Useless Tags

Although META tags can be useful for giving instructions to search engines and web browsers, not all of them are so useful. In fact, quite a few are completely useless. The useless tags (listed here for completeness) are usually created by overly-pretentious authoring software adding META tags just to impress the user.

If you add any of these tags to your web pages, I will mock you without hesitation or mercy.

Some of these tags are mystery tags spotted on high-profile sites like IBM. Most of those tags are probably for use with proprietary authoring or indexing systems, and therefore useless to anybody else. They're only mentioned here for the sake of completeness and rumor-control.

name="Abstract"

This tag is almost an urban legend. Seriously. There are hundreds of pages recommending name="abstract" listed in Google, but not a single damn one of them can name a search engine that uses Abstract. So far as I can tell, the only public engine using Abstract is Fireball.de, who accept it as a synonym for Description. Obviously, you're better off just using Description, since that's what all the other engines read.

Abstract also shows up in the source code of many pages authored by IBM, suggesting it's used internally there. I suspect the usage by IBM was combined with somebody's wishful thinking and transformed into pseudo-fact by dimwitted web advisors who copy-and-paste their advice from each others' sites.

I'm serious about the copy-and-paste accusation. I've seen countless cases (like this one) of sites "borrowing" the same inaccurate information. It's pathetic. (Even more pathetic is the fact that those sites think an abstract is "a one line sentence". Clearly some people have never seen an abstract in print.)

Reference: none

name="Agent markup Language Version"

This tag was used by the Agent Markup Language, a semantic markup system for web pages that was never widely used. Content is a string identifying which version of the language was used to markup the tagged page.

Reference: AML Agent Markup Language (via the Wayback Machine)

name="Author"

I'm pretty sure this one is useless, but at least two search engines say otherwise, so I've moved discussion of Author to the page about META tags for search engines.

Reference: Navigator Gold Authoring Guide

name="Classification"

Apparently introduced in the authoring tool of Netscape Navigator Gold, this tag was meant to categorize a page for indexing by a Netscape Catalog Server. Since I'm apparently the only person who ever found that explanation in the documentation for Navigator Gold, most people have used Classification as a synonym for Keywords (as WebSite Complete does), or ignored it all together. Gigablast started using Classification in 2003, but hasn't explained their implementation of it very well.

Reference: Navigator Gold Authoring Guide

This implementation of the META tag is used by pretentious software (and pretentious authors) to identify the copyright holder for the tagged page. I hate this tag. It's a waste of space, people. It's redundant if you're in a Berne Convention nation, and probably useless if you're not, since it's invisible to normal users. If you want to put a copyright statement on your page, put it in the BODY.

Reference: none

name="CreatedAppVer"

Used by WebSite Complete for something, but I'll be damned if I know what.

Reference: none

name="Division"

This tag was used by Alaska Department of Environmental Conservation to identify which department division was responsible for the tagged page. They seem to have stopped using this tag.

Reference: DRAFT Standards for web documents on DEC's website

name="EKBU"
name="EKdocType"
name="EKdocOwner"
name="EKdocTech"
name="EKpostDate"
name="EKreviewDate"

Spotted in the source code of various pages at Eastman Kodak, these META tags are probably used by Kodak's proprietary authoring and indexing systems. They're completely useless to anybody who doesn't work for Kodak.

name="Formatter"

Apparently, this was used by early versions of Microsoft FrontPage, in place of (or in addition to) Generator.

Reference: none

name="Generator"

The content for Generator is an unqualified (freeform) text string identifying the software program that created the tagged web page. I'm not sure which software first put this on a web page, but I know it goes back at least as far as Netscape Navigator Gold 3.0.

Sadly enough, one academic study says Generator is the most common META tag on the Web.

Reference: none

name="GoogleRank"

This one was spotted by a poster at Webmasterworld. The clear consensus is that anybody who uses GoogleRank has no idea how META elements or Google work.

References: Search Engine Forums: META NAME="googlerank"

name="IBM.country"

Spotted on various pages at IBM.com. Content value is a two-letter ISO abbreviation for the country, probably the country of authorship.

Reference: none

name="Owner"

Spotted on various pages at IBM.com; not to be confused with the http-equiv="Owner" used by MOMspider. Sometimes the content is an e-mail address and sometimes it's an unqualified name accompanied by geographic location. Presumed to be used by IBM's internal indexing and maintanence systems.

Reference: none

name="Microsoft Border"
name="Microsoft Theme"

Used by Microsoft software for something, obviously, but I don't really care what.

Reference: none

name="Revist-After"

Another urban legend of META tagging, this value is probably a mangled version of SearchBC's revisit tag. Like revisit, this tag is supposed to tell search engine robots how often to recrawl a page. Also like revist, this tag doesn't work.

One clear sign that this tag isn't real: The sites recommending it can't agree on the the content syntax. Some say it's a whole number indicating the requested re-crawling interval in days, while others say the words "days" needs to appear after the number.

The other clear sign that this tag is worthless? Nobody can conclusively cite an engine that obeys it (except for NetInsert, who treats is a synonym for revisit). So there.

Reference: none

name="Section"

This tag was used by Alaska Department of Environmental Conservation to identify which department program was responsible for the tagged page. They seem to have stopped using this tag.

Reference: DRAFT Standards for web documents on DEC's website

name="Security"

Spotted on various pages at IBM.com, always with the content value of public. Presumbably, anything with a different content value isn't meant for public consumption.

Reference: none

http-equiv="SHOE"

This tag indicates that the tagged web page uses the Simple HTML Ontology Extensions. So far as I know, no software ever implemented those extensions, so I'm declaring this tag useless.

The value of content indicates what version of SHOE is used in the page.

Reference: SHOE 1.1: Indicating a SHOE-conformant Document

name="Source"

Spotted on various pages at IBM.com, with various content values. The content value appears to refer to a design template used for the tagged page, so this META tag is probably used by some authoring software.

Reference: none

name="Template"

Used by various Microsoft authoring products to identify a predesigned template used to build the tagged page.

Reference: none

name="ThemeName"

Used by Go Daddy Software's Website Complete authoring software to identify which set of theme templates the page author used.

Reference: none