38 replies to this topic

[Randy]

You forgot a word in there Michael.

You should have said I suspect that both Google and Yahoo! wrongly concluded at some point ...

Reality is it's the same type of mistake pretty much every webmaster makes on their own site(s). And if you get set in your ways, never testing alternatives, you wouldn't know it was a bad conclusion based upon faulty data.

Who knows for sure why they did it in the beginning. It could be simply the way it was always done because back in the day their snippet grabbers weren't all that great. Those automated processes have certainly gotten much better than they were back in the 90's.

But on the other hand I can see how someone might make an assumption that unconnected editoral comments might be the way to go. And that might even be right if sites never changed. Unfortunately for them that's just not the way the world works.

[St0n3y]

Flawed concluding.

I separated my robots tags:

<META NAME="ROBOTS" CONTENT="NOYDIR">
<META NAME="ROBOTS" CONTENT="NOODP">

Got spidered by Yahoo and now its not showing Yahoo Directory title/description. I'll keep track and see what Google does.

[Michael Martinez]

Flawed concluding.

I separated my robots tags:

<META NAME="ROBOTS" CONTENT="NOYDIR">
<META NAME="ROBOTS" CONTENT="NOODP">

Got spidered by Yahoo and now its not showing Yahoo Directory title/description. I'll keep track and see what Google does.

Do please let us know what Google does. I'll be interested to see if they grab the second tag or not.

[Jill]

Well it certainly doesn't bother Yahoo, Michael.

[Michael Martinez]

But "not bothering Yahoo!" doesn't tell us whether it will work. I don't think people should be using duplicate meta tags, but if a search engine will not only tolerate them but also uses the duplicates, that will help everyone to know. Think of all the possible site reviews where people might say, "You have duplicate robots tags and Google ignores the second". What if we're wrong to say that?

I'd like to know.

[qwerty]

Makes sense to me.

FWIW, I'm testing the other setup. I have a site that's listed in both ODP and the Y directory, and I've added the single robots tag:

HTML

<meta name="robots" content="noodp, noydir">

Google's been using the meta description for the snippet on searches for the site name for some time, so I won't see a change there, but MSN uses the ODP description and Yahoo has been using their directory description for years.

Neither one has picked up the change yet, and it's been about three or four days so far.

[Jill]

Think of all the possible site reviews where people might say, "You have duplicate robots tags and Google ignores the second". What if we're wrong to say that?

Most likely, anyone who has ever said that is probably wrong. I haven't tested it, but I'd be very, very surprised if that were the case.

I'm certainly interested in seeing what others tests show, but I think I already know the answer.

[St0n3y]

Still waiting to see a current cache date but I want to note that I already had the NOODP tag so with the addition of the NOYDIR (first) we'll see if Google ignores the second and starts using ODP again.

[WebGeek182]

Anyone have info on how this affects interactions with other ROBOTS directives such as NOARCHIVE, NOFOLLOW, and NOINDEX?

[qwerty]

It would be superfluous with noindex, since if you're making sure they don't index the page, you needn't worry about what they'd use for the snippet if they were to index it. As far as the others go, I'm assuming that if it works to combine noodp and noydir, it should be fine to add noarchive or nofollow. I can't test that, as the page I'm doing this with needs to be archived and followed.

[WebGeek182]

Right....meant to put INDEX. Thanks for the input. Let us know how the results turn out.

[Jill]

Right....meant to put INDEX. T

No need to tell the engines to index. They do that as the default.

[qwerty]

Official word from Google:

Multiple content values
We recommend that you place all content values in one meta tag. This keeps the meta tags easy to read and reduces the chance for conflicts. For instance:

<META NAME="ROBOTS" CONTENT="NOINDEX, NOFOLLOW">

If the page contains multiple meta tags of the same type, we will aggregate the content values. For instance, we will interpret

<META NAME="ROBOTS" CONTENT="NOINDEX">
<META NAME="ROBOTS" CONTENT="NOFOLLOW">

The same way as:

<META NAME="ROBOTS" CONTENT="NOINDEX, NOFOLLOW">

Makes sense to me. It doesn't cover how Yahoo or MSN will deal with it, but I doubt it's going to be an issue.

[Jill]

That's a good post with good info.

[Michael Martinez]

Yes. Shame on me for not having noticed that before.