Recently went from page 1 to page nowhere after renaming my photo gallery pages to descriptive file names.
About 150/300 filenames were suddenly returning 404 after being deleted and replaced by new descriptive filenames. The spider can find that overnight. I recreated all 300 old filenames and used meta refresh on all 300 of the photo files. After several crawls it's still showing the old files as 404. It sure didn't find my redirects overnight.
I experimentally didn't redirect a few and used the URL removal tool for those. That worked just fine and quickly. But meta-refresh seems to be pretty useless so far. I used meta-refresh on most of them because I thought they might have backlinks that I want to keep.
Do I need also a noindex,nofollow on each of these old filename files to get the spider to see it? What a mess this is! There is no content at all in the old filename files, just the meta-refresh set to 0 and empty body. Is nofollow okay, or is the meta-refresh considered a link that should be followed? You see how crazy I am. I would offhand say no, that's not a link, but who knows?
I saw here that Phil said a simple meta-refresh won't cut it.
Please tell me what to do! This is a nightmare. I did much work to redo those gallery pages and upload that monster gallery on 56K, and now I have no ranking since August 4 because of it.
(Obviously, you replace [myoldpage] and [mynewpage] with filenames that you are using.
Notice that after the [Redirect 301], you state only the filename (or subfolder/filename) of the file you are redirecting from. This MUST be in or below the current folder.
Also note that the .htaccess file can be created/edited with any plain text editor, but you must ensure that the filename is [.htaccess] and there is no filename extension like .txt stuck on the end. (Many editors will default to that and you have to remove the ending if it appears.)
Before you invest too much time and effort in this, try it with just ONE URL in case your hosting company does not allow you to use .htaccess at all.
If you cannot use the .htaccess at all, you could check out whether you can use PHP to issue the 301.
If you are limited to HTML with no server-side scripting language such as PHP or ASP available to you, I have run out of ideas.
> Recently went from page 1 to page nowhere after renaming my photo > gallery pages to descriptive file names.
> About 150/300 filenames were suddenly returning 404 after being > deleted and replaced by new descriptive filenames. The spider can find > that overnight. I recreated all 300 old filenames and used meta > refresh on all 300 of the photo files. After several crawls it's > still showing the old files as 404. It sure didn't find my redirects > overnight.
> I experimentally didn't redirect a few and used the URL removal tool > for those. That worked just fine and quickly. But meta-refresh seems > to be pretty useless so far. I used meta-refresh on most of them > because I thought they might have backlinks that I want to keep.
> Do I need also a noindex,nofollow on each of these old filename files > to get the spider to see it? What a mess this is! There is no content > at all in the old filename files, just the meta-refresh set to 0 and > empty body. Is nofollow okay, or is the meta-refresh considered a link > that should be followed? You see how crazy I am. I would offhand say > no, that's not a link, but who knows?
> I saw here that Phil said a simple meta-refresh won't cut it.
> Please tell me what to do! This is a nightmare. I did much work to > redo those gallery pages and upload that monster gallery on 56K, and > now I have no ranking since August 4 because of it.
If all of Robbo's ideas don't work (try those first!) you could always go back to the 404 option + removal request for those pages that you haven't done it for yet. The problem then would be outdated links of course and there's really no easy way to go around correcting those other than emailing people who have linked there.
Google seems to keep the old files in their system for quite a long time. I'm still waiting (6 months +) for google to stop trying to access a few files which we linked to for a day but then replaced with a trailing slash version....
> (Obviously, you replace [myoldpage] and [mynewpage] with filenames > that you are using.
> Notice that after the [Redirect 301], you state only the filename (or > subfolder/filename) of the file you are redirecting from. This MUST > be in or below the current folder.
> Also note that the .htaccess file can be created/edited with any plain > text editor, but you must ensure that the filename is [.htaccess] and > there is no filename extension like .txt stuck on the end. (Many > editors will default to that and you have to remove the ending if it > appears.)
> Before you invest too much time and effort in this, try it with just > ONE URL in case your hosting company does not allow you to > use .htaccess at all.
> If you cannot use the .htaccess at all, you could check out whether > you can use PHP to issue the 301.
> If you are limited to HTML with no server-side scripting language such > as PHP or ASP available to you, I have run out of ideas.
> Robbo
> On Aug 28, 10:39 am, carole wrote:
> > Recently went from page 1 to page nowhere after renaming my photo > > gallery pages to descriptive file names.
> > About 150/300 filenames were suddenly returning 404 after being > > deleted and replaced by new descriptive filenames. The spider can find > > that overnight. I recreated all 300 old filenames and used meta > > refresh on all 300 of the photo files. After several crawls it's > > still showing the old files as 404. It sure didn't find my redirects > > overnight.
