It has long been the belief that sound enterprise applications die hard, even if at the expense of being technologically backward—especially those that feature solid vertical functionality that "does the job" and that have large install bases worldwide. What is meant by "backward" here is technology that has monolithic code on two-tier client/server architectures, if not even more outdated than that (see Architecture Evolution: From Mainframes to Service-oriented Architecture).
In most cases, loyal and conservative customers prefer to stay with their workhorse product almost indefinitely, and in many cases, it even makes economical sense for the vendor to continue support for the aged product (see Support for Old Releases: Good for the User, but Is It Good for the Vendor?) Thus, in some instances, we have endorsed the pledges of some vendors to keep their legacy products alive (see SSA Global—the Right Product Strategy).
At the Inforum 2007 annual user conference, amid so many upbeat and seemingly "sexier" announcements, we learned about some former independent software vendors' (ISVs') bad product development and customer support policies, and that Infor has recently acquired these products. While these former ISVs (MAPICS and SSA Global in particular) were preaching "no sunset" (no discontinuation of product enhancements, support, etc.) promises to the outside world, behind the scenes they were in fact discontinuing many products and persuading customers to migrate to the next-generation releases. Making such a change would often entail great leaps over multiple consecutive product releases.
More specifically, after initial assurances from the vendor that it would continue providing support for all acquired products, the former SSA Global's subsequent (and simplified) product strategy changed drastically. The diverse products were converged into one of two major next-generation products: UNIX/Java 2 Enterprise Edition–based solutions (see Understand J2EE and .NET Environments Before You Choose) and IBM System "i"–based solutions (formerly AS/400 and iSeries—see The Blessing and Curse of Rejuvenating Legacy Systems).
A similar situation took place at the former MAPICS, where products were lumped either into the IBM System "i" camp or the Microsoft .NET camp (see Subtle [or Not-so-subtle] Nuances of Microsoft .NET Enablement). The former Pivotpoint product on the Oracle database, the SyteLine product version on the Progress Software database, and the OpenEdge development platform (see Frontstep Ups the .NET Ante) were all discontinued, with barely a qualm.
Little thought or effort (if any at all) was given to ongoing enhancements and support for older product releases (some of which are still used by thousands of relatively happy—albeit disconcerted—users, and still in need of ongoing upgrades), especially not to how to enable smoother migrations to the "famous, best-ever" upcoming next-generation products.
The first "generous" options granted to customers were to either stay on the old release or to migrate to the "latest and greatest" release at the customer's pace. But eventually, customers suspected the vendor's initial subtle persuasion would turn into an outright "migrate or else" attitude.
Ironically, even those next-generation, service-oriented architecture (SOA)–enabled products were not that cleverly or thoroughly designed either. Namely, just professing support for the IBM WebSphere or Microsoft .NET stack, without seriously rewriting or at least breaking the monolithic code into manageable, self-contained software components (services), is not the way to modernize any product. The underlying code has to be painstakingly revamped to harness these proprietary SOA infrastructure platforms. And never mind the hefty price tag to own these platforms or the learning curve for developers that have long been well versed with other technologies (e.g., the procedural IBM RPG [report program generator] language versus object oriented Java or Microsoft C#).
These vendors apparently had neither the guts nor the wherewithal to take the SOA plunge (see Rewrite or Wrap-around Old Software?), and their professed "WebSphere/.NET will help everything" approach has certainly not lived up to the customers' expectations (see Contributing to the Rejuvenation of Legacy Systems in the Enterprise Resource Planning Field).
Infor's Damage Control
However, disseminating bad news and negativity is the not the intention of this article. Indeed, the annual user event left us with the impression that Infor has been "hitting on all cylinders," with new software license sales, new product deliveries, and technological rejuvenation. In fact, in its five years of existence as a private company, Infor has become a $2.3 billion (USD) business, with over 9,200 employees in nearly 120 offices worldwide, and with 1,100 channel partners. Given these numbers, the vendor has become a major competitor to SAP and Oracle in many enterprise-level software deals (see The Enterprise Applications "Arms Race" to Be Number Three).
