Nearly all search engines use spiders (also known as robots, their original name) to go out and scour the Web looking for Web pages. These search engine spiders then bring the data back to be indexed by the engine.
Since roughly 1996, individual meta commands have been used on individual Web pages to modify how search engine spiders behave. The most useful of these commands are fairly universal and respected by almost all search engines.
Here are some of the more popular ones and reasons you might want to use them (or not).
<meta name="robots" content="index">
This meta command is one of the most common ones used—and it is also the least necessary. It tells search engine spiders to come on in and put the page in their index. However, all search engines do this by default anyway.
Basically, if you want to put it in there for fun, be my guest; but this command is not giving you any special treatment. All search engines will index your page, unless you specifically tell them otherwise.
<meta name="robots" content="follow">
The follow command is different from the index command. It basically requests that the search engine spiders follow the links that are on a particular page. Again, however, this piece of code is completely unnecessary, because all search engines are going to follow the links on a page, unless otherwise directed.
<meta name="robots" content="noindex">
The noindex command is the opposite of the index command; it tells search engine spiders not to index the content of a page. It's important to note, however, that search engine spiders will still follow the links on a page that uses only this command.
When not used for legitimate purposes, this tag can be dangerous, because it can put you at risk for penalization by most, if not all, search engines. That's because you can use a noindex tag to hide pages with multiple links that you don't want visitors to see but do want all search engines to index.
There are, however, some legitimate uses for the noindex command. For example, if you have a dynamic site and you've created static pages to replace some of your dynamic pages, which can make them easier for search engine spiders to access, you might put a noindex tag on the dynamic version.
As Google notes in its Webmaster Help Center:
Consider creating static copies of dynamic pages. Although the Google index includes dynamic pages, they comprise a small portion of our index. If you suspect that your dynamically generated pages (such as URLs containing question marks) are causing problems for our crawler, you might create static copies of these pages.
In such cases, it is acceptable to use the "no index" command on the dynamic version of the page so that your content will not be treated as duplicate. You are not tricking search engines, you're just redirecting them.
<meta name="robots" content="nofollow">
This tag tells search engine spiders that it's OK to go ahead and index a page and list it but that they shouldn't follow any of the links that are on the page. This can be useful if, for example, you had some partners that requested a link on your site that you felt obligated to give, but you wanted to hold onto as much Page Rank as possible.
Now this is of course between you and your personal god... but you would be able to in effect have a partners page, add the nofollow attribute to the meta tags, and basically not pass on any of your Page Rank to any of the sites to which you are linking. The nofollow command in effect tells all search engines that this is the end of the line.
<meta name="robots" content="noindex,nofollow">
Obviously, noindex and nofollow are powerful tags—and in combination they can make a page and the subsequent pages to which it links invisible to nearly all search engines. This combination command tells search engine spiders, "Do not read this page; do not follow any of the links on this page; do not include this page in your index."
This command has its beneficial uses. For example, it can be placed on pages on a site that have duplicate content for legitimate reasons. A Web site might have both a page for the United States and a page for England that cover the same product with exactly the same content. However, nearly all search engines would see this as duplicate content and could devalue both pages. So placing this command on one of them means that search engine spiders will walk on by and you won't be penalized.
<meta name="robots" content="noarchive">
Finally, almost all search engines today, including Google and Yahoo, offer a cached version of a site alongside its listing that provides a snapshot of what the page used to look like. The noarchive tag, therefore, is available to be used when there is content on your Web site that is time-sensitive and which you might not necessarily want search engine spiders to cache for people to have access to later.
For example, a business might run a one-time special that has a ridiculously low price to drum up some business while things are slow. The business would want to be able to shut that sale down as soon as sales are back up to a solid level. However, it is conceivable that someone could click on the cached version of the business's site, see the old deal that was out there, and insist on getting it for themselves.
By using the noarchive tag, you are telling search engine spiders, in effect, "This page is subject to frequent changes, and I don't want my visitors to have access to some of this content at a later time."
The commands discussed here are just a few of many, and new ones are being added frequently. While nearly all search engines support these commands, some don't. The ones in this article, however, are fairly universally understood by search engine spiders, no matter where they originate. As more universal commands are introduced, I will write about them.