What Clyde Street can teach us about inquiry

Across 2,051 indexed Clyde Street records, recurring patterns were not hard to find.

What the evidence would not support was one Keith Lyons method.

That left a harder question: were the recurring patterns fragments of one method, or different practices that only sometimes travelled together?

A small example shows why that distinction matters.

In February 2015 Keith was following batting partnerships in the Cricket World Cup. After the opening games, a century partnership and a total of 300 runs had accompanied victory. Then Ireland chased down the West Indies’ 304. Keith did not simply add the fifth match to the series. He changed what he was willing to say: the combination might have been necessary in the earlier games, but it was not sufficient to guarantee success.

The interesting thing is not that Keith was always willing to change his mind. The corpus does not support that kind of personality claim. What it does support is more observable: in some inquiries, evidence, challenge, people or changed conditions produced a visible change in the work.

Across the corpus, the answer was not one method but a repertoire of recurring intellectual practices, with boundaries that stopped superficially similar things being called the same.

This article asks what another person might learn from those observable practices.

These are not eleven rules for thinking. They are not a competency framework, and the study did not test whether teaching them improves performance. They are transferable in a more modest sense: each one can be turned into a judgement a reader can inspect, try and challenge in their own work. The practices can also deepen: the practice stays; the judgement becomes harder as the stakes and consequences rise.

The simplest way to read them is to keep two questions together:

What is the practice?

And what looks similar without actually being the same thing?

1. Show how you got there

In a 2014 post about the Netherlands–Costa Rica penalty shoot-out, Keith did more than report that Tim Krul had been substituted into goal and that the Netherlands won. He placed several kinds of material next to one another: the coach’s explanation, Simon Gleave’s interpretation, the relevant Laws of the Game, his own viewing of the broadcast, Krul’s account of his preparation and Keith’s record of where the penalties had gone in an earlier match.

That does not make the account beyond dispute. It does something more useful: it gives another person enough of the route to see how the account was assembled and where they might disagree.

The recurring practice across Clyde Street was not simply “include data” or “add links”. It was making enough of the construction, evidence handling or comparison visible for the inquiry to become inspectable.

The boundary matters: a data table, chart or finished display is not automatically an inspectable inquiry. A reader may be able to see the result while still being unable to see how selections, transformations or comparisons produced it.

Could another person see enough of how I selected, transformed and compared the evidence to challenge me intelligently?

2. Let evidence or challenge change the inquiry when it warrants it

The Ireland–West Indies example is useful because the change is small and precise. Keith did not abandon the inquiry. He weakened the claim.

Elsewhere, criticism changed a visualisation or introduced a new monitoring variable. Across the strong cases, what changed was not always the same thing. Sometimes it was claim strength. Sometimes the representation. Sometimes the question, method, evidential use or practical direction.

That variation is important. “I learned something” is often too vague to tell us what happened intellectually.

The boundary is equally useful: new information is not the same as a changed inquiry. A new result can simply be added to an existing series. Feedback can be acknowledged without altering the work. A criticism can be heard and rejected. Change counts only when something material in the inquiry actually changes.

What, specifically, changed because of what I encountered?

A second question follows:

Did the evidence change the claim, the question, the method, the representation, the action, the public object — or nothing yet?

3. Repair the word or representation when a mismatch is demonstrated

In January 2017 Keith changed a post title after Neil Lanham objected that “artefacts” misrepresented materials he understood as organic and alive. Keith replaced it with “treasure”.

Elsewhere, a Brumbies–Kings draw exposed an outcome Keith’s existing win/loss notation could not represent, so he changed the notation.

The fuller Clyde Street journey appears in “When a word stops fitting”. Here the transferable point is narrower: sometimes the thing being described exposes a mismatch in the words or categories being used.

But the boundary protects the practice from becoming “keep inventing new terminology”. Novel wording is not automatically a repair. Imagining a better phrase is not the same as changing the representation. Broad conceptual change is not automatically language repair.

What has actually shown me that this term, category or representation no longer fits?

The useful habit is not constant renaming. It is learning to recognise when the existing description is forcing the case into a shape it no longer supports.

4. Use experience and position to locate the inquiry, not settle it

In a #UCSIA15 post, Keith used seven years living in rural Braidwood — its community, distance and limited broadband — to think about how an open course might work.

The experience located the inquiry; it did not settle the design. The fuller Clyde Street treatment appears in “When does your own position become part of the analysis?”

Across Clyde Street, biography, professional history, place and lived experience sometimes did analytical work in this way. They helped shape what Keith noticed, compared, questioned or bounded.

The boundary is straightforward but important: autobiography is not automatically analysis. A memory, personal connection or first-person passage can be present without materially changing the inquiry.

What is my own position helping me notice here — and what would I still need to know from somewhere else?

