How to Eat an Elephant

Posted on February 22, 2012 by

Have you ever heard the question, “How do you eat an elephant?” The answer is, “One bite at a time.” We talked with Scott Bowen about how an iterative approach has changed software development.

“Waterfall development is a serial process that starts with gathering requirements, moves to designing and development, and then testing to make sure that it works properly. If you’ve worked in a waterfall development environment, you usually spend a lot of time early in the process to get all of the customer requirements down on paper, interpreted correctly. Any software product I’ve worked on over the past fifteen years that has employed a waterfall method has had significant challenge to properly document all the user requirements up front and then demonstrate working code after a long cycle of development that meets user requirements. Often, the first time you demonstrate a product to a customer, they may like a little of it, but much of what is demonstrated is not at all what they expected. So much interpretation happens between getting requirements and executing the development. At least in the first demo or two you are likely to hear, “I don’t like that. It doesn’t work for me.”

Agile software development is an iterative development process, using collaborative teams of developers, testers and project managers working together to understand the customer’s requirements, build an iteration of working pieces, demonstrate those to a customer, and then from the customer feedback, refine the development for the next sprint. Following this process, you get quick feedback to allow you to inspect and adapt, and you’re constantly working from a prioritized backlog of work. As you evolve the product iteratively toward the release date, you have this ability to be flexible, making changes to align the product with the customer’s vision for the product, as a customer’s idea of what they want often evolves, as well. You can adapt to changes more quickly, and more efficiently.

The specific environment or goals of a development project ultimately determine what method will work, however, I’ve had great success with Agile approaches like Scrum in developing a software product that stays tightly aligned with user requirements while empowering and motivating talented developers.. Instead of focusing a team’s effort upfront on so much documentation to understand and interpret a customer’s requirements, making assumptions that may be proven incorrect down the road, Scrum allows you to spend your resources from the very beginning on building a working product. The conversation that happens when you demonstrate an early working increment of a product results in a lot of useful feedback that is hard to get from a waterfall methodology.”

Although you may never need to eat an elephant, an iterative approach is a very good idea for software development. What are your experiences in Agile software development?

Contact Scott Bowen by email at ScottJB@vmc.com or 877.393.8622.
Posted in Software Development and Testing | Tagged , , , , , , , , | Leave a comment

What if Darwin was a Data Center Manager?

Posted on February 21, 2012 by

Charles DarwinWhen you’ve been around something for 25 years, you’ve likely seen a lot. If you’ve lived in the data center world for the last 25 years, the evolution has likely changed everything about the way you design, build, run and support your data centers. We talked with a resident expert who has been involved in every aspect of the data center business for a while now, and here’s what Chuck Hobart had to say about that evolution.

“Over the last 25 years that I’ve been doing this, the challenges in designing and building a data center have been very broad across the spectrum. Probably one of the biggest challenges is the rapid evolution of the data center. Seven to eight years ago you had 7U servers as a standard, which were large servers that you could only fit six into a rack. It was fairly easy to design the power and cooling. Today, that standard has grown to 128 servers (cores) inside a rack, which changes and complicates the power, cooling, connectivity, among other things. I’ve seen issues around the types of data centers, including weight loads, power, cooling, connectivity, security, and accessibility, all before the first server goes in.

A recent trend is that people are now weighing the value of building a data center, at all, because even before the first server is installed; you have to cover of the costs of the physical space, facilities and utilities and maintenance costs. There have been new developments in the design of facilities in stages, where the first stage is operational while the second stage is being constructed and the third is in design. The stages are basically isolated from each other until the last minute when they link back together, allowing for ramping and faster utilization.

With cloud computing, we’re getting even more challenges as these data centers are becoming “mega structures”. This is because there is an economy of scale and a better ROI with a larger facility. On the other side of the spectrum, there’s much innovation with the development of things like “pods”, which are isolated data centers within existing data center structures. There are also “containerized” data centers, where the pre-configured infrastructure is in a trailer out in the open, and a tractor trailer pulls up and drops off a load of servers that plugs into a large power cable and network fiber, then you’re up and running in a few hours. These have been use in places where military operations or emergency relief takes place.

Data centers cover quite a wide gamut anymore, and each presents unique challenges”.

