CitconWiki - User contributions [en]

In the age of the vibe coder, Kata == waste

2025-09-27T17:26:06Z

Gareth Tomlinson: It probably is. Don't force people to improve - provide a framework to want to be better and support those that do by Toyota Kata and targeted coaching. This may include kata, or not.

'aka How to keep newer developers engaged'

"We introduced coding katas as a motivation for return to office"

Are katas pointless now, when we can just ask a LLM to do it.

Q: what motivates people to practice basic skills?

Jeff: People have (in general) never been doing Katas. A select minority do it because they actually want to improve. Another minority do it because they are (socially) rewarded.

Evolutionary advantage in having only a few "experimenters" and a few "laggards", and most of the population just doing what already works

People will behave according to what they think the company wants and what they believe they will be evaluated on, not necessarily what the boss says he values.

Don't expect people to practice in an artificial situation. Allow them, to get better while doing real work. Note that this requires that there is a feedback mechanism.

Recommendation to practice the Toyota kata.

=== Ideas for engagement ===
People have different learning styles. Some people find use in repetition. Some find use in having extra time to think about real life hard problems.

* Ivan ran an international conference on post-modern programming. He ran a "scrap heap challenge" workshop where attendees used whatever means to solve a problem. He was minded to run it again for a new toolset.
* "Do katas in production (code)" with "Don't mock anything you don't own".

=== Valuing improvement ===
Evaluating/ranking people based on "impact".
"Are you an under-performing senior, or are you an over-payed mid-leven engineer?"

[https://www.amazon.co.uk/Drive-Surprising-Truth-About-Motivates/dp/1594484805/ Daniel Pink - Drive]
[https://academics.hamilton.edu/documents/themundanityofexcellence.pdf The Mundanity of Excellence]

People are motivated by autonomy, purpose and mastery.

[https://www.youtube.com/watch?v=OqmdLcyES_Q Talk: "Greatness" David Marquet]

People might not generically "want to be better", but they want to do challenging work.

=== Ways to great a desire to improve ===

Bespoke coaching is high value, high bandwidth, but high cost. See: [https://en.wikipedia.org/wiki/Taguchi_loss_function Taguchi loss function]

People need tough hard problems to solve.

Do your developers have "problems" to solve, or do they just have "tasks"?

Senior/junior pair programming value does not come from the senior "correcting" the junior, but from the junior observing the seniors best practice.

"incremental marginal gains"

When people start pairing, they start spending a lot more time actually coding.

In the age of the vibe coder, Kata == waste

2025-09-27T16:48:05Z

Gareth Tomlinson: Created page with "*aka How to keep newer developers engaged* "We introduced coding katas as a motivation for return to office" Are katas pointless now, when we can just ask a LLM to do it. Q..."

*aka How to keep newer developers engaged*

"We introduced coding katas as a motivation for return to office"

Are katas pointless now, when we can just ask a LLM to do it.

Q: what motivates people to practice basic skills?

Jeff: People have (in general) never been doing Katas. A select minority do it because they actually want to improve. Another minority do it because they are (socially) rewarded.

Ivan ran an international conference on post-modern programming. He ran a "scrap heap challenge" workshop where attendees used whatever means to solve a problem. He was minded to run it again for a new toolset.

Evolutionary advantage in having only a few "experimenters" and a few "laggards", and most of the population just doing what already works

Maybe we should just be doing more interesting things?

"Do katas in production (code)"
"Don't mock anything you don't own".

Different learning styles. Some people find use in repetition. Some find use in having extra time to think about real life hard problems.

People will behave according to what they think the company wants and what they believe they will be evaluated on, not necessarily what the boss says he values.

Evaluating/ranking people based on "impact".
"Are you an under-performing senior, or are you an over-payed mid-leven engineer?"

People are motivated by autonomy, purpose and mastery.

Talk: "Greatness" David Marquet

People might not generically "want to be better", but they want to do challenging work.

Getting back to the original topic: Don't expect people to practice in an artificial situation. Allow them, to get better while doing real work. Note that this requires that there is a feedback mechanism.

Toyota kata

Bespoke coaching is high value, high bandwidth, but high cost.

[Taguchi loss function](https://en.wikipedia.org/wiki/Taguchi_loss_function)

People need tough hard problems to solve.

Do your developers have "problems" to solve, or do they just have "tasks"?

