W3C home > Mailing lists > Public > public-vocabs@w3.org > March 2015

Re: Collection / Container / List (+Pagination)

From: ☮ elf Pavlik ☮ <perpetual-tripper@wwelves.org>
Date: Mon, 09 Mar 2015 10:43:34 +0100
Message-ID: <54FD6B46.2090204@wwelves.org>
To: Martin Hepp <martin.hepp@unibw.de>
CC: "Wallis,Richard" <Richard.Wallis@oclc.org>, W3C Web Schemas Task Force <public-vocabs@w3.org>, Melvin Carvalho <melvincarvalho@gmail.com>
On 03/09/2015 08:00 AM, Martin Hepp wrote:
> In the notion of the underlying GoodRelations model, a product can be any tangible or intangible entitity on which some rights can be offered to transfer, for free or any compensation, e.g. monetary amounts, bundles of rights on other objects (as in barter trade), or even vague promises like "good karma".
> 
> So yes, a product is essentially a role that a thing can take in an offer, but offer is meant much broader than in traditional e-commerce settings.
> 
> But still I would not put container or anything else that is not rigidly bound to the product role as a subtype of product. Rather put is somewhere else in the hierarchy and use multi-typed entities.
+1!

> 
> By the way, I have on my long agenda plans for better support for a) configurable products and variants and b) shopping baskets / detailed wishlists
> 
> Note that bundles are already supported in offers via schema:includesObject and schema:TypeAndQuantityNode. I think that for containers and shopping baskets, we should try to reuses schema:TypeAndQuantityNode and reuse/extend schema:ItemList.
Interesting, so here an Offer itself also acts as container, since we
don't have another URI for container, one distinct from the Offer URI

I try to brainstorm Thing and Collection as well Thing/Collection in
image attached to this tweet
https://twitter.com/elfpavlik/status/574193664974520320

Imagine a container with 100k elements
1. you may want to have distinct URI just for collection of all the members
2. you most likely want to page it

LDP paging may come useful but if we don't use distinct URI for
container. Flow for person browsing it using UI (web browser), and LDP
client may work different. Because of setting HTTP Header
'Prefer: return=representation; max-triple-count="500"'
http://www.w3.org/TR/ldp-paging/

We had conversation with Melvin Carvalho about using foaf:Group, I
pointed to him that looking at W3C Community Grup page
e.g. https://www.w3.org/community/rsp/
we find

1. Group resource https://www.w3.org/community/rsp/ (e.g. foaf:Group)
 a) it includes group details
 b) it includes count Participants (83) but only includes 10 of them
2. Group Members resource https://www.w3.org/community/rsp/participants
 a) it includes all 83 members
 b) for 2000 members it would provide intuitive way to page it

Just some thoughts + real world examples

Cheers!


> 
> 
> Martin
> 
> PS: Attached is an example of modeling bundles in offers.
> 
> 
> <div itemscope itemtype="http://schema.org/Offer" itemid="#offer" >
>   <span itemprop="name">Special offer: 1 cellphone + 2 travel chargers</span>
>   <link itemprop="http://schema.org/businessFunction" 
>         href="http://purl.org/goodrelations/v1#Sell" />
>   <div itemprop="http://schema.org/includesObject" itemref="p1"></div
>   <div itemprop="http://schema.org/includesObject" itemref="p2"></div
> <div>
> 
> <div id="p1" itemscope itemtype="http://schema.org/TypeAndQuantityNode">
>   <meta itemprop="amountOfThisGood" content="1">
>   <meta itemprop="unitCode" content="C62">
>   <div itemprop="typeOfGood" 
>        itemscope itemtype="http://schema.org/Product" itemid="#cellphone">
>     <span itemprop="name">ACME Cellphone BulkyTalky</span>
>     EAN: <span itemprop="http://schema.org/gtin13">1234567890123</span>
>   </div>
> </div
> <div id="p2" itemscope itemtype="http://schema.org/TypeAndQuantityNode">
>   <meta itemprop="amountOfThisGood" content="2">
>   <meta itemprop="unitCode" content="C62">
>   <div itemprop="typeOfGood" 
>        itemscope itemtype="http://schema.org/Product" itemid="#charger">
>     <span itemprop="name">ACME Travelcharger</span>
>     EAN: <span itemprop="http://schema.org/gtin13">2345678901234</span>
>   </div>
> </div>
> 
> 
> --------------------------------------------------------
> martin hepp
> e-business & web science research group
> universitaet der bundeswehr muenchen
> 
> e-mail:  martin.hepp@unibw.de
> phone:   +49-(0)89-6004-4217
> fax:     +49-(0)89-6004-4620
> www:     http://www.unibw.de/ebusiness/ (group)
>          http://www.heppnetz.de/ (personal)
> skype:   mfhepp 
> twitter: mfhepp
> 
> 
> 
> 
> 
> On 06 Mar 2015, at 17:52, Wallis,Richard <Richard.Wallis@oclc.org> wrote:
> 
>>
>> My understanding of Product, within the context of schema.org, is broader than that.  Something that is result of an action or process.  Not necessarily having an economic role.
>>
>> Pushing that definition to an extreme I would suggest that a lump of rock is a product - maybe the result of a geological process, or divine intervention.   
>>
>> The Container subtypes I suggest are all physical items, that a person/organisation would/could have purchased or produced to put things in.  Pushing to the other extreme someone or a software application would make a directory/folder to place files within.
>>
>> Perhaps a slight rewording of the description of Product might help, if folks agree with my understanding that is ;-)
>>
>> ~Richard
>>
>> On 6 Mar 2015, at 16:25, ☮ elf Pavlik ☮ <perpetual-tripper@wwelves.org> wrote:
>>
>>
>>> Hi Richard,
>>>
>>> On 03/06/2015 04:47 PM, Wallis,Richard wrote:
>>>> New types:
>>>>
>>>>    *   Container  - a subType of Product
>>> To my understanding it means:
>>> "Every instance of a Container is also an instance of a Product"
>>>
>>> I really don't understand where this comes from?!
>>>
>>> http://schema.org/Product
>>> "Any offered product or service. For example: a pair of shoes; a concert
>>> ticket; the rental of a car; a haircut; or an episode of a TV show
>>> streamed online."
>>>
>>> Which sounds to me like something having clear economic (or e-commerce)
>>> role. Which I don't think includes all the possible instances of Container.
>>>
>>> Thanks for clarifying little more your reasoning behind it :)
>>>
>>
> 
> 



Received on Monday, 9 March 2015 09:44:01 UTC

This archive was generated by hypermail 2.3.1 : Monday, 9 March 2015 09:44:01 UTC