| 9:17 pm on Sep 19, 2003 (gmt 0)|
If the old pages don't exist anymore, use the 404 as it will eventually disappear from the Google index as Google is usually good about HTTP response codes.
If the pages are just changing location, path, etc use the 301 & make sure either way, you encourage people to link to the *new* location, and replace links to the old location.
From experience, the 'in bound link thing' is the only caveat with 301 redirects & google for changing the location of content.
Oh, and Welcome to WebmasterWorld!
| 5:14 am on Sep 20, 2003 (gmt 0)|
If you want to make sure you don't lose any visitors, I'd use a 301. I have a custom 404 that says the page you requested has been moved due to a site redesign, and I've seen several people hit that page and leave, instead of browsing.
But if you want the search engines to drop the old pages as soon as possible, I'd use a 404. Google's pretty good about it either way, but Inktomi obeys a 404 much much faster than a 301.
| 8:48 am on Sep 20, 2003 (gmt 0)|
I currently had the same problem. Content rested the same, filenames changed from kw1kw2.htm to kw1-kw2.htm and similar. I currently use an automatically generated .htaccess (as the static pages themselves are also automatically created) to redirect 301 the old to the new ones. Google follows the 301, but the results are not yet visible. During my regular searches I recognized a mix between the old and the new names ranking equally (well) in the serps.
I consider this method better, because I will not let any user run into whatever error page. It is already bad when he needs to click again to get to the appropiate page. I though have no idea on google long-time behaviour with this, i.e. when will I be able to remove the .htaccess controlling the redirects. Actually, there are no glitches.
[edited by: bull at 9:57 am (utc) on Sep. 20, 2003]
| 8:52 am on Sep 20, 2003 (gmt 0)|
I may want to split off a section of my web to its own domain. If I do that with a 301, will the new domain automatically get credited with the existing (external) links to that section and thus also the PR?
| 12:11 pm on Sep 21, 2003 (gmt 0)|
|I may want to split off a section of my web to its own domain. If I do that with a 301, will the new domain automatically get credited with the existing (external) links to that section and thus also the PR? |
It should do, yes. I've recently moved a whole site over to a new domain and used 301 redirects from the old domain, and the PR was transfered within a few weeks.
| 4:13 pm on Sep 21, 2003 (gmt 0)|
> Inktomi obeys a 404 much much faster than a 301.
I've not had the pleasure of that experience. Slurp is the only bot I notice that still regularly returns to lick over 404's on pages which haven't existed for months, even years in some instances.
I've finally just accepted that ol' slurpie is gonna lap around the empty bowl indefinitely.
| 6:11 pm on Sep 21, 2003 (gmt 0)|
The best way in PHP:
header('HTTP/1.1 301 Moved Permanently');
| 12:36 pm on Sep 24, 2003 (gmt 0)|
Thanks Guys. Sorry I haven't been back to comment, but it has been busy. Thanks again, it looks like the 301, even though it might be more work, seems like the route to go.
| 5:04 am on Oct 5, 2003 (gmt 0)|
Using a normal .htaccess with
Redirect permanent oldfile http*//filenew
Googlebot came back this night. She is very intelligent, did not pick a single
oldfile any more. Slurp does occasionally pick old ones though.
| 9:04 am on Oct 5, 2003 (gmt 0)|
|She is very intelligent, did not pick a single oldfile any more. |
My mileage DOES vary. Googlebot came quite regulary for pages of the old domain, always getting the redirect. From the new domain, only the index was crawled every day. PR was successfully transferred to the new domain (as shown by the Toolbar).
Then suddenly the new domain dropped out of the index. PR still there, Googlebot still crawling the index page. No deep crawls at all.
| 4:47 pm on Oct 6, 2003 (gmt 0)|
I've just started using 410 = Gone, as I saw suggestion it's faster at removing pages.
404 is like a customer service bod who just shrugs ignorantly. Google, as I understand it, comes back a few times and then gives up, in the same way a real world visitor might just dispair at customer service that can't find it's own pages.
Better to direct visitors, be they bots or not, using 301 or kill it with 410 otherwise.