Senior/junior pair programming value does not come from the senior "correcting" the junior, but from the junior observing the seniors best practice.

"incremental marginal gains"

When people start pairing, they start spending a lot more time actually coding.

CITCONEurope2025Sessions

2025-09-27T16:42:02Z

Gareth Tomlinson:

{| class="wikitable"
|-
! Room
! 10:00 - 11:00
! 11:15 - 12:15
! 12:30 - 14:00 (Lunch)
! 14:00 - 15:00
! 15:15 - 16:15
! 16:30 - 17:30
|-
| Presentation Area (6th Floor)
|
|
|
|
|
|
|-
| Space 1 (7th Floor)
|
| [[What is CBT and how it can help you?]]
|
|
|
|
|-
| Space 2 (7th Floor)
| [[HardCoreGit]]
|
|
|
|
|
|-
| Space 3 (7th Floor)
|
|
|
|
|
| [[Code until you die / older female devs]]
|-
| Kaizen Room (7th Floor)
| [[How can we accelerate change adoption in users?]]
| [[Vibe Coding with Glamorous Toolkit]]
|
|
|
| [[In the age of the vibe coder, Kata == waste]]
|}

How can we accelerate change adoption in users?

2025-09-27T12:51:54Z

Gareth Tomlinson: /* What does good look like for operators of a system? (support / sales etc) */

== How can we accelerate change adoption in users? ==

=== Context ===

This discussion was raised in response to a concern about the ever increasing pace of change which we ask users of systems to deal with. Since the 50s people have been exposed to automated systems delivered via computers, with the reach and scope rapidly expanding with the dot-com bubble and advent of web 2.0. At this stage it's almost infeasible to consider life without the small lump of metal, plastic and glass in our pockets which control and gate-keep access to everything from banking to train tickets.

Arguably this change has outpaced the evolution of the species and societies already, and we're in the middle of heaping on a whole new world of exponential change through the powers of generative AI and LLM technology.

What can we do, and what should we do to help users adapt and adopt the change coming really quickly towards them.

=== What are we observing right now ===
* AI is being spread around organisations creating systems like peanut butter spread over everything.
* People are fearful of the new technologies
** Also a fear of "AI coming for our jobs"
* We as early adopters should share our personal experiences to help sooth concerns
* There is an erosion of trust due to AI slop. This could be generated by:
** Bad actors
** Legitimate actors exploiting technology mindlessly
** Perceived or real benefits from Social Media networks (i.e. influencer income)
** Recognition / ego boosts through engagement by others
* There is persistent and pervasive change fatigue
* A great example of fear is the self driving car - it may be demonstrably safer but the perception is any accident is not tollerated
* "You can't mimic emotional intelligence in AI" [ED NOTE: Although I might challenge that notion! ;-) ]
* Use LLMs for inspiration - not trusting their output yet

== Key themes ==
=== Trust ===
* The erosion of trust due to poorly curated content / interactions leads to fear (ie spam / phishing).
* Fear drives people to shutdown and hide as part of a human response
* As thought leaders and creators we are motivated to offer reassurance to help drive success.
* We need to create a place users feel safe
* We need to create a safe space for users to feel safe
* We need to consider security implications in our systems

=== Get good at change to win ===
People get fearful through churn and specifically churn fatigue.
==== What does good look like for system creators ====
* Early adoption
* Willingness to experiment - "feel the fear and do it anyway"
* Seek lessons learned and iterate on them to better
* Continuous iterative improvement
* Insist on fast feedback to stakeholders
* Communicate what we found and what we did about it ("Heard it - did it" table)
* Management of change as a standing agenda item for all meetings

==== What does good look like for operators of a system? (creators / support / sales etc) ====
* 'KEEP CALM AND CARRY ON'
* Help people here for the $ find the next place to get their $ - make changing role easy for the less passionate
* Natural attrition leads to leaner, stronger, more engaged workforce
** Focused on users / UX
** More engaged with the customer need / want
** Consider the parallel with test automation - fewer QA more focused on finding new issues vs people just blindly running through test scripts manually

==== What does good look like for consumers of systems ====
* What's in it for me?
* Pivot fear to trust
** Clear communication
** Trust is currency - push this agenda with the busiess
** Transparency with consumers - be very very clear what the expectations are - don't lie
* Earn trust through feedback
** Build directly into the UX
** Follow up on feedback (See "Heard it - did it" table)

=== Communication of need ===
* The way system creators interact with tech stack is changing
* "Coding" will give way to communication and change management
* 'Q'uality x 'A'doption = 'E'ffectiveness

=== It's on us not to let the implementation details (and perils) of the new tooling leak into the user experience ===
* Doing this well is a competitive advantage
* Change but don't change perceptably

=== Participants ===
(in order of appearance)
[[Gareth Tomlinson]],
[[Kemal Yilmaz]],
[[Melissa Daul]],
[[Antti Pohjonen]]

How can we accelerate change adoption in users?

