Duplicate Content Problem with Renamed Pages?

[shady]

I have a number of optimised pages on my site which are indexed in google.

These are the format make_widgetname.html

I have renamed these to widgetname_make.html

I have left the original pages on the site, but the links now point to the new pages.

As make_widgetname.html is a clone of widgetname_make.html, should I be worried in case google thinks of this as duplication of information?

[Brett_Tabke]

Yes, they very well could be seen it as dupe content. You have to remove the old pages or put in redirects where they were to the new content.

[shady]

Thanks Brett, I'll get to it!
They are html pages, so should I simply use meta HTTP-EQUIV="REFRESH" or is there a more suitable way to do it?

[sun818]

If you have Apache, you can also place a redirect line inside .htaccess. This will take precedence over any HTML file on the server.

[Brett_Tabke]

I'd do as sun818 suggests. If you don't have that, a meta refresh should suffice. Make sure to put MOVED in the title of the old pages.

[piskie]

Incidentaly, I have a theory (non-proven) that Google and Co prefer red-widget.htm to red_widget.htm

[shady]

sun818 - I do have apache - how would I code this in the .htaccess (it is multiple files). Is this telling the spider that the page has moved rather than just serving up a different page?

Brett - In addition to the .htaccess mod, I should also simply replace the contents of the TITLE tags MOVED

Thank you both for your help!

p.s. Piskie - thanks but I'll try not to complicate things further at the moment. How did your theory come about, btw?

[sun818]

Brett's answer is more complete :)

[edited by: sun818 at 7:07 pm (utc) on Nov. 5, 2002]

[Brett_Tabke]

If you have the old content in one directory:

Redirect 301 /oldserverpath/ http://www.domain.com/newurlpath/

If it is a specific file:

Redirect 301 /fred.html http://www.domain.com/foo/boo/fred.html

Old server path first, and the URL second. (first part is your directory path and the second a full url)
You can use quite a few before it's a problem. I have one with 250 urls in it that redirect.

[eplus]

If there are a lot of these pages you can use a custom error page and a bit of php to redirect people on the fly.
You just have to use something like
<code>
$url=explode("/",$REQUEST_URI);
if ($url[1]=="widgetpage.html"){
header("Location: [yoursite.com...]
}
if ($url[1]=="widgetpage2.html"){
header("Location: [yoursite.com...]
}else {
header("Location: [yoursite.com...]
}
</code>
as it stands this isn't much more efficient than editing your .htaccess file (probably a bit more time consuming) but if you have used a logical renaming system that can be automated this might be the way forward. It also gives you another way of doing it if you can't use .htaccess but do have custom error pages.

I believe this is as acceptable to google as .htaccess but I'm sure I can rely on other to correct me if I'm wrong.
If you go with this but would like some help on php try devnetwork.net forum, a valuable resource.

[piskie]

Shady
I did some housekeeping changes, nothing drastic mainly background colours and images to change visual apearances.

I changed the red_widgets.htm to red-widgets.htm as I went.

The plan was to optimise after the visual makeover was complete because it was a new client who wanted to see a quick visual result. However on the next upgrade, several pages in the 20 to 30 positions became 18 to 25 positions.

One reason I changed the filenames was to see if Google would drop deep pages if not found and deep crawl their replacements and then index the new pages seamlessly.
The answer is in this case it was seamless except for the dance period.

The observation on file name seperators was an incidental factor and I would like to repeat it sometime in a more controled way with no other simultaeneous changes.

[shady]

I've had googlebot recrawl (and crawl and crawl) my site before making these changes. Could I be in trouble?