Data centers will continue to be a critical part of the infrastructure chain, however, they will continue to change and evolve to better serve the needs of users. What trends are you seeing in the data center business?

Contact Chuck Hobart by email at ChuckH@vmc.com or 877.393.8622.
Posted in Data Centers | Tagged , , , , , , , , | Leave a comment

What Can Be So Bad About A Messaging Migration?

Posted on February 14, 2012 by

In the early 14th century, the “Black Death” wiped out nearly half of Europe’s population. These days, messaging upgrades and migrations performed by internal IT teams cause serious problems or fail to deliver expected results about half of the time. So, among messaging IT directors and managers, those projects may be just about as welcome as Bubonic Plague. Today, we begin a series of blogs by two VMC experts, Engagement Manager Kris Honkola and Program Manager David Cherry, on the trials and tribulations of messaging migrations and ways to be successful with them. To begin, we asked Kris to shed some light on what makes these IT assignments so fearsome.

“Upgrades and migrations are not very difficult in a sheer technical sense. Really, they’re much simpler than other integration and IT projects a director will manage. But, I’ve seen them deliver a fatal blow to many IT directors’ careers for a couple of reasons.

First, major upgrades and migrations just don’t happen often. Maybe every ten years – once or twice during an IT person’s career – he or she is responsible for a migration. With that kind of infrequency and time span between experiences, it may be that they’ve changed employers or that technology has changed so significantly that whatever best practices they learned haven’t been retained or just aren’t relevant now. That lack of continuity in history and experience alone creates a risk that an IT manager would rather not face.

The second reason migrations can be nerve-wracking is due to their high visibility. If something goes wrong, it’s likely that it will impact the company’s top management. Just recently, I witnessed such a case. Some potential minor impacts of email server maintenance were not communicated to the users ahead of time. When the issue occurred, of course it was the company executives who were caught off guard and affected most. Invariably that’s what happens, or at least what IT directors fear…that whatever goes wrong will take a Murphy’s Law kind of turn. It won’t create problems for a part-timer or the janitor; it will create problems for the people at the top – the people who have the power to fire the IT directors.

This example gets to one of the two biggest reasons a migration goes wrong: comprehensive understanding of the business impact of the changes being done. We’ll cover that further, as well as the second major reason, next time.”

If there is a server upgrade or migration on your horizon, what can you do to look forward with greater insight instead of with dread?

By email, contact Kris Honkola at KyostiH@vmc.com or David Cherry at DavidCher@vmc.com. Both may be reached by phone at 877.393.8622.

Posted in Data Centers, Project Management, Enterprise Collaboration, Knowledge Management, Consulting | Tagged , , , , , , , , , , , , , , , , , | Leave a comment

Good Cloud Migration Decisions Hang in the Balance

Posted on January 24, 2012 by

Unbridled enthusiasm isn’t always for the best. Getting caught up in the excitement without balancing it with reality may not work out quite the way we have in mind. Although VMC Engagement Management Chuck Hobart shares in the excitement of Cloud computing’s potential, he discourages companies from putting the cart before the horse in making a move to the Cloud. We asked why.

“Before you can make a good decision about the Cloud, you should clearly understand the current costs to run your IT organization. You need that baseline to determine whether there’s potential for a meaningful return on your investment. If there’s no ROI for going into the Cloud, then don’t go there.

Because IT is not most companies’ reason for existence – they’re a meat packing company, a manufacturer, a pharmaceutical company – they’re usually focused instead on the ROI of their core business. IT costs are considered a ‘necessary evil,’ so they tend not to view those costs in an ROI framework, but just accept them as an unavoidable cost of doing business.

So, even though the Cloud can be a powerful, cost-effective and beneficial environment, it’s important not to put the cart before the horse. Only after you’ve researched your current IT costs and issues are you at the right point to ask, ‘Now that we understand what our costs are, what can we do to reduce them?’ Although one option is going into the Cloud, another option is to rearchitect and streamline without going to the Cloud.

If you try to move into the Cloud without understanding what you currently have, you’re going to fail and it will be very costly. One insurance company I know tried to do that. So far, their cost overruns are more than a million dollars. The IT organization was given a directive to go into the Cloud, and they were told to meet schedules that made it impossible to do an upfront analysis of their current state. They also underestimated the complexity of the migration process, because early hype about migrating to the Cloud made it sound far simpler than it really is. There can be a lot of benefit to doing it, but it’s still a complex process. So, if you bypass the ‘due diligence’ of accurate assessment and planning, you’ll end up with cost overruns and a non-functional system. You may even put your core business at risk.

