iiinova — where your digital self keeps evolving

Henry Garner @hendroid.io · 25/06/2025

Ben Shneiderman, author of the book 'Human-Centered AI', argues that the choice between human control and computer automation is a false dichotomy. They are actually orthogonal dimensions creating four distinct regions.

A chart showing “computer automation“ on the x axis and “human control” on the y axis adapted from Ben Shneiderman‘s ‘human-centered AI’ framework. The top right, corresponding to high human control and high computer automation, is labelled “reliable, safe and trustworthy“.

100

Henry Garner @hendroid.io · 20/06/2025

Phronesis is making a comeback, baby!

Google n-grams chart showing steady growth in usage of the word phronesis since 1970

030

Henry Garner @hendroid.io · 18/06/2025

In his book Human-Centered AI Ben Shneiderman argues that "interface designs that are consistent, predictable and controllable are comprehensible, thus enabling mastery, satisfaction and responsibility."

A point of view photograph of me reading the book Human-Centered AI, with the section that I quote highlighted in the text.

180

Henry Garner @hendroid.io · 05/04/2025

No one who’s applied optical character recognition at scale will be surprised to learn that Pokemon was quite popular in the early 18th century.

A screen grab from Google Ngram Viewer showing that ‘pokemon’ was almost as popular 1700-1750 as from 2000+

An example of ‘pokemon’ showing an OCR misinterpretation of ‘possession’ written in copperplate gothic font.

2111

Henry Garner @hendroid.io · 07/01/2025

I’m just going to leave this here

An imagined cover for a book titled “The Tao of Claude” showing a large monochrome illustration of a Shoggoth wearing a Winnie the Pooh mask flying a children’s kite with a Yin-Yang symbol on it.

1221

Henry Garner @hendroid.io · 17/12/2024

Words showing the largest unexplained statistically-significant increase in the scientific literature 2024 vs 2020

A table containing the following words, their occurrences per million in 2020 and 2024, and their corresponding % increase:
delves
delved
delving
showcasing
delve
boasts
underscores
comprehending
intricacies
surpassing
intricate
underscoring
garnered
showcases
emphasizing
underscore
realm
surpasses
groundbreaking
advancements
aligns

181

Henry Garner @hendroid.io · 13/12/2024

This is a very good article. The phony comforts of AI scepticism www.platformer.news/ai-skeptics-...

This is the ongoing blind spot of the "Al is fake and sucks" crowd. This is the problem with telling people over and over again that it's all a big bubble about to pop. They're staring at the floor of Al's current abilities, while each day the actual practitioners are successfully raising the celling.

4181

Henry Garner @hendroid.io · 12/12/2024

I’ll be adding “respects the laws of physics” to all my specifications from now on learn.microsoft.com/en-us/azure/...

• Hardware agnostic: Qubits in quantum algorithms aren't tied to a specific quantum hardware or layout. The Q# compiler and runtime handle the mapping from program qubits to physical qubits.
• Integrates quantum and classical computing:
The ability to perform classical and quantum computations is essential in a universal quantum computer.
• Respects the laws of physics: Q# and quantum algorithms follow the rules of quantum physics.
For example, you can't directly copy or access the qubit state in Q#.

0100

Henry Garner @hendroid.io · 10/12/2024

Always good to be reminded of BARTScore arxiv.org/abs/2106.11520

Bart Simpson writing the following line repeatedly on a chalkboard “Treat evaluation as a generation task”

050

Henry Garner @hendroid.io · 10/12/2024

Beyond cookie-cutter evals: Eugene’s rigorous and accessible into to custom AI assessment metrics is an excellent read eugeneyan.com/writing/evals/

Illustrative line charts showing ROC curves and histograms showing Jensen-Shannon divergence on a benchmark dataset pre- and post-finetuning.

140

Henry Garner @hendroid.io · 09/12/2024

Shiny! The newly released Llama 3.3 LLM leads the LiveBench ranking for instruction following¹, beating Claude 3.5, GPT-4o, OpenAI o1, and you can run it on your local² machine. > ollama run llama3.3 livebench.ai#/?IF=as

The LiveBench leaderboard showing llama-3.3-70b-instruct-turbo in the leading position by average instruction following performance

371

Henry Garner @hendroid.io · 06/12/2024

Me reading the comments in your code

Photo of a hotel towel rail with two notices stuck to the wall nearby. One reads CAUTION VERY HOT RADIATOR in large text. The other reads “towel rail no longer operational due to our new heating system”.

