Microdata Vocabularies

After reading the Microdata article on MDN:

I noticed that the article mentions that developers can make/use their own microdata vocabularies, instead of using the popular, schema dot org.

I wanted to know more, such as:

  1. How do I make my own vocabulary? (I want to learn, I know there are other huge ones). Is there any further reading I can go to. How do I know the format in which to write it, how to structure it, just be plain JSON hosted on a site?, etc.

  2. Does it matter to search crawlers whether you use a less popular vocabulary? In other words: does using a smaller, unknown library not have the same impact as a bigger one like the schema project?

  3. Lastly: my understanding, so far, is that you have a namespace type of thing that the bot can refer to in order to understand text/images from your webpage and what relationships these text/images have with other data on the web. And so, this means the vocabulary you refer it to should have many outward links to relevant items for each “itemtype”. Is my undertstanding so far correct? Or does a microdata vocabulary not need such a hyperlink system?

Thank you for the help.

Also I do know people think that it is easier to just use the Schema vocabulary, and want me to just use that. However, I want to learn how they work (not just how to use them), how the crawlers understand vocabularies, and properly understand their scale and what it takes to make one.

EDIT: added link to the article

Hello @lightblue

could you give a link to the topic

and have a nice day

Hi, the article is on MDN. The link is:

let me ask @Rafael_Green

do you have any experience about that

thanks and have a nice day both of you

vocabulary (itemtype) is just an identifier (usually set to a url because urls are unique)
when using your own identifier, you don’t have to include anything in the content pointed by the address you define (though you may want to provide the user with help information in that address either in prose or a formal language such as RDF)

for more information see this: https://www.w3.org/TR/microdata/#sec-navigation-timing

1 Like

@justsomeone and @Rafael_Green, the help is very much appreciated!

I realize I was overthinking this, and your answer helped alot. The URI is not needed to understand the data. So it is a just a namespace for the itemtype identifier, nothing useful needs to be there (although you should put an interface for people to understand each itemtype). Clears alot up for me.

1 Like