Because of cases like this, I encourage people to slow down and even be a bit skeptical about the Cloud. They should take a close, rational look at what is hype and what is reality – and not just in general terms, but specific to their company, their business goals and their IT environment. But, despite some of the hype that’s out there, if the ‘buzz’ gets people asking whether they should move to the Cloud, it has inherent value. Whether their answer is ultimately ‘yes’ or ‘no’ isn’t the point. The question’s value is that it prompted them to evaluate and understand their current IT environment better. Before, they had no idea what IT was costing. Now, not only can they make a more intelligent decision about whether the Cloud is a viable option; they have a great knowledge base for addressing other IT questions, too.”

The success of a future Cloud migration hangs in the balance of understanding your current state. What can you do to make a well-grounded decision that won’t overburden your available resources?

Contact Chuck Hobart by email at chuckh@vmc.com or 877.393.8622.

Posted in Cloud Computing, Consulting, Data Centers | Tagged , , , , , , , , , , | Leave a comment

Building Timeliness into Your Digital Snack Collection

Posted on January 11, 2012 by

To keep contents relevant and encourage repeat visits, museums round out their permanent collections with new acquisitions and special exhibits. They provide us with a selection of both the timeless and the timely. VMC’s Michael Pruitt explains how timing and selection factor into building a digital banquet – a coherent collection of digital snacks.

“On average, it takes about three months to create a typical technical manual, so we use that as a benchmark for development of an initial digital banquet. Digital snacks should provide a quicker time to market. If we start exceeding the time required to publish a manual, then the snacks aren’t providing the right business value.

But, there’s a difference in how and when you release your digital banquet versus a technical manual. For the manual, you pull all the documentation together in a linear way, and you don’t print or release it until it is complete, until every chapter is done. A digital banquet is much more flexible. You still need to have a sufficient collection of core elements – snacks – to achieve some sort of critical mass before you release, but it doesn’t have to be comprehensive. That is, for a manual, if one of your content contributors hasn’t finalized a certain chapter, it delays your release. You can’t print it until all the chapters are done. An initial digital banquet could be published without that chapter – without the equivalent digital snacks it would contain. You’d just add that chapter’s worth of snacks later, as they were ready.

Manuals also tend to get treated as being ‘done’ when they’re released. Less attention is given to keeping them current. The producers move on to the next project, the next manual. But, a digital banquet and the snacks it contains are never really finished. It’s an iterative process in which you continue to add, update or remove content to keep the collection of snacks relevant and timely. How long and how much you continue to add really depends on the needs of the information consumer. At some point, the digital banquet becomes more permanent and less fluid. That’s based on diminishing returns – but what’s key is that the point of diminishing returns is determined from the consumers’ perspective – when they see less need or value. In marketing terms, digital snack content delivery is based on ‘pull-through’ – consumer demand – instead of a ‘push’ model in which the producers put the information out there on their own terms.”

A special exhibit gives people a reason to go back to a museum, even if they’ve visited many times. What are you doing with content to keep your information consumers coming back for more?

Michael Pruitt is an Engagement Manager at VMC Consulting. You can contact Michael at MichaelPru@vmc.com or 877.393.8622.

Posted in Consulting, Knowledge Management, Mobile Services | Tagged , , , , , , , , , | Leave a comment

The Theory of Relativity Applied to Service Desk Staffing and Training

Posted on January 5, 2012 by

“Put your hand on a hot stove for a minute, and it seems like an hour. Sit with a pretty girl for an hour, and it seems like a minute. That’s relativity.” Einstein’s explanation of his theory makes sense in other realms besides physics. Time is also an important dimension in determining the relative merits of service desk onshoring versus offshoring. We asked service desk expert Andrew Vloedman to enlighten us.

“When I got into this industry in the mid-’90s, a level two network engineer was making $32 an hour. Today, though, the U.S. market is flooded with people who are capable of doing this work. So, rates in the U.S. are dropping, while rates offshore are increasing due to the labor force competition I mentioned. Do people in Costa Rica make what they do in the U.S.? No. But, is the scale sliding up on one end and down on the other? Yes. Slowly, this pay scale convergence will drive service desk business back to North America. There just won’t be the kind of per-hour labor cost differential that made offshoring so appealing before.

