Data integrity is the #1 focus in this step. Lost data is absolutely inexcusable. Every project stakeholder and production resource needs to ensure that they’re focused on a quality approach to data access and management.
An analogy would be the postal service. There’s standard mail; it requires a stamp, and you only know that it was delivered when the recipient tells you that it was delivered. If the mail didn’t get there, there’s really no way of knowing where that piece of mail is, what happened to it, and if you’re ever going to get it back. Sometimes the mail is a little bit more important and you need to verify that the mail was delivered. In these cases, we have opportunities to verify the delivery by using certified mail, FedEx, UPS, etc.
In mobile control applications, we need to make sure that we’re thinking of our user data being delivered to our information systems in the same way that we think of our mail. It’s important to ensure that our user’s data is not lost in transit. For this, there are mechanisms that ensure that it gets delivered. These include response based transits, API level authentications, and other low level socket communication protocols. These ensure that we never lose data.
Automation and control systems often produce and use data differently than other application types. Many times, there is risk of information overload.For example, If we are using a text-message based delivery of information for controls that are delivering information 5 times a second, that’s going to be too much information for people that have these devices in their pocket. It’d be pretty overwhelming to these users, and not exactly a very good use case.
Some controls measure data information thousands of times a second, and it would literally be impossible to send all this data to mobile or tablet devices as it comes in. In these cases, we need to develop business logic algorithms which report only the changes in data that would require user interaction.
If we have a range for the type of data that outputs to the control device, we only report then that data comes out of this particular standard deviation. As you can see, data access and management is a critical component to understand before deciding “how to build” the application. Whether straight forward APIs are used to access data, or a technique is needed to directly access specific hardware, this step will often drive application architecture and the other steps that follow.
