Slashdot Mirror
Declaring The Death of Metatags
theduck writes "Andrew Goodman of Traffick.com pleaded for someone to announce the end of metatags (at least with respect to trying to skeeve good search engine ranking). and Danny Sullivan, Editor of The SearchEngineReport obliged. Personally, I've resisted using them for years, but convincing clients that they're not worth the effort has always been difficult. Does anyone (except porn sites) actually use them anymore?"
Well, I know the post was except for porn sites, but the reason that porn sites use 'em is because they work! Nobody knows search engines more than porn site owners. Part of what got me this listing was good meta tags. Porn sites rule the web as far as traffic and profitability. When in doubt, do what to porn sites do.
Meta tags are used a lot... there's widespread knowledge of so-called "google bombing".. Google pops up some of its search results based on the content between an A HREF tag, as you can read about here: Google Bomb...
Much like security, I think this is the kind of thing that hackers and tinkerers will always find a way to exploit. The question is who can stay ahead in the race?
The theory of relativity doesn't work right in Arkansas.
we weren't trying to spoof search engines, but rather feeding them into our own search engine (a licensed copy of altavista), and using them to have a correct synopsis and ranking.
of course, other search engines ignored them, and quite often we used custom tags, but when working with 3rd party software, they were (and are, since the dotcom still somewhat exists) a godsend.
The Canadian Federal Government is in the process of sinking millions of dollars of taxpayers' money in what they're calling the "Common Look and Feel Project". One of the biggest money-pits in the project has been the metatagging. Amusingly,the standard they're pushing has changed mid-stream often enough that very few federal employees know what their target is. (Kind of a microcosm of what's happened on the Internet as a whole...)
IMO what metatags are good for is supplying synonyms that you don't want to have to put into your text.
For example, a webpage might be about "OOP Criticism". However, searchers may not think to use the word "criticism", and instead look for "OOP complaints", "OOP skeptics", etc.
"OOP criticism" and "OOP skepticism" are pretty closely related. But text indexing or link indexing probably would not be able to make the connection.
Thus, they have legit uses IMO. Sure, they are abused, just like any other technology, including word indexing an link tracing.
A search engine should use *multiple* approaches IMO. Better yet, allow one to select the weights of each one for a given search. Have drop-down boxes with numbers from 0 to 9 on which to select the weightings given to links, text, and metatags.
Table-ized A.I.
I guess if the only value you see to these tags is as a way to manipulate the search engine results, then yeah, maybe a case could be made to do away with them. But meta tags can be used for a whole lot more -- other people mentioned using them to refresh or redirect pages, but there are other goodies too. For example, I encourage my developers to drop this onto each page: "name='developer' content='Employee Name'" -- it's an ego stroke for developers to be able to show that off to their friends. Also, the copyright can be put into a meta tag. Why? Because it isn't visual, so all the clueless newbies who copy the site with a GUI tool will fail to remove that tag. We catch a few people that way, although only the most stupid.
For a while, at Borland, I had a pretty low-end (but working) content-management system, where I put an expiration date into a meta tag along with an author name, and then had a Perl script that flagged any out of date file and emailed the author. This was brute-force Perl recursing through the htdocs folder and reading in each file, so it wasn't database-backed, but in 1995 my boss thought it was hot. Nowadays there are better ways to do most everything, and meta tags are not required for much, but they are still a very useful option, and allow for some creativity -- regardless of search engines.
My Greasemonkey scripts for Digg &
I publish a photo gallery and have relied upon keywords to describe what's pictured but not necessarily mentioned in a photo's caption. This appears to work with Google from what I can tell. The same keywords are used by my site's internal search engine, so I have to think of and store them anyway. I would be happy to change if there's a better way.
www.cgstock.com
Better than metatags, IMO, is Googlebombing -- i.e. making a bunch of sites point to yours.
I actually managed to pull off a wholey unplanned yet quite effective googlebomb in the last few months. A side project of mine, Quizilla, has ome feature where it give you HTML coede to past into your weblog. Well, since Quizilla is a free service , I put an advetising string in that HTML, "brought to you by Quizilla", with a link to the site.
Well, through some circumstances that got really popular really quick and people were pasting a lot og this HTML into their pages.. and what happened when Google indexed all those pages?
Instant Googlebomb.
I'm kinda sad I wasn't selling anything, or else I'd be rich.
