You put something into the world.
A post. A presentation. An analysis. A piece of research.
It is public now.
Does that mean it is finished?
Or can the published thing itself remain open to change?
In November 2008, Keith Lyons was writing about a week in the open online course CCK08.
The post, CCK08: Week 10 Wild Flower Garden, moves through a remarkable amount of material: papers on open educational resources, course discussions, other participants' posts, connectivism, networks, teaching, virtual worlds, social media and ideas Keith intended to follow later. He described himself as somewhat removed from the week's live activity, catching up with “the parts I saw read and heard”.
Near the end, something else happens.
Keith added a postscript.
And in that postscript he explained a practice:
“I revisit my posts after first posting to include other posts that come my way from alerts.”
Then:
“Today I lost my postscript!”
The detail is almost comic, but the distinction underneath it matters.
The post had already been posted.
That had not made it untouchable.
New material could arrive later and be incorporated into the public object.
In this post, publication did not end the possibility of addition. Keith described revisiting already-posted material when new items reached him through alerts.
That does not mean every Clyde Street post was endlessly revised.
It does not tell us how often readers noticed changes.
And it does not mean that leaving work open is always preferable to declaring it complete.
What it gives us is one clear encounter with a different possibility:
putting work in public and treating it as finished are not necessarily the same decision.
What exactly remains open?
That distinction is easy to blur.
Most of us continue thinking after we publish something.
We may write another article.
Produce another analysis.
Teach the subject again.
Return to the same question months later.
But continuing to think about a subject is not quite the same as allowing an existing public object to change.
Keith's 2008 post makes the difference unusually visible.
He did not merely say that another idea might influence a future post.
He said that he revisited posts after first posting to incorporate material that subsequently reached him.
Something already public remains capable of alteration.
That raises a practical question for anyone who produces work in public.
When is publication a stopping point?
When is it simply the moment at which other people are first able to encounter something that may still change?
And if the object does remain open, what kinds of change belong inside it rather than in a new piece of work?
Clyde Street does not give us a rule.
A later example makes the problem more concrete.
A presentation that changes before the talk
In October 2016, Keith was preparing a Smart Talk called Cores and Edges in High Performance Sport.
The presentation was already available on Google Slides.
Keith noted that it had received visitors during the two days before he wrote the Clyde Street post.
Then he changed it.
He wrote:
“Today, I have added some data without labels or scales to prompt a discussion about where edges are.”
He also added visual material intended to stimulate thought about locating ourselves in a world of performance, and he added conclusion slides so that people finding the presentation from that morning onward would have the full presentation.
The sequence matters.
The presentation was not being completed privately and then released once everything had settled.
It was already public.
People had already reached it.
And Keith was still altering what they could encounter there.
That makes the presentation a useful second case because the changes are not merely corrections to a typo or formatting repair.
New material was being added around the forthcoming discussion.
But again there is an evidential boundary.
Keith wrote:
“I think this access transforms the opportunities I have in the talk to build upon people's reflections about the content.”
That was Keith's assessment of the opportunity created by advance access.
The post does not show that visitors sent him reflections.
It does not show that audience responses caused the changes he made.
It does not demonstrate that the resulting talk was better.
And it does not establish that an evolving presentation is inherently superior to a fixed one.
What we can see is simpler.
A public presentation already being visited remains under active development before its live use.
Public access had begun before Keith stopped changing the presentation.
Public does not have to mean settled
There is a temptation to turn examples like these into a celebration of unfinished work.
That would go too far.
Some public objects may require stable versions or clear records of change. In other situations, continued alteration may be part of what the object is for.
The Clyde Street cases do not tell us which arrangement is right in every context.
They make the questions harder to ignore.
Publication is often treated as a boundary:
before publication, work can change;
after publication, the work is done.
But Keith's postscript and the evolving Smart Talk show two occasions where that boundary does not hold neatly.
The object is public and still changing.
That raises a more useful question than whether openness or revision is inherently good:
What kind of object am I making, and what does publication commit me to keeping stable?