> > I experimentally didn't redirect a few and used the URL removal tool > > for those. That worked just fine and quickly. But meta-refresh seems > > to be pretty useless so far. I used meta-refresh on most of them > > because I thought they might have backlinks that I want to keep.
> > Do I need also a noindex,nofollow on each of these old filename files > > to get the spider to see it? What a mess this is! There is no content > > at all in the old filename files, just the meta-refresh set to 0 and > > empty body. Is nofollow okay, or is the meta-refresh considered a link > > that should be followed? You see how crazy I am. I would offhand say > > no, that's not a link, but who knows?
> > I saw here that Phil said a simple meta-refresh won't cut it.
> > Please tell me what to do! This is a nightmare. I did much work to > > redo those gallery pages and upload that monster gallery on 56K, and > > now I have no ranking since August 4 because of it.
> (Obviously, you replace [myoldpage] and [mynewpage] with filenames > that you are using.
> Notice that after the [Redirect 301], you state only the filename (or > subfolder/filename) of the file you are redirecting from. This MUST > be in or below the current folder.
> Also note that the .htaccess file can be created/edited with any plain > text editor, but you must ensure that the filename is [.htaccess] and > there is no filename extension like .txt stuck on the end. (Many > editors will default to that and you have to remove the ending if it > appears.)
> Before you invest too much time and effort in this, try it with just > ONE URL in case your hosting company does not allow you to > use .htaccess at all.
> If you cannot use the .htaccess at all, you could check out whether > you can use PHP to issue the 301.
> If you are limited to HTML with no server-side scripting language such > as PHP or ASP available to you, I have run out of ideas.
> Robbo
> On Aug 28, 10:39 am, carole wrote:
> > Recently went from page 1 to page nowhere after renaming my photo > > gallery pages to descriptive file names.
> > About 150/300 filenames were suddenly returning 404 after being > > deleted and replaced by new descriptive filenames. The spider can find > > that overnight. I recreated all 300 old filenames and used meta > > refresh on all 300 of the photo files. After several crawls it's > > still showing the old files as 404. It sure didn't find my redirects > > overnight.
> > I experimentally didn't redirect a few and used the URL removal tool > > for those. That worked just fine and quickly. But meta-refresh seems > > to be pretty useless so far. I used meta-refresh on most of them > > because I thought they might have backlinks that I want to keep.
> > Do I need also a noindex,nofollow on each of these old filename files > > to get the spider to see it? What a mess this is! There is no content > > at all in the old filename files, just the meta-refresh set to 0 and > > empty body. Is nofollow okay, or is the meta-refresh considered a link > > that should be followed? You see how crazy I am. I would offhand say > > no, that's not a link, but who knows?
> > I saw here that Phil said a simple meta-refresh won't cut it.
> > Please tell me what to do! This is a nightmare. I did much work to > > redo those gallery pages and upload that monster gallery on 56K, and > > now I have no ranking since August 4 because of it.
> Recently went from page 1 to page nowhere after renaming my photo > gallery pages to descriptive file names.
> About 150/300 filenames were suddenly returning 404 after being > deleted and replaced by new descriptive filenames. The spider can find > that overnight. I recreated all 300 old filenames and used meta > refresh on all 300 of the photo files. After several crawls it's > still showing the old files as 404. It sure didn't find my redirects > overnight.
> I experimentally didn't redirect a few and used the URL removal tool > for those. That worked just fine and quickly. But meta-refresh seems > to be pretty useless so far. I used meta-refresh on most of them > because I thought they might have backlinks that I want to keep.
> Do I need also a noindex,nofollow on each of these old filename files > to get the spider to see it? What a mess this is! There is no content > at all in the old filename files, just the meta-refresh set to 0 and > empty body. Is nofollow okay, or is the meta-refresh considered a link > that should be followed? You see how crazy I am. I would offhand say > no, that's not a link, but who knows?
> I saw here that Phil said a simple meta-refresh won't cut it.
> Please tell me what to do! This is a nightmare. I did much work to > redo those gallery pages and upload that monster gallery on 56K, and > now I have no ranking since August 4 because of it.
Hi Craig, A 301 redirect is a redirect that appears with HTTP status response 301 in the server access log, so a suggestion would be to check in the server access log what is the HTTP status response for a specific request, for a specific URL and a specific user agent.
They won't appear like a 301 in the response, however, Google does TREAT them like one. I moved a domain about a year ago with zero time redirects and it worked, but it wasn't nearly as fast as a 301, over a month.
