Phoenix 2019 - 8^th to 11^th April, 2019

This session will talk about the MultiValue database and application design — what works, what is optimal, and how it should and shouldn’t be done. Since MultiValue Databases and files allow developers to combine structured and non-structured data, as well as for developers to create their own structures, many developers design their data files differently. Due to this flexible way of storing data, developers do not always design new files and structures optimally. Developers may also have inherited sometimes chaotic designs in an application that has been modified extensively.

Whether you are a new developer or an experienced developer, join us in this session to help understand the why and how. Or just impart your own knowledge and understanding.

If you you are new to working with MultiValue databases and need to access, extract, or share data with the MultiValue platforms in your enterprise, then this session is a must. This session will help developers who are more familiar with other development environments understand the MultiValue database and data structures. If you are a .NET, PHP, or Python developer and you are trying to explain your data requirements and needs to your MultiValue developers and administrators, then join us to learn the common terms and design practices used in the MultiValue applications.

Experienced MultiValue developers will learn how best to describe their MultiValue systems and applications to new developers in order to help them come up to speed with less effort on both you.

It is easy to add an index to a UniVerse file to increase performance, but are you aware of the system performance issues of doing this? Don’t want to wait for a select to run during month-end processing? Create a trigger. Want instantaneous retrieval of sales history based on region, date, dollar amount and phase of the moon? Create a trigger.

This session will talk about the pros and cons of adding indexes to UniVerse files, and different ways to address them.

Those of you who work in the retail environment know all too well the anxiety an upcoming holiday season brings. A lot of companies put all systems on a “frozen” status for some number of weeks or even months prior to the beginning of the Holiday “crunch.” The thought of a system outage during Black Friday, for example, is enough to cause staff members to start losing sleep, developing stress-induced illnesses, and plummeting morale weeks in advance.

But now that everybody knows everything about anything bad that happens (it’s called The Internet), we see that software catastrophes are not limited to the Holiday Season. On any given day you can read about an airline accidentally selling thousands of dollars of First Class tickets to prime locations for a fraction of the regular cost losing thousands of dollars in fares and lost profits due to fuel costs.

So what went wrong? Didn’t they test their software/hardware changes and improvements? Think about this, if you knew the software for the robotic assisted surgery machine had been tested with the same standards your company uses for changes to your order processing system, would you still want that cardiac bypass done using it?

We’ve talked about automated testing, test-driven development, etc. in previous Conferences, and in this one, too. This session will address not just methodologies but also the Human side of testing. How do we write tests – automated and manual – That work. We will also discuss how to start preparing your current legacy code for automated testing. (Hint – don’t throw out a gazillion lines of proprietary legacy code and start over from scratch. Or not even by buying a "package."