Notes on data, AI, IT
and security
No marketing fog. The way I think about real problems with founders and managers.
Public cloud SLA: what it says and what it does not
A breakdown of where the provider's responsibility ends and the customer's begins - and why this matters before an incident, not after.
Warehouse robots: the count should go beyond FTE to flow predictability
The economics of warehouse automation are not just about replacing headcount. The real gain is SLA, traceability, and operational stability.
Logs as a data source, not garbage: what you can see before you have a SIEM
How to treat logs as operational material - for diagnostics, audit, and analytics - even without a specialised platform.
Self-service BI: how not to turn reporting freedom into a contradiction factory
Self-service analytics only works with a shared metrics vocabulary and data trust - otherwise every department arrives at the meeting with its own version of the truth.
NoSQL without the hype: what it solves and what it only complicates
A framework for choosing NoSQL through three real dimensions: consistency, load, and cost of maintenance.
MDM versus the ERP zoo and local spreadsheets
Why most BI project failures are actually master data failures - not analytics failures and not tool failures.
Backup after virtualisation: why old schemes no longer hold
Having backup copies and being able to restore from them are two different things. Virtualisation changes both, and old backup approaches often create the illusion of protection where there is none.
Virtualisation is not the cloud: where the practical boundary lies
A hypervisor makes servers more flexible, but it does not make infrastructure service-oriented. The difference between virtualisation and cloud, and why it matters for business decisions.
Data quality before analytics: why dirty master data breaks any BI
Dashboards and BI tools produce answers that are exactly as good as the data underneath them. Until reference data and master records are in order, visualisation only makes the disorder look convincing.
Hadoop in business terms: when a cluster makes sense and when a DWH is enough
Hadoop does not replace a data warehouse and is not the answer to every large-volume question. A breakdown of which tasks justify a cluster and which ones do not.
Open data and APIs: why your company should think like a platform now
An API is not a trendy interface and not a way to look technical. It is an integration discipline that determines how easily your systems will work together a year from now.
After Stuxnet: ICS segmentation is no longer optional
How to ground industrial control system security in real assets, contractors, maintenance windows, and a minimum isolation baseline.