> Hi Craig, > A 301 redirect is a redirect that appears > with HTTP status response 301 > in the server access log, > so a suggestion would be to check in > the server access log what is > the HTTP status response for a specific > request, for a specific URL and a > specific user agent.
> Cristina.
> On Aug 28, 1:23 pm, cass-hacks wrote:
> > I've read both Sebastian and JohnMu saying that a 0 second redirect is > > taken the same as a 301, although I haven't tried it myself.
His current server environment isn't going to allow for a 301.
To answer your first question: If you noindex,nofollow your meta redirect, then the bot is never going to make it to the old pages. Leave the redirect in place for the time being.
If you want to 301, then you will need to switch server environments. Since you're using .mac, I assume you have your own server made from an old Mac using DSL or cable modem to link it to the internet. For a few months, move your site to another server and let Google catch up to the 301 changes then move back to your Mac.
> Hi Craig, > A 301 redirect is a redirect that appears > with HTTP status response 301 > in the server access log, > so a suggestion would be to check in > the server access log what is > the HTTP status response for a specific > request, for a specific URL and a > specific user agent.
> Cristina.
> On Aug 28, 1:23 pm, cass-hacks wrote:
> > I've read both Sebastian and JohnMu saying that a 0 second redirect is > > taken the same as a 301, although I haven't tried it myself.- Hide quoted text -
I suppose one way to check how redirection via meta refresh tag works is to look which URL is cached by Google and what is the content of the cached copy.
I think that redirection at server level (when possible) is preferrable to redirection via refresh meta tag.
> I suppose one way to check how > redirection via meta refresh tag works is to > look which URL is cached by > Google and what is the content of the cached copy.
> I think that redirection at server level (when possible) > is preferrable to redirection via refresh meta tag.
Hi Webado, You mention a regular html link to the new page in the old page with the refresh meta tag.
It is not clear how bots follow the meta refresh tag and a zero delay can be considered different from a non-zero delay.
A non-zero delay is the usual set-up for the meta refresh tag, and it is intended for people, not bots, and people can see during the non-zero delay that they are redirected to the link you mentioned, and if the redirection via refresh meta tag does not work they can click on that link and go to the new page.
A meta refresh tag with zero delay is intended for bots, not for browsers used by people, and I think that the intended redirect should be done at server level instead, with clear HTTP status response 301 or 302.
> Isn't it advisable to also have a simple, regular html link in the > body of the page with the meta refresh?
> On Aug 28, 9:29 am, cristina wrote:
> > I suppose one way to check how > > redirection via meta refresh tag works is to > > look which URL is cached by > > Google and what is the content of the cached copy.
> > I think that redirection at server level (when possible) > > is preferrable to redirection via refresh meta tag.
Inded doing any of that redirection at the server level is ideal, but when that option is not available (like on most free hosting acounts), keeping a page with meta refresh AND an html link will help robots and visitors alike. Not everybody allows meta refresh (when I use Firefox I disable it for troubleshooting) and if robots don't follow it, at least they will follow the html link.
> Hi Webado, > You mention a regular html link to the new page > in the old page with the refresh meta tag.
> It is not clear how bots follow the meta > refresh tag and a zero delay can be considered > different from a non-zero delay.
> A non-zero delay is the usual set-up for > the meta refresh tag, and it is intended > for people, not bots, and people can see during > the non-zero delay that they are redirected > to the link you mentioned, > and if the redirection via refresh meta tag does not work > they can click on that link and go to the new page.
> A meta refresh tag with zero delay is intended > for bots, not for browsers used by people, > and I think that the intended redirect should > be done at server level instead, with > clear HTTP status response 301 or 302.
> Cristina.
> On Aug 28, 2:36 pm, webado wrote:
> > Isn't it advisable to also have a simple, regular html link in the > > body of the page with the meta refresh?
> > On Aug 28, 9:29 am, cristina wrote:
> > > I suppose one way to check how > > > redirection via meta refresh tag works is to > > > look which URL is cached by > > > Google and what is the content of the cached copy.
> > > I think that redirection at server level (when possible) > > > is preferrable to redirection via refresh meta tag.
Oh and when I've had to use meta refresh it's always been with a 0 delay anyway. For visitors who do not disable meta refresh that will be an instant transfer, and for those who do, it doesn't matter, all they can do to continue is click the link.
> Hi Webado, > You mention a regular html link to the new page > in the old page with the refresh meta tag.
> It is not clear how bots follow the meta > refresh tag and a zero delay can be considered > different from a non-zero delay.
> A non-zero delay is the usual set-up for > the meta refresh tag, and it is intended > for people, not bots, and people can see during > the non-zero delay that they are redirected > to the link you mentioned, > and if the redirection via refresh meta tag does not work > they can click on that link and go to the new page.
