I has bee often discussed if, among other things, you
use a meta refresh with a time delay of zero, then
Google might consider this similar to a 301 redirect.
I saw a post that contained this:
~ Webmaster Tools has started reporting meta refreshes as redirects
now,
~ so it will complain about them when found in the sitemap file. I can
~ vouch for that.
My questions are:
What are these indications? Does this just apply to the URLs
in sitemaps and info in GWMT?
Or
Is there evidence in the actual crawling, indexing and ranking
where the meta refresh redirect has changed or
has a more consistent behaviour somehow as of late?
Apologise for the typo,
I has bee often discussed if
should be
I has been often discussed if
Also,
To be a bit more clear in the meta refresh inconsistencies,
Some sites that use them on the index or home page
where they use a timer of Zero, Include a regular HTML link
to the new page plus _sometimes_ other details like meta robots
noindex seem to get PageRank transferred but some others do not.
(At least IME previously).
What I am asking is if there is a new definite position Google
will take on usage of these type of meta refresh redirects?
If so, where is this change evident? Does this mean that
now all cases of meta refresh with time of 0 or something
else are assured to be definitely considered as a 301
redirect or does some ambiguity in consideration
(or configuration) still exist?
Thanks,
Abracadabra
On Mar 2, 8:20 am, abracadabra wrote:
> I has bee often discussed if, among other things, you
> use a meta refresh with a time delay of zero, then
> Google might consider this similar to a 301 redirect.
> I saw a post that contained this:
> ~ Webmaster Tools has started reporting meta refreshes as redirects
> now,
> ~ so it will complain about them when found in the sitemap file. I can
> ~ vouch for that.
> My questions are:
> What are these indications? Does this just apply to the URLs
> in sitemaps and info in GWMT?
> Or
> Is there evidence in the actual crawling, indexing and ranking
> where the meta refresh redirect has changed or
> has a more consistent behaviour somehow as of late?
i am confused with this issue due to this article in netmechanic
http://www.netmechanic.com/news/vol4/promo_no15.htm which states:-
"AltaVista is the most strict: sites with the refresh attribute set to
less than 30 seconds have been banned as spam."
So are meta refreshs now acceptable to search engines or not?
Another article http://www.w3.org/QA/Tips/reback gives fairly
reasonable reasons why not to use them
I touched on meta refresh type redirects back in my posting on meta
tags at http://googlewebmastercentral.blogspot.com/2007/12/answering-more-pop... . In general, we go not recommend using this kind of redirect. If you
need to redirect users to a page, it is preferred to use a clean
server side redirect.
As Silverstall mentioned, some search engines (AltaVista is mentioned
in that article) have seen meta refresh type redirects with a short
time to redirect as a spam-signal. Personally, I find the longer time
periods problematic as well, at least for the user. Who expects to be
redirected to a different page after 30 seconds? It's just
confusing :-).
So, if you can, make sure that you use a clean and accurate server
side redirect.
Years ago when a friend was trying to give up smoking a hypnotherapy
site exposed her to 30+ seconds of watching a clock pendulum swing.
The page then refreshed into a flash movie with a soft voice
explaining the perils of smoking. She thought it was excellant up to
the point when she realised it had knackered her browsers back button
- i never forgot her telling me "luckily i lit a cigarette and calmed
down."
I reported that Google has started to act quickly on mmeta refresh as
a method of redirection (almost) on par with a 301 redirection.
I use it to redirect from one site where it's impossible to use any
other form of redirection (and isp hosting account, so no domain name
involved) to a new site with a domain name. The old site was very well
indexed, though with some dead wood as well due to numerous server
glitches in the past.
