Architecture Drives The Roadmap

Your future state architecture really drives the roadmap to Fusion Applications. Gardner McKay once said "If you don't know where you're headed, how will you know when you get there?" A future state architecture is simply a depiction of where you're headed.

The picture above depicts the logical future state architecture JPL's New Business Systems' ("NBS") implementation of Fusion Applications. For clarification, NBS currently consists of the Oracle E-Business Applications used at JPL plus a number of custom applications. In our future state, the E-Business Applications will be replaced by Fusion Applications. This may not be how things finally work out, but it's my best attempt based on the information I have today.

Take a minute to consider the picture. One point should be readily apparent: this is pretty complex stuff we're dealing with, if only because the architecture consists of many layers. Each of those layers has many "moving parts".

One of the significant purposes in depicting an architecture is to share ideas with others. In sharing complex messages, it's my opinion (and just my opinion) that complex messages are best communicated by breaking the architecture down into digestible chunks. So, in order to get the message across to my audience, I've broken down the roadmap effort into layers - in essence, I've created a roadmap for each architectural layer depicted above. That layer-by-layer roadmap is what we'll be diving into for most of the remainder of this series.

NEXT UP: Before You Dive In...

Detailed Roadmap to Fusion Applications - Intro

I often hear complaints from Oracle Application customers about the difficult of strategic IT planning in the face of an absolute dearth of available information on Fusion Applications. I’ve written previously about the balance Oracle must strike between sharing and withholding information, so I don’t plan to rehash that subject here. What I do hope to do here, through a series of “deep dive” articles”, is share what I consider to be a pretty good and detailed roadmap from my own shop that may serve as an example to help you in your own roadmap planning. This post is the first in that series.

Caveats

Some caveats to keep in mind for this series of articles:

1. Your organization is very different from mine. Your needs, considerations, planning and results may vary.
2. Any representation or speculation made is strictly my own opinion, which does not represent the opinion of the Jet Propulsion Laboratory or Oracle in any way.
3. I am the father of seven children, so I have no money...suing me would be a huge waste of your time and effort.
   

Background

JPL’s ERP system is currently running on version 11.5.10.2 of the Oracle E-Business Suite. We’ve implemented Financials (with the exceptions of Receivables and Fixed Assets), Project Accounting, Payroll, HRMS, Purchasing, iProcurement, and Discrete Manufacturing.

Some Assumptions

Some major assumptions in building JPL’s roadmap to Fusion Applications are:

• Applications Unlimited won’t last forever - it's likely that all Oracle Applications customers may have to move to Fusion Applications at some future time.
  
• Oracle will continue to support EBS 11i as a “jump off point” for migration directly to the first two or three releases of Fusion Applications.
• The “Five Years Plus Forever” approach of the Unlimited Support program will continue through at least 2015.

Why Fusion Applications?

In past articles, I’ve written about the need of each Oracle customer to determine their technology needs based on their business needs. At JPL, we’re planning to migrate from EBS to the Fusion Applications for three business-based reasons:

1) Integration and Orchestration: we have many internal customers and partners that utilize different technology stacks: PHP + MySQL, .Net, and Cold Fusion + SQL Server are some examples. We also have compelling needs to integrate the various data repositories and to orchestrate business processes across these multiple technology stacks. The Fusion Applications hold the promise of loose coupling through standards-based integration, which will ease the integration and orchestration of business processes across various technology stacks and data repositories.

2) Flexible User Interface: we’re often faced with customer requirements for user interfaces that go far beyond the capabilities of OAF and the E-Business Suite. The Fusion Applications, which will allegedly utilize ADF, will allow us a greater trade space for tailoring our UI before crossing into the dark and dangerous world of unsupported customizations.

3) Better Conceptual Fit: at JPL, we manage our enterprise in terms of business processes. At its core, EBS is organized around vertical application modules: HR, PA, GL, and so on. Fusion Applications will be organized around business processes: Procure-to-Pay, Hire-to-Terminate, etc. Fusion Applications seem be a better conceptual fit for the process-oriented way we think about and manage our enterprise.