2025-09-27T12:51:37Z

Gareth Tomlinson: /* What does good look like for system creators */

== How can we accelerate change adoption in users? ==

=== Context ===

This discussion was raised in response to a concern about the ever increasing pace of change which we ask users of systems to deal with. Since the 50s people have been exposed to automated systems delivered via computers, with the reach and scope rapidly expanding with the dot-com bubble and advent of web 2.0. At this stage it's almost infeasible to consider life without the small lump of metal, plastic and glass in our pockets which control and gate-keep access to everything from banking to train tickets.

Arguably this change has outpaced the evolution of the species and societies already, and we're in the middle of heaping on a whole new world of exponential change through the powers of generative AI and LLM technology.

What can we do, and what should we do to help users adapt and adopt the change coming really quickly towards them.

=== What are we observing right now ===
* AI is being spread around organisations creating systems like peanut butter spread over everything.
* People are fearful of the new technologies
** Also a fear of "AI coming for our jobs"
* We as early adopters should share our personal experiences to help sooth concerns
* There is an erosion of trust due to AI slop. This could be generated by:
** Bad actors
** Legitimate actors exploiting technology mindlessly
** Perceived or real benefits from Social Media networks (i.e. influencer income)
** Recognition / ego boosts through engagement by others
* There is persistent and pervasive change fatigue
* A great example of fear is the self driving car - it may be demonstrably safer but the perception is any accident is not tollerated
* "You can't mimic emotional intelligence in AI" [ED NOTE: Although I might challenge that notion! ;-) ]
* Use LLMs for inspiration - not trusting their output yet

== Key themes ==
=== Trust ===
* The erosion of trust due to poorly curated content / interactions leads to fear (ie spam / phishing).
* Fear drives people to shutdown and hide as part of a human response
* As thought leaders and creators we are motivated to offer reassurance to help drive success.
* We need to create a place users feel safe
* We need to create a safe space for users to feel safe
* We need to consider security implications in our systems

=== Get good at change to win ===
People get fearful through churn and specifically churn fatigue.
==== What does good look like for system creators ====
* Early adoption
* Willingness to experiment - "feel the fear and do it anyway"
* Seek lessons learned and iterate on them to better
* Continuous iterative improvement
* Insist on fast feedback to stakeholders
* Communicate what we found and what we did about it ("Heard it - did it" table)
* Management of change as a standing agenda item for all meetings

==== What does good look like for operators of a system? (support / sales etc) ====
* 'KEEP CALM AND CARRY ON'
* Help people here for the $ find the next place to get their $ - make changing role easy for the less passionate
* Natural attrition leads to leaner, stronger, more engaged workforce
** Focused on users / UX
** More engaged with the customer need / want
** Consider the parallel with test automation - fewer QA more focused on finding new issues vs people just blindly running through test scripts manually

==== What does good look like for consumers of systems ====
* What's in it for me?
* Pivot fear to trust
** Clear communication
** Trust is currency - push this agenda with the busiess
** Transparency with consumers - be very very clear what the expectations are - don't lie
* Earn trust through feedback
** Build directly into the UX
** Follow up on feedback (See "Heard it - did it" table)

=== Communication of need ===
* The way system creators interact with tech stack is changing
* "Coding" will give way to communication and change management
* 'Q'uality x 'A'doption = 'E'ffectiveness

=== It's on us not to let the implementation details (and perils) of the new tooling leak into the user experience ===
* Doing this well is a competitive advantage
* Change but don't change perceptably

=== Participants ===
(in order of appearance)
[[Gareth Tomlinson]],
[[Kemal Yilmaz]],
[[Melissa Daul]],
[[Antti Pohjonen]]

How can we accelerate change adoption in users?

