The term “sustainable” appears to be very popular in its use, but not very well defined – if you do an Internet search of the term, the first page of results will yield several different definitions.  So, in an effort to avoid confusion, we will define what we mean:

• The care taking and maintenance of a sustainable solution should be maintainable by local staff, providing good jobs for local citizens.
  When we install a new system, we will typically start the project by handling many of the tasks involved in getting it into full production, but at the same time, training local staff to perform installations, upgrades and provide at least first-tier help desk services.  At the point when local staff become capable of performing these duties, Visual Software will drop back to advisory and second-level help desk roles and providing updates to the software as needed.
  • This chart shows a typical “mix of effort” for a large-sized project that includes cloud-based software, training of local staff, and implementation of our solar energy package. The blue areas represent work performed by Visual Software staff and the orange area represents work performed by local staff.
  • At the beginning of the project, the software is configured, the cloud hosting is set up and training of local staff begins.  As the implementation progresses, local staff will become proficient in the installation of solar kits and provisioning of accounts.  Visual Software involvement will become less necessary as the project continues and reach the point where the implementation is almost exclusively being managed by local staff.
• Applications in the solution should be easily adaptable to meet current needs of the customer (country).
  All VSI products are highly configurable, allowing the customer to change:
  • The data fields in each “object” (student, school, clinic, patient, book, etc.) can be modified to meet current needs without making any changes to the application’s code.
  • The “look and feel” of the application can be changed to meet the customer’s design requirements without changing any application code.
  • The data exchange standard (education, healthcare, agriculture, immigration, etc.) referenced can be changed as necessary without changing application code.
  • The site navigation can be modified for different user roles (admin, student, patient, teacher, etc.) without changing application code.
  • The foundational framework can be used with existing databases without changes to the application code.  These databases can be hosted using a wide range of providers, including Oracle, SQL Server, MySQL, PostGreSQL, Access, DB2, Sybase or many others.
• Applictions should interoperate with other new and pre-existing applications in real-time, securely, and automatically. 
  The foundation software for all our applications is a platform for exchanging information between disparate applications in real-time.  This consists of a high-speed message switch, capable of transferring millions of messages hourly between connected applications, and an adapter that is able to enable existing applications to interoperate with other applications without requiring changes to the original application.
• The solution should, once installed, continue to work for several years without requiring physical maintenance (in most cases).
  We accomplish this in several ways:
  • Application software is hosted in a countrywide “private cloud”.  This allows us to update the software as necessary without issuing individual updates to all remote locations.  The only software required remotely is an Internet browser, which gets automatically updated as changes are made to the browser software.
  • Physically, our system is designed to run “low voltage/low power”, eliminating the parts of a typical solar power system that have the shortest lifespan and produce the most heat.  Defined in the US Military specification MIL-HDBK-217, it is widely accepted that for each reduction in temperature of 10°C, the lifespan of electronic components is doubled.
• The cost of the solution should decrease over time, so that once the project is complete, remaining ongoing costs are affordable by the country.
  There are a few things that factor into this:
  • The application software license cost should reduce to a “maintenance level” by the end of the implementation.  We have structured the cost of the application licenses to be relative to Visual Software’s involvement in the process, reducing to a minimal level after the third year when (typically) the country will be responsible for ongoing costs (special funding may have been available during implementation that runs out once complete).
  • Attention should be paid to the Internet access costs.  Many countries outside of the US still pay for Internet services by GB transferred and do not have fixed-price contracts with unlimited data included.  These costs will need to be paid from operating budgets after implementation, so it is important to keep these costs as low as possible.  To address this we have spent considerable time optimizing our applications so that a minimum of data is transferred.

Optimized for Low Bandwidth

To illustrate this optimization, we performed a number of tests, browsing to existing sites and recording information about how long it took to load the page and the amount of data transferred when the page loaded.

These results vary between the popular web sites (the fastest site doesn’t always equate to the least information being sent to the browser), but the ratio between the popular sites and Veracity are similar (Veracity transmits much less information to the browser).

This may not seem significant, but when multiplied by the number of requests for data sent in a billing cycle by everyone, combined, the differences become far more pronounced.

Why is this important?

Frequently, countries (or other organizations) receive special assistance (grants, aid, etc.) to implement a new system, but that financial help will only be for the implementation phase.  Eventually, the organization will need to pay for the operating costs itself.  In Africa, Internet customers pay for their wireless Internet data by the gigabyte; there are no “unlimited data plans.”  In Africa, as of May 2019, data rates for 10GB of data range from USD $8.37 (Egypt) to $279 (Equatorial Guinea).  Prices continue to go down, but they are not close to the price of data in the US.

Using an average price per 10GB of data, we can compare Veracity against a highly optimized site (Google, blank search box).  Assuming 2 million page requests daily and a low-average cost/GB (in bulk), the following chart illustrates costs accessing various popular Internet sites (for comparison purposes only):

Although data rates will continue to go down, chances of Internet Providers changing the model for charging outside of the US from “per GB” to “Unlimited” are still low.

Scalable Technical Model

All Visual Software products are designed to be scaled to support very large user communities, as evidenced by our ability to support customers such as London, UK, the Australian states of Victoria, New South Wales, Tasmania, South Australia, Queensland and eventually, the entire continent.  Some of our guiding principles in the development of our products is and has been:

• Make sure the architecture is simple and flexible.  The application code itself may be complex in making sure that the transportation of data is secure and very fast, but as these component pieces are finished and tested, they make a complex problem (like transport security) look as if it were a simple task.
• Select the appropriate tools to support a scalable solution.  For example, we only use enterprise-level databases to support the interoperability platform and our vertical applications. These can typically scale across several servers and, for disaster recovery, across several locations.  Examples of Enterprise class database servers are Oracle, Microsoft SQL Server, Sybase and DB2.
• Continually test performance with very large datasets – millions of patients or students, tens of thousands of schools or clinics, hundreds of millions of school attendance or assessment records.  In doing this testing, look for bottlenecks and work to improve the performance of those sections.

Scalable Implementation Model

As we begin working with countries to implement education, healthcare or interoperability projects, we plan each project in phases:

1. Short “proof of concept” either with testers and simulated data or in 1-2 selected schools or clinics.  This is to give the customer a sense of comfort with the functionality of the system.
2. Short “pilot” with 50-100 schools or clinics.  This is to work out all implementation details, such as delivery of equipment (solar panels, laptops, etc.), provisioning accounts, etc..
3. Full implementation.  This will typically run 1-5 years, depending on the number of schools (or medical facilities) being deployed.