Training is also something in which costs and benefits are relative. For example, we onshored one service desk from Central America. There, locals received eight weeks of technical training after two weeks of language and cultural training. That’s a total of ten weeks of training – non-phone time – whereas, we were up and running here in the States with ten days of training…and that included the soft skills and VMC training, as well as the customer-specific training.

What you’ll find, like near our San Antonio call center, is an abundance of talented people who have worked in this business, who already have the technical background and who are available – especially in this economy. You don’t have to teach them how to reset a password. You only have to train them on the nuances of this specific tool for this specific customer. The overall skill set is already there.

At minimum, you’re looking at reducing your training time by half, usually more, with North American agents. Now, if you add up total costs, applying the wage differential against training time, maybe your overall training costs are a wash. But, if you look at your ability to scale, your ability to react to attrition, shrinkage, ramp-ups due to volume, et cetera, you don’t have that ten-week lead time to get up and running in the U.S. Plus, you have greater flexibility in matching resources to demand, which also helps with risk reduction. So, again, there are those costs that show up on the bottom line; but there are also those unintended consequences to consider. With an onshore desk, you’re not failing for ten weeks while you’re waiting for people to be able to get on the phone. You can respond much faster.”

Time may be money, but relatively speaking, it’s even more than that in the service desk world. It’s also flexibility and risk reduction. How could you make the most of your time in service desk operations?

Contact Andrew Vloedman by email at AndrewVl@vmc.com or 877.393.8622.

Posted in Customer Care | Tagged , , , , , , , , , , , , , | Leave a comment

A Fresh and Flexible Framework for Digital Snack Delivery

Posted on January 4, 2012 by

The abundance of choices in a buffet lets us pick and choose just what we like and have a hankering for. We don’t have to wait for our order to be placed and prepared, and we’re not constrained by a menu offering a handful of entrees and our choice of two sides. Last time, VMC’s Michael Pruitt mentioned the need for a unifying framework for digital snacks. He elaborates here about this “digital banquet.”

“The digital snack pivots around the idea that we give consumers information the way they want it, but we still present it as a coherent collection. So, a blog or a video clip might be a digital snack, but we group and organize those into a “digital banquet.” Think about going to a buffet restaurant. You have lots of options. Yet, those options are grouped at various stations or tables. You go to one station for entrees. Desserts are grouped at a separate station. There’s a salad bar. A buffet uses an overriding framework that gives you all the parts that make up a meal. Subgroups of those elements – our digital snacks, metaphorically – are presented logically. This isn’t a constraint for what you put on your plate. It just facilitates finding what you need. If you want to build just a salad or stick to only dessert and coffee, you know where to find those elements. What you choose is up to you.

Even though a blog is a digital snack, a bunch of blogs without the overriding theme – the digital banquet it’s a part of – is fragmented. For example, if I’m a service tech out in the field working on a tractor, I can’t just go to a tractor blog and quickly find what I need. There’s no coherent structure to help me navigate easily to the right information for a specific troubleshooting issue. It would be like going to that buffet and finding a piece of pie sitting next to the green beans. Or, in tech world terms, it would be like developers looking for a shiv to solve a programming problem. If they Google it, they pull up too much information, and there’s no quick way to know whether the various pieces of code are reliable, the latest version, or even functional.

The digital banquet, on the other hand, uses a themed approach. For example, technical information for fixing a Mercedes would be grouped into a Mercedes-Benz digital banquet that contained a set of digital snacks that the information consumer knows are current and maintained. It’s not as free-form as a blog world or Google. There’s just enough structure to find the specific information needed for the task at hand. It’s how we enable information consumers to make efficient, useful choices without forcing them to sift through a bunch of information that isn’t current or relevant.”

A digital banquet gives people both the freedom and framework to quickly find and digest just what they need. In content delivery, are you offering your audience options or forcing them to stick to your menu?

Michael Pruitt is an Engagement Manager at VMC Consulting. You can contact Michael at MichaelPru@vmc.com or 877.393.8622.

Posted in Consulting, Knowledge Management, Mobile Services | Tagged , , , , , , , , , | Leave a comment