“To the man who only has a hammer, everything he encounters begins to look like a nail.”
In the case of OSS, our hammer is the computer monitor, keyboard and mouse, which are the weapons of choice for interacting with the OSS tools.
But are they the best I/O (Input / Output) devices for performing each of the different CSP operation tasks? A data entry operator will have vastly different needs than a NOC operator or a contracts (ie SLA) manager or a field service technician or an OSS application administrator, etc. Have OSS vendors considered custom I/O tools for the needs of each type of operator?
For example, any old OSS tool can register an alarm, show it in an alarm list, allow operators to acknowledge an alarm and create a ticket. In this case, they are designed to handle the most common tasks.
But where an OSS can really prove itself to be remarkable is how it assists an operator handle the outlier, such as when an alarm storm hammers the network.
“Noise” removal, information processing, understanding of alarm / network relationships, and root cause identification are the key functionality when handling these (hopefully) rare events. Are keyboard and mouse the best devices to assist with the speed of processing available information, recall of historical events and ultimately fault-prognosis? Are these tools even tapping into the best of our senses for triage and decision-making?
Would touch and gesture computing allow faster prognosis? Would collaborative gesture computing speed prognosis even further? What would the interface look like?
Like the factory foreperson of old, could they listen to the machines and be able to identify harmonics that meant something was about to fail (eg a bearing was starting to rattle)? Could we design an audible sense of the network to generate harmonics that indicate when different parts of the network are within expected thresholds, but different harmonics when nearing or exceeding thresholds? Could these sounds give us predictive capabilities? If we have surround-sound, could we be spatially aware about where anomalies are in the network?
Has anyone heard of a behavioural analyst or an industrial designer being used to create the GUI and I/O interfaces for an OSS product? If not, why not? I know I haven’t.