| 8:31 pm on Jan 26, 2003 (gmt 0)|
|Anyone care to provide some feedback on the above tags and why they are being used? |
For the most part they're being used because someone heard or read somewhere that they work, and most people don't spend the time or go to the effort to test such things themselves. Instead, "well it might help, and it probably won't hurt, so..."
Of course, it's one thing when that approach is taken by a general-purpose website but quite another when it's taken by someone marketing their SEM/SEO services, skills, and knowledge.
And of course, as you point out, some of these things actually can do some harm -- both directly and indirectly.
| 8:35 pm on Jan 26, 2003 (gmt 0)|
|And of course, as you point out, some of these things actually can do some harm. |
JayC, actually the only harm I see being done is credibility if that even applies here. Yes, many marketers are following information guidelines that were somewhat mythical in their time to begin with.
I cannot honestly say that any of the above are harming a websites visibility other than the excess code. My belief is that the major SE's do not give weight to any of the above.
| 8:50 pm on Jan 26, 2003 (gmt 0)|
<meta name="revisit-after" content="5 days">
Let me clarify something with the above tag. There is a common use for this when it comes to internal indexing for a large site or one that is updated frequently and runs its own spider.
| 9:05 pm on Jan 26, 2003 (gmt 0)|
I believe it is Ink's Slurp spider whose default behaviour is (or was) "index,nofollow". Therefore, I include "index,follow" on my pages where it matters.
The "revisit-after" tag was once reportedly used by one small directory in Canada. Today, spiders decide when they want to visit us, and not the other way around. A good alternative to using this tag in an intranet environment would be for the spider to check last-modified dates by doing HEAD requests instead of fetching the pages.
The main problem (as I see it) with all these extra tags is that they push the indexable content down, and may therefore lower its relevancy-value, and that they otherwise just take up space.
| 9:12 pm on Jan 26, 2003 (gmt 0)|
|Someone help me to understand why... |
The biggest problem obviously is mis-information... and a major part of this is print.
If an author had million copies of his book published in 1996 and sold half it's likely that copies are still being circulated even today in some discount bookstore.
This industry segment has moved more rapidly than most.
| 9:18 pm on Jan 26, 2003 (gmt 0)|
|I believe it is Ink's Slurp spider whose default behaviour is (or was) "index,nofollow". Therefore, I include "index,follow" on my pages where it matters. |
jdMorgan, you can change the robots-terms to "all" and save a few k of space. Can someone confirm if Ink still uses this rule for indexing?
fathom, it's not only books, but there are thousands of web sites out there that advocate the misuse of these tags.
| 9:23 pm on Jan 26, 2003 (gmt 0)|
>Can someone confirm if Ink still uses this rule for indexing?
No, it doesn't. I removed all robots tags at the start of INK PFI and have been deep crawled fine by INK for non PFI pages.
Getting back to the rest of your comments, surely this shows the very young nature of the SEO/SEM industry. Most people starting out go and look at the top 50 sites and try and take what they think are the 'cleverest' things they see on those sites and incorporate them into their own. It is only hard experience, particularly in reasonably competitive areas and after a couple of penalties/bans that you really start concentrating on being careful of what you do and the useless 'short cuts' get thrown out the window.
I honestly thought of cloaking my main site a couple of years ago, to put all these garbage tags (and a few more) in a human visitor version to perpetuate the myths that these 'tricks' work - but thought that maybe that could be a trick that might just turn around and bite me in the behind!
[edited by: makemetop at 9:33 pm (utc) on Jan. 26, 2003]
| 9:30 pm on Jan 26, 2003 (gmt 0)|
<meta name="robots" content="all">
Okay, we have a confirmation from MMT that Ink no longer uses the index,nofollow rule. If anyone is an expert with Ink, it's MMT, so we have an authoritative answer. ;)
Any other reasons why someone would use the above tag other than those listed above? I'm really on a mission to set the record straight when it comes to basic metadata information.
P.S. This does not apply to the robots-terms of noindex, nofollow, or none.
| 11:13 pm on Jan 26, 2003 (gmt 0)|
Can you also confirm that Ink used to default to index,nofollow? I'm just trying to determine if my memory is correct, or if my Alzheimer's is kickin' in again... Or maybe it was a different (but top-ten) SE that used that default? (It's been three years or more since I looked at this issue.)
| 11:20 pm on Jan 26, 2003 (gmt 0)|
>Ink used to default to index,nofollow...
