The problem with session IDs in URLs is that you present Infinite Duplicate Content to search engines.
Google, for example, has a 'crawl budget' assigned to a site, and you do not want to waste that by having it load what is essentially the exact same page over and over again just because it has hundreds of different URLs that it responds to, each with just a change in the session ID number. Nor do you want to split your Pagerank over multiple alternative URLs for each page.
Each page needs to have just one 'canonical' URL that it responds to, with all other alternatives redirecting to the canonical form. Sessions should be implemented in cookies these days, not as a parameter in a URL.
These articles contain a lot more information:
http://www.webmasterworld.com/google/3691518.htm
http://www.webmasterworld.com/google/3718246.htm
Other people, looking at their backlink listings in Google WebmasterTools, where their own website has an incoming link from some other site that runs DokuWiki, will see the URL of the DokuWiki page that their link is supposedly on, appear to change to a new URL in the listings every few days as Google finds new URLs for the same old content and drops the old URLs from their list. After some time their process can seemingly give up and fail to list the other site as even containing the outgoing link. It's likely one of their Signals of Quality measurements is kicking in - the same content being found on a different URL every time they visit is a bad sign as far as searchengine spiders are concerned.
Several years ago, several search engines actively avoided spidering URLs with session IDs in them. One signal of a session ID in a URL came from the name of the parameter: especially 'session', 'id', 'sessionid' and so on.