Documentum and Alfresco – Open Source Interface – Preview of HPI 2.1

We are working with a couple of large clients on the latest update for our High Performance Interface (HPI). The 2.1 release will have an all-new interface as well as numerous functions and features not available in the previous release. This post will present a preview of the HPI 2.1 release.

HPI 2.1 – Configuration, JavaScript and REST

HPI was originally created as a configurable web interface for our clients looking for Documentum Webtop alternatives. Over five years ago, the initial interface incorporated some of our best practices from an interface perspective with configuration via XML files as well as support for both Documentum and Alfresco. In 2013, we determined that we wanted to not only update the infrastructure of HPI but move away from XML files for configuration. HPI 2.1 now contains an administration module that provides the ability to configure the entire application. Additionally, configuration updates take affect right away and do not require a server restart. Some of the major items that are configurable in the HPI 2.1 admin include:

• Object Type Configuration – what object types are exposed, type and attribute labels, hidden attributes, etc.
• Search Configuration – what objects are searchable, which attributes to display in search criteria, search results view options (grid, list, and/or table).
• Action Configuration – what actions are available when users click on a document or folder. Some example actions include: checkin/checkout, annotate, folder notes, notify, as well as Active Wizard specific actions among others.

Another major component to the upgrade was moving away from a Java-based front end. HPI 2.1 is a JavaScript application that communicates with our OpenContent services layer entirely using REST. For our technical readers, HPI 2.1 utilizes Backbone.JS, Twitter Bootstrap, and jQuery. Using this framework provides for a more modern user experience as well as more flexible deployment options. For example, HPI can now be deployed to a simple web server such as Apache communicating with OpenContent on the application server tier.

Performance

Our non-technical readers may have glazed over or skipped the last paragraph, but there are some important practical ramifications for business users. The modern framework allows HPI 2.1 to perform significantly better than our previous versions. Users no longer need to wait for the server to render an entire page before being presented with the view. In HPI 2.1, we can display the page quickly and then have data fill in the background.

Along with physical performance, HPI 2.1 still contains many of the best practices that improve performance by minimizing user error. For example, many ‘kitchen sink’ ECM applications (ex: Webtop, D2) contain complicated menu and sub-menu layouts. In HPI, users get actions for documents upon view, and those actions are tailored to the user’s permissions as well as the business process.  A ScreenCam is still available in our learning zone comparing WebTop, D2, xCP and HPI in regards to Search and Contributor interface differences.

Drag and Drop – Adding Documents and Emails

Long-time Documentum users have always struggled with importing documents into the system. With HPI, we strove to keep the interface “browser neutral” and avoid customer complaints about having the correct version of Java installed on their browser. (Documentum users all struggled with the UCF Applet).  With the introduction of IE 10 as well as continued user adoption of Chrome, Firefox and Safari from our clients, HPI 2.1 will feature a multi document import with drag and drop.
The new function will allow users, when adding a document, to drag multiple documents to a “drop zone” to quickly add documents to a folder.

One unique aspect to HPI includes documents inheriting attributes from the folder automatically. The interface provides the ability to add individual attributes to each document before finishing the import.

Emails can be dragged from Outlook in the same manner. All emails are stored in the repository at .MSG files, complete with attachments. HPI will index the documents as emails with Subject, To, From attributes, render the email to PDF as well as break off the attachments as separate files (but are still contained in the .MSG original file).

Scanning from the Desktop

The import module also has the capability of scanning multiple documents directly from the add document screen, (See Post and Demo). HPI uses Dynamsoft’s Image Capture Suite to drive any TWAIN-compliant scanner as well as to OCR the resulting scan for full text indexing. See the previous post for more information as well as a demo.

Dashlets – Landing Page

HPI 2.1’s dashboard functionality is being updated as well. Dashlets will be customizable in the HPI admin. Currently we have two dashlets – Saved Search and Inbox. The Saved Search dashlet simply executes the same saved query each time. The Active Wizard uses this dashlet to quickly display the Forms that the logged in user owns. In the future, we will be adding dashlets to the core product, but the dashboard also provides us the ability to build custom dashlets for our clients.

Summary

HPI 2.1 continues provides an powerful alternative for Documentum and Alfresco customers looking for open source alternatives to improve user performance.