Don't worry, you are not wrong - that used to be the INK default. Though I'm not sure how strictly it used to adhere to it.
| 12:27 am on Jan 27, 2003 (gmt 0)|
<!-- Keyword Stuffed Comments -->
That's the one that I'm really interested in seeing some comments on. There are some rather visible names out there that advocate the use of this tag in a search engine marketing strategy.
I know there are a bunch of you reading this that use comments tags for this purpose. Would you care to share with me how they are performing and which SE's are indexing them?
| 12:43 am on Jan 27, 2003 (gmt 0)|
I had used them in the past basically repeated the description tag making the first few words a different keyphrase.
Did it help or improve anything - don't believe so (that was a couple of year ago as well).
One interesting point - which occurred on removing them, was I had accidently cut only half of the comment tag on one page.
Months later noticed that the Google snippet for the page ended in --> (and verbatim), and upon investigation found the open ended tag in the header.
Amazingly the page was ranked #1 on a couple of keywords/phrases(although don't really know if this had anything to do with that).
I pondered the idea for a while - posted it in WebmasterWorld but never got any feedback - and forgot about it.
I suspect they don't hurt anything overall - but don't help either... just weigh the page down.
| 2:18 am on Jan 27, 2003 (gmt 0)|
W3C - 3.2.4 Comments [w3.org]
Hey fathom, I too took that route back in the days because that's what everyone was doing. You learn through trial and tribulation. You quickly adapt to industry trends and make the necessary adjustments.
Some less experienced coders are at risk when using the comments tag. If you have improper syntax, you can end up commenting the entire page after the tag, I've seen it happen.
I've been running a few tests today to check the indexibility of the comments tag and I'm coming up with zero, nadda, zilch. It would be nice if there was an authoritative resource who could confirm that the comments tag does not get indexed and never has. At least not by any of the major SEs.
Again, my goal is to hopefully dispell some of the metadata myths that have been in our industry since the beginning. Hopefully by the time this thread has run its course, we'll have a better understanding of metadata and how to properly use them when applicable.
On a side note, the <!-- Comments Tags --> are not considered metadata. They are html markup. Used typically by developers to assist in organizing, providing instructions for, and navigating code.
<!-- Comments Tags --> are also used in some wysiwyg editors as a means to execute program specific code. For example, FrontPage includes...
<!-- webbot bot="include" u-include="http://www.sample.com/footer.htm" tag="body" -->
| 9:03 am on Jan 27, 2003 (gmt 0)|
|Can you also confirm that Ink used to default to index,nofollow? |
Yes, I'd heard the same thing, and I used to use index,follow on all pages I wanted spidered. Sometime last year I heard that this was no longer necessary, and I dropped it and have observed that Ink eventually does get around to spidering all my pages. Good to have it also from an official source.
As for the comment tags, I've never believed these worked, and I've searched for unique strings in my comment tags and not found them.
The way I've seen people use most keyword type strings is all wrong anyway... They're usually presented as unfocussed wish lists containing everything imaginable. If the engines did look at them, many of the lists I've seen would actually be diluting a page's density for target terms.
At this point, I use title (I think not technically a meta tag) and description. I don't always bother to use meta keywords, but when I do I try to keep them short and focussed.
I find the meta keywords tags on existing pages useful, though, to get an indication what prospective clients wanted to target. Along with the page itself, it's a quick indicator of the sophistication of what I'll be dealing with.
| 10:18 am on Jan 27, 2003 (gmt 0)|
|1. What does this tag do in today's SEM environment? |
<meta name="robots" content="all">
2. Here's a good one, how can you control how frequently the robots visit?
<meta name="revisit-after" content="5 days">
3. Here is one of the most abused areas...
<!-- Keyword Stuffed Comments -->
4. Here's another that I'm finding too many instances of...
<meta name="abstract" content="Keyword Stuffed Listing">
1. This tag does pretty much nothing unless the content is noindex or nofollow.
2. I can think of about 30 search engines that once used this tag. it was abandoned to my knowledge over 6 years ago.
3. There was a time when you couldn't surf the web without coming across a SEO page that told you to do this. SEO's back then used to think that was all you had to do until search engines got wise. You can't find these pages now because they have either gone, been changed are don't get listed in the search engines because the page uses that technique.
