Crawlable navigation

[temi]

I'm I correct in assuming that search engine spiders cannot crawl JavaScript. Ajax is a variant of JavaScript, does does it mean search engine crawlers do not crawl Ajax as well?

While on the subject, does anyone have a list of stuff search engine crawlers can crawl or cannot crawl?

 

[OldWelshGuy]

Ajax is pretty much useless if not cded properly. So the answer is maybe

Temi, rather than have a list of what spiders will and will not spider, it is better to say that if links use 'href=' , then they can be spidered and carry link juice.

Google WILL read javascript, and also flash, and WILL pull out urls as well. BUT they will only be added to the google 'to be crawled' database and when they get crawled the page where the urls were found will not pass on any link benefit (juice).

One point here, many people confuse the term 'link juice' to mean PageRank (and it it pisses me off). Link juice means a HOSt of things,
PageRank
Anchor text link benefit
Authority trust benefit
to name 3.

If we are not careful, we will end up with the term 'link juice' going pretty much the same way as 'seo' in that people will be using it without really understanding what it means. Personally I DESPISE the term SEO as I do not optimise search engines, I am a web marketer, I make sure website are the best they can be for both users and the search engines, I make sure the sites can be found by as many routes in as possible. So SEO is a load of crap to describe what I do, as the search engines are just a PART of the overall package.

E.G. I got one of my clients a half page spread in the Sunday times colour suppliment for free, as I wrote a good informative quirky piece for them and I got it picked up by the PROPER press release agencies.

[ Rant over ]

 

[temi]

Thanks very much OWG, I must confess I get link juice muddled up as well, thanks for your clarification. I will add your description of link juice to my SEO scrap post

 

[smeagain]

I always assume that unless it is a Text Link or a Linked Image and Not a JS Rollover, then it is unlikely to be crawled by ALL the Spiders. Sometimes this means that I am taking a "Better Safe than Sorry£ approach, but that suits me. If you use something that makes all your pages for you and that will include links and code etc then generally you have little control over the "Goodness" of what you end up with without lots of customisation. I am glad I produce all my own stuff including CMS so I don't have this worry.

 

[Bagi Zoltán]

Temi, from my personal experience, Ajax need to be used only at posting forms, and where crawling is not an aspect, but loading time.

 

[deluxdon]

Temi, from my personal experience, Ajax need to be used only at posting forms, and where crawling is not an aspect, but loading time.

I am second that.

DON.

 

[Gorkfu]

it is better to say that if links use 'href=' , then they can be spidered and carry link juice.

This is not always 100% true. If a link starts to looks like this at the beginning:

<a href="javascript...

Then it most likely won't get indexed.

That's why I always tell people to never use JavaScript for linking unless it is outside of the href in the <a> link tag. Using JavaScript events outside the href is much better than using JavaScript inside the href. Like this example.

<a href="http://site.com" onclick="dothis"></a>

Although a lot of time when using Ajax the links might be set up to pages that don't really need to be indexed. Mainly because those pages are used inside of other pages.

Basically to sum up... if you roll over a link and it starts with "http://" your good to go... If it starts with, "javascript:" It will most likely certainly not be indexed.

 

[Gorkfu]

Temi, from my personal experience, Ajax need to be used only at posting forms, and where crawling is not an aspect, but loading time.

That's all I use AJAX for. Although, there are many sites that have those tabbed displays that load content on the tabs instead of taking you to a new page. I think this is a nice feature for any site. However, if you use something like that you should create pages that contain the content also. That way search engines can see the data. Search engines still can't pick up AJAX linked code as far as I know.

 

[OldWelshGuy]

This is not always 100% true. If a link starts to looks like this at the beginning:

<a href="javascript...

Then it most likely won't get indexed.

That's why I always tell people to never use JavaScript for linking unless it is outside of the href in the <a> link tag. Using JavaScript events outside the href is much better than using JavaScript inside the href. Like this example.

<a href="http://site.com" onclick="dothis"></a>

Although a lot of time when using Ajax the links might be set up to pages that don't really need to be indexed. Mainly because those pages are used inside of other pages.

Basically to sum up... if you roll over a link and it starts with "http://" your good to go... If it starts with, "javascript:" It will most likely certainly not be indexed.

I agree 100%, and thought I had made that clear in my post (obviously not) Google will pull the url out of javascript, and add it to the 'pending' indix, but the link will not carry any link juice. If you use javascript for navigation, then you need to have the complete set of links in the noscript area.

Apologies if I didn't make myself clear enough :sorry:

 

[Gorkfu]

Google will pull the url out of javascript, and add it to the 'pending' indix

I don't think Google can always pull a link though... I think they look to see if there's a http:// or a www. string before actually pulling it. I've seen JavaScript that only refers to a a page without the host in front. Unless Google takes the page and slaps the host they are currently crawling in front of the link. That's just a guess on my part.

 

[OldWelshGuy]

We are now going down a slippery road.

I agree that if a site is not using absolute linking then it will cause problems. That said you should use absolute over relative links anyhow.

Google indeed does take relative links to mean on the same site. I had a problem recently where google was getting confused by relative linking and load balanced servers. I can't say too much as it is confidential, but safe to say that thanks to relative links, google got totally confused and completely screwed up the target location, assigning them to a completly different domain name.

My advice is always use absolute links.

I agree with you again though that they need to see the http &/or www. to know it is a link I did say that google would 'pull the url' . ../page.html really isn't a url it is a relative page address.