S3Model is a solution to a foundational problem in managing the context of data to produce information. The problems lie in the ability to communicate the full meaning of data, as information between applications. Regardless of the function of that application, as well as the physical and temporal distance from the original data collection.
According to McKinsey the Internet of Things market will be worth $11T by 2025. That is, with semantic interoperability. Without semantic interoperability it could be worth as much as 60% less.
That is just the IoT field. But it puts a dollar figure on the importance across all industry segments. Especially in cross-domain interoperability settings.
Bad data can kill.
In most cases, it just means that poor or incorrect decisions are made that do not kill people. However, what causes and what defines poor quality data in today’s electronic information-rich world?
During the height of application-centric, SQL database usage across enterprises data errors was being reported as high as 30% and broader organizational databases a study found 50% - 80% of criminal records in the US were erroneous. This period was at a time we could easily measure data quality. In 2017, the trend is to dump everything into a data lake and sort it out later for decision making. We expect data quality has gotten worse instead of better.
Some reasons for poor data quality:
Excessive amounts of data are collected leading to less time to do it, and “shortcuts” to finish reporting.
Many manual steps moving figures, summing up, and other manual operations, between different paper forms.
Unclear definitions and wrong interpretation of the fields to be filled out.
Lack of use of information no incentive to improve quality.
Fragmentation of information systems can lead to duplication of reporting.
Missing or poorly thought out or poorly implemented constraints at data entry time.
Weak or missing validation in the application and when transferring data between systems.
When determining that certain data should be captured for later re-use there are specific considerations that should be made. From the very small “what is the datatype of the data” to the very broad “what does this data mean in this context”? As well as why do we want to capture it now? What are the rules or guidelines in place that govern what this data means? Is this data captured at specific locations only? Is there a unit of measure from a possible set of units? As well as many other probing questions.
When these questions are answered, the data becomes information. This information is what humans use to make decisions. Simple data points are not very useful without the surrounding context. Since we build computers to emulate our decision-making process, then we must be able to encode this context in a way that the computer can interpret and process.
It is impractical to encode this context in every existing and future application that might need to process this information. S3Model provides the capability to record and share this contextual information using standards-based technology.