With more than 70,000 customers and its ongoing strong performance, Infor believes its strategy, which is to match the domain expertise and business-specific functionality of the niche, best-of-breed provider, together with the stability and scale of a global provider (see The Impact of the "Assembler Strategy" in the Enterprise Applications Field), has been validated in the market. Infor has reported record growth based on strong demand from both new customers and existing customers, that seem to be acquiring other complementary solutions from Infor's burgeoning product portfolio.
For one, brand new customers are selecting Infor: over 1,700 new client logos have been added to the vendor's install base in the past year. Some skeptics might point out that most of these new accounts are small deals, as they mostly involve emerging markets, where merely showing up is the key to winning the (very small) contract. True, the days of blockbuster deals are long gone, and "showing up" can be practiced by any competitor. But not that many competitors can claim new-deal success like Infor can, nor can they claim a growth of about 15 percent in new license revenue.
Secondly, Infor's existing customers are extending the use of their current solutions. In the last year or so, Infor's customer base has accounted for 10,000 license sales that add additional users, sites, and modules to their solutions.
Last but not least, existing customers are expanding their investments in terms of functional footprint. Over 1,000 customers acquired new, complementary solutions, ranging from enterprise asset management (EAM), demand planning, the Infor SupplyWeb supplier relationship management (SRM) product, workforce scheduling, warehouse and transportation management, corporate performance management (CPM) and analytics, customer relationship management (CRM), strategic network optimization (SNO), product lifecycle management (PLM) for process manufacturers (the Optiva PLM product), and so on.
Infor's Open SOA Strategy
The trends noted above occurred after Infor announced its ambitious SOA strategy in March 2007. The strategy is based on the principles of openness and heterogeneity (i.e., the customer chooses the platform, the time frame, etc.); on leveraging existing applications; on being free of charge (i.e., free outside of the standard maintenance contract, as long as the embedded enterprise service bus [ESB] will interconnect and orchestrate processes amid Infor's disparate applications. Connecting to non-Infor applications will entail additional license fees, which is fair enough); and on adhering to the enablement of extended-enterprise business processes.
For more details and concepts behind Infor Open SOA, see Two Stalwart Vendors Discuss Platform Approaches (Wars). An even more detailed description of how Infor plans to deliver pragmatically on its multiyear SOA project deserves an entirely separate research article.
Infor has kept its promise to some customers by rolling out the Infor Open SOA road map at Inforum 2007. The road map gave customers insight into reasonably paced SOA-enablement plans and into the long-term direction of the vendor's existing solution. Infor has published statements about its direction and has held design reviews with many customers to ensure that they understand the approach and that it will serve their businesses. Most of these customers are of the opinion that Infor has made significant progress on Infor Open SOA, as well as to its adherence to the vendor's "Three E's" (enrich, extend, and evolve) strategy.
The "enrich" part of the strategy refers to the adding of value to current products and projects. Infor has released nearly 100 product upgrades and feature packs over the past year or so. It is important to note that there is no forced march imposed upon customers here; these feature packs can be enabled or disabled by turning the "on" or "off" switches in a parameterized setup. An example of this "enrich" part is Infor's delivery of two new versions of the Infor ESB, and the inclusion of these versions in some of its product releases. Another example is Infor's first feature pack release of the former Workbrain human capital management (HCM) solution, now rebranded as Infor HCM Workforce Management.
The "extend" part of the strategy refers to the extending of the footprint and Open Applications Group Interoperability Standards (OAGIS)–based interoperability within Infor's portfolio of applications, in order to meet the growing complexity of global supply chains. Customers will receive ongoing SOA-based integrations, enabling them to extend their current solutions and build a foundation for future SOA capabilities. The initial integrations are focused on enabling applications to use standardized business object documents for linking business process flows across applications in a publish-subscribe, event-driven, and decentralized framework, all via Infor ESB. A good example is Infor ERP SyteLine, which is now interoperable through Infor Open SOA with Infor's EAM (former Datastream) and SRM (SupplyWeb) solutions. This integration should give discrete manufacturers better capabilities for manufacturing processes, purchase order collaboration, and asset-tracking.
