Forum Moderators: open
I got into a conversation yesterday about wether or not to use absolute (ie www.example.com/product/index.htm) or relative (ie "/product/index.htm") when linking to pages in a menu.
Does it make in difference in terms of spreading pagerank?
Does anyone know if it makes any difference at all? What difference it makes and why?
Help us resolve our differences and get back to normal life.
Cheers
John
[edited by: pageoneresults at 1:07 pm (utc) on April 22, 2004]
[edit reason] Removed URI Reference - Please refer to TOS [/edit]
There are some practical plusses and minuses to each, of which I imagine you are well aware (chiefly "ease of maintenance: and simplicity of coding vs. making sure a page printed from your site displays a full URL and that the in-site links on that HTML page someone saves really work, rather than pointing to a non-existent file on their computer)
But when it comes to the main SEO question --is a page rewarded (e.g., with a higher pagerank in Google) for absolute links to other pages on the same site? --all I seem to find is speculation.
Theoretically, I understand the conjecture that absolute links are favored (at least in Google). But consider that the SE's have to first resolve all the relative links to absolute links anyway, and I'm not so sure. And if it is true, as widely thought, that Google recognizes sites on the same serverand can penalize tricks such a set of sites tries to play on the SE's, shouldn't Google also be able to figure out that specific links on a page are to pages on the same site(even if each page is treated "independently")?
My gut feeling, until I see hard evidence, is that any SEO gain from absolute links is marginal. The decision of which to use should be based on the sorts of clearcut reasons mentioned above.
Let's say your official domain name is comprised of two or more words, without hyphens, e.g., "jjconstruction.com". But, for those who think/guess at "jj-construction.com" you also purchase that variant and direct the server to serve the same pages to anyone who makes that mistake.
So, someone gets to your site via the variant name. When they follow a relative link they will get "jj-construction.com/subpage.html". So too will the spiders that visit you. And that's the version that will end up listed in the SE (at least for those pages -- you may get a mixture across SE's and even within the same SE).
But if you wish to have the correct listing of those pages(sans hyphens) in the search engines, the only way you can assure that(in this scenario) is to use absolute links with the correct domain name.
Worth considering, I think.
It would be useful to know what weight if any, google or other SEs put on internal relative linking as opposed to absolute.
It would also be useful to know if the googlebot or other spiders are more readily willing to follow absolute or relative linking when indexing new pages and domains.
I wonder!
John
There are some recent developments in search engines, that make absolute addressing more attractive than ever.
This is an older article at Search Engine World but it still applies in today's SEM environment.
Add up the five points above, and absolute addressing wins hands down. Some folks think absolute is harder to maintain, but I don't.
But if you wish to have the correct listing of those pages(sans hyphens) in the search engines, the only way you can assure that(in this scenario) is to use absolute links with the correct domain name.
Not quite true. Using that method won't prevent your homepage from being indexed under the wrong domain. And since the homepage is usually the focus of the majority of inbound links, losing it to a different domain can be quite costly. (This is a huge problem with Yahoo right now).
The best way to handle having multiple domains pointing to your site is to set them up as 302 redirects. That way, no one (including bots) is ever served a page from a domain that you don't want indexed.
But absolute urls are still a good idea, because it helps protect you from redirect/dns hijackers.
The best way to handle having multiple domains pointing to your site is to set them up as 302 redirects.
WG, can you expand on the 302 a little bit? I've always recommended a 301 for multiple domains pointing to a site. That way the pointed domain is not indexed. With a 302, which is a temporary redirect and instructs the userAgent to maintain the original URI, wouldn't that cause some possible duplicate content issues?
The requested resource resides temporarily under a different URI. Since the redirection might be altered on occasion, the client SHOULD continue to use the Request-URI for future requests. This response is only cacheable if indicated by a Cache-Control or Expires header field.
This is an older article at Search Engine World but it still applies
The five-year old article you cite makes some good points in favor of absolute links(several of which I alluded to in my first post).
But when it comes to search engine ranking--the question raised in this thread-- the argument is significantly weaker:
My feeling is that the absolute links make it into the link counting algos of a couple of (remaining nameless) search engines. . . .
Note that this is simply a "feeling". The argument that follows does not present evidence of any search engine actually counting absolute links differently. And that is what we're really after here, isn't it?
Not quite true. Using that method won't prevent your homepage from being indexed under the wrong domain.
I did not say otherwise. Please re-read the line. I wrote:
if you wish to have the correct listing of those pages (sans hyphens) in the search engines [emphasis added]
I do not deny that there is still a problem of the first page visited (whether home page or some other) being indexed with an alternate domain name. I simply did not address the matter. My only point I was making was that using absolute links could, in such cases, be of some help,not that it could solve the whole problem.
I redid a sustainable agriculture site in 2002 that gets visitors from places like rural Namibia and Brazil. Keeping bandwidth low and page load times minimal was critical. At that time, our tests suggested that relative URLs loaded faster in some browsers. Probably not so true nowadays... Does anyone know for sure?
My rule is, if it increases SEO and doesn't inconvenience the user, do it. But if it does inconvenience the user even slightly, think very carefully before implementing it, even if SEs like it.
