The Gilbane Report: Volume 9, Number 4

Open Source Content Management: A Parallel Universe?

Download a PDF version of this article

Read the news for this issue.

Open Source Content Management: A Parallel Universe?

It is typical that "enterprise" software applications originate as highly customized solutions designed and built in-house, often with assistance from integrators or consultants. As word gets out about the benefits similar applications are built, and eventually software vendors build commercial products. As vendor competition increases, more functionality is built into products, prices come down, and some features creep into infrastructure software, further reducing prices and forcing high-end vendors to expand their solutions to maintain business models that require large sales. Enterprise content management will progress through this same cycle. To ensure successful content management strategies it is critical you consider how this cycle will affect your existing and planned implementations.

Open Source software adds another dimension to this complex cycle. Open Source software consists of much more than Linux and Apache. There are all kinds of Open Source software tools available, including Open Source content management systems and databases and application servers that can be used to build content management systems. This month we welcome back Sebastian Holst, who provides advice on how to think about Open Source software in the context of the evolution of enterprise content management software.

Open Source Content Management: A Parallel Universe?

This report will examine the likely impact of Open Source (OS) software on the content management market and introduce some of the early OS-Content Management Systems (OS-CMS).

OS-CMSs are providing viable alternatives to a small but growing number of organizations and their influence is on the rise. Understanding if and when an OS-CMS is appropriate has the potential to save an organization a significant amount of money. Having said that, the rapidly expanding scope of "content management" and the caveats that come with open source software may still temper your enthusiasm.

The open source software approach to development has won legions of supporters, generated millions of lines of code and been the subject of numerous white papers and analyst reports. Content management systems are firmly established as essential building blocks of e-business and other information intensive operations. Yet, the final chapter of these two stories is far from having been written. Interestingly, the interaction between these two phenomena sheds light on the essential characteristics of both.

The Essential Characteristics of Open Source Software

Open Source refers to software distributed under a license that permits free distribution and requires open availability of the source code. Examples of Open Source licenses include: the Gnu Public License (GPL), the BSD license, the MIT X License, the Artistic License, and the IBM Public License. Open Source licenses should meet the Open Source Definition, which is described at http://opensource.org/osd.html. The core requirements include:

Free Redistribution

Good News: This requirement is critical to maintaining a focus on long-term quality by removing the incentive to derive works for immediate sale. But: Developers with high value and proprietary intellectual property are not likely to incorporate their work in this forum.

Source Code License

Good News: Programs cannot easily evolve unless the source code can be enhanced. This ensures that all aspects of an open source product can be enhanced on an as needed basis. But: You have licensed source code.

Derived Works

Good News: You are free to modify, experiment with and enhance your licensed product. But: You might be required to modify, experiment with and enhance your licensed product.

Integrity of The Author's Source Code

Good News: Authors and maintainers can develop solid reputations and set clear expectations as to the scope of their responsibilities. But: Who is qualified to evaluate the quality and reputations of the authors and maintainers?

No Discrimination Against Persons or Groups

Good News: This is intended to be a meritocracy.

No Discrimination Against Fields of Endeavor

Good News: This is designed to ensure that no commercial use can be excluded. The intent is to include, not quarantine commercial developers. There are a number of other clauses that protect against other types of license traps as well.

License Must Not Contaminate Other Software

Good news: Distributors of open-source software have the right to make their own choices about their own software.

Open Source licenses guarantee everyone has the right to read, redistribute, modify, and use the software freely. Under many Open Source licenses modifications of existing software must be distributed under the same license as the original software and all source code to any changes or improvements must be made available to the public as well.

The terms of the Open Source license are crafted to encourage development by loosely affiliated developers rather than a single software vendor. Typically, "project maintainers" who address technical or end-user requirements guide Open Source software development.

Commercial opportunities include the packaging, distribution and support of Open Source software. For example, Linux vendors such as Red Hat, Caldera, SuSE, and others package existing Open Source software and provide a more convenient way for their customers to access and install the software.

