One drawback of search in previous versions of SharePoint was that almost everything had to be managed from the central administration page, which meant that search was managed at the farm level.

SharePoint 2013 changed that by adding most of the search settings from the farm level to site collections and sites (SPWebs). As SharePoint 2013 is offered as a cloud service (through Office 365), and cloud users have no access to settings in the farm-administration level, this was a welcome change that both cloud and on-premise site administrations can take advantage of.

Let's have a look at what settings are available for us to administrate; these are shown in the following screenshot:

We will discuss these settings in detail in Chapter 2, Using the Out of the Box Search Components, but for now just keep in mind that a site administrator can configure the search experience on his/her site in ways that were reserved exclusively to farm administrators in previous versions.

In addition, Microsoft introduces a new crawling mode, continuous crawl. Continuous crawl helps to keep the search index as fresh as possible by crawling SharePoint sites (and only SharePoint sites) every 15 minutes, by default; we can change this value using PowerShell, as shown in the following snippet:

$ssa = Get-SPEnterpriseSearchServiceApplication
$ssa.SetProperty("ContinuousCrawlInterval",<minutes>)

The value we use for <minutes> is the number of minutes between crawling.

When running, the crawler gets changes from the crawled SharePoint sites and pushes them to the content processing component, which will process the new content on the fly.

By enabling the continuous crawler, items appear in the search almost immediately after being crawled.

If there is one change in SharePoint 2013 Search that just pops to the eyes, it is the new and fresh user interface (UI). If you worked with SharePoint 2010 search, you'll remember the following screenshot, showing a search-results page:

By looking at the preceding screenshot, we can see that it sports a pretty simple UI. We have textual refinements on the left side; predefined search scopes for websites and people (All Sites and People) on top, and a main, simply styled results area without grouping or categorization of results.

To customize the way the results are shown, we had to use XSL/XSLT, which is quite a messy and unattractive way to design.

Fast forward to the present day. The following screenshot displays how the results page looks like in SharePoint 2013:

Now that's quite different, isn't it? The UI is modern thanks to the use of HTML and JavaScript templates. Instead of messing around with XSLT, we now have display templates to design our results using languages we already know and love: HTML, JavaScript, and CSS.

Take a look at the refinement panel on the left. While we still have textual refiners, we also have graphical ones, such as a scroller for dates.

We have new out-of-the-box search scopes—Videos and Reports—and results are grouped by their types; for example, the first two results are discussion items from a discussion board.

The biggest and most notable change, however, is the new hover panel. Whenever we hover over a search result, we are presented with a floating panel containing additional information about the hovered item. As SharePoint 2013 seamlessly integrates with Office web apps, any Office document we hover over will show a preview of its content in the new hover panel. The most important thing about the hover panel, however, is that we, as developers, have complete control over the content of this hover panel. Just like search results, the hover panel is also controlled by HTML, CSS, and JavaScript.

We will discuss all of these new and exciting customizations features in detail in Chapter 4, Customizing the Look.

As mentioned earlier, SharePoint 2013 Search took the best features of SharePoint Search and FAST and improved them. As such, SharePoint 2013 uses new and improved ranking models to determine which items are to be displayed and what would be their rank (the order in which they are displayed).

The key to successfully determine the relevancy of search results is to satisfy the intent of the person who issues the query. Let's explain this statement with an example; say I'm performing a search for Apple. Now, did I search for apple the fruit or Apple the technology company?

SharePoint 2013 Search continuously tracks and analyzes search usage to determine how content is connected, how often an item appears in search results, and which search results people click in order to continuously improve the relevance of items to the search query. So, if I clicked on a lot of fruit-related results, the search engine will assume I was looking for apple the fruit, and not the technology company.

We will discuss these new features in Chapter 2, Using the Out of the Box Search Components, and Chapter 3, Using the New CSOM and RESTful APIs.

With this new release of SharePoint, Microsoft made changes to the search-development model. The old SOAP web service (ASMX) has been deprecated alongside the SQL query syntax that we could use to query against SharePoint data.

But, just like the the old saying goes, "out with the old and in with the new", we get some new features to play with to replace the ones that are gone.

We will discuss all of these topics in detail in Chapter 3, Using the New CSOM and RESTful APIs, and Chapter 5, Extending Beyond SharePoint.

Now that we have a general idea about what's new in SharePoint 2013 Search, let's go ahead and discuss the architecture that makes all of this happen.