In my other post I have explained the new search architecture of SharePoint 2013. In this post I want to describe some new features of Search Administration. Creating the Search Service Application is the same as with SharePoint 2010. After creating the service application, the search administration looks somewhat like this:
Note: this article is written based on SharePoint 2013 Preview and functionality can still change when the RTM is released.
Out of the box the following connectors are available:
The list is the same as with SharePoint 2010, but there are some enhancements. E.g. when crawling HTTP sites a new type of authentication is available: anonymous.
One obvious thing when creating a new content source is that the option of Continuous Crawls is available.
This only applies for SharePoint content sources. When you crawl continuously every 15 minutes the crawler gets changes from SharePoint sites and pushes them to the Content Processing Component. The interval of 15 minutes is the default value, but you can change is with the PowerShell cmdlet Set-SPEnterpriseSearchCrawlContentSource. Because of changes how the index is created and stored items can appear within seconds of going through the Content Processing Component.
This is new in SharePoint 2013, but actually it is the combination of Federated Search and Scopes. You create and use a result source to specify a location to get search results from, and to specify a protocol for getting those results. In comparison, in SharePoint Server 2010, you specified a location and a protocol by creating a federated location.
With a result source, you can also restrict queries to a subset of content. You can use Query Transformations for returning certain content. In SharePoint 2010 this was called a search scope. In the next example I have opened the pre-defined result source Local Video Results and you can see the query transformation.
When clicking on the Launch Query Builder, you can create or edit the search query. You can even test it!
Query Rules and Search Schema are other features that are new or improved. Search Schema is the improved Managed Properties management. You can create managed properties to store entities and use these properties later with search refiners. These entities can be words or phrases, such as product names. To specify which entities to look for in the content, you can create and deploy your own dictionaries.
Query Rules allow you to have search requests from a user triggering multiple queries and multiple result sets. All matching Query Rules can generate results and the query component will organize the results for rendering to the user. With the Technical Preview Query Rules didn’t work properly, so I hope to explore it with the public beta.