How this forum works

This forum is part of the library, not a comment section attached to it. The same standard applies here as to anything else published on this site: say where it came from, say what it does not cover, and make it possible for someone else to check.

What a good post looks like

• Name the source. A series, a document, a record identifier — something another reader can open. "I read somewhere" is the start of a question, not of an answer.
• Give the unit and the period. A number without them cannot be compared, and comparison is usually the point.
• Say what you are unsure about. Stating the limit of what you know is a contribution. Concealing it wastes everyone who reads after you.
• Separate the estimate from the measurement. If a figure was modelled, projected or imputed, say so where you use it.

Corrections

If something in this library is wrong, post it in Corrections with the record identifier and what you expected to see. Corrections are treated as contributions rather than complaints, and they are acted on in public: you will be able to see what changed and when. A library that is hard to correct is a library that stays wrong.

Languages

English and French are both first-class here. Post in whichever you think in. Nobody should switch language to be taken seriously, and a reply in the other language is a normal thing rather than a problem.

What gets removed

• Claims presented as fact with no source, after a request for one has gone unanswered.
• Personal attacks. Disagree with the argument, including sharply; leave the person out of it.
• Promotion unconnected to a question being discussed.
• Anything that would identify a private individual who has not chosen to be identified.

Who moderates

The ADG editorial desk. Removals are logged, and a removed post's author is told why. Moderation here is about evidence and conduct — it is not an editorial position on whose conclusions are correct.

What ADG owes you in return

• Sources on every figure the library publishes, including the inconvenient ones.
• Stated limitations on every analytical document.
• Corrections made in public, with the record showing what changed.
• Open licences and machine-readable data, so you can check the work yourself rather than trusting it.