Why Ask GGi is a library, not a search engine

22 September 2026

Jaco Marais on how the great collaborative works were made, and what the Ask GGi editorial framework asks of every contribution

Points raised:

- Ask GGi is a curated library, not search

- Collaboration strengthens knowledge without requiring complete consensus

- Editorial judgement turns many contributions into coherent shared knowledge

- Every contribution must meet Ask GGi’s four editorial tests

 

A library is not a heap of books. The difference is not size but coherence: somebody decided what belonged, how it related to everything else, and what the whole thing was for. Ask GGi is a library in that sense. It draws on more than 17 years of GGi's published work, and it is one thing rather than a collection of things.

That distinction has a long history, and it is worth pausing on, because the works that have shaped how societies think were almost none of them written by one person at one desk.

Made by many, read as one

The Hebrew Bible was composed over something close to a millennium, by many hands, and became a single object only through centuries of argument about what belonged inside it. When King James I commissioned an English translation in 1604, the method chosen was deliberately collective: six companies of translators, roughly 47 scholars, each company reviewing the others' drafts before anything was settled. The 1611 result reads as one voice. It was not.

The Mahābhārata, one of the Sanskrit epics of ancient India, is attributed to Vyāsa and runs to something like 100,000 verses, accumulated across centuries. When scholars at the Bhandarkar Oriental Research Institute in Pune set out to establish a critical edition — a project that ran from 1919 to 1966 — they worked from over a thousand manuscripts, there being no single authoritative copy. The poem existed in the relationship between its versions.

The Talmud, the central text of Rabbinic Judaism, goes further and puts the disagreement on the page. The Mishnah was redacted around 200 CE; the Gemara that discusses it was compiled over the following centuries; and the printed page most students use places Rashi's eleventh-century commentary on one side and the Tosafists' on the other, wrapped around a text they are arguing with and with each other about. Dispute is not a flaw in the design. It is the design. The tradition's own formulation of this, from a debate between the schools of Hillel and Shammai, is that "these and those are the words of the living God" — two incompatible readings, both held.

Shakespeare wrote alone rather less often than we imagine: Henry VIII with John Fletcher, parts of Timon of Athens with Thomas Middleton, Pericles with George Wilkins. And the thing we call 'Shakespeare' exists because two colleagues from his company, John Heminges and Henry Condell, gathered 36 plays into the First Folio of 1623, 18 of which had never been printed. Without that act of editorial assembly, half the canon would have gone.

The Oxford English Dictionary is the clearest case of all, because its collaboration was designed. The Philological Society proposed it in 1857. Under James Murray, the Reading Programme recruited volunteers across the English-speaking world to send in quotation slips showing words in use; they arrived in their millions. The first fascicle appeared in 1884, and the first edition was completed in 1928 — 71 years for a work whose whole method was that no one person could possibly read enough.

What these have in common is not committee work. It is that many contributions were made coherent by an editorial function: somebody deciding what belonged, how it related, and what the whole was for.

Why some groups think better

Management research has spent the last three decades arriving at much the same place by a different route.

Amy Edmondson's 1999 study of hospital teams produced the finding that has since become unavoidable: the best teams appeared to make more errors because they were the ones prepared to report them. What distinguished them was psychological safety, which she defined as "a shared belief held by members of a team that the team is safe for interpersonal risk taking". Not comfort, and not agreement — the confidence that speaking up will not cost you.

Mayer, Davis and Schoorman's model of organisational trust, published in 1995, explains what makes that confidence rational. They identify three things people assess in deciding whether to make themselves vulnerable to another party: ability, benevolence and integrity. Competence alone is not enough; nor is good intent. A body of knowledge people will actually rely on has to demonstrate all three — that it knows the subject, that it is trying to help rather than to sell, and that it will say the same thing on Friday as it said on Monday.

Edward de Bono's Six Thinking Hats, from 1985, is the practical companion to both. Its insight is that argument is an inefficient way to think, because it forces people to defend a position rather than examine one. By having a whole group wear the same hat at once — information, then caution, then benefit, then feeling — it separates the modes of thinking that ordinarily collide. Different perspectives stop being obstacles and become an asset.

Enough common ground

The thread running through all of this is that collaboration does not require consensus. It requires enough common ground to make something better together.

That is the honest description of how GGi's knowledge was built. Our contributors have not agreed with each other. Chairs and chief executives, special advisers, academics, regulators and practitioners have written for us from positions that cannot all be simultaneously right, and the resulting body of work is stronger for it. What made it one thing rather than a pile was a shared commitment to public purpose, and an editorial function that decided what belonged.

That function now has a name and a set of rules. We call it the Ask GGi editorial framework, and it holds every contribution to four tests. Does it meet the Nolan Principles, the standards in public life our clients are themselves held to? Does it show mature judgement rather than easy certainty? Is it written in plain English, in language that brings people together instead of setting them against each other? And does it develop the thinking further, in the spirit of the King reports, revised again and again because governance does not stand still? Nothing joins the library without passing all four.

Sharing and caring at a societal level is, in the end, the thing our clients, our contributors and we ourselves have in common. It is why people give their time. It is what the library is made of.

Ask GGi is that library, made askable. You are not searching a heap. You are asking something that was assembled on purpose.

Try it → good-governance.org.uk/ask-ggi

 

References

  • de Bono, E. (1985) Six Thinking Hats. Little, Brown.
  • Edmondson, A. (1999) 'Psychological Safety and Learning Behavior in Work Teams', Administrative Science Quarterly, 44(2), pp. 350–383.
  • Mayer, R.C., Davis, J.H. and Schoorman, F.D. (1995) 'An Integrative Model of Organizational Trust', Academy of Management Review, 20(3), pp. 709–734.
  • The Mahābhārata, critical edition (1919–1966). Bhandarkar Oriental Research Institute, Pune.
  • King V
  • Nolan Principles
  • Shakespeare, W. (1623) Mr. William Shakespeares Comedies, Histories, & Tragedies (the First Folio), ed. J. Heminges and H. Condell.
  • Babylonian Talmud, Eruvin 13b.
  • Oxford English Dictionary, first edition (1884–1928). Oxford University Press.

Meet the author: Jaco Marais

Partner and director of knowledge, brand and voice

Email: jaco.marais@good-governance.org.uk Find out more

Prepared by GGI Development and Research LLP for the Good Governance Institute.

Enquire about this article

Enquire
Contact us