Join Jen Kramer for an in-depth discussion in this video Appreciating the value of the semantic web, part of Learning Web Semantics.
Way back, when the web first got started, there were lots of websites, just as there are today. The way you got from one website to another was you had to find links on that particular website. And those links connected the websites together across the web. Here's lynda.com's homepage from roughly 1996 and you can see down here in the bottom in the navigation there's a link here for links. And so this was a way that we would get between websites. There were no search engines, it was really hard to find anything.
Fortunately, around 1994, there was a wonderful invention by WebCrawler and Lycos. The were some very early full text commercial search engines. And these were some of the very first websites that would search all of the words in a web page. Previously search engines had been able to search just titles or a few words here and there. But the revolutionary development with Lycos and WebCrawler was that they would search all of the words on the web page. You can now type in words and phrases into the search box and the search engines would return pages containing those words or phrases.
The search engines really come quite a long way in 19-years but its still fundamentally working the same way. You'll type some words or a phrase into the box for these search engine and you get back some matches for that word or phrase. However, you might have noticed that search is starting to change a little bit. I've gone into google and I've done a search for peanut butter and fruit dip recipe. So I've got lot and lot of results here, most of which, look like recipes, its pretty clear that I could click on any of these and I'd go to a page that would produce a recipe.
But notice the two results that are highlighted. These two results have a rating, you see the star rating that's there? They have a certain number of reviews, they have an amount of time that it will take to make the recipe and even the calorie count. How do those results look different relative to all other result on the page? Well, its because they marked up their contents in a very well defined way that's known to Google. Google knew that these are recipes with ratings and time and calorie counts. So it included all of that information in the search results.
Like wise, if I did another Google search result for the Indigo Girls and I got a bunch of results for Indigo Girls. And their music and so forth but one of the things you will notice here is the result for last dot fm the very first result here on this page. For this particular search result, you'll notice that it's listed some tracks here underneath. So we have information about the Indigo Girls and we also have some tracks that play snippets of their music. The key to having Google be able to provide specialized results like this.
Based on the type of information that's being displayed, is making sure that your content is marked up in such a way that Google can read and display it accordingly. This is done through semantic markup in HTML5. When we talk about semantic markup, we talk about markup that has meaning rather than markup defining the presentation or the look of the website. Semantics then takes on two complimentary approaches. One is to look at the mark up used in plain old HTML5, which I'll do in the next chapter.
We'll look at tags that have meaning behind them like, headers, footers, articles, asides and more. I'll also look at links and their relationships, which are also very important for semantics. The second approach is the standardized way of presenting a certain type of content online. Events, recipes, people, places, these are all common information on the web. Through the use of micro data or RDFa lite, we can add some additional HTML5 attributes to our markup to standardize the way this information is presented.
So, let's get started looking at the new HTML5 semantic tags.
- Using HTML5 header, nav, and footer tags
- Addressing related content with aside and mark tags
- Defining internal and external link relationships
- Defining next and previous relationships
- Introducing schema.org
- Using itemscope and itemtype to add meaning
- Understanding the difference between RDF, RDFa, and RDFa Lite