Hilary Rosen's speech was about her love of money and her desire to roll around naked in a pile of money.
I think the main issue with "keyword" metatags is that they're completely unreliable for search engine use, since it's easy to abuse them by stuffing them with terms that users search for that aren't necessarily related to the content of your page. Fine, I think that's obvious. Nobody's really going to argue that one.
:P
The "description" metatag is still EXTREMELY useful, though. Even if a search engine doesn't use the metatags for ranking purposes, it can still use the "description" metatag to display a nice human-readable summary of the page. Often search engines just display the first N characters of text on the page and use that for a summary, which usually is not a good or readable summary for the site.
The problem with Google is that it seems to randomly use the "Description" metatag sometimes, but not others. Here's an example. Notice how the second "Anime Expo 2002 at Bootyproject" link has a nice readable summary under it, but the first one doesn't. (It may have changed between the time I posted it and the time you view it, who knows) Which makes no sense to me, because if you look at the source for each of the two pages, the metatag information is identical for both pages. I don't get it, I dunno if Google's just a little broken in that respect, or if I screwed something up. Sorry to pimp my own site there... it's just an example I'm obviously quite familiar with.
But anyway, when search engines and authors use the description metatag properly (ie, the search engine doesn't use it for ranking, and the author takes the time to write a nice summary), it's pretty nice.
OtakuBooty.com: Smart, funny, sexy nerds.
however, thay should be limited to 25 characters. This way they would need to be relevent and precise to get proper ranking.
The Kruger Dunning explains most post on
Is there a better way than using meta tags for indexing your site? What if your pages are mostly composed of graphics? Adius
It does no just have to be search. You can use properly organized metadata to do thigs like associating related content together. This can be done more reliably with author specified metadata.
This summer I wrote a perl module called FileMetadata (available from CPAN) that collects metadata from files. I have used it to ease content management headaches on my website. Each HTML (XHTML) file has metadata that is used to advertise it on my site's index pages. I have ideas for more nifty things that can be done with metadata but as always time is finite.
It's rather silly to proclaim the death of meta tags just because they can be manipulated. The content of web pages are also being manipulated to obtain a better rating in search enginges. Should we stop using content as well? Blank web pages everywhere, because that is the only way to be sure that they actually contain what search engines promise us?
It would probably be far more useful to begin black listing sites who try to divert traffic their ways by means of "lies". Something along the line the anti-spam lists that are in use for email.
First of all, just because "keywords" tags can be fraudulently specified, doesn't mean that they are useless. I can publish pr0n in a book titled "Undergraduate Physics"; does that make book titles useless? The fault is not in the "keywords" tag; the fault is in naively trusting unverified data. It's okay to put lollipops from the store in your mouth, but it's not okay to do the same with lollipops that you pick out of the gutter.
OK, my turn now. I wish somebody would call a moratorium on printing an entire webpage in a teensy weensy font. I have carefully specified my default font size, because that is the size which is most appropriate for reading long pages of text on my monitor with my eyes. It's okay to make stuff smaller if it's supposed to be "the fine print", but for whole articles, please use the default font size.
"It sure was strange to see something on Usenet about me that didn't involve Klingon gang rape." -- Wil Wheaton
In contrast to nearly about everyone else on /., I'm going to stick my neck out and say that I appreciate good meta tags.
If I'm on a slow link, I get to see a brief description of the page and then decide if I want to go to it. And if I'm on a slow link I disable flash, scripting, etc. and set cache to a small amount.
It also helps that I use a different browser for slow links. =) (Nope, not IE, Mozilla or Opera.)
I put spamtrap addresses in META TAGs, links to wpoison pages, etc... Lots of fun.
I know I use them (BurntMail)...
Last I checked, a lot of the big-guys still use them as well (Cisco, RedHat, Microsoft, Mandrake, and SourceForge for example)
PR
If the HTML-standard had imposed a limit on the number of meta-keywords a webmaster may enter for her page (say 10 max), webmasters would have been forced to think about which words they were including. It's the perceived lack of scarcity of resources that prevents a healthy "keyword-economy" from developing.
In my opininion it would still be possible to turn this thing around. If a couple of big search engines plastered an announcement all over their sites: "We only look at the first ten uniqe meta-keywords", things might change for the better.
Being well balanced is overrated. -- John Carmack