2025-09-27T12:51:06Z

Gareth Tomlinson: /* Key themes */

== How can we accelerate change adoption in users? ==

=== Context ===

This discussion was raised in response to a concern about the ever increasing pace of change which we ask users of systems to deal with. Since the 50s people have been exposed to automated systems delivered via computers, with the reach and scope rapidly expanding with the dot-com bubble and advent of web 2.0. At this stage it's almost infeasible to consider life without the small lump of metal, plastic and glass in our pockets which control and gate-keep access to everything from banking to train tickets.

Arguably this change has outpaced the evolution of the species and societies already, and we're in the middle of heaping on a whole new world of exponential change through the powers of generative AI and LLM technology.

What can we do, and what should we do to help users adapt and adopt the change coming really quickly towards them.

=== What are we observing right now ===
* AI is being spread around organisations creating systems like peanut butter spread over everything.
* People are fearful of the new technologies
** Also a fear of "AI coming for our jobs"
* We as early adopters should share our personal experiences to help sooth concerns
* There is an erosion of trust due to AI slop. This could be generated by:
** Bad actors
** Legitimate actors exploiting technology mindlessly
** Perceived or real benefits from Social Media networks (i.e. influencer income)
** Recognition / ego boosts through engagement by others
* There is persistent and pervasive change fatigue
* A great example of fear is the self driving car - it may be demonstrably safer but the perception is any accident is not tollerated
* "You can't mimic emotional intelligence in AI" [ED NOTE: Although I might challenge that notion! ;-) ]
* Use LLMs for inspiration - not trusting their output yet

== Key themes ==
=== Trust ===
* The erosion of trust due to poorly curated content / interactions leads to fear (ie spam / phishing).
* Fear drives people to shutdown and hide as part of a human response
* As thought leaders and creators we are motivated to offer reassurance to help drive success.
* We need to create a place users feel safe
* We need to create a safe space for users to feel safe
* We need to consider security implications in our systems

=== Get good at change to win ===
People get fearful through churn and specifically churn fatigue.
==== What does good look like for system creators ====
* Early adoption
* Willingness to experiment - "feel the fear and do it anyway"
* Seek lessons learned and iterate on them to better
* Continuous iterative improvement
* Insist on fast feedback to stakeholders
* Communicate what we found and what we did about it ("Heard it - did it" table)
* Place management of change as a standing agenda item for all meetings

==== What does good look like for operators of a system? (support / sales etc) ====
* 'KEEP CALM AND CARRY ON'
* Help people here for the $ find the next place to get their $ - make changing role easy for the less passionate
* Natural attrition leads to leaner, stronger, more engaged workforce
** Focused on users / UX
** More engaged with the customer need / want
** Consider the parallel with test automation - fewer QA more focused on finding new issues vs people just blindly running through test scripts manually

==== What does good look like for consumers of systems ====
* What's in it for me?
* Pivot fear to trust
** Clear communication
** Trust is currency - push this agenda with the busiess
** Transparency with consumers - be very very clear what the expectations are - don't lie
* Earn trust through feedback
** Build directly into the UX
** Follow up on feedback (See "Heard it - did it" table)

=== Communication of need ===
* The way system creators interact with tech stack is changing
* "Coding" will give way to communication and change management
* 'Q'uality x 'A'doption = 'E'ffectiveness

=== It's on us not to let the implementation details (and perils) of the new tooling leak into the user experience ===
* Doing this well is a competitive advantage
* Change but don't change perceptably

=== Participants ===
(in order of appearance)
[[Gareth Tomlinson]],
[[Kemal Yilmaz]],
[[Melissa Daul]],
[[Antti Pohjonen]]

How can we accelerate change adoption in users?

2025-09-27T12:49:45Z

Gareth Tomlinson: Build trust and get good at change management to win. Keep calm and carry on, don't let the disruption of change leak into the UX. Be transparent about the expectations a user can have and give them a way to feedback easily and for them to see actions on their feedback.

== How can we accelerate change adoption in users? ==

=== Context ===

This discussion was raised in response to a concern about the ever increasing pace of change which we ask users of systems to deal with. Since the 50s people have been exposed to automated systems delivered via computers, with the reach and scope rapidly expanding with the dot-com bubble and advent of web 2.0. At this stage it's almost infeasible to consider life without the small lump of metal, plastic and glass in our pockets which control and gate-keep access to everything from banking to train tickets.