3291

Henry Garner @hendroid.io · 04/12/2024

Mind blown by this impressive paper using genetic algorithms to discover more energy-efficient LLM architectures arxiv.org/abs/2411.17800

Diagram from the paper showing the fundamental operations of STAR evolution (akin to other evolutionary optimization algorithms): 1) initial population, 2) assess, 3) pair, 4) recombine, 5) mutate and 6) repair

1285

Henry Garner @hendroid.io · 29/11/2024

Catbench: definitely the cutest way to develop your embedding vector intuition github.com/tanelpoder/c... #AI

Screenshot of a heatmap visualisation of hundreds of embedding vectors overlaid with a mouseover tooltip containing a single cute n fluffy kitty photo

290

Henry Garner @hendroid.io · 29/11/2024

This, but for AI prompt-to-SQL tools and little Bobby Tables

Top panel: figure steps on a rake which smacks them in the face
Bottom panel: figure does impressive skateboard trick on a rake which smacks them in the face

190

Henry Garner @hendroid.io · 27/11/2024

You automate evaluation of your LLM outputs, right?

Anakin and Padme meme with four panels.
Anakin: We evaluate our system
Padme: With automated tests?
Anakin: …
Padme: …

0120

Henry Garner @hendroid.io · 26/11/2024

The ‘strangler pattern’ is a popular way to modernise legacy systems incrementally. Success relies on corresponding organisational and cultural shifts, but I don’t know of a similarly apt name to describe these. Do you? ‘Agile’ seems too broad to be useful. martinfowler.com/bliki/Strang...

Threaded through all of this is broader organizational change. Legacy systems become rigid and brittle because the design thinking and organizational processes that produced them built them that way. If there's no change in organizational culture and leadership, the new systems will end up in a similar mess. So we also need to introduce new development practices, including reorganizing the development organization and its connections with the wider business in recognition of Conways Law.

1100

Henry Garner @hendroid.io · 25/11/2024

As a systems engineer, I seek out simple designs as a prerequisite for reliability. Users do care about reliability, but sometimes they care about being able to run SimCity more. www.joelonsoftware.com/2000/05/24/s...

Windows 95? No problem. Nice new 32 bit API, but it still ran old 16 bit software perfectly. Microsoft obsessed about this, spending a big chunk of change testing every old program they could find with Windows 95. Jon Ross, who wrote the original version of SimCity for Windows 3.x, told me that he accidentally left a bug in SimCity where he read memory that he had just freed. Yep. It worked fine on Windows 3.x, because the memory never went anywhere. Here’s the amazing part: On beta versions of Windows 95, SimCity wasn’t working in testing. Microsoft tracked down the bug and added specific code to Windows 95 that looks for SimCity. If it finds SimCity running, it runs the memory allocator in a special mode that doesn’t free memory right away. That’s the kind of obsession with backward compatibility that made people willing to upgrade to Windows 95.

1165

Henry Garner @hendroid.io · 25/11/2024

Highly recommend you read this beautiful, thought-provoking & funny article on contraptions (and other helicopter-like things) contraptions.venkateshrao.com/p/contraptio...

Zoologists seem to make a distinction between animals that can be domesticated, like normal cats, dogs, and cows, and ones that can at best be tamed, like orange cats, honey badgers, lions, tigers, and elephants.

I don’t know how solid the distinction is in zoology, but I think it applies to technologies.

4296

Henry Garner @hendroid.io · 23/11/2024

Research into advanced AI reasoning mostly just proves why we all have anxiety arxiv.org/pdf/2411.14405

We introduced a reflection mechanism by adding the phrase: “Wait! Maybe I made some mistakes! I need to rethink from scratch.” at the end of each thought process.

1142

Henry Garner @hendroid.io · 22/11/2024

Thinking about this taxonomy of AI failures from the “Fallacy of AI Functionality” paper. The first question I ask when thinking about a new system: “who’s going to get hurt when this breaks, and how badly?” arxiv.org/abs/2206.09511

Impossible tasks: conceptually impossible, practically impossible.
Engineering failures: design failures, implementation failures, missing safety features.
Post-deployment failures: robustness issues, failure under adversarial attacks, unanticipated interactions.
Communication failures: falsified or overstated capabilities, misrepresented capabilities.