Migrating Directly From 11i To Fusion Apps

At JPL, we have yet to identify a significant business benefit from moving to R12 (again, you may decide otherwise because your business is different from ours) – so our plan at the moment is to stick with 11i until we migrate to Fusion Applications (more details later on how we’ll do this). However, if we learn something in the future that causes us to reconsider, “Plan B” is to upgrade to R12 on Fusion Middleware. This backup plan will at least allow us to achieve the integration and orchestration benefits.

NEXT UP: Architecture Drives The Roadmap

Oracle AppCast

My interview by Cliff Godwin for the Oracle AppCast podcast series was released today. Even though I did some radio work in my younger days, I've never gotten over the odd feeling of hearing my own voice in any type of broadcast...

What's In It For Me?

In my role as a Co-Chair of the OAUG Fusion Council, I have the opportunity of listening to Oracle Applications users from around the globe. In listening, I consistently hear the same types of questions about Fusion Middleware and Fusion Applications: “What will all this spiffy technology do for my enterprise? Will it lower my costs? Increase my sales? Will it allow me to improve the quality of the products or services I deliver? Can I run my business smarter or more efficiently? How about shortening my delivery or product development lead times?” To summarize, customers are asking “what’s in it for me?”

Any C-level executive will tell you that the days of buying and implementing technology for technology’s sake are long gone. Technology has to be justified in business process terms: better quality, faster deliveries, lower costs, or increased sales.

The Challenge

I’ve recently had the opportunity to work with Oracle’s Markus Zirn, a co-author of the BPEL Cookbook. Markus is now a Sr. Director of Product Management for Fusion Middleware and is very focused on getting the value proposition message of Fusion Middleware out to Oracle Applications users. Markus recently pointed me to an Oracle whitepaper, Bringing SOA Value Patterns To Life. It’s a pretty good paper, well worth the read, which summarizes three value propositions that served as a foundation for the BPEL Cookbook:

• SOA-based integration
• Modern, composite application development
• Modernizing legacy and mainframe applications
  

To an IT professional, these three points sound good and make sense. However, explaining their value to a CFO who neither knows nor cares about the difference between BPEL and butter, these value propositions by themselves offer no traction.

Selling Fusion To The Business People

I recently presented a 5-year technology roadmap to an executive team in my own company. Many of the folks in the audience were business people, most of whom care passionately about the outcome or net benefit, but don’t care much about the details of the technology or the tools. A crucial component in the roadmap is a migration from the Oracle E-Business (11.5.10.2) to Fusion Applications, with a move to Release 12 on Fusion Middleware as a backup plan. The move will cost us in terms of capital, equipment and labor. Selling the benefits of the technology in terms that mattered to them was one of the toughest challenges in developing the roadmap. So, for this audience, the value propositions consisted of the following points:

• We have an important need to improve the quality of business intelligence available to our managers, so that we can make better strategic and tactical decisions about operating our enterprise. The significant obstacle to be addressed is the fact that we have independent silos of data throughout the institution. Moving to a Fusion architecture will allow us to consolidate those data silos and deliver integrated, tailored business intelligence on a real-time basis to the people making the decisions.

• We also have an important need to automate critical business processes across different technology stacks. For example, consider our employee termination process. When an employee leaves, we have a complex business process that cuts across our HR department (using Oracle E-Business), our office property group (using an Oracle database but not E-Business), our security office (using MySQL and PHP), and our projects business systems (built on both Cold Fusion and .NET technology). Most of this integration is currently done via “sneaker net” – somebody walking to physical locations with papers in hand. Moving to a Fusion architecture will allow us to maximize the automation of integration between the various technology stacks, freeing up our “sneaker net” people to do other, higher-value tasks.

• Quite a bit of our component manufacturing is performed by our supplier community. Fusion architecture will allow us to automate the flow of communication, products, services, invoices, payments, etc. up and down the supply chain, regardless of the technology utilized by our various suppliers.