That question travels well beyond research. Analysts, teachers, coaches and practitioners often know a setting from the inside. The useful discipline is to treat that proximity as a resource for noticing, not as privileged access to the answer.

5. Move inquiry into realised practice before claiming practice changed

In November 2014, three months before #UCSIA15 began, Keith described work already being done to build the course. A welcome video had been produced. An open Moodle forum was being prepared. Students would use ePortfolios, Keith had begun constructing his own Mahara example, course documentation was being organised in Google Docs and the evolving course content was being mapped.

Some of the language remained prospective — Keith was still hopeful about what the course might become — but the infrastructure was no longer only an intention. Things had been made.

That distinction recurs across the corpus. Inquiry sometimes moved into situated practice, but the study repeatedly refused to treat a plan, aspiration, training event or proposed action as if the consequence had already occurred.

Intention is not realised practice.

What can I point to that I actually did differently?

There is a second boundary too. Action does not prove that the preceding inquiry was correct. A realised change tells us that something happened. It does not, by itself, establish effectiveness.

6. Follow people for where they lead the inquiry, not just for what they said

In September 2013 Keith received a Twitter alert from Darrell Cobner. Darrell was sharing Harold Jarche’s writing about narrating work. Keith followed that lead into questions of narration and narrative form, revisited dozens of his own posts carrying a narrative tag, returned to Donald Polkinghorne’s work, and then began considering tools that might support public narration and peer acknowledgement.

The intellectual significance is in the route.

Darrell’s alert did not merely become another citation. It changed where Keith looked next.

That kind of person-led traversal recurs in Clyde Street. People sometimes acted as directional resources into new sources, questions and comparisons. But ownership matters: Jarche’s ideas remained Jarche’s; Polkinghorne’s remained Polkinghorne’s; Keith owned what he subsequently did with them.

The boundary is easy to overlook in networked writing. A long list of names, links or conference encounters is not automatically evidence that another person redirected the inquiry.

After this encounter, what am I now looking at that I would not otherwise have examined?

And alongside it:

Which part of this reasoning belongs to me, and which part arrived from someone else?

7. Keep some public work open to revision when the object and context warrant it

Clyde Street sometimes shows publication and completion separating.

In one early post Keith revised material already public as new items arrived; later, public access to a presentation began before he stopped changing it.

The key distinction is whether the existing public object remains open to alteration or the inquiry continues in something new. The fuller journey appears in “When does publication not mean the work is finished?”

Another post or instalment is not automatically revision of the existing public object.

Is the continuing inquiry located inside the object already made public, or in what comes next?

The transferable lesson is not that everything should remain permanently editable. Stability, versioning and revisability are judgement questions. Readers are entitled to know what they can treat as settled enough to use and what remains open to change.

8. Make limits, absence or incompleteness change what you are willing to claim

Clyde Street contains plenty of cautious language, but caution alone was not enough for the study to count a limit as intellectually consequential.

The stronger cases are those in which uncertainty or incompleteness changes the permissible conclusion. In posts about review technology, for example, imperfection, trade-offs and inconclusive evidence are retained as constraints on what the technology can decide. Elsewhere, Keith explicitly marked what an indicator did not capture rather than allowing the indicator to stand for the whole phenomenon.

This is more demanding than adding “there are limitations” near the end of an account.

Generic caution is not a claim-constraining limit.

If this limitation disappeared, would I be willing to claim something stronger?

If the answer is no, the caveat may not really be doing any work.

A second useful question is:

What evidence would allow me to strengthen the claim — and what would remain unresolved even then?

9. Borrow from distant domains only when the borrowed idea does work on the problem

Clyde Street repeatedly crosses domains. Keith brought material from education, sociology, safety, art, technology, history and elsewhere into questions arising in sport and learning.

The strong cases are not strong because the connection is surprising. They are strong because the imported material changes what becomes visible.

One example uses the design of airline safety videos as a resource for thinking about pedagogy. Another brings climate prediction and art forensics into performance-data work. Elsewhere, ideas about place helped Keith think about sport facilities and learning.

The boundary is crucial because cross-domain writing can become decorative very easily.

Analogy is not evidence. Thematic resemblance is not necessarily analytical transfer. An unusual connection is not valuable merely because nobody expected it.

What does this borrowed idea let me distinguish, question or see that I could not see as clearly before?

If the answer is only “it reminds me of something”, the connection may still be interesting. It has not necessarily earned analytical weight.

10. Make what you create usable, not merely available

In the build-up to #UCSIA15, Keith assembled videos specifically to stimulate discussion about informatics and analytics. The point was not only that participants could access the clips. The material was organised around something they were expected to do with it: interpret, discuss and continue the inquiry.

Across other cases, Clyde Street shows participation routes, facilitation structures, definitions, procedures and contextual material being assembled so another person might act with the resource rather than merely receive it.

That leads to two boundaries.

Availability is not usability.

And designed usefulness is not the same as actual uptake or effectiveness.

