Can Search Engines Read "incorrect" urls?

[nicole.healthline]

I know that ideally a url should be something of the nature domain.com/topic, but if the url contains additional characters, for example, domain.com/topic?keyword, can the search engines still understand the complete words in the domain? Even though there are additional "incorrect" characters? Or do they stop "reading" once they find odd characters?

Thanks!

[john4math]

A few other things to note for having parameters in URLs:

• In Google Webmaster Tools and Bing Webmaster Tools, you can instruct the search engines to ignore certain parameters, so that they'll treat domain.com/topic?keyword and domain.com/topic as the same page (if ?keyword doesn't change the page content)
• You can also place the rel=canonical element on pages.  So you could set domain.com/topic?keyword to rel canonical to domain.com/topic to pass its pagerank along.

[oznappies]

Search engines will read all your parameters unless you tell google with webmaster tools what parameters to ignore. This can cause an issue with the url like domain.com/topic?keyword&somefield then pages that include the keyword and other parameters will share the link juice. So, if you have 10 options of somefield you will get ~1/10 value per page indexed.

So, it is better for you to use rewrites to include your keyword in the url and then mark parameters to not be indexed in Goggle etc.

[reilly3000]

Search engines can read most characters in a URL string, but specifically & generally refers to a variable in a script which doesn't typically have much valuable information regarding what a page may be about. Sometimes those variables may be the topic of a category of a shopping cart, so I have to imagine that information could be taken into account, but for long urls like the following it is hard to believe everything is factored into the URL's relevance to the keyword: http://www.google.com/search?q=long+url+string&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a

Search engines index the whole URL and if there is keyword rich content that can definitely help, both from having the keyword bolded in the snippet (CTR WIN!) and a possible bump in the page's relevance to the keyword.

[Nemek]

In general search engines are able to identify keywords in the URL even if they are i.e. a parameter that follows a "?" or other non-alphanumeric character. They might not treat it as an equally strong signal as when the keyword is a part of the file name, subdomain or domain name though. Hope that answers your question.