August 23, 2009 By J. Schwan
When it comes to IT, there is a multitude of low-cost, creative ways to make resources available. Which ones are worth considering? Is there a more efficient development approach? How can you effectively combine resources with other government districts? The clear need is to evaluate low-cost resources that save money in the short term and provide proven solutions that are advanced and secure enough to avoid long-term pain. This article delves into practical examples you can start using today to save money, speed development and deliver higher-quality solutions.
For the past 10 years, many technologies we've been using reached wide-scale adoption and essentially became commoditized. It's baffling how many private and public organizations continue to invest large amounts of money for technologies that have reached mainstream status. These technologies aren't necessarily offering the best value for the investment anymore. They simply help run the institution. In these cases, consider other options to cut costs by using reliable low-cost or free solutions.
Commoditized software often comes in the form of open source technology projects. I'm not saying all open source projects are commoditized technologies, but many open source software (OSS) projects spring up because a group of talented developers are tired of paying for services that can be cloned easily.
Good open source projects have strong communities behind them, and in some cases, build better features and offer better support than the technologies they set out to reproduce. The key to picking an open source project is not basing it on the product's current features, but by the strength and viability of the community behind it.
The community can be measured by the speed of innovation (i.e., how quickly product releases reach shippable status), the support forums (i.e., how mature and active the discussion forums are), and the number of active installations currently in production environments. If support, service and commercial product businesses have been built around open source solutions, you can feel comfortable that the project has some legs and will be around for the foreseeable future.
Some commoditized software morphs into a software-as-a-service (SaaS) model. New and existing vendors are packaging solutions that don't come with a sticker price, but instead a subscription fee. Many of these solutions have free programs available for a limited number of users (great for small to medium-size municipalities or districts) or with limited feature sets, which in some cases are all an organization needs.
It's even more important to analyze the viability of SaaS vendors, since the technologies are relatively closed compared to open source solutions. When evaluating SaaS solutions, it's important to understand a vendor's funding/cash position, subscription base and the openness of its platform, in case you want to migrate to a different solution later.
In today's market, where the cost and time to build software have been greatly reduced, almost every common function of an organization has an open source or SaaS option to support it -- from office productivity and collaboration platforms to customer relationship management.
Henry Thiele, director of technology for Illinois High School District 207, recently replaced a Novell-based e-mail solution with Google Apps. "Every teacher in the district has a District 207-branded Gmail account. When we moved from Novell to Google Apps, we gained a ton of features without having to spend a dime," he said, "not to mention mailbox storage went from 254 MB to 7 GB."
These are by no means the only options, but all are definitely worth considering as low-cost alternatives. Open source software and SaaS often provide 80 percent or more of the proprietary solution's functionality, which accounts for 99 percent of what's actually required.
Municipal CIOs can also introduce many of the free SaaS social-media tools as a way for leaders to connect with constituents. Mark Bradford, co-founder of the new media marketing company ChirpUp.com, has a message for public CIOs: "Municipalities have an opportunity to apply many of the new SaaS and OSS technologies -- like blogs, Twitter and social-networking sites -- to connect with constituents. In an era when communities need help, governmental transparency expectations are high and budgets scarce, municipalities have a chance to get creative and find new, efficient ways to humanize interactions with citizens. There are free tools available to mayors, councilmen and aldermen for authoring, publishing and promoting their message."
Some municipalities have already started to use this medium. "Look at Chicago Mayor Richard Daley's new YouTube channel," Bradford said. "He's taking his message directly to Chicago citizens and, perhaps, national leaders. There is no mediation or dilution by mainstream media journalists -- no color commentary, no editorial by the broadcasters. He's speaking directly to us."
Commoditized hardware solutions are now offered in the form of cloud computing infrastructure services. Many vendors -- like Amazon, Google, Microsoft and IBM -- are starting to build massively scalable data centers (much like power plants), and are offering their computing resources in a "pay by the drink" model. Instead of institutions paying for an entire server and all of the business continuity, disaster recovery and administration fees that come with it, they can pay by the amount of storage used per month or by the central processing unit power they consume. In other words, you can pay for infrastructure services the same way you pay for water or electricity, with all of the business continuity, disaster recovery and support costs baked in.
The birth of these services is relatively recent, and there are still many questions concerning security, service-level agreements and privacy, all of which are being handled in slightly different ways by each vendor. There's bound to be some standardization in these areas as the technologies mature, but there are ways to take advantage of these services in the short term -- if not as alternative production environments for nonmission-critical applications, then certainly as test environments for new and existing applications. The associated hardware and networking costs for test environments often mirror the cost of production environments. Cloud computing is a great way to mitigate those costs.
