Posts adapted from thirty years of internal editorials, edited for an external
audience. All posts show their original writing date; they were published here
starting in August 2026.
I have always used spaced en-dashes – that's not a sign of AI at work.
- 7 October 2026Bloodlines What tagging given names tells us about company history (and about data modeling)
- 2 August 2026Are we helping book authors and editors? Filling coverage gaps doesn't make a good data product if the underlying model is built for journals
- 2 August 2026Is an editorial an editorial? Model the content, not the words – the central lesson of content modelling, applied to document types
- 2 August 2026Twenty years of Tag by Tag The printed XML documentation that standardised scientific publishing – and what comes after it
- 2 August 2026A fine line between data cleansing and data laundering The principle of faithful exhibit capture – and why correcting the input is the wrong answer
- 2 August 2026Content modelling = modelling the content The central lesson from thirty years of practice – model what the content says, not what the world should look like
- 2 August 2026Profile-ready entities Why technical entity definitions and user-facing profile pages don't always map one-to-one
- 2 August 2026Abstracts are en vogue again Decades of neglect in the secondary database – now amplified by AI
- 2 August 2026Validation by the producer, verification by the consumer A principle learned the hard way in the 1990s – and still worth repeating
- 2 August 2026CP/LD, and a bit of history A new NISO standard prompts a look at thirty years of content standards – and what comes next
- 2 August 2026Acting on errata A broadcast correction from the paper era – and why databases should actually apply it
- 2 August 2026Keep an eye on the prize The arguments used to justify each technology transition are usually things we could already do – the real reasons lie elsewhere
- 2 August 2026How data products lead the way for quality Measuring capability quality instead of data product quality is how we fool ourselves
- 2 August 2026The mother of all metrics How long does it take to find a data error? That number is too low – and it reveals a deeper problem
- 2 August 2026The beauty of minigraphs A worked example showing what goes wrong with a single XML file – and how minigraphs fix it
- 2 August 2026Data mottos worth repeating Some short rules that have guided content and data work for decades – and why ISO standards often violate them
- 2 August 2026Content modernization starts with content, not systems The general mindset is towards systems modernization – but users are waiting for modernized content
- 2 August 2026The two dimensions of content Analytics and presentation impose completely different requirements on the same data
- 2 August 2026Minigraphs A canvas plus annotations – a better way to structure scholarly content
- 2 August 2026Exhibit A How a PhD photograph explains probabilistic knowledge representation