- From: Jarno van Driel <jarnovandriel@gmail.com>
- Date: Tue, 20 May 2014 01:02:12 +0200
- To: Niklas Lindström <lindstream@gmail.com>
- Cc: Thad Guidry <thadguidry@gmail.com>, Jarno van Driel <jarno@quantumspork.nl>, "martin.hepp@ebusiness-unibw.org" <martin.hepp@ebusiness-unibw.org>, Laura Dawson <Laura.Dawson@bowker.com>, Vicki Tardif Holland <vtardif@google.com>, W3C Web Schemas Task Force <public-vocabs@w3.org>
- Message-ID: <CADK2AU3NcG54Ebog-PJTsREihqHA-J495DPzC6opH5zrcArcwg@mail.gmail.com>
> > "What is the procedure for deciding what the "main entity" of something, > specifically a page, is?" Good question! 1] If one declares the WebPage (which is implied anyway) and links, let's say, a Product to it by means of @about then you get no Product Rich snippet. 2] If one declares the WebPage but doesn't link the Product on the page to it, then Google renders a Product Rich snippet. Which for me means Google looks at other top-level entities to generate Rich snippets for than WebPage. 3] If one does not declare the WebPage (which is implied nevertheless) and only marks up the Product, Google renders a Rich snippets Now I can go on like this quite long since if I've done hundreds of these, with all kinds of entities, but in all of those one thing is consistent, the order in which entities are linked most definitely has influence on how entities get treated in regards to the snippets in the SERPs. So when I markup pages I am forced to take this into consideration, else I'll be without clients faster than ice melts in the sahara. In practice for me this means I need to markup such that the entity I want to be eligible for a Rich snippet is the top-level entity of a page, and that all the rest is 'nested' within it. Now as for @about, I always thought @about sets the Subject of an Object. So if I've got that right than: <body itemscope itemtype="http://schema.org/WebPage"> <div itemprop="about" itemscope itemtype="http://schema.org/Product ">...</div> </body> means the WebPage is the Object and the Product is it's Subject and probable also is the reason why this doesn't result in a Rich Snippet/Experience. Now when I look at my example of MedicalProcedure, expressing: <div itemscope itemtype="http://schema.org/MedicalProcedure"> <span itemprop="about" itemscope itemtype="http://schema.org/VideoObject ">...</span> </div> Just feels completely wrong to me, as the VideoObject isn't the subject of the MedicalProcedure. It's simply an extra piece of info but not it's subject. 2014-05-20 0:37 GMT+02:00 Niklas Lindström <lindstream@gmail.com>: > What is the procedure for deciding what the "main entity" of something, > specifically a page, is? That should be made explicit by linking, using > about, from the web page to, in this case, the MedicalProcedure. Will > Google understand that? > > Cheers, > Niklas > > > > On Mon, May 19, 2014 at 11:17 PM, Jarno van Driel <jarnovandriel@gmail.com > > wrote: > >> Making the 'about' relation isn't all that difficult, add an @itemid to >> the MedicalProcedure, link the VideoObject to it, and Voila: >> >> <div itemid="medical-procedure" itemscope itemtype=" >> http://schema.org/MedicalProcedure"> >> <span itemprop="name">Tummy Tuck</span> (<span >> itemrop="alternatename">abdominoplasty</span>) >> <span itemprop="video" itemscope itemtype=" >> http://schema.org/VideoObject"> >> <link itemprop="about" href="medical-procedure"> >> <link itemprop="embedURL" href=" >> https://www.youtube.com/watch?v=xjC_dCvVB8s"> >> ... >> </span> >> </div> >> >> But if the MedicalProcedure doesn't have a property to link the >> VideoObject to it, then the end result is the the VideoObject gets treated >> as the main entity, which it's not. The Product/MedicalProcedure is. >> >> >> 2014-05-19 23:10 GMT+02:00 Jarno van Driel <jarnovandriel@gmail.com>: >> >> Thad, if I get you right you suggest I write: >>> >>> <div itemscope itemtype="http://schema.org/VideoObject"> >>> <meta itemprop="name" content="How to get a flat stomach? tummy tuck >>> or exercise?"