AJAX and SEO

[dak888]

Hello all,

First, I'm very new to this AJAX stuff and I'm putting things together with Dreamweavers new Spry widgets. They had a great demo of dynamically updating a page using an xml file. Their demo page is here http://labs.adobe.com/technologies/s...cts/index.html

I got it working fairly easily for what we want to do but I'm not sure how readable the page is. By utilizing the xml file for data, my page is pretty much completely blank (when you view the source) for spiders. This makes it a not very attractive option for us. Here is our test page: http://www.continentalwindowfashions.com/spry-fun3.html

So my question, are xml files readable when used in this manner? I know the general rule of thumb is that if you can't view it in the source, the spider can't read it but I didn't know if maybe I was missing something or there was another approach that we could take to be able to use these dynamic possibilities.

Thanks all.

DaK

[kgun]

AJAX and SEO or XML and SEO. There is a link with the following heading

"AJAX and SEO"

if you scroll down the second link in my signature.

[dak888]

Excellent link! Looks like I've got some research to do.

DaK

[simonwheatley]

Hi Dak,

If you can get hold of Firefox, you can see what Search Engines see fairly easily by following this process:
* Install the Web Developer extension
* Turn off all CSS
* Turn off all JS
* Turn off all images

If it's loaded by JS, spiders can't read it, if it's words in an image, spiders can't read it, etc etc...

What you might be looking for is something that gracefully degrades... so the site is still usable when you turn off JS/CSS/etc. Maybe keep all the products as separate pages, with all the info in the source, but once a user has landed on one page and chooses another one, allow them to do that with JS/AJAX? (Does that make sense?)

Good luck.

Simon

[kgun]

Note, code like JS, PHP and ASP is not indexed by SeBOTS.

Extensible Markup Language (XML) documents should be easier to index than HTML documents, since a XML document need to be well-formed, that is no open and badly nested tags are allowed. If you use DreamWeaver, there is an inbuild XML validation tool (File menu + Check Page + Validate as XML).

In XML documents, you can make your own tags, and for that reason you should have a good title tag at the start. The markup can only be indexed if it is a real unique page.

There are at least three tools you should be aware of:

1. Search Engine Spider Simulator
   
   "A lot of Content and Links displayed on a webpage may not actually be visible to the Search Engines, eg. Flash based content, content generated through javascript, content displayed as images etc.
   
   This tool Simulates a Search Engine by displaying the contents of a webpage exactly how a Search Engine would see it.
   
   It also displays the hyperlinks that will be followed (crawled) by a Search Engine when it visits the particular webpage".
   
2. Site Crawler, Indexer and Search engine
   
   Scroll down and click on the link "Search Engines (External)" and then on "Site Crawler, Indexer and Search engine."
   
   "Crawls and indexes a web site placing the indexed words into a database. A user control is then used to search the database displaying the results as a list of urls with title and extract in a similar format to popular search engines".
   
3. rpsoft 2000 site crawler, webmaster software
   
   "Basic Applications:
   Rpsoft 2000 Site Crawler is a helpful and powerful utility for webmasters! For sites that are simply constructed for their links, this rpsoft 2000 site crawler can help with a number of common webmaster challenges, such as: checking for reciprocal links, general site searches, email address finding, meta tag help, external link count and keyword density".

[simonwheatley]

Hi Kgun,

I agree that a well formed XML webpage (XHTML) may [1] be easier for a robot to understand... but I think the point here is that the page content is not there on initial load, but is pulled in via JS as an XML file from the server. This means that the robot, which doesn't have JS, doesn't see the XML file, it sees nothing other than the navigation and other mundane page furniture.

The Search Engine Spider Simulator you link to above looks like the tool to check this with.

Cheers,

Simon

[1] Having said that, the robots have to deal with such trash markup in most of the web that I doubt it makes a great deal of difference to them.

[kgun]

I agree that a well formed XML webpage (XHTML) may [1] be easier for a robot to understand... but I think the point here is that the page content is not there on initial load, but is pulled in via JS as an XML file from the server. This means that the robot, which doesn't have JS, doesn't see the XML file, it sees nothing other than the navigation and other mundane page furniture.

My bolding.

That is implicit in my answer above, if you study it in more detail. I bolded code.

Note there is no standard way to present a XML document in browsers.

Exmple:
Are you prepared for web 2.0 programming?


AJAX is real easy but because it relies on JS, there are a lot of browser related problems. My first post in this thread may also be of interest.

The AJAX library, zXML 1.0.2 that is mentioned there may save you hours of coding since it has includes for cross browser compatibility that is important when you use JS.

An AJAX web application shall degrade gracefully. That is one of the purposes of that library.

[dak888]

What are your feelings on hidding and showing div's with css and js? This would bypass the use of the xml file and I would probably be able to pull off a similar effect. Worst case scenerio is the customer turned off js (which was less than 1/2 percent in the month of April) and it displays the whole page...

DaK

[simonwheatley]

What are your feelings on hidding and showing div's with css and js?

I think this is worse from an SEO perspective... the page is targetted (in content terms) at your whole product range, rather than individual pages targetted at each of your products. However it would be a quicker experience for your customers. If you do it, make sure the page(s) are comprehendable without JS or CSS (maybe not pretty, but comprehendable).

[moredial]

So my question, are xml files readable when used in this manner? I know the general rule of thumb is that if you can't view it in the source, the spider can't read it but I didn't know if maybe I was missing something or there was another approach that we could take to be able to use these dynamic possibilities.

What a lot of circles everyone is going round in. Go back to basics:

Whenever you use javascript, make use of the NOSCRIPT tag to display what you need to display to those with js disabled/unavailable and this also gives you the chance to give to search engines whatever they need to see.

As with all web design, keep it simple and follow the rules and you will have a good page every time.