W3C home > Mailing lists > Public > public-vocabs@w3.org > April 2012

Re: proposal for new classes Activity and Action

From: Daniel Dulitz <daniel@google.com>
Date: Thu, 19 Apr 2012 13:02:09 -0700
Message-ID: <CACWrOGZunz--aicAuvCpOR0H2oMS4P6ee7Q408ydiY7OKZ9CnQ@mail.gmail.com>
To: Bob Ferris <zazi@smiy.org>
Cc: public-vocabs@w3.org
Thanks Bob. Those links provide useful context.

I think a key difference in this proposal and the Event Ontology is that,
in order to express a factor in this proposal, one needs to define a result
("product") that names the factor as an instrumentality.

And a key difference in this proposal and the Counter Ontology is that in
this proposal the "counter" isn't present in the schema -- it is behind the
scenes, creating aggregateCounts, but never existing as an item on the page.

I think those differences in perspective are appropriate given schema.org's
focus on representing what is visible (counts, results/products). If there
are immediate lost use cases, please mention them now so we can go in with
our eyes open.


On Thu, Apr 19, 2012 at 00:50, Bob Ferris <zazi@smiy.org> wrote:

> Hi Daniel,
> I really like your proposal - it looks really well thought through. I
> co-developed a similar Semantic Web ontology called the Counter Ontology
> [1] some time ago that can be slightly utilised together with the Event
> Ontology [2]. So you may also have a look at these ontologies (if you are
> not already aware of them ;) ) to get further inspiration for your
> schema.org extension proposal.
> Cheers,
> Bo
> [1] http://purl.org/ontology/co/**core#<http://purl.org/ontology/co/core#>
> [2] http://purl.org/NET/c4dm/**event.owl#<http://purl.org/NET/c4dm/event.owl#>
> On 04/19/2012 04:05 AM, Daniel Dulitz wrote:
>> Hello,
>> The three of us (Will, Jason, and I) have an idea for improving
>> schema.org <http://schema.org>'s representation of activities, and we'd
>> like to start a discussion about the merits of our proposal. We've tried
>> to adopt some of the core insights of ActivityStreams while remaining
>> compatible with the (many) "implied activities" that can be drawn from
>> schema.org <http://schema.org> items in general.
>> The proposal may be found at
>> https://docs.google.com/**document/d/**1VYZ9FmN7Vl2PzzR1kX3KZgfdPK-**
>> qUfxR3z12NC3i5V0/edit<https://docs.google.com/document/d/1VYZ9FmN7Vl2PzzR1kX3KZgfdPK-qUfxR3z12NC3i5V0/edit>
>> . Comments are turned off on the document so the discussion will be on
>> this list.
>>  From the top of the document:
>> """
>> Historically (in earlier schema.org <http://schema.org> versions), the
>> http://schema.org/**UserInteraction <http://schema.org/UserInteraction>type was focused on reporting
>> aggregate interaction counts, despite being a subclass of Event. Over
>> time it has shifted more to representing individual interactions. This
>> transition has not been smooth; the original intent is still reflected
>> in Example 1 for UserInteraction, which uses subtype names as part of a
>> structured text interactionCount. It is also reflected in the names of
>> the UserInteraction subtypes, which are plural, as well as the
>> duplication between Comment/text and UserComments/commentText.
>> UserInteraction subclasses UserComments, UserCheckins, and UserTweets,
>> when interpreted as individual interactions, contain creative content
>> that would benefit from many of the properties of CreativeWork. Other
>> subtypes of CreativeWork, such as Review, lack corresponding subtypes of
>> UserInteraction. AggregateRating has its own type, while other types
>> lack corresponding types to represent aggregates.
>> To resolve this tension, this proposal:
>>  * creates a new type, Activity, to represent a single activity, with
>>    consistent properties (across all activity instances) to structure
>>    the grammar of the activity;
>>  * creates a new type, Action, with subtypes that describe classes of
>>    “verbs” within the grammar;
>>  * creates a new type, AggregateActivity, to represent an aggregation
>>    of activities e.g. in CreativeWork/interactionCount;
>>  * adds a new property Thing/action to indicate an action that may be
>>    performed on a thing; and
>>  * deprecates UserInteraction and its subclasses, and AggregateRating,
>>    in favor of the new types.
>> """
>> Looking forward to your thoughts,
>> d
Received on Thursday, 19 April 2012 20:02:59 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:48:45 UTC