Session Intro: Practical tips, tricks, and workarounds for search-friendly architecture.
- Microsoft Stack
Including IIS, ASP.Net, Silverlight, Microsoft SQL Server
- LAMP Stack
Including Apache, PHP, Ruby, Flash/Flex, mySQL
- CMS Considerations (such as .NET Blog Engine, AxCMS, Wordpress, Movable Type, Drupal, Joomla)
Jeff is up first. He talks about SEO tweaks for Apache:
- URL Canonicalization - need to decide on 'non-www' or 'www' and then
redirect to main. Jeff shows some Apache code to accomplish this which you
would add to the .htaccess file.
- 301 Redirects - he shows some code to accomplish 301 redirects in Apache,
again in the .htaccess file.
- SEO Friendly URLs - clear concise URLS. Easiest method to employ is CakePHP format.
For more info on this, check Apache
documentation.
Next, Jeff talks about PHP which has been around since 1997. It is extremely
popular and easy to code. One SEO tweak Jeff provides is a method to remove the
PHPSessionID where a long session ID may be inserted in the actual URL or pages.
Another tweak has to do with 404 error pages. Make sure it returns a 404 header
instead of 200.
He moves on to MySQL which has been around since 1995. Only SEO tweak you can
really accomplish is related to performance. He says you can do lots to optimize
MySQL performance for quicker crawling by search engines so in the end they can
crawl more pages.
Finally he provides a quick overview on web frameworks which are typically very
good for SEO. CakePHP, Symfony, Zend, and Ruby on Rails are good web frameworks
to use.
Next up is Colin who will talk about ASP.net. Issues with ASP.net are similar to
what Jeff talked about - URL canonicalization, custom error pages, meta data
management, performance and crawlability.
Colin shows actual screenshots of how to deal with all these issues in IIS 6.
While I cannot reproduce those here, you can very likely find resources and
screenshots online. One thing to keep in mind include making sure error pages
are set up correctly or they could return a 200 header response to search
engines, meaning they are okay when in reality they are not there.
Next up is Duane who will talk about Adobe Systems. He shows us how search
engines are doing a better job diving into Flash. However, you can't really
force search engines to adopt technologies sites use, so what is one to do? He
talked about using MVCs to deliver multiple data to various sources (i.e. users,
bots, etc.). Adobe is working to help search engines index their technologies.
Next up is Nikhil. He is going to focus on indexability. This is crucial when
building rich web applications where you have single pages that fetch additional
data using XMLHTTP based on user actions. You therefore have to add indexable
content back into the page. He is not endorsing serving up different content
than what a user would see but rather content a search engine can understand.
Using site maps and strategic navigation will help in the discoverability of
content.
He shows us how to use the div tag and the "display=none" to place alternate
content in the html in a media rich application. The only way a user will see it
is if they don't have the rich media plug-in. For example, if the application is
Flash and a user does not have a Flash plug-in, they will see static content. If
they do have Flash plug-in, they will never see static content but a search
engine will.
Q&A:
Here is a recap of "some" (not all) of the questions that were asked and
answered.
- With alternate content in html, should you also build in a link structure?
Duane advises that it is best to have one set of data and to avoid having duplicate sets of data.
- Does offering alternate views as discussed in the session label you as a
cloaker? Will you get dinged?
Duane won't answer for engines but says in his experience, he has not had any problems with it. Vanessa pointed out that if you are serving content specifically to bots such as Googlebot, you could get dinged. Better to serve content for users like in Flash example above. In other words, okay to serve search friendly content but I think IP delivery would raise a red flag.
Note: In this session, a lot of screenshots of
actual code were shown which is almost impossible to capture in a live blogging
environment. You might be able to get those from the speakers themselves.
PowerPoint presentations are also available to SMX Advanced conference
attendees.
Session coverage by David Wallace - CEO and Founder SearchRank.