Examples of widely used Open Source software

Many of the leading Internet and networking software programs are Open Source software:

• Linux, the operating system
• Apache, a Web server
• Sendmail, a Internet mail server
• BIND, a Domain Name Server (DNS) platform
• Samba, a cross-platform networking server
• Perl, a scripting language
• Mozilla, the browser from Netscape/AOL
• MySQL, a database

The important trait that all of these OS programs share is that they are VERY general. There are no wildly successful Open Source molecular modeling software packages or financial derivative trading packages. For the Open Source development method to thrive, large numbers of relatively dedicated developers need to come together over relatively long periods of time. This requirement self selects out specialized or highly complex technologies where the development communities are relatively small. This will prove to be a very important consideration when investigating the growing interest in Open Source content management.

The Essential Characteristics of Content Management

It is important to distinguish between content management systems and content management solutions. A solution combines a system (traditionally this has been synonymous with a content management product) with customizations, the underlying infrastructure and integration with other content-centric systems such as authoring tools, application and syndication servers, and other delivery facilities.

Figure 1 illustrates the relationship between a content management system and a complete content management solution. The software that comprises a content management solution can be placed in one of four categories:

• The underlying infrastructure is broadly adopted, low cost (on a per user and per transaction basis) and forms an almost invisible platform upon which the rest of the solution stack is based.
• The content management system: Commercial software that adds significant value to the underlying platform justifying a relatively high cost (on a per user and per transaction basis).
• Integration services: Custom development, integration and configuration of premium products to bridge the gap between an individual site's requirements and the generic capabilities of the underlying commercial software.
• Creation and distribution facilities: These are other segments of the content lifecycle that include the creation, distribution and syndication of content outside of the content management solution stack.

Figure 1 also illustrates the natural product feature creep that occurs over time as features migrate from custom extension to premium product and finally to an underlying infrastructure service. For example, the degree to which a DBMS cannibalizes today's CMS functionality and the pace that CMS vendors incorporate the work of professional services into their CMS products can be modeled within this framework.

Product managers inside commercial CMS vendors know that in order to justify a license premium, their offerings must consistently add significant value to the underlying platform (DBMS, application servers etc.). CMS vendors take great care to incorporate the right features at the right pace to assure that significant value-add is preserved even as the underlying infrastructure suppliers cannibalize the features of many of yesterday's CMS offerings.

This is the underlying dynamic that has driven the expansion of the content management systems and it is not unique. Email, Internet chat software, and file management software all debuted as custom software, have been licensed as relatively expensive applications and are now fairly standard components of today's operating systems. The vendors that initially licensed these applications either moved on to other applications or, like the applications themselves, were devoured (typically this can be read as acquired).

As we will see, applications developed under an Open Source Software license live outside of this traditionally closed ecosystem. At the very least, OSS CMSs add a significant new dynamic and are very likely to dramatically impact the behavior of the traditional CMS vendor community.

Feature Creep of Content Management Infrastructures

There are numerous underlying infrastructure components that support a content management system, but the one that most actively cannibalizes the CMS is the DBMS.

Figure 2 charts the evolution of the DBMS as it has expanded its information modeling functionality and its information management functionality. This is a reasonable comparison as the management functions such as security, transaction management, schema evolution, etc. get increasingly difficult as the information model is made more robust. Therefore, DBMS architects often make conscious decisions to simplify one facet of their system in order to extend or optimize the other. It is a useful comparison because it also shows that the market overwhelmingly prefers trusted management functionality over enhanced modeling capabilities. The early relational products were ridiculed by the then dominant network (CODASYL) DBMS vendors (you couldn't even model an organization chart let alone a parts assembly with the relational model!). The buying public decided that elegant modeling could reduce coding efforts and improve design, but management facilities were essential to running a business. Many years later, the early Object-Oriented (OO) DBMS vendors had visions of conquering the then multibillion-dollar DBMS market due their revolutionary information model. Again, the market spoke unambiguously in favor of predictable and scalable platforms.

