I've been working on an Arena "search connector" control that talks with our Microsoft Search Server 2008 (SS) in order to provide a seamless search integration with our Arena powered website.  Ultimately, I think we're probably going to be bailing on this new product...

It's been a challenge to get the SS webservice to work with anonymous requests.  This may be in part due to the fact that SS is built on Microsoft's SharePoint platform.  Derek said this (see picture) is what happens when you install the free Search Server RC.  Yikes!  That was a new, clean server...  Then there is the matter of trying to manage the application in three layers (IIS, SharePoint, and Search Server) and Bil Simser's blog entry illustrates only the smallest aspects of the madness -- the rest must be seen to in order to believe. After spending a day making various recommended config changes, Derek and I were still getting "HTTP status 401: Unauthorized".
IIS Madness

The other issue has to do with the inability to control what page content is indexed by SS when the site is crawled.  With a Google Mini you can wrap content in your page (such as your navigation bar's content) with these tags:

<!-- googleoff: index --> <!-- googleon: index -->

...in order to exclude that from being indexed to your page.  Microsoft's Search Server does not have such a feature.  Instead they only have a way to exclude pages based on URL path name patterns. This is fairly critical issue for us due to the way our current site's navbar works.  All the page titles are in our navbar but are hidden from view unless the user clicks on the menu-twisty.  Unfortunately they are still 'visible' to SS.  I guess I could AJAXify our navbar, but the top level items would still be indexed on each page.  Or I could Flash or Silverlight it... or find a better search service.