This is a true standards-based approach designed for diverse IT environments, as it reduces the costs and complexity usually associated with adding or upgrading solutions. Among the 20 or so similar product integrations is a new business-specific analytics module for Infor SCM Warehouse Management called Labor Forecasting, which is based on Infor Performance Management solutions. Also, a good example of a the functional footprint extension is the latest release of Infor SCM Demand Planning, the supply chain solution that mitigates the risk and complexities of managing dynamic supply chains.
Finally, the "evolve" part of Infor's Open SOA strategy follows along the lines of "future-proofing" products (developing products that will not become obsolete for some time). Again, customers are heavily involved here via numerous design sessions so that new enhancements are easy to implement and to use. To that end, after the quiet acquisition of the ClearUX tools and intellectual property, Infor has been building role-based home pages (with built-in role-based security levels and data access) that use Web 2.0 technologies such as Asynchronous Java and XML (AJAX) or Adobe Flex. This gives users personalized views of information that is important to their role, and in a "rich-client" metaphor.
Using the Infor Open SOA framework, these role-based home pages also enable "enterprise mashups" (composite applications) of data from Infor and non-Infor systems, so that users can intuitively view and access information. With about 140 devised roles currently in the library, Infor has delivered the first generation of role-based home pages in Infor ERP LN (formerly Baan), with the next generation of featured home pages to be delivered in mid-2008.
Infor has conducted its own research, and the vendor realizes that users want to deploy thin clients and browser-based user access with the flexibility to use many user interface (UI) technologies, including mobile ones. Infor is thus pursuing a flexible, zero-footprint UI solution, using state-of-the-art technologies such as AJAX, Flex, and other open standards.
Infor's UI strategy is also centered on providing role-based, personal interfaces that enable broader business processes across traditional functional silos in the business. Furthermore, these role-based home pages will be enabled by Business Information Services, another upcoming SOA component that will help enterprises capture and consolidate data in a centralized, secure database for reporting processes. Reporting services will come with pre-built analytics and will support the ability to drill down to the original data source from the user's home page. These services will also support both Infor and non-Infor applications.
New, Cherry-picked SOA Components
Further along the "evolve" lines of Infor's "Three E's" strategy, the vendor is accessing its vast intellectual property to develop more universal (overarching), business-specific solutions to create new, self-contained components based on native SOA technologies, which are interoperable by design. These new components will be "cherry-picked" (selected precisely) to tap business-specific functionality and industry experience from Infor's global development team, to provide functionality that will work with software from both Infor and third parties.
For example, the upcoming multi-books accounting module is an SOA component that should give global companies the ability to conform to multiple, country-specific accounting standards and currencies. The module can either run concurrent with an existing general ledger (G/L) system, or serve as the primary accounting module. The idea behind the multi-books accounting capability is that it is to work alongside financial management systems to help companies cast their financials in multiple ways. If, for instance, a corporation has an operation in China or Brazil, and it has to follow these governments' rules on what one precisely refers to local accounting concepts, like "salary," "wage," "value-added tax" (VAT), or "sales tax," how do users get a system without having to rip and replace what they already have in order to work in China, Latin America, the US, and Europe?
To that end, Infor has designed a new component that can run independently of enterprise resource planning (ERP) systems, and that can coexist with other systems to manage financials through the component and ensure compliance.
The multi-books accounting and other SOA components (e.g., pricing, contracts, and promotions; actual costing; multi-echelon inventory control, planning, and optimization [with capabilities like distributed order management, vendor managed inventory, etc.]) are slated for delivery at the end of 2008. Given that SOA cannot exist without master data management (MDM) and SOA governance tools (which register and manage components' life cycles), Infor also promises the availability of such public and federated components, again at no additional license charge.
This is part one of the three-part series Ambitious Plans and Promises: An Enterprise Software Provider Keeps Its Word. Part two takes a more in-depth look at Infor's Open SOA, as well as customer response to the vendor's latest solutions.