Of course, consumers don't like to choose, they want it all and the RDBMS vendors have moved to incrementally enhance their modeling capabilities by (you guessed it) cannibalizing many of the most generalizable features of content management. First came enhanced datatypes including data blades and data cartridges from Informix and Oracle respectively. Next came numerous interfaces, filters and data processors that provided persistent storage and various levels of validation of OO-based and XML content. Most recently, the tight bundling of application servers and file system and hierarchical storage management support have been bundled into "the DBMS." These developments have seriously cut into the functionality of the basic CMS system. As such, one can see CMS vendors responding by accelerating enhancements to minimize the importance of these commoditized (low value) capabilities.

Feature Creep of Content Management Systems

Feature creep of premium applications expresses itself as specialization or expansion. Specialization focuses on increasingly complex applications that have traditionally required significant integration services, examples include document workflow, portal management, and customer relationship management. Expansion focuses on moving "up" or "down" the information supply chain incorporating creation, publishing and syndication capabilities as part of a more comprehensive platform.

Figure 3 maps a variety of specialized solutions that are derivative or at least highly dependent upon content management functionality. As leading DBMS systems have incorporated increasing amounts of basic CMS functionality, the premium application vendors have either specialized their content management functionality, e.g. knowledge management or expanded their functionality, e.g. CRM.

The specifics on how these content management variations differ from one another is beyond the scope of this paper. However, what this high level graphic should convey is that content management is a dynamic category where innovations, value propositions and user requirements are evolving in multiple directions simultaneously and at a rapid pace.
In order to succeed, Open Source CMS projects must find a sustainable niche within this very competitive software market segment.

Open Source CMS Projects

Today's Open Source CMS projects are growing in the same "parallel universe" as their other OS project peers. They do not need to buy into the rat race of feverish functional growth to drive market expansion. Rather, they tend to focus on a growing community of users who have relatively stable and well-defined content management requirements. The following are just three good examples of OS-CMSs that are achieving growing acceptance in the marketplace today.

ACS Content Management System

Excerpted from www.arsdigita.com/products/cms
ACS Content Management System (CMS) is a component of the ArsDigita Community System that provides functionality for managing Web site content. ACS CMS allows establishment of clearly defined roles and documented workflow for all forms of web content.

ACS CMS provides tools that facilitate the design, authoring, review, approval, aggregation and deployment of web content, the maintenance, monitoring, revising, and eventual archiving of published content, and reporting and analysis. ACS CMS offers a set of web site content services, including:

• Type definition
• Item creation and editing
• Common storage
• Associations
• Versioning
• Categorization
• Searching
• Reporting

The ArsDigita web site lists 60 web sites that have deployed using the open source without paying for support or service and 20 sites that have chosen to pay fees for service and/or support.

Cofax

Excerpted from www.cofax.org
Cofax is a Web-based text and multimedia publication system. It was designed to simplify the presentation of newspapers on the Web and to expedite real-time Web publication. At Knight Ridder, it is serving newspaper content for several of the papers, including the largest papers, the Philadelphia Inquirer and Philadelphia Daily News and smaller papers like the Wilkes-Barre Times Leader.

Cofax CMS is a web based Content Management System used to edit and manage the Online Editions of 16+ Newspapers. Xmultra is a flexible data ingest application that integrates with legacy news publication systems & classifieds systems at the newspapers.

Cofax is a framework of software modules and tools to manage and deliver news content from different sources. It was developed using Java, database, and XML technologies.

The Cofax web site identifies 13 newspaper web sites and 2 other web sites that are powered by Cofax.

Midgard

Excerpted from www.midgard-project.com
Midgard is an Open Source (OS) content management system based on Apache and MySQL. The system uses PHP as its scripting language. While Midgard will always implement an OS development to publishing solution, future releases will include APIs for implementing commercial applications.

