Martyn Richard Jones

Our man in Spain.

A Coruña, 13th December 2025

A Dozen Data Wishes for Christmas 2025 — and a Sharper, Smarter New Year

As 2025 winds down, the tinsel comes out. It’s tempting to imagine a quieter, saner digital ecosystem on the horizon. Call it festive optimism or a data professional’s annual catharsis. Here are twelve wishes for 2026. They are equal parts hard-edged realism, FT Weekend reflection, and Wired-grade futurism.

1. A cleaner information commons.
Here’s to a year when social platforms and consultancy mega-blogs finally stop amplifying nonsense. This will not be achieved with heavy-handed censorship. Instead, intelligent filtration systems will detect hype, hallucination, and outright grift from the usual posers and miracle-cure merchants. A future where credibility wins, not theatrics.

2. Less data whining, more engineering.
Data quality matters, but it can’t remain the universal scapegoat for every failed AI project. In 2026, may we replace the ritual complaint cycle with honest diagnostics and proper design. “Bad data ruined my AI” is the new “the dog ate my homework”, and it’s time we graduated from that.

3. A little humility in quantum and AI punditry.
If you barely understand entanglement or transformer architectures, maybe don’t headline the conversation. The world could use fewer instant experts and more people willing to say, “I don’t know—teach me.”

4. Less hype, more substance in data architecture.
Data lakehouse evangelists and mesh-as-panacea prophets have had their turn at the megaphone. Time to retire the breathless buzzwords and return to first principles: coherence, clarity, and actual performance.

5. A renaissance for data warehousing.
For analytics, statistics, and visualisation, the classic warehouse still solves real problems with elegance. It’s not retro; it’s foundational. In 2026, may organisations remember that durability, governance, and structure aren’t optional.

6. Clearer professional boundaries.
A data engineer is not a Java generalist. A data engineer is not a data scientist. A data engineer is not an analyst. A data engineer is not a statistician. Each discipline is distinct. Let 2026 be the year enterprises finally staff accordingly.

7. A literary break that actually breaks big.
May a sharp-eyed literary agent see the potential in the work of Martyn Jones. He is also known as Martyn de Tours and Martyn Bey. May the agent deliver the kind of deal that shifts culture and bank balances.

8. A real data quality authority.
Not another toothless working group, but a proper body of knowledge, plus accreditation that means something. Companies deserve standards that are auditable, actionable, and enforced, because trustworthy AI requires trustworthy data.

9. Seamless data integration at last.
Silos, legacy sprawl, and multi-cloud entropy currently throttle AI’s potential. The dream: mature fabrics and mesh-hybrids that virtualise access, eliminate redundant ETL drudgery, and enable true zero-copy federation. Real-time, enterprise-wide data without the firefighting.

10. Governance and quality built-in, not bolted on.
In 2026, let observability, lineage, anonymisation, and ethics checks be systemic. Automated. Invisible. A default, not an afterthought. Only then can “trustworthy AI” evolve from marketing slogan to operational reality.

11. Architectures that respect unstructured and multimodal data.
Lakes shouldn’t deteriorate into swamps every time someone uploads PDFs or video logs. We need platforms with vector search, semantic layers, and streaming ingestion as native capabilities. These are the fuel for GenAI without kludges and duct tape.

12. Semantic ontologies and knowledge graphs everywhere.
This is the bridge the enterprise desperately needs: a unifying layer that turns scattered information into coherent knowledge. Ontologies that stitch together structured and unstructured content, reduce hallucinations, and support both human reasoning and agentic intelligence.

---

These twelve wishes aren’t just festive musings, they’re a roadmap for a saner, sharper, more intelligent data future. If 2026 delivers even half of them, we’ll be better off than we were this year.

Thanks for reading, and here’s to a bold, brilliant New Year ahead.