Among the major players in this space, Amazon is definitely the most mature, but IBM, Google and Microsoft are hot on its heels with more advanced solutions coming available.
For projects that involve some level of custom development work, an institution can fall back on a traditional "waterfall" type of methodology for planning and delivering the solution. This involves doing heavy upfront scoping, planning, analysis and design before starting development or coding. These planning phases are done in an attempt to uncover all the stones and predict how long the development effort is going to take, as accurately as possible.
The results of this methodology have been mixed at best. A recent study by the Standish Group showed that 82 percent of all IT projects were either failures or were considered challenged (i.e., over budget, late or didn't deliver on promised scope). The top reason for these unsuccessful projects was deemed "requirements change and scope management." These statistics illustrate the true reality: You can spend as much time as you want planning, but the world is changing while you're doing it.
Another interesting metric uncovered by the Standish Group is that only 20 percent of features built are either "always" or "often used," while 64 percent of features built are "rarely" or "never used." In other words, almost two-thirds of what folks are building is not adding any value to the customer or constituent of the application.
Another issue with waterfall-based development is that many stakeholders will try to fix a project's scope, timeline and budget. In these cases, the only thing that's left variable is quality, the result being many waterfall-based projects suffer from support issues and quality control problems long after the product is built.
The agile development methodology attempts to address these issues by turning software development on its head. The goal of agile development is to deliver "shippable," high-quality code that is providing some level of value to the organization in tight (usually monthly) iterations. Each month, the team collaboratively sits down with the functional stakeholders to determine what will be delivered in that month's "sprint," or iteration. The team, including the stakeholder, has a "stand-up" (a short meeting) every day to discuss questions, reset priorities, demonstrate new functionality and resolve any issues the team encounters. The functional stakeholder decides at the end of every sprint, whether the product is ready to be released to the constituents.
The agile project's results are gratifying to both the technical team and the functional stakeholder. Both see a real product developed and delivered every month. And unlike waterfall projects, if a project's budget is killed or the project must be put on hold after three months, an agile project has three months of working code versus a waterfall project's three months of documentation that will quickly become outdated. With agile, as the institution's needs change, the project's priorities can be adjusted on a monthly or daily basis.
In these times of tight budgets, it's time to stop planning and start building. Agile development provides just enough process to encourage collaboration, shared accountability with your stakeholders, and most importantly, delivery.
Many public institutions have exactly the same needs. Citizens want to pay their bills, sign up for classes, register for events, file for permits and receive public safety alerts online. Yet in the absence of quality packaged solutions (or open source/SaaS) solutions, individual governments with limited budgets will individually build their own proprietary, often low-quality, solutions. It's bewildering how many different municipalities, counties, states, universities and other public institutions have rebuilt the same system a hundred different ways.
The reality is, if these governments and institutions reach out to their peers before embarking on their next custom-development initiative, they may be able to pool their budgets and resources to build a higher-quality solution.
Gordon Eaken, the director of information systems for Hoffman Estates, Ill., is an advocate of intergovernmental collaboration. "One shared service we use is the state of Illinois' ePay Web site, which we plug into to do all the processing of online payments," he said. "Payments get processed through a state-level agency. The program offers some major cost savings and it leaves the security issues to someone else to manage."
But collaboration doesn't have to stop at software. Municipalities also can share their infrastructure environments. "Northern Illinois University has a campus nearby with a huge pipe to the Internet, with bandwidth we could never afford." Eaken said. "So we laid fiber down around town to connect to NIU's pipe. It turns out some local hospitals and surrounding municipalities were able to plug into our fiber. Now we have shared costs for our [wide-area network] and our Internet access, which saves the taxpayers a lot of money."
The mantra of open source cannot apply better to government institutions: Build as a community, for the community, thereby reaping the rewards of being a community. For your next IT initiative, think about working together to build something better.
We're all being asked now to tighten our belts. To succeed in this economic environment, we should find creative ways to stop paying license and support fees for commoditized services, stop using old methodologies to build new solutions, and stop building in a bubble. By embracing open source, SaaS, cloud computing, agile development and intergovernmental collaboration principles, we can stop wasting money on services that provide little value to our constituents and save our dollars for who's really going to help us deliver -- our people.
You may use or reference this story with attribution and a link to