1132

Henry Garner @hendroid.io · 21/11/2024

Thanks for reminding me of this gem

Sometimes people use "respect" to mean "treating someone like a person" and sometimes they use "respect" to mean "treating someone like an authority" and sometimes people who are used to being treated like an authority say "if you won't respect me I won't respect you" and they mean "if you won't treat me like an authority I won't treat you like a person" and they think they're being fair but they aren't, and it's not okay.

050

Henry Garner @hendroid.io · 20/11/2024

It doesn’t matter how many years of experience I gain

A photo of Ken from the Barbie movie looking troubled with the caption “My job… it’s just puzzle”

130

Henry Garner @hendroid.io · 19/11/2024

Claude’s sensory overload is so relatable

Alert box containing the following text: Temporarily Defaulting to Concise Responses
We're experiencing high demand. Chat more with Claude using concise responses or switch back to full responses.

2121

Henry Garner @hendroid.io · 14/11/2024

existentialcomics.com/comic/102

Epictetus was an ancient Roman Stoic philosopher. Like the Stoics before him, and most of the Greek thinkers, Epictetus's thought was closely tied to how to live the "good life". This meant being virtuous and free from pain. For Epictetus, how to do this was to come to the realization that anything outside of your control must be accepted, and couldn't harm you unless you let it. He meant this in quite an extreme way, even saying that the loss of loved ones can't and shouldn't cause you suffering, because their death was part of their existence all along (i.e. they were always going to have died, so when it finally comes it should be taken simply as a part of reality). He even wrote more practical guides to how this might be achieved, such as imagining to yourself that they die every day, so when it comes to be you are mentally prepared for it. Of all the Greek philosophers, the Stoics were probably closest to Buddhism (although there is no evidence that either knew of each other), in that they emphasize the acceptance of reality in order to eliminate suffering.

While Nietzsche admired both the Stoics and Buddhism in some ways, he was also critical of both, claiming that they were nihilistic, life denying philosophies (although his understanding of Buddhism in particular was probably not very good, due the materials available to him). Not only that, but he claimed they were inherently contradictory. He said they didn't actually "accept things as they were", but willfully strode towards the Stoic or Buddhist good, which was contructed by them.
Schopenhauer was one of the western philosophers most influenced by Buddhism (although like Nietzsche, he probably didn't have the most accurate account of it available to him). He mostly accepted their ideas about how suffering was caused by desire. However, unlike the Stoics and Buddhists, he was much more pessimistic about there existing a solution to this situation. He felt that ultimately humans were creatures entirely composed of a will towards something, and that would always cause suffering when it wasn't fulfilled, or boredom when it was. He thought then that humans must suffer, and was one of the first to commit to the antinatalist position, claiming that humans were better off never existing at all.
Sorry! is a board game where the goal is technically to get all your pieces to the "home" area, but most of the enjoyment comes from inflicting suffering on your opponents by knocking them back to the start at the last possible moment.

020

Henry Garner @hendroid.io · 13/11/2024

A rare night out with Natalie tonight to see “What If?” live. Looking forward to pondering the imponderables for pleasure rather than business 🤔 www.fane.co.uk/randall-munroe

000

Henry Garner @hendroid.io · 13/11/2024

Leave peeking to the professionals

A person in hi-viz with the word PEEK written across the back peers into a roadside cabinet

030

Henry Garner @hendroid.io · 11/11/2024

Notepad becomes AI enabled blogs.windows.com/windows-insi...

The Clippy next to a speech bubble saying “It looks like you’re trying to create a little bit of confusion”

120

Henry Garner @hendroid.io · 10/11/2024

See? This is how they do it, man 🤨

010

Henry Garner @hendroid.io · 10/11/2024

Totally. cf the Jock/Creep Theory of Fascism www.unpopularfront.news/p/the-jockcr...

110

Henry Garner @hendroid.io · 10/11/2024

Statler: “Santa’s coming to town!” Waldorf: “Great, I hope he brings a gift… like a new act!” Statler: “Yeah, this one’s been on the naughty list for years!” Both: “Ho ho ho ho!”

110