Arguably this change has outpaced the evolution of the species and societies already, and we're in the middle of heaping on a whole new world of exponential change through the powers of generative AI and LLM technology.

What can we do, and what should we do to help users adapt and adopt the change coming really quickly towards them.

=== What are we observing right now ===
* AI is being spread around organisations creating systems like peanut butter spread over everything.
* People are fearful of the new technologies
** Also a fear of "AI coming for our jobs"
* We as early adopters should share our personal experiences to help sooth concerns
* There is an erosion of trust due to AI slop. This could be generated by:
** Bad actors
** Legitimate actors exploiting technology mindlessly
** Perceived or real benefits from Social Media networks (i.e. influencer income)
** Recognition / ego boosts through engagement by others
* There is persistent and pervasive change fatigue
* A great example of fear is the self driving car - it may be demonstrably safer but the perception is any accident is not tollerated
* "You can't mimic emotional intelligence in AI" [ED NOTE: Although I might challenge that notion! ;-) ]
* Use LLMs for inspiration - not trusting their output yet

== Key themes ==
=== Trust ===
* The erosion of trust due to poorly curated content / interactions leads to fear (ie spam / phishing).
* Fear drives people to shutdown and hide as part of a human response
* As thought leaders and creators we are motivated to offer reassurance to help drive success.
* We need to create a place users feel safe
* We need to create a safe space for users to feel safe
* We need to consider security implications in our systems

=== Get good at change to win ===
People get fearful through churn and specifically churn fatigue.
===== What does good look like for system creators ====
* Early adoption
* Willingness to experiment - "feel the fear and do it anyway"
* Seek lessons learned and iterate on them to better
* Continuous iterative improvement
* Insist on fast feedback to stakeholders
* Communicate what we found and what we did about it ("Heard it - did it" table)
* Place management of change as a standing agenda item for all meetings

==== What does good look like for operators of a system? (support / sales etc) ====
* 'KEEP CALM AND CARRY ON'
* Help people here for the $ find the next place to get their $ - make changing role easy for the less passionate
* Natural attrition leads to leaner, stronger, more engaged workforce
** Focused on users / UX
** More engaged with the customer need / want
** Consider the parallel with test automation - fewer QA more focused on finding new issues vs people just blindly running through test scripts manually

==== What does good look like for consumers of systems ====
* What's in it for me?
* Pivot fear to trust
** Clear communication
** Trust is currency - push this agenda with the busiess
** Transparency with consumers - be very very clear what the expectations are - don't lie
* Earn trust through feedback
** Build directly into the UX
** Follow up on feedback (See "Heard it - did it" table)

=== Communication of need ===
* The way system creators interact with tech stack is changing
* "Coding" will give way to communication and change management
* 'Q'uality x 'A'doption = 'E'ffectiveness

=== It's on us not to let the implementation details (and perils) of the new tooling leak into the user experience ===
* Doing this well is a competitive advantage
* Change but don't change perceptably

=== Participants ===
(in order of appearance)
[[Gareth Tomlinson]],
[[Kemal Yilmaz]],
[[Melissa Daul]],
[[Antti Pohjonen]]

CITCONEurope2025Sessions

2025-09-27T12:17:19Z

Gareth Tomlinson:

{| class="wikitable"
|-
! Room
! 10:00 - 11:00
! 11:15 - 12:15
! 12:30 - 14:00 (Lunch)
! 14:00 - 15:00
! 15:15 - 16:15
! 16:30 - 17:30
|-
| Presentation Area (6th Floor)
|
|
|
|
|
|
|-
| Space 1 (7th Floor)
|
| [[What is CBT and how it can help you?]]
|
|
|
|
|-
| Space 2 (7th Floor)
| [[HardCoreGit]]
|
|
|
|
|
|-
| Space 3 (7th Floor)
|
|
|
|
|
|
|-
| Kaizen Room (7th Floor)
| [[How can we accelerate change adoption in users?]]
| [[Vibe Coding with Glamorous Toolkit]]
|
|
|
|
|}

Gareth Tomlinson

2025-09-25T05:47:39Z

Gareth Tomlinson: Created page with "A relative newcomer to the joy and liberation which all XP practices bring. [https://uk.linkedin.com/in/gareth-tomlinson LinkedIn]"

A relative newcomer to the joy and liberation which all XP practices bring.

[https://uk.linkedin.com/in/gareth-tomlinson LinkedIn]