> >>> <span itemprop="about" itemscope itemtype=" >>> http://schema.org/MedicalProcedure"> >>> [...] >>> </span> >>> </div> >>> >>> This makes the main-entity the VideoObject which, as a result, has >>> Google render a VideoObject rich snippet. >>> >>> Yet in this case the main entity is a Product/MedicalProcedure and the >>> aim is to get the Product rich snippets (which an MTE can have, as I have >>> the examples for it). And therefor the main-entity can't be a VideoObject >>> and has to be the Product/MedicalProcedure: >>> >>> <div itemscope itemtype="http://schema.org/Product >>> http://schema.org/MedicalProcedure"> >>> <meta itemprop="name" content="How to get a flat stomach? tummy >>> tuck or exercise?"> >>> <span itemprop="video" itemscope itemtype=" >>> http://schema.org/VideoObject"> >>> [...] >>> </span> >>> </div> >>> >>> >>> 2014-05-19 22:57 GMT+02:00 Thad Guidry <thadguidry@gmail.com>: >>> >>> Jarno, >>>> >>>> It's simpler than that... just use: >>>> >>>> http://schema.org/about >>>> >>>> Someone's CreativeWork (a video somewhere out there) is ABOUT the >>>> subject of "abdominoplasty". >>>> Where that's CreativeWork is in video form...and might or might not >>>> have an author, creator, and who knows...the video might even get an Award >>>> someday, etc. which all of those are properties of CreativeWork. >>>> >>>> Happy hacking! >>>> >>>> >>>> >>>> On Mon, May 19, 2014 at 3:54 PM, Jarno van Driel <jarno@quantumspork.nl >>>> > wrote: >>>> >>>>> I wonder, if the 'itemprop-reverse' addon for Microdata gets accepted, >>>>> would this than suffice: >>>>> >>>>> <span itemprop-reverse="about" itemscope itemtype=" >>>>> http://schema.org/VideoObject"> >>>>> <link itemprop="embedURL" href=" >>>>> https://www.youtube.com/watch?v=xjC_dCvVB8s"> >>>>> <meta itemprop="name" content="How to get a flat stomach? tummy >>>>> tuck or exercise?"> >>>>> ... >>>>> </span> >>>>> >>>>> Because if so, than maybe we don't need to do anything at all. >>>>> (thinking out loud, again) >>>>> >>>>> >>>>> >>>>> On Mon, May 19, 2014 at 10:45 PM, martin.hepp@ebusiness-unibw.org < >>>>> martin.hepp@ebusiness-unibw.org> wrote: >>>>> >>>>>> I would not recommend using hasPart as a generic property for >>>>>> relatedness. >>>>>> >>>>>> >>>>>> On 19 May 2014, at 22:38, Jarno van Driel <jarno@quantumspork.nl> >>>>>> wrote: >>>>>> >>>>>> > It's more the other way around. So in this case I want to express >>>>>> the main entity on a page, a MedicalProcedure, has a video giving >>>>>> additional information about the procedure. Something like this: >>>>>> > >>>>>> > <div itemscope itemtype="http://schema.org/MedicalProcedure"> >>>>>> > <link itemprop="sameAs" href="http://www.freebase.com/m/01_mbc >>>>>> "> >>>>>> > <span itemprop="name">Tummy Tuck</span> (<span >>>>>> itemrop="alternatename">abdominoplasty</span>) >>>>>> > <span itemprop="description">A Tummy Tuck, or abdominoplasty, >>>>>> is the most effective way to...</span> >>>>>> > <span itemprop="howPerformed">In summary the Tummy Tuck >>>>>> procedure takes 3-4 hours of...</span> >>>>>> > >>>>>> > <span itemprop="hasPart" itemscope itemtype=" >>>>>> http://schema.org/VideoObject"> >>>>>> > <link itemprop="embedURL" href=" >>>>>> https://www.youtube.com/watch?v=xjC_dCvVB8s"> >>>>>> > <meta itemprop="name" content="How to get a flat stomach? >>>>>> tummy tuck or exercise?"> >>>>>> > ... >>>>>> > </span> >>>>>> > </div> >>>>>> > >>>>>> > I started this thread by suggesting to add @video to Thing but that >>>>>> quickly evolved in the idea for a 'general' or 'related' property instead, >>>>>> and then halted. >>>>>> > >>>>>> > Now I have a website where many MedicalProcedure, Service and >>>>>> Article entities have additional videos about those entities, yet only >>>>>> Article has properties like @associatedMediao and @video. Both >>>>>> MedicalProcedure and Service do not have the properties to express this. >>>>>> Which still leaves me questioning what I should if I do not want the >>>>>> VideoObject to be main entity. >>>>>> > >>>>>> >>>>>> >>>>> >>>> >>>> >>>> -- >>>> -Thad >>>> +ThadGuidry <https://www.google.com/+ThadGuidry> >>>> Thad on LinkedIn <http://www.linkedin.com/in/thadguidry/> >>>> >>> >>> >> >
Received on Monday, 19 May 2014 23:02:40 UTC