So according to Google, the correct way to indicate that a page has moved is to serve a HTTP 301 response. Great but what about those of us using ISP storage?!?!?
I’ve noticed two things that I’ve taken steps to try and fix. The first is that the old blog comes up in search results but the new blog doesn’t. I have updated all the old blog pages to point to the equivalent page on the new blog. I can’t do a 301 but with the links in place the new blog should get seen by Google. I’ll retire the old blog eventually but I wanted to give it some time so that people can update bookmarks and such.
The second thing I’ve noticed is that adding in all the feed links means the feed .xml files are coming up in Google results. Yuck. I’ve added rel=’nofollow’ to all of the feed links to hopefully prevent this. I don’t have access to robots.txt on either of my blog sites so I can’t fix it using robots.txt (again, the “correct” way is not possible).
Given the time it takes to update Google’s cache I’d say the old blog will have to stay up for at least a few more weeks. Luckily I haven’t changed ISPs so there’s no hurry to do that but it’s still annoying to me that I can’t fix it properly.
Update 17 July 2007
That’s… interesting.
With just my redirection bits stuck into my old pages Google has shifted over to my new location on blogspot. It’s as if I had included 301s except that I didn’t.
So if you want to move a page in Google and you can’t put up a HTTP 301 page, put content like this on instead.
This goes in the header, it redirects automatically (for most people anyway).
<meta http-equiv="refresh" content="5;url=NEW_URL">
This goes in the body. It gives people a real link to click if the redirect doesn’t work.
You will be redirected to <a href="NEW_URL">NEW_URL</a> in 5 seconds…
At least… that’s what I used. Ensure that you do this for every page and ensure that NEW_URL is set correctly. I used a quickly hacked-together perl script to tweak all my HTML files to have this.