Another post is not necessarily a revision
A 2020 Clyde Street post helps mark the other side of the distinction.
In Learning from the Final, Keith examined a Big Bash League match between Melbourne Stars and Melbourne Renegades.
The result reversed the outcome of the 2019 BBL Final.
Keith brought the two matches together.
He included his ggplot of the new game and his Excel plot of the 2019 Final. He compared partnerships, wicket losses and the performances of the two teams, and considered what the contrast might reveal about what had happened since the previous final.
The earlier final was being used inside a new 2020 comparison.
This source does not document Keith altering the earlier public analysis itself. Instead, the earlier material becomes part of a new comparative post.
That is important because otherwise almost any continuing public inquiry could be described as revising earlier work.
A person writes one post today and another next week.
An analyst publishes one match report and then another.
A researcher revisits a topic in a new paper.
A teacher changes the subject of next year's class.
Those may all represent continuing inquiry.
They are not automatically revisions of an existing public object.
Learning from the Final therefore gives this trail a useful boundary.
Continuing the inquiry is not necessarily the same as keeping the public object itself open.
Where does the change happen?
That distinction can be easy to miss because, from the maker's perspective, everything may feel connected.
You are still working on the same subject.
Still gathering evidence.
Still changing your mind.
Still producing things.
But from the reader's perspective, there is a practical difference between:
encountering a document that may itself change,
and encountering a sequence of separate documents through which the inquiry continues.
The first raises questions about the status of the object in front of you.
The second creates a trail through several objects.
Neither is automatically better.
They simply organise public inquiry differently.
Keith's 2008 post makes change visible inside an already-published post.
The 2016 presentation remains publicly accessible while material is still being added.
The 2020 cricket post instead creates a new object that places earlier and later material into comparison.
Those sources do not give us a three-part method.
They do not establish a progression from blogging to slides to analytics.
They do not show Keith becoming increasingly committed to one model of open scholarship.
And they do not justify treating unfinishedness as a philosophy.
They give us three cases through which to inspect one problem:
Where is the continuing inquiry located — inside the object already made public, or in what comes next?
Why this matters now
Imagine releasing a piece of analysis to colleagues.
A new result arrives the following morning.
Do you change the shared document?
Add a dated note?
Create a new version?
Leave the original untouched and produce a follow-up?
The answer depends partly on what other people need the object to be.
The same problem appears when a teacher shares course material that continues to develop, when a researcher maintains a public notebook or dataset, or when a practitioner circulates a working document that others may already be using.
The Clyde Street cases do not tell us at what point further change becomes inappropriate.
They do make the status of the public object part of the judgement.
Different public objects may require different expectations about stability, versioning or change.
The Clyde Street cases do not supply a universal rule for those choices.
The useful distinction is not:
finished work good / unfinished work good.
It is:
What have I made public, and what relationship to future change does that object now have?
A public object can have a history
Perhaps that is what the Clyde Street examples make easiest to see.
A public object does not always have to appear in one final form.
It can have a history.
Material can be added.
An argument can remain open to continuation.
A presentation can change while people are already encountering it.
But that history only becomes analytically useful if we keep different things separate.
A revised object is not the same thing as a new object about the same subject.
A willingness to alter something is not evidence that every public object was altered.
Continued inquiry is not proof of deliberate revisability.
And revisability does not establish that the eventual work was better.
Those boundaries matter because otherwise revisability can become a romantic description rather than a useful judgement problem.
The strongest Clyde Street cases suggest something more modest.
Publication can sometimes be an entry point into an object's public life rather than the final moment in its construction.
But whether that is appropriate depends on what the object is doing, what readers need from it and what kind of change is being made.
The useful question is therefore not whether work should always remain open.
It is:
Question to carry
When I put work in public, what am I treating as finished — and what am I still prepared to change?
Sources followed
Keith Lyons, “CCK08: Week 10 Wild Flower Garden” — 08/11/15
Keith Lyons, “Cores and Edges (5)” — 16/10/16
Keith Lyons, “Learning from the Final” — 20/01/07