Sometimes the correct answer is never to transfer
The structure assumes handover is the goal. Where the receiving organisation has no path to the capability, forcing a transfer produces an orphan with an owner's name on it.
5 min read
Singular State / Perspectives
Perspectives
Practical thinking on what to build, how to make it work, and what it takes to own it.
Stabilization: reducing immediate operational risk
Stabilization addresses a named weakness in current operation. Isolation, recovery, and dependency repair reduce different risks and leave different constraints.
4 min read
Synthetic, masked, sampled, production: what each data strategy costs you in evidence
Not four rungs of one ladder. Three are construction choices that overlap and one is an operating context, and synthetic does not mean anonymous.
5 min read
The application has moved, but the data center cannot close
How shared infrastructure and recovery dependencies can keep a source facility in the operating path.
3 min read
The evaluation harness is the deliverable
The system can be rebuilt; the encoded definition of working cannot. Ship the harness first, and test it by making a deliberately wrong answer fail.
4 min read
The full cost of retaining or replacing an application
Comparing whole-life costs against a consistent baseline, including coexistence and the conditions for realizing savings.
3 min read
The pilot as a sales tactic
A cheap pilot is a qualification instrument for the vendor. The diagnostic is what evidence and permitted artifacts remain if the buyer purchases nothing further.
5 min read
The receiving team is the deliverable
Invert the frame: the software is the byproduct and the capability is the product. Every practice that eases transfer is a training programme with an artifact attached.
5 min read
The run-cost handoff nobody scheduled
A transferred system arrives with a bill and no budget line, because the fiscal year that funded the build did not fund the operation. Six things move on six different dates.
5 min read
The separate support lifecycles inside one application
A deployment combines product, runtime, development-tool, operating-system, and component support commitments. A single supported label cannot describe them all.
4 min read
Third-party risk management is the gate your prototype actually has to pass
A business arrangement can exist with no contract and no payment, so the free pilot is already inside scope. And some obligations survive the relationship ending.
5 min read
Throwaway or evolutionary: decide before the first commit
Whether prototype code survives is a decision per component, made in advance, and separate from whether the learning survives. Neither always-rewrite nor demo-implies-retain holds.
5 min read
Transition assistance clauses: the exit you negotiate at the start
Termination for convenience is a settlement mechanism, not a free exit. And staff are participants with rights, not artifacts that transfer with the repository.
4 min read
TSB’s 2018 migration: data transfer and service acceptance
What the documented TSB case establishes about the boundary between migrated data and working banking services.
4 min read
Virtualizing equipment-control software and preserving physical behavior
Distinguishing a recoverable virtual machine from validated behavior at the equipment boundary.
3 min read
Walking skeleton, tracer bullet, strangler fig
Three patterns that build a thin end-to-end path before breadth. Two of them overlap and the third is a different kind of thing, because it has to keep an incumbent running alongside.
5 min read
What an unattended RPA bot depends on
The identities, sessions, selectors, and recovery rules behind unattended desktop automation.
3 min read
What automated source conversion actually changes
Separating syntax transformation, runtime support, behavioral equivalence, and future maintenance.
3 min read
What belongs in an application inventory?
An application inventory records deployed systems, their business roles, dependencies, and the evidence behind each attribute. Unknown values remain distinct from observed facts.
4 min read
What makes a system legacy?
Legacy status describes a constraint on continued operation and change. Age, support status, and treatment choice are separate assessments.
4 min read