Forum Moderators: open
1. What gets indexed, the frameset part or the <noframes> part (assuming both exist)?
2. Does Googlebot follow frame references (ie. "<frame src="http://...."
3. Does Googlebot follow links within the <noframes> part of the page?
4. If the answer is yes in 2 and/or 3 above, does the frameset page require a minimum PR level in order for Googlebot to follow such links?
5. Does PR pass through a "<frame src..." reference? (For example, if a framset page has PR5, with a single frame reference, without <noframes> tags, does the frame source get the benefit of the PR5 "backlink"?)
6. What PR effect, if any, occurs as a result of "frame respositioning" -- scripts that ensure that a frame is displayed only within its parent frameset? Do such "redirects" pass all the PR to framset page (and none to the framed page)? Does it matter whether the script is client-side or server-side?
1. Typically, the page that calls the framset and also the pages referenced in the frameset.
2. It didn't used to but it does now.
3. Yes
4. Not in any special way because it is in a frames page.
5. I would say that it probably does but this is extremely difficult to verfiy.
6. The same effect as with any other redirect. If it's javascript google won't even follow it, if it's server-side then it depends on may other factors.
Google has indexed my frameset but its cache for that page is completely blank.
Tthe page will appear to be blank in the cache - look at the source and you will see your frameset there. It looks blank because the framset code has to go between the <head> and <body> tags. Google's code at the top of the page breaks the frames code. It's nothing to do with PR, just simple HTML ;)
This question has already been asked and answered on this page. Use javascript and Google will not even know the redirect is there because it doesn't try and read javascripts (currently).
can I get a copy of this script from someone
thanks!
Try something like this
if(top.frames.length>0) {
void(0);
} else
top.location.href = "theframepage.html"
[tech.irt.org...]
I butchered these sources for the code I used on a site earlier this week. It's not very adaptable, but it is fairly compact. Here's the email that I send out to my colleagues:
When frames sites are indexed on search engines, the individual frames are listed on the results pages, and can be clicked through to by the user without the site loading in the navigation frames, resulting in orphaned pages, a stranded user and no business for the client. Combat the problem with this code:
Firstly, make sure your frames are named, and that you call the frameJammer() function using the onLoad event. This will usually be in your index.html page:
<frameset rows="50,50" onLoad="frameJammer()">
<frame name="topFrame" src="topframe.html">
<frame name="bottomFrame" src="bottomframe.html">
</frameset>
This example has 'topFrame' as the navigation frame, and 'bottomFrame' as the content frame. The orphaned pages that we would expect to be listed on search engines should actually be loaded into 'bottomFrame'. Also, you need to add the frameJammer() function into the <HEAD> section of index.html:
function frameJammer()
{
var pageURL = location.search;
if (pageURL.indexOf("~bottomFrame")!= -1)
{
pageURL = pageURL.substring(1);
var tilde = pageURL.lastIndexOf('~');
var pagePath = pageURL.substring(0,tilde);
var pageFrame = pageURL.substring(tilde+1);
eval("top."+pageFrame+".location.replace('"+pagePath+"')");
}
}
This fuction assumes the frame 'bottomFrame' is being used. If you are using a different name, you will have to change the reference in the function (but leave the tilde intact).
In the <HEAD> section of every content page (i.e. page that should load into 'bottomFrame'):
if (window.name!='bottomFrame' &&!((self.innerHeight == 0) && (self.innerWidth == 0)))
top.location.replace('index.html?'+location.pathname+location.search+'~bottomFrame');
Once again, 'bottomFrame' is referenced. Change if needed.
How does it work?
The code in the content page checks to see if it's loaded into 'bottomFrame' (i.e. correctly positioned in the appropriate frameset). If not, it calls the index page, and sends a reference to itself (path to the file and any parameters) and a reference to 'bottomFrame' for the frameJammer function to access.
The frameJammer() function checks to see if it needs to do anything (is '~bottomFrame' in the refering URL?), and if so, splits the parameter string back into its component parts, and uses the parts to load the page into the frameset correctly.
If you need to trap more than one frame, just expand the second line of the frameJammer() function:
if (pageURL.IndexOf("~bottomFrame") ¦¦ pageURL.IndexOf("~topFrame"))
"GET /?internal-page.html~mainframe HTTP/1.0" 200 13249 "-" "Googlebot/2.1 (+http://www.googlebot.com/bot.html)"
1. internal-page.html is wrong syntax for this page should be Internal-Page.html
2. Actually what it pulled was the index page, '13249' kb is for that.
I have never seen Googlebot doing this on my site before.