> A meta refresh tag with zero delay is intended > for bots, not for browsers used by people, > and I think that the intended redirect should > be done at server level instead, with > clear HTTP status response 301 or 302.
> Cristina.
> On Aug 28, 2:36 pm, webado wrote:
> > Isn't it advisable to also have a simple, regular html link in the > > body of the page with the meta refresh?
> > On Aug 28, 9:29 am, cristina wrote:
> > > I suppose one way to check how > > > redirection via meta refresh tag works is to > > > look which URL is cached by > > > Google and what is the content of the cached copy.
> > > I think that redirection at server level (when possible) > > > is preferrable to redirection via refresh meta tag.
A HTML link is a good idea as some people block META redirects. As for 301/302 vs META....that's just fine and dandy advice to a person using a server that supports them. Unfortunately, not everyone has that option. Setting up severs attached to DSL and cable modems is becoming quite popular however a good portion of the people setting them up know very little about the differences in server technology (and then there are those that pay no attention to the proper maintenance of the server and get hacked).
> Hi Webado, > You mention a regular html link to the new page > in the old page with the refresh meta tag.
> It is not clear how bots follow the meta > refresh tag and a zero delay can be considered > different from a non-zero delay.
> A non-zero delay is the usual set-up for > the meta refresh tag, and it is intended > for people, not bots, and people can see during > the non-zero delay that they are redirected > to the link you mentioned, > and if the redirection via refresh meta tag does not work > they can click on that link and go to the new page.
> A meta refresh tag with zero delay is intended > for bots, not for browsers used by people, > and I think that the intended redirect should > be done at server level instead, with > clear HTTP status response 301 or 302.
> Cristina.
> On Aug 28, 2:36 pm, webado wrote:
> > Isn't it advisable to also have a simple, regular html link in the > > body of the page with the meta refresh?
> > On Aug 28, 9:29 am, cristina wrote:
> > > I suppose one way to check how > > > redirection via meta refresh tag works is to > > > look which URL is cached by > > > Google and what is the content of the cached copy.
> > > I think that redirection at server level (when possible) > > > is preferrable to redirection via refresh meta tag.
> Inded doing any of that redirection at the server level is ideal, but > when that option is not available (like on most free hosting acounts), > keeping a page with meta refresh AND an html link will help robots > and visitors alike. Not everybody allows meta refresh (when I use > Firefox I disable it for troubleshooting) and if robots don't follow > it, at least they will follow the html link.
> On Aug 28, 10:06 am, cristina wrote:
> > Hi Webado, > > You mention a regular html link to the new page > > in the old page with the refresh meta tag.
> > It is not clear how bots follow the meta > > refresh tag and a zero delay can be considered > > different from a non-zero delay.
> > A non-zero delay is the usual set-up for > > the meta refresh tag, and it is intended > > for people, not bots, and people can see during > > the non-zero delay that they are redirected > > to the link you mentioned, > > and if the redirection via refresh meta tag does not work > > they can click on that link and go to the new page.
> > A meta refresh tag with zero delay is intended > > for bots, not for browsers used by people, > > and I think that the intended redirect should > > be done at server level instead, with > > clear HTTP status response 301 or 302.
> > Cristina.
> > On Aug 28, 2:36 pm, webado wrote:
> > > Isn't it advisable to also have a simple, regular html link in the > > > body of the page with the meta refresh?
> > > On Aug 28, 9:29 am, cristina wrote:
> > > > I suppose one way to check how > > > > redirection via meta refresh tag works is to > > > > look which URL is cached by > > > > Google and what is the content of the cached copy.
> > > > I think that redirection at server level (when possible) > > > > is preferrable to redirection via refresh meta tag.
> > Inded doing any of that redirection at the server level is ideal, but > > when that option is not available (like on most free hosting acounts), > > keeping a page with meta refresh AND an html link will help robots > > and visitors alike. Not everybody allows meta refresh (when I use > > Firefox I disable it for troubleshooting) and if robots don't follow > > it, at least they will follow the html link.
> > On Aug 28, 10:06 am, cristina wrote:
> > > Hi Webado, > > > You mention a regular html link to the new page > > > in the old page with the refresh meta tag.
> > > It is not clear how bots follow the meta > > > refresh tag and a zero delay can be considered > > > different from a non-zero delay.
> > > A non-zero delay is the usual set-up for > > > the meta refresh tag, and it is intended > > > for people, not bots, and people can see during > > > the non-zero delay that they are redirected > > > to the link you mentioned, > > > and if the redirection via refresh meta tag does not work > > > they can click on that link and go to the new page.
