Security is applied and practised, not titled, in the real world: Organisational, individual, departmental, government and representative failures.
Security is applied and practised, not titled, in the real world: Organisational, individual, departmental, government and representatives failures

Security is applied and practised, not titled, in the real world: Organisational, individual, departmental, government and representative failures.

Security remains a perpetual, unfinished and adaptive practice, poorly represented by divisive, siloed and role-based behaviour(s).

That is, in order to provide security, it requires holistic consideration of the threats, assets, environments, technology, social factors, organisation/entity and freedom of action or constraints.

Therefore, security can not be solved by one person, discipline or perspective because all hazards require all hands, disciplines, perspectives and considerations.

This division, myopic view and division present gaps and exploitable vulnerabilities and play security actors, protectors, representatives and disciplines off against each other... not the adversary, threat, danger or hazard. As a result, a better approach (read comprehensive, inclusive and excluding nothing) is the provision of security, not a job title, role, responsibility, budget, audit or other factorial aspects involved in the assessment, design, implementation, delivery and maintenance, of a secure or safe environment.

A simple enough premise, it seems overly simplistic but remains the primary Achilles heel and exploitable weakness for human adversaries, criminals, bad actors and organised groups with deliberate, malevolent intent.

In other words, thieves, hackers, bad actors, terrorists and criminals are focused on the prize(s), outcomes, obstacles and means required for success.

Threat actors DO NOT CARE about your org chart, budget allocation/cost code, job title, department, agency, pedigree or where you went to school.

The bad actors benefit from division, poor communication, duplication, inefficiency, rivalry, confusion and other organisational separations.

No alt text provided for this image
Security is applied and practised, not titled, in the real world: Organisational, individual, departmental, government and representatives failures
Organisational structures and resourcing should follow the analysis of the opportunity and threats, not drive the solution(s).

Particularly when it comes to security, as a dominant threat actor, consideration is that of intelligent, adaptive, resourceful and purpose-driven human adversaries that will intentionally circumvent 'controls', security measures and practices.

Safety and Security Convergence Frameworks. Enterprise Security Risk Management. Tony Ridley. Security Science
Not a single org chart to be seen
Security convergence and Business Continuity. Book Review. Tony Ridley MSc CSyP CAS MSyl M.ISRM. Security risk resilience safety and management sciences
Another year, another year of talking about convergence when divergence remains the norm and the preferred state.
You'd be hard-pressed to find a business or asset class that doesn't encompass varying socio-technical factors. Therefore, socio-technical solutions and security practices are required.
Security Management. Convergence. Critical Infrastructure. Tony Ridley MSc CSyP CAS MSyl . Security risk resilience safety and management sciences
Division, duplication and internal rivalry manufacture insecurity and opportunity for threat actors.
Convergence remains challenging, especially when governments, regulators, administrators, boards and executives tend to have a 'get away with' approach to a 'grudge cost' such as security. Make money, make a profit... maybe security is considered later, only after we make a bigger annual profit.

In other words, safety/security first rarely takes pride of place at the top of any business budget, investment or plan.

Convergence. Security Management. Tony Ridley MSc CSyp MSyl Security Risk Resilience Safety and Management sciences
Weeds grow in the cracks... as to vulnerabilities and means of access and exploitation by threat actors.
In sum, if you start your security thinking, planning and approach with 'security of...' as opposed to <label> security, you will find a remarkable change of framing and approach.

That is, you will perceive the problem from all aspects and representative agents, as opposed to attributing artificial and bounded rationalities to complex, wicked, network and nascent threats, risks and vulnerabilities. Try it. Try it with a generalist, and watch how they change their approach to 'security'. Perhaps you and your organisation should too.

Tony Ridley, MSc CSyP FSyI SRMCP

Security, Risk, Resilience, Safety, & Management Sciences

Security Management Risk Management Crisis Management

Risk, Security, Safety, Resilience & Management Sciences

To view or add a comment, sign in

More articles by Tony Ridley, MSc CSyP FSyI SRMCP

Insights from the community

Others also viewed

Explore topics