Web Search

What is Web Search?

Web search typically refers to the ability to find websites and webpages through a search engine. For example, web search is what websites like Google and Bing offer. More generally, web search is a function a search engine may have of searching the world wide web for information, of searching for for text, images, and videos embedded in the content of webpages, and for other pertaining metadata.

Main Features

Links to The Results: as web search is performed on the web, and things on the web—both webpages and images—are addressable by their own unique URLs, the search engine results page will generally feature a link to each result. That's because all you need to make a link is the URL of the thing being linked to, and all the results are going to have URLs, so you can link to all of them.

Titles for Results: all webpages are written in a code language called HTML. In HTML, you can specify the title of a HTML comment with the <title> tag. This means there is a standard way for webpages to specify their titles, which is what shows on the tab of your web browser. Web search engines make use of this same field to display the title of a webpage result. Images on the other hand do not have a similar title field, which is why web image search doesn't get titles for images. Videos don't really have them either, but in practice all web video search does is find webpages that contain videos, so the title of the webpage is assumed to be the title of the video.

Snippets: web search that has text queries will also generally feature a snippet marking where the words typed in the query appear in the result. There is a standard field in webpages where authors can specify a description for the webpage itself. Sometimes this field is used instead of a snippet.

Stale Results: there are trillions of webpages on the web, so it's not possible to search through them all in real time. A cached version in an index is used instead, which means that a result found by web search because it contains a word according to the index may no longer contain that word when you actually check the current version of the webpage. The content of each of these trillions of webpages may change at any moment, which means that for a web search engine to be up to date, it will have to regularly download trillions of webpages and compared the current version to the last version they saw to check if they have changed or not in order to update their index. All of this takes a lot of resources, which is why we have very few web search engines in the market today that can search the entire web.

Favicons: webpages often have an icon called a favicon that is displayed on their tab in a web browser. Web search engines may choose to display this icon beside the results.