> > > A meta refresh tag with zero delay is intended > > > for bots, not for browsers used by people, > > > and I think that the intended redirect should > > > be done at server level instead, with > > > clear HTTP status response 301 or 302.
> > > Cristina.
> > > On Aug 28, 2:36 pm, webado wrote:
> > > > Isn't it advisable to also have a simple, regular html link in the > > > > body of the page with the meta refresh?
> > > > On Aug 28, 9:29 am, cristina wrote:
> > > > > I suppose one way to check how > > > > > redirection via meta refresh tag works is to > > > > > look which URL is cached by > > > > > Google and what is the content of the cached copy.
> > > > > I think that redirection at server level (when possible) > > > > > is preferrable to redirection via refresh meta tag.
I still think the only way to check redirection via meta tag is to see which URLs involved in the redirection by meta refresh tag are indexed and cached by Google and if possible to check server access logs to see if the redirection via meta refresh tag is followed.
There is a difference between a bot following a meta refresh redirection and following the link in the body of the page. If it follows the redirect only one of the URLs will appear in the index, if it crawls the page and follows the link to the new page then the bot could index both URLs. Also I am not sure what kind of redirect will be from a meta refresh tag with zero delay, will it be considered like a permanent redirect HTTP status 301 or temporary redirect HTTP status 302?
Is redirection via refresh meta tag considered in the same way (like a redirection 301 or 302 or no redirection) by all(-ish) search engines and by all datacenters?
Well, guys I'm back after a few hours sleep. Moving to the apache server on my modem hookup temporarily doesn't sound good - and I could get hacked, and I don't keep my computer on 24/7 either. I tried .htaccess and .Mac informed me in an alert that I can't begin a filename with "." because it's used by their system. I've already read in their forums that .php isn't allowed either.
Actually, I don't why people would link to those individual gallery pages. What I've seen is that they hotlink to the .jpgs themselves in some online forums. I had a "Please don't hotlink. Download the picture instead" notice. Renaming the files and jpgs breaks their hotlinks, and that's good. Even linking to an individual gallery page is a large file transfer. I was getting 99% file transfer for the 2/ week period a few months ago from .Mac and a warning letter that they take me offline for the period if it hits 100%. That prompted me to go ahead and rename, which need to be descriptive names anyway. It was odd that Google now has approx. 150/300 show as 404. Maybe the crawl has only picked up the first half so far.
What about keeping the 0 meta-refresh I have now on all 300 files (which truly makes them 200 - even if the bot hasn't been by and noticed it) and adding robots "noindex,nofollow" to each file, so that it will qualify for the URL Removal tool? The tool seems to work and rather quickly. Some of the new files are showing up in the index.
When I did this a year ago, Sebastian said just meta-refresh 0 would do it, and it took a couple of weeks to get back to page 1 in Google. But this is a big directory, whereas last year I just renamed 10 files. The changefreq is yearly, but doesn't the bot also look at the last modified file date? (.Mac does deliver that in the http header). They are all modified this month.
Well, guys I'm back after a few hours sleep. Moving to the apache server on my modem hookup temporarily doesn't sound good - and I could get hacked, and I don't keep my computer on 24/7 either. I tried .htaccess and .Mac informed me in an alert that I can't begin a filename with "." because it's used by their system. I've already read in their forums that .php isn't allowed either.
Actually, I don't why people would link to those individual gallery pages. What I've seen is that they hotlink to the .jpgs themselves in some online forums. I had a "Please don't hotlink. Download the picture instead" notice. Renaming the files and jpgs breaks their hotlinks, and that's good. Even linking to an individual gallery page is a large file transfer. I was getting 99% file transfer for the 2/ week period a few months ago from .Mac and a warning letter that they take me offline for the period if it hits 100%. That prompted me to go ahead and rename, which need to be descriptive names anyway. It was odd that Google now has approx. 150/300 show as 404. Maybe the crawl has only picked up the first half so far.
What about keeping the 0 meta-refresh I have now on all 300 files (which truly makes them 200 - even if the bot hasn't been by and noticed it) and adding robots "noindex,nofollow" to each file, so that it will qualify for the URL Removal tool? The tool seems to work and rather quickly. Some of the new files are showing up in the index.
When I did this a year ago, Sebastian said just meta-refresh 0 would do it, and it took a couple of weeks to get back to page 1 in Google. But this is a big directory, whereas last year I just renamed 10 files. The changefreq is yearly, but doesn't the bot also look at the last modified file date? (.Mac does deliver that in the http header). They are all modified in August 2007.