Midgard addresses the publishing challenge by providing tools that are based on content organization which creates a content tree that's similar to a Yahoo style directory. Layout is separated from content by a style system that enables HTML developers to focus on site design while writers focus on articles. The SiteGroup (SG) tool separates the single MySQL database into multiple virtual databases and manages access control. When many developers work on one site, each one is assigned membership to a SiteGroup.

Code developers can directly embed Midgard objects; content, users and styles into PHP applications with the Application Programming Interface (API) which makes calls to the Midgard library. Redundant coding is controlled with CodeSnippets which are shareable objects.

The Midgard site identifies 35 sites that are powered by Midgard.

OS-CMS and Content Management Solutions

The previous OS-CMS examples are representative of the kinds of CMS systems that are getting some traction in the marketplace today. Understanding where they fit into the commercial content management landscape is an important step in predicting where they are likely to succeed and the kinds of organizations that might be able to consider an OS-CMS today or in the near term future.

Figure 4 illustrates the impact of the Open Source model on enterprise solution ecosystem. The major changes are:

• The feature lists highlighted are not as specialized or advanced as those currently being touted by commercial software vendors.
• The underlying infrastructure is not cannibalizing the CMS layer nor is the CMS layer cannibalizing the integration layer, rather new OS projects build upon - but are distinct from - earlier OS projects. This is a difference that can be directly attributed to the Open Source methodology. The lack of ownership removes the imperative to stretch the market category beyond its natural boundaries.
• Integration/customization is no longer distinct from the application/system layer - again, a direct result of OS methodology. Customizations and derived works are also open and available to the broader community.
  Implications of the new Open Source "ecosystem"
  The following changes in the nature of content management solutions are likely to flow from the Open Source development approach.
• The modular components that result from cooperative OS projects as opposed to competitive product development may ultimately result in a solutions architecture that is functionally organized rather than being defined by ownership and licensing terms of particular product suites.
• Clearer evidence of applicability and fit to specific applications will be generally available due to the distribution of extensions and customizations associated with each deployment. This is contrasted with the divergent terms and conditions of ownership between infrastructure, systems and professional services.
• While more logically organized and distributed, Open Source CMS systems are likely to lack cutting edge features that may prove highly valuable to many organizations. Whether these are associated with advanced search algorithms, multimedia support, cutting edge encryption, digital rights management or 3D projection onto plasma paper (a future future), organizations that see content management as a competitive advantage will likely require the advanced functionality of the commercially licensed content management system.

Is Your Organization a Good Candidate for an Open Source Solution?

The following list is a minimum set of questions that every organization considering an Open Source CMS should consider.

• Content management functionality required: this is a fairly straightforward criterion to evaluate. After all, OS solutions are available for inspection and evaluation.
• Support required: Is there a vendor willing to sell a support contract? Are the terms consistent with the other support contracts your IT organization already has in place? What is their financial viability?
• Technical skill of in-house staff: Does your organization have individuals in-house or under contract that can evaluate and enhance the areas of the CMS that will require enhancements?
• Mission criticality: Ask yourself "what if?" What other organizations are "powered by" and do they have the same level of dependence?
• Anticipated life span of deployment: Most deployments of software solutions do not last more than a couple of years. Is this project an exception? The community that is behind the OS-CMS is not likely to evaporate completely in 12-24 months, but how confident are you that there will be such a community in 60 months?

If you can step through these questions and feel comfortable, then there is certainly no immediate or obvious reason that OS-CMS alternatives should not be considered alongside the traditional CMS vendor list. There is certainly a potential that you may simplify your deployment and save some significant licensing fees.

Open Source Content Management Systems will prove to be a lasting option and a viable one for a small but growing community of users. However, these systems will not prove to be a serious threat to the "old guard". The only vendors that will be threatened are those that fail to acknowledge this viable alternative.

Sebastian Holst
sebastian.holst@artesiatech.com