I have replaced every page from the old site with:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://
www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<title>Page moved</title>
<meta http-equiv="Content-Type" content="text/html;
charset=iso-8859-1">
<meta name="robots" content="noindex,follow" />
<meta http-equiv="Refresh" content="0; URL=http://www.newsite.com/ somepage.html">
</head>
<body>
<p>Page moved to <a href="http://www.newsite.com/somepage.html">http://
www.newsite.com/somepage.html</a></p>
</body>
</html>
Note the pages have a plain title, no keywords, descriptions nor any
content but a meta refresh and a plain html link. Trying to keep a
very low profile there.
To stimulate the re-crawling of the old site so that these meta
refresh redirects can be applied faster, I built a text sitemap of all
the pages on the old site, each of which is getting redirected in this
manner to the new site. I submitted the text sitemap on behalf of the
old site. Also of course submitetd an xml sitemap for the new site.
This involves about 160 urls.
After 1 week I started seeing urls appearing in the index for the new
site and some dropping out from the idnex for the old site. After 2-
weeks about 1/3 had been moved over the new site's index listing and
removed from the index for the old site.
Also I have seen messages in Webmaster Tools in relation to the old
site, complaining that the sitemap contains too many redirects. So the
meta refrsh rediretion does work.
All the while the old urls are dropping out of the index from the old
site and the new site is getting mroe and more of its urls indexed.
So I will state that yes, meta refresh done the way I did it works as
well (or nearly as well) as a server-side 301 redirection would.
I cannot say whether PR is getting transferred yet. I have seen
appearing for the new site. Not sure if it's because the site owner
took care to advise the linking sites to change the links or if ot
happened as a consequence of the redirection by meta refrsh working.
Under the circumstances I am very pelased with the results. Unexpected
and unhoped for, really.
> Years ago when a friend was trying to give up smoking a hypnotherapy
> site exposed her to 30+ seconds of watching a clock pendulum swing.
> The page then refreshed into a flash movie with a soft voice
> explaining the perils of smoking. She thought it was excellant up to
> the point when she realised it had knackered her browsers back button
> - i never forgot her telling me "luckily i lit a cigarette and calmed
> down."
its intresting because what you are doing is using the noindex tag
before the refresh however unless i have got it wrong you are
redirecting from an old site that you want de-indexed anyway. I think
the spam penalities used by Alta Vista and maybe google would apply to
the originating URL so my geuss is that would risk the PR of the old
url if the refresh was percieved as a spam attempt.
How do you get round the problem of a user pressing the "back" button
as IE6 and possibly other browsers make the refresh would work again,
and the user would bounce forward
The back button works fine and it takes you to the page where you
were before you went to the page which has the meta refresh on it.
You are describing problems normally seen in javscritp redirects.
I use noindex because I don't want the old page indexed - in fact I
want it deindexed asap so the new one can get indexed.
> its intresting because what you are doing is using the noindex tag
> before the refresh however unless i have got it wrong you are
> redirecting from an old site that you want de-indexed anyway. I think
> the spam penalities used by Alta Vista and maybe google would apply to
> the originating URL so my geuss is that would risk the PR of the old
> url if the refresh was percieved as a spam attempt.
> How do you get round the problem of a user pressing the "back" button
> as IE6 and possibly other browsers make the refresh would work again,
> and the user would bounce forward
Don't confuse NOINDEX with NOFOLLOW. The "noindex,follow" robots meta
tag or X-Robots-Tag instructs search engines to follow links as well
as the meta refresh, and Google passes PageRank from links at least (I
guess this kind of redirect passes PR, even without a single on-the-
page link that would pass all the link juice the redirecting page has
accumulated to the link destination).
Sebastian
> its intresting because what you are doing is using the noindex tag
> before the refresh however unless i have got it wrong you are
> redirecting from an old site that you want de-indexed anyway. I think
> the spam penalities used by Alta Vista and maybe google would apply to
> the originating URL so my geuss is that would risk the PR of the old
> url if the refresh was percieved as a spam attempt.
> How do you get round the problem of a user pressing the "back" button
> as IE6 and possibly other browsers make the refresh would work again,
> and the user would bounce forward
