What is the difference between curation and aggregation?

Aggregation gathers content automatically on a mechanical criterion: source, keyword, popularity. Curation adds a human selection and a comment that situates the resource. An RSS feed aggregates; a newsletter explaining why three articles out of thirty are worth your time curates.

Is content curation a copyright problem?

It is lawful as long as it stays within short-quotation limits: a brief excerpt, a clearly identified source, and original commentary that justifies the reuse. Republishing a full article, even with credit, falls outside that. In France, article L122-5 of the Intellectual Property Code sets the boundaries.

Should I curate publicly or just for myself?

The two differ in rigour, not in method. Public curation forces you to state the angle, which mechanically improves the selection. Private curation feeds a personal knowledge base and tolerates rougher notes.

What is content curation?

Content curation is the practice of selecting, contextualising and sharing content produced by other people around a stated editorial angle. It differs from aggregation by adding human judgement: the curator explains why a resource made the cut.

What exactly are we talking about?

Content curation means selecting resources produced by other people, framing them with your own commentary, and publishing them for a defined audience. The word is borrowed from museums: a curator is the person whose job is to choose what gets exhibited and to write the wall label that goes with it.

That origin says exactly where the value sits. Nobody visits a museum to see every painting in storage; they go to see the ones a curator judged worth showing together. Curation without an angle is just tidying.

What separates curation from aggregation?

	Aggregation	Curation
Selection	Automatic, by criterion	Human, by judgement
Volume	High	Low by design
Value added	Gathering	Context and ranking
Rejection rate	None or low	90% and up
Example	RSS feeds, Google News	An edited newsletter

Rejection rate is the sharpest of these signals. A curator who publishes everything they read has not curated. The work sits overwhelmingly in what does not make the cut, and that is also the part the reader never sees.

How to curate: five steps

Define the angle before the sources. One sentence saying who you are writing for and about what. Without an angle, selection degenerates into a pile of things you happened to enjoy.
Build a wide intake. Fifteen to thirty sources followed regularly, favouring primary sources over second-hand commentary.
Filter on usefulness. The right question is not "is this interesting" but "what will my reader do with it". That shift eliminates most candidates.
Write the label. Two to four sentences saying what the resource delivers and what it does not cover. This is the part that separates a curator people follow from an account that posts links.
Credit and link. Author name, original source, direct link. It is both an ethical requirement and the legal condition for short quotation.

What is the legal framework in France?

Curation rests on the short-quotation exception set out in article L122-5 of the French Intellectual Property Code. Three cumulative conditions apply: the quotation must be brief relative to the work quoted, the source and the author's name must be clearly stated, and the quotation must serve a critical, educational or informational purpose — it cannot substitute for the work.

Hyperlinking raises no such difficulty: in the Svensson ruling of 13 February 2014, the Court of Justice of the European Union held that linking to freely accessible content is not an act of communication to the public requiring authorisation.

What are the common failure modes?

Curation without rejection. Publishing twenty links a week transfers the sorting work back to the reader, which cancels the service.

The summary that replaces reading. A curator who paraphrases an article without reading it through propagates misreadings. You can spot it in commentary that only restates the standfirst.

No stated position. A neutral note — "interesting piece on X" — adds nothing. Readers follow a curator for judgement, including when it is critical of the thing being shared.

Curation without memory. Publish, then forget. Curation that does not accumulate in a searchable space repeats the same work every cycle and builds no cumulative expertise.

How does curation feed a personal knowledge base?

Public curation and a personal knowledge base share the same upstream work: read a lot, keep little, understand what you keep. They diverge only at the output. Doing both separately means paying for the selection work twice.

Synthiz sits at that junction. Followed sources — RSS feeds, YouTube channels, podcasts, articles, PDFs — are transcribed and synthesised, then filed into a thematic "memory" where recurring concepts link to one another with sourced quotations attached. The sorted material then serves equally well as raw input for a publication or as a private resource, without doing the triage twice.