A presentation can be online and still leave the reader unsure what to do next. A resource can be carefully designed for onward use without any evidence that people used it. People can use it without evidence that it helped.

What can another person actually do next because of how I have organised this?

Then ask:

Which claim do I have evidence for — I designed it for use, somebody used it, or it helped?

Keeping those claims separate is one of the most practically useful distinctions in the whole study.

11. Return to earlier work when changed conditions give you a reason to do new work on it

In June 2014 Keith compared goal scoring in the 2010 and 2014 FIFA World Cups. He did not simply reproduce the 2010 material. He wrote that he had reworked the earlier data, changing the earlier quarter-of-game treatment into 15-minute blocks so the two tournaments could be compared.

Other strong cases return to earlier presentations, analyses or practices under changed evidence, technological or educational conditions and perform a materially new operation on them.

The boundary keeps this from becoming a celebration of repetition.

Reuse is not the same as reworking. A previous analysis can be used as a baseline without being materially altered. A post can be resurfaced, linked or republished without the earlier intellectual object changing.

What has changed now, and what new work does that change require me to do on the earlier object?

Sometimes the answer will be “none”. Changed circumstances alone do not oblige us to rewrite the past.

The same practices can deepen

These are not undergraduate, postgraduate and professional tiers.

The same practice becomes more demanding as the consequences of the judgement increase.

Showing your working can begin with naming sources and explaining steps. In consequential work it may require exposing transformations, ownership, uncertainty and rival interpretations so that others can challenge the judgement before acting on it.

The practice stays. The judgement becomes harder.

Four harder lessons from looking across the corpus

The eleven practices are recurrent. The next four lessons have a different status. They come from the way the whole evidence base behaved, including failures, countercases and relationships that the study refused to turn into stronger claims.

1. Let failure remain failure

An early 2009 plan for a weekly public synthesis was disrupted. Keith later noted that he had managed only two editions and that, despite his intentions, his divergent reading had sent him elsewhere.

It would be easy to rewrite that episode as a polished story about adaptation and resilience. The source record does not require that story.

A useful question is simpler:

What did not work here, and have I allowed the record to say so?

Failure does not have to become a virtue in order to remain informative.

2. Conditions matter without automatically becoming the explanation

Technology changes what is possible. Platforms disappear. New evidence becomes available. Professional environments shift.

The corpus contains cases where changed technology enabled a genuinely new operation on earlier work, and others where a platform change led mainly to preservation or migration. The condition mattered in both. The intellectual consequence was different.

A useful question:

Is the changed environment the thing that altered the thinking, or the condition under which a separate change became possible?

This matters especially when new technology is tempting us to explain everything around it.

3. Co-occurrence is not a method

Some Clyde Street episodes contain several of the practices in this article at once. One case may involve an encounter with another person, a changed inquiry, a revised public object and a practical consequence.

That can make a compelling narrative.

It does not establish a universal sequence.

The whole-corpus study repeatedly found combinations that were unique, unresolved or too sparse to support a recurrent relationship. The evidence therefore resists turning a good cluster into “Keith’s method”.

A useful question:

Am I turning a compelling combination into a process simply because the combination makes a good story?

4. Visibility is not development

Later digital work often leaves richer traces. Methods, data, links and revisions may be easier to recover. Earlier work can be thinner, harder to locate or preserved under different conditions.

That creates a methodological trap: what becomes easier to see can look as though it became more sophisticated.

The study did not support that developmental conclusion.

A useful question:

Am I comparing intellectual change, or merely a change in what the record allows me to see?

What actually changed?

One question survived almost every part of this work:

Question to carry

What actually changed?

Not: did I read something new?

Not: did somebody disagree with me?

Not: did I publish again?

Not: did a new technology arrive?

What changed in the inquiry?

The claim?

The question?

The method?

The representation?

The action?

The public object?

The earlier work?

Or nothing yet?

That question does not provide a method. It does something better. It slows down the move from an interesting event to an inflated conclusion.

The whole-corpus study of Clyde Street did not reveal a single intellectual recipe. It revealed a repertoire: ways of exposing inquiry, letting it change, repairing representations, using experience carefully, following people and distant ideas, preserving limits, carrying thinking into practice and usefulness, keeping some public work revisable and returning to earlier work when conditions genuinely warranted it.

Just as importantly, it revealed the borders around those practices.

A chart is not automatically an inspectable inquiry.

An exception does not automatically destroy a pattern.

Autobiography is not automatically analytical use of position.

A name, link or citation is not automatically evidence of people-led inquiry.

Availability is not usability.

Action is not effectiveness.

Repetition is not revision.

Caution is not automatically constraint.

Those distinctions are part of what makes the practices transferable. They give us something more useful than slogans: questions with which to examine our own work.

Keith’s work is the case. The judgement about what travels from it belongs to the reader.