4. All I have to say is Altavista. Altavista was the major se that used this tag. It was an extended description tag. It's now redundant.
| 12:28 pm on Jan 27, 2003 (gmt 0)|
|It would be nice if there was an authoritative resource who could confirm that the comments tag does not get indexed . . . |
Well I know AltaVista has spoken to this in their FAQs:
|What types of Web pages does AltaVista not index? |
Comments, that is, text between <! -- and --> symbols in the source code;
I would safely assume that the other SEs . . . those that count anyway . . . follow suit.
| 4:24 pm on Jan 27, 2003 (gmt 0)|
jlr1001, nice find! I haven't read AV's FAQs in quite some time. It's nice to see they are doing what they can to dispell the myth too! And listed right after that FAQ is ths...
|13. How much information on my page is indexed? |
AltaVista indexes all words in your document except for comments. It indexes all text, ALT text for images, links (hrefs and images), anchors, titles, description, applet and ActiveX object names, the page URL, the host name, and the domain name. The crawler cannot read words that are imbedded within graphics, so these words are not indexed.
| 4:26 pm on Jan 27, 2003 (gmt 0)|
I should clear up something above by jlr1001. AV's statement regarding what types of pages it won't index may be misleading if you don't add this...
|12. What types of Web pages does AltaVista not index? |
AltaVista's crawler does not index the following types of content:
AV will still index your page, it just ignores the <!-- Comments Tags --> like any respectable SE would.
| 5:10 pm on Jan 27, 2003 (gmt 0)|
I wondered about the whole comment tag thing. I remember 6 months ago, when I was first starting to research SEO, finding a site with 181 keywords stuffed primarily in the comment tags.
This was an individual web designer who had somehow successfully managed to get his site up to the top of the pack. Near as I can tell, he did it through stuffing keywords in his comments. He was up there with quiet a few of the big boys and that was the only thing I could think of that could possibly have gotten him there.
Wish I could remember how I found him in the first place, and very glad I didn't try to emulate him! ;)
| 6:45 pm on Jan 27, 2003 (gmt 0)|
RE: comments tag
Three years ago, when I used to work for a then-successful SEO company, we did put keywords in the comment tags. We didn't overstuff (in fact, we kept it pretty lean and simply started off the comment tag with the important keyword or k. phrase).
Because we were using other strategies as well (rich content, linking, etc.) the sites placed fairly well. Since then, however, it's my conclusion that the comment tags are negligible today. And that if sites are doing well (despite their over-stuffed tags), it's because they've established history (ranked well in the past for other reasons). But I'm open to seeing that disproved here.
| 6:53 pm on Jan 27, 2003 (gmt 0)|
|As for the comment tags, I've never believed these worked, and I've searched for unique strings in my comment tags and not found them. |
So far, we have one member who has stated that they've not been able to find unique strings in their comments tags when performing searches.
I was hoping by now we would have seen some of the more senior members (geeky types) jump in and let us know what they think. Those who watch the spiders closely and know what they index.
Personally, I don't think they were ever indexed by the SE's. I don't have any proof of that, but logic tells me that it just isn't so. Think about it, look how much abuse there is now. If any weight were given to the comments tags, you'd see more authoritative resources giving detailed explanations on how to use the tag effectively.
I think somewhere in history, someone misconfigured their comments tag as fathom pointed out earlier. That broken tag got indexed and the myth began. That person then posted something somewhere and the myth just snowballed from there.
P.S. If Brett or any of the other administrators here jumped in and said that the comments tag is weighted in the algo, thousands would flock to their sites and start inserting them! ;)
That is how strong the authorities in our industry are!
| 10:11 pm on Jan 27, 2003 (gmt 0)|
|Your search - sitename-menu-relative-leftnav-txt - did not match any documents. |
| 10:21 pm on Jan 27, 2003 (gmt 0)|
>unique strings in their comments tags..
For the sake of proving this once and for-all. I will now add a comment tag stating that 'makemetop is a silly twit' to one of my websites (not my main one - in case someone wants to play around). The site is in PFI and should be spidered by all search engines (including Google because I will choose a 'freshed' site) within 7 days.
If anyone finds a site with this text indexed AND in a comment tag and sticky mails me with the URL, I will tell them if it is correct or not. This will then be verified by checking the WHOIS information - so no possibilty of cheating :)
My bet is it won't be - but like pageoneresults - it would be nice to see :)
| 10:24 pm on Jan 27, 2003 (gmt 0)|
There once was a time when comments tags were indexed by a few search engines. Those were the early days when developers were still learning and defining the "rules".
From a programmer's prospective, it's very easy to detect and omit these tags from playing into a search engine algorithm.
| 10:27 pm on Jan 27, 2003 (gmt 0)|
pageoneresults, a search for '<!--DWLayoutTable-->' returns an uninspiring set of results which includes archived copies of HTML emails for some reason?
|Your search - 'makemetop is a silly twit' - did not match any documents. |
Just checking! I'll try this as well (different comment of course :)).