Rather than simply search the text of the website content, Higher Logic's search functionality gathers information on all content on the platform, including the files and attachments uploaded to the library, discussion posts, blogs and more. Searches recognize and reflect user-specific security permissions - only results available to the user are returned in search results. Users who have not logged in will access content with view permissions set to public.
There are two search experiences on Higher Logic websites, each generated with a unique control:
1) Common:FacetedSearch:Search (faceted search)
With its release in February of 2016, faceted search represents a deeply enriched search experience. Check boxes (facets) reward progressive narrowing/expanding of search results. The Higher Logic product line presents dozens of possible facets. Clicking into a content "Category" will expose facets relevant to a particular genre of content. For example, facets for specific users and content types are exposed with click on the User Content category. This control allows for community-specific search and content types to be excluded or included in search results. Other points of configuration are limited.
2) hlcommon:Search (legacy search)
Once the primary search experience, the legacy search remains available. It continues to be the primary method for configuring search-specific content displays and achieving similar business needs. It supports multiple points of configuration consistent with establishing advanced search interfaces, feed-like displays of content at page load, and other business needs.
The search page - page code: search - is the primary search page for any website. A search input on the main navigation bar opens search results on this page. With its release in February 2016, faceted search is configured to be used on the main search page. The legacy search control, hlcommon:Search, can be configured to be utilized on this page if desired, but this is not recommended.
The page with page code: search requires having a search control as its primary configuration. Any other page can host an additional instance of either search control. Where a page is community specific - either as a community landing page or community microsite - both search controls exclude items from outside the community. This default can be overcome with configuration.
Both search controls have points of configuration to remove/include various types of content and specify a community for items to return. The legacy search control has more points of configuration allowing it to meet multiple needs, including display of content lists at page load. See the Legacy Search: Search Parameters and Legacy Search: Content Lists for configuration instructions. See the Faceted Search page for information on configuring faceted search.
Search-specific URLs and Operators
Search provides a distinct URL for the results of every search. In faceted search, use of facets will update the URL. This makes it possible to send an email with a single link for a search you want to share, or to set a search URL to open on click from a website's navigation. This also enables products available through Search to be better promoted through Google Adwords and similar services. The easiest path to achieve a search-specific URL is to perform a search and then copy the URL from the browser address bar. It is also possible to craft a search-specific URL directly in the URL query string.
Similar to supporting variables in the URL, the search input respects specific syntax as shortcuts to return specific lists of content. Perhaps the most widely-known search operator is the use of quotation marks to specify a specific multi-character, multi-word string. Quotes and other search operators are supported by both legacy and faceted search inputs.
See the Search: Operators and Search-specific URLs page for lists of syntax and variables supported.
Search Results and Indexing
Relevance scoring is determined by an algorithm that weighs value based on number of times a term is used and where a term is used within a piece of content. The relevance scoring algorithm serves all clients and cannot be modified. It is based the following core factors (in descending order or weight):
- Titles – the names of the indexed objects, such as the subject of the discussion post
- Descriptions – the body or description of the object, such as the library entry description or the body of a blog post
- Text in other fields – comments are a common third-level text input that is included in relevance scoring
- Descendants – content on the Higher Logic platform often exists within a larger framework, such as a discussion thread or an attachment to a post within a thread
- Boosts – the following characteristics are also possible sources of increased relevance scores
- Time – the more recent the more the boost
- Quantity – the more the searched term is used the higher the boost
- Direct match – if the searched term is a direct match in content, not a hit based on stemming, there is a slight boost. Stemming is achieving a hit for derivatives of term. “Friendly” being a hit for “friend” through stemming.
Content available to a user is indexed by search. Every save or send for indexed content initiates indexing/reindexing for the content. This includes content associated with all Higher Logic products including the Resource Library. The text of text-readable files uploaded/attached to a Library Entry are indexed. PDFs, txt files, and Microsoft Office documents have their contents indexed by search and searches that include terms or strings included in the bodies of these files result in inclusion in search results. See the complete list of file types supported by the Higher Logic Resource Library product on the Resource Library Overview page.