AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

Are we happy with schema.org
 
 

Some of you might have seen this article by Karolina:

http://www.chatbots.org/conversational/agent/google_bing_yahoo_semantic_search_avatar_schema/

I was wondering whether chatbot developers are happy with this development, and I’m curious after the first results!

 

 
  [ # 1 ]

Yes, I am happy that the major engines have agreed on a single specification.

Some of us have created our own schemas to try to represent information. The problem with most micro-formats is that they were designed to focus on a very limited subset of data (FOAF, Vcard, etc.).
Schema.org provides a standard hierarchy that we could use : http://schema.org/docs/full.html
It also provides for a standardized method to extend this basic schema set.

 

 
  [ # 2 ]

The suggestion that computer science is dead, seems to give rise to many attempts to reanimate it.

 

 
  [ # 3 ]

Doesn’t this require for the webmasters to want their site to be easily scraped by bots? Don’t know if that’s always the case.

 

 
  [ # 4 ]

For people with personal web pages, like mine, the addition of this additional information tagging won’t net any real benefits, as near as I can tell. Not to mention that, at this point in time, the addition of “itemscope” and “itemtype”, as shown in this example:

<div itemscope itemtype="http://schema.org/Movie">
  <
h1>Avatar</h1>
  <
span>DirectorJames Cameron (born August 161954)</span>
  <
span>Science fiction</span>
  <
a href="../movies/avatar-theatrical-trailer.html">Trailer</a>
</
div

would tend to throw warnings in all of the currently used markup validators (e.g. http://validator.w3.org/, etc.), making it seem that the page is no longer “standards compliant”, which is technically true. For myself, that part alone will prevent me from “taking advantage” of this, since I’m very particular about my pages being able to honestly display valid markup logos.

For most commercial websites, however, this can help to make their site more “target oriented”, sharpening the focus on the correct demographic range they wish to reach, thus improving the site’s chances of success in reaching the intended audience. This is going to affect SEO for nearly every site out there, at some point, so many companies (the vast majority, I’d imagine) would almost “have to” make use of this micro-format as soon as possible, or risk seeing their positions in the major search engines suffer.

At least, this is how I see it. smile

 

 
  [ # 5 ]

The HTML5 specification includes a description of microdata, a new markup standard for specifying structured information within web pages.

“itemscope” is part of HTML5 : http://www.w3.org/TR/html5/microdata.html
http://googlewebmastercentral.blogspot.com/2010/03/microdata-support-for-rich-snippets.html

Jan Bogaerts - Jun 25, 2011:

Doesn’t this require for the webmasters to want their site to be easily scraped by bots? Don’t know if that’s always the case.

Mostly this is targeted for businesses and others who want to show up in search engines.

I am exploring keeping some of my bot’s knowledge in a HTML format.

 

 
  [ # 6 ]

I don’t believe this is the correct way to build a global machine readable semantic knowledge system.

Actually, structured semantic data can not help the machines get more intelligence to understand real natural language.

The industry put so much resource to build semantic data on manually on Internet in the past years. However, there is no any practical system can build the semantic bridge between machine and natural language data.

We should put more resource to teach the machines to understand unstructured text, in instead of building so much structured data.

 

 

 
  [ # 7 ]

Nathan, I agree that this is a crutch for the search engines. But, it might help crowd sourcing NLP by surrounding unstructured text with the semantic equivalent. A smart search engine could use the mark-up as a resource to improve NLP much in the same way Google translate learned languages.

 

 
  login or register to react