>>> Maybe this won’t generate much interest, but I have obtained my own fork of the Schema.org GitHub repository and also have set up a Google App Engine project to be share it publicly. I take Phil Barker’s point that one should “first make sure that schema.org is the best vocabulary for this type of information, e.g. by thinking about use cases that fall within the scope of its mission.” I really don’t know the answer to that. Hopefully I can generate more conversations about this in the future. Thank you, Phil, for the links to Richard Wallis’ blog posts and videos.
>>> 
>> You're welcome. 
>> 
>> This discussion has already lead to a suggested improvement in schema.org, so there is clearly some overlap between your interests and its scope. I like Richard's parallel to TouristAttraction.
>> 
>> I think there are also issues around Periods, Events and historical reference points that could be unpicked.
>>  
>>> Regarding “they already know those differences,” I think search engines would not know that “Ætna groan” in the passage below refers to the 1669 eruption of Mount Etna, a “NaturalEvent” (as opposed to a satiric reference to a really firey, angry queen at her coronation) without markup:
>>> 
>>> Nor with more heavy strokes could Ætna groan,
>>> When Vulcan forg’d the Arms for Thetis’ Son.
>>> 
>>> —Poems on Several Subjects, by Stephen Duck (1730)
>>> 
>> 
>> This is a good use case. This probably isn't the right forum to go into details of addressing it, but by way of illustrating a point ... [I think you mentioned microdata at one point]
>> Nor with more heavy strokes could 
>>   <span itemprop="mentions" itemscope
>>         itemtype="Event" 
>>         name="1669 eruption of Mount Etna">
>>      Ætna groan
>>      <link itemprop="sameas" href=
>> "https://www.wikidata.org/wiki/Q2411998"
>> >
>>      <meta itemprop="location" content="Mount Etna">
>>      <meta itemprop="startDate" content="1669-03-08">
>>   <span>,
>> 
>> When Vulcan forg’d the Arms for Thetis’ Son
>> Follow the sameas URL and you will see that I cheated, but adding the relevant eruption would be possible, and what I did link to illustrates how machine readable information can be provided beyond the schema.org markup. I have been minimal in my description of the event in the inline schema.org markup, there could be a lot more there if required/useful.
>> 
>> Phil
>> 
>> -- 
>> Phil Barker. http://people.pjjk.net/phil
>> PJJK Limited: technology to enhance learning; information systems for education.
>> CETIS LLP: a cooperative consultancy for innovation in education technology.
>> 
>> PJJK Limited is registered in Scotland as a private limited company, number SC569282.
>> CETIS is a co-operative limited liability partnership, registered in England number OC399090
>> 
> 
> ....................................................
> Allison Muri
> Department of English
> 
> Arts 418
> University of Saskatchewan
> Saskatoon, SK, Canada
> ph: 306.966.5503
>

Received on Tuesday, 12 June 2018 07:55:56 UTC