Use cases

These are a few use cases that examine how the Semaphore product is installed/configured in particular installation types.

Content Management System (CMS) integration

Content Management Systems (CMSs) and Enterprise Content Management Systems (ECMs) are now often seen as critical to the business. As content is added to these systems it is important it is correctly managed and maintained. Semaphore can aid in this process by ensuring this content is consistently tagged with terms from your business vocabulary so that it can be easily managed.

Functional requirements

A typical Semaphore CMS integration provides two main pieces of functionality:

1. Tagging of content as it is added to the system with tags from the model - This can be done in an automatic, assisted or manual way based upon how much is required of your users. “Automatic” refers to tags assigned without any user intervention (which may be required for all existing content on initial configuration in any case), “assisted” is where Semaphore assigns tags but the user can adjust them as they feel fit and “manual” requires the user to manually assign any tags from the model they deem appropriate for the content added. Commonly the “tags” assigned to content are put into columns or fields associated with the document in the CMS. Note that this information can be used to drive workflow mechanisms as content is added or updated in the CMS – For example, filing documents in the most relevant section of the CMS based upon the Semaphore tags assigned.
2. Ability to search for documents assigned specific model tags - Once tags are assigned you will want to be able to use these tags to retrieve the relevant content – For example, “Those documents tagged with ‘Budget 2014’ and ‘Information Technology’”.

The specific implementation to provide this functionality potentially involves:

1. Model synchronization - For those CMS products supporting the definition and management of controlled vocabularies the Semaphore model can be synchronized into these stores so that the terms the model contains can be used in the CMS.
2. Modifications to the user interface
   • Allow model tags to be selected when content is added/modified
   • Allow “batch” classification of existing content
3. Modifications to the search interface
   • Display model tags assigned to content in search results
   • Allow filtering of content based upon model tags - Using, for example, our “SAYT” functionality that displays model terms as a user types in a search box, or providing “facet filtering” where terms present in the results are selectable to allow further refinement of the search when selected.
   • Display of any additional Semaphore information required - For example, you may wish to display additional information stored in the Semaphore model when the user is searching on a term.

Architectural requirements

To support this type of implementation Progress typically recommends that the architecture is physically split as follows:

• Content Management System Server(s)
• Semaphore “Semantic Enhancement Server” Server(s) - Where the “Semantic Enhancement Server” software is installed.
• Semaphore “Classification Server” Server(s) - Where the “Classification Server” software is installed.

The Semaphore software is split in this way so that the different functional elements do not interfere with each other - For example, someone classifying a document (using Classification Server) should not impact the performance of retrieving a list of terms from the model (using Semantic Enhancement Server).

Semaphore CMS integrations

It should be noted that Progress itself provides following shrink-wrapped CMS integrations:

• Semaphore for OpenText (Content Server)
• Semaphore for SharePoint (Online, 2010, 2013, or 2016)

However the following CMS integrations have also been implemented:

• SiteCore
• RedDot
• EPI Server

Additionally a “CMIS toolkit” is available that facilitates integration in other versions of OpenText, Documentum, Alfresco and others supporting the CMIS specification (see Content Management Interoperability Services for details regarding CMIS and supported ECMs).

If you want further details regarding any of these integrations contact Progress.

For more details on integrating Semaphore into a CMS see the Semaphore Generic Integration Guide.

Search integration

In an age where content is found in a myriad of different locations adding a search engine to allow search of content across these information silos is often critical to getting the right information at the right time. Semaphore can assist in this process by ensuring that the content, regardless of it’s original source and format, is tagged with the correct business vocabulary so that it can be easily found in the search process.

Functional requirements

A typical Semaphore search-only integration provides two main pieces of functionality:

1. Tagging of content as it is added to the search engine with tags from the model - As content is crawled for inclusion in the search results you will want to have it automatically classified by Semaphore or, if the content has previously been tagged by Semaphore in the system being crawled, extract those existing tags to use those. Commonly the “tags” assigned to content are put into columns or fields associated with the content in the search engine.
2. Ability to search for documents assigned specific model tags - Once tags are assigned you will want to be able to use these tags to retrieve the relevant content – For example, “Those documents tagged with ‘Budget 2014’ and ‘Information Technology’”

The specific implementation to provide this functionality potentially involves:

1. Modification to the crawl mechanism
   • Classify content using Semaphore as it is “crawled” by the search engine
   • Convert existing Semaphore tags for use in the search engine
2. Modifications to the search interface
   • Display model tags assigned to content in search results
   • Allow filtering of content based upon model tags - Using, for example, our “SAYT” functionality that displays model terms as a user types in a search box, or providing “facet filtering” where terms present in the results are selectable to allow further refinement of the search when selected.
   • Display of any additional Semaphore information required - For example, you may wish to display additional information stored in the Semaphore model when the user is searching on a term.

Architectural requirements

To support this type of implementation Progress typically recommends that the architecture is physically split as follows:

• Search Engine Server(s)
• Search Front End Server(s) - If technology/software separate from that which is provided by the search software is used to deliver the interface a user uses to perform the search then it is recommended that a separate environment be used to host this so that the interface display/rendering does not effect the other components of the solution.
• Semaphore “Semantic Enhancement Server” Server(s) - Where the “Semantic Enhancement Server” software is installed.
• Semaphore “Classification Server” Server(s) - Where the “Classification Server” software is installed.

The Semaphore software is split in this way so that the different functional elements do not interfere with each other - For example, content being crawled and classified (using Classification Server) should not impact the performance of a user using the interface to retrieve a list of terms from the model (using Semantic Enhancement Server).

Semaphore search integrations

It should be noted that Progress itself has implemented the following search integrations:

• Semaphore for MarkLogic
• Semaphore for Solr

If you want further details regarding any of these integrations contact Progress.