• Many of our users could perform their jobs more efficiently if we changed the application forms. Although we have made some inroads here lately, Fusion Applications will allow us to modify the user interface in ways that make most sense for our specific users.

• We manage our enterprise according to horizontal business processes. Fusion Applications will be organized around horizontal business processes rather vertical application modules. With Fusion Applications, Oracle is finally coming around to our way of thinking. Our ERP software will now be a much more accurate model of the way we do business.

These are just a few examples from the presentation I made. In a nutshell, rather than attempting to explain the technology to the business people, I focused on end-state use cases tailored for my specific company. You'll also note that, without too much stress on the brain, you can relate all these use cases back to one of the three basic value propositions from the BPEL Cookbook I mentioned earlier. My approach worked – my audience understood and I “closed the deal.”

Summing Up

In considering whether a move to Fusion Applications and/or Fusion Middleware makes sense for your enterprise, keep in mind that the technology is not an end unto itself but is a means to an end. So start by figuring out what end or ends you’re attempting to achieve. Build the use cases and attempt to make the arguments in terms of higher quality, increased throughput or lower costs…things that matter to business people. That’s how you’ll know whether it makes sense for you. That’s also how you’ll sell it to your business people if you decide that moving does make sense for you...and without their buy-in, you don’t have a project. In short, the first step in moving to a Fusion architecture is figuring out what’s in it for you!

Metalink Search In Your Browser Search Box

This is an example of good news spreading fast...

Catching up on my research last night, I noticed an article at Steven Chan's blog about a new search browser plug-in for Oracle Metalink users. Steve's article pointed me to the Oracle AppsLab blog, where Jake Kuramoto has written about a nifty Metalink search plug-in for Firefox and IE.

Why do I think so highly of Jake's plug-in? Three reasons:

1. It's built on the OpenSearch standard, which means it will behave just like the other search tools in my browser's search box.
   
2. Because my browser is Firefox 2, I can perform context-sensitive searches in Metalink without disrupting my E-Business session.
   
3. Jake developed this plug-in and is freely sharing it just to make life better for the user community. The blog article actually includes pointers and instructions for installing and using the plug-in, including a path to a screencast demo. Folks, this is the spirit of Web 2.0 at its best.

This little plug-in is turning out to be a great productivity tool and the "cool" factor is pretty high - just today, I've had several people in my shop look over my shoulder and ask how they can get the plug-in for their own use. I'm grateful to Jake for building and sharing, and to Steven for bringing this plug-in to my attention. So, stop reading here and go get the plug-in yourself!

Ready For Your 15 Minutes Of Fame?

In 1968, artist Andy Warhol stated that “In the future, everyone will be world-famous for 15 minutes.” If you’re an Oracle Applications User's Group ("OAUG") member and you have a compelling story to tell about your use of Fusion Middleware technology, your opportunity for that 15 minutes of fame may be at hand.

The International Oracle User’s Council (“IOUC”), in cooperation with Oracle, is sponsoring a series of podcasts on the experiences of Oracle Applications users with Fusion Middleware. The podcasts will be recorded over the next two to three months. OAUG is sponsoring members with relevant stories for participation in this program. If you are interested in sharing a relevant story, leave a comment here before July 13 with a return email address – I’ll provide you with instructions on submitting your story for consideration by the IOUC.

And, by the way, if you use Oracle Applications but you're not an OAUG member...you're missing out on some great value.

An Oracle Support Blog

My recent comments on Oracle Support generated quite a bit of energy with readers of this blog, so I thought you all would be interested in a blog dedicated to the subject. Chris Warticki, a Sr. Customer Support Manager with Oracle, moderates an Oracle Support blog here.

Chris and I talked about Oracle Support's desire for more user feedback over dinner together during last year's Oracle OpenWorld. He and most of the Oracle Support folks are looking for constructive criticism, regardless of whether it's good news or bad news, so that they can identify areas that need improvement. Chris' blog is, in part, an attempt to collect some of that constructive criticism.

I encourage you all to read Chris' blog and share your perspectives with him.