There is an argument, one that has real support among American voters that the geopolitics of Europe or indeed most of the world is no business of America’s and, in particular, that America need not lift a finger to prevent Ukraine being devoured by Russia. If that is what President Trump believes, then he should say so. He should not pretend that he is “wants everyone to stop dying” or that he is pro-peace and anti-war, or that he is trying in any realistic way to negotiate a settlement.

President Trump would like glory of being a great deal-maker. But he will fail, because he will not allow America to do its part. The outlines of a peace have been clear since at least late 2022:

• Ukraine and the world recognise Russian sovereignty over Crimea other parts of Eastern Ukraine.

• Ukraine and the world agree that it will never enter NATO.

• But West provides some other powerful guarantee of Ukraine’s security.

One can argue that the West has no duty to deliver on that last point – but they cannot also credibly claim to want peace. Like many aggressive powers throughout history Russia has a long track-record and demonstrated willingness to invade its neighbours and to ignore old agreements.

An agreement without a hard-power security guarantee is a license for Russia to lick its wounds and continue the war under conditions of its own choosing. The shape of that guarantee would probably be a well-fortified and armed Ukraine, along with western European troops and the economic might of America’s military industrial complex.

In the past, America has used its own troops to make lasting peace in countries like Germany, Japan and South Korea. Indeed, South Korea is a good model for the kind of peace that America could forge. Even David Sacks, a vocal critic of assistance to Ukraine seems to see this

Mr Sacks is an intelligent and well informed man, but he has somehow blinkered himself to the point he doesn’t seem know that American troops defend the Korean border to this day.

Perhaps it’s because America’s security guarantee to Korea works so smoothly that Americans forget it even exists. But this is also an example of the self-deception required to equate appeasing Russia with supporting peace. Mr Sacks is part of the chorus that takes as given that somehow NATO stared the war, and the 2014 revolution in Ukraine was an American coup d'état.

The “great leadership” with which Eisenhower brought peace to Korea is not what President Trump is offering. He is instead making an isolationist choice not to lead. Trump nonetheless hopes to gain the laurels of a peacemaker by cornering Ukraine into agreeing to phony simulacrum. Support him in this if you wish, but do not pretend that you are on the side of peace.

I hope this is not just another Twitter storm, but the generation-long atrocity of the state-condoned Pakistani rape gangs of Britain is once more getting attention. This time, it might even be enough to force Britain to change course and that has some wondering why the sudden new interest in a tragedy that has been well known for decades.

In one sense, this was inevitable. This scandal is not a revelation; there have been journalistic investigations and even government inquiries exposing it. Sometimes there have been prosecutions. Yet the gangs have not been busted, and the earth from which they spring has not been salted. Infuriatingly, it remains dangerously politically incorrect to complain about it. People might have been too afraid, too demoralised and simply too busy getting on with their lives to talk about it, but that just means there is a huge population inversion of furious people just waiting to start lasing in unison.

But what has triggered the cascade which had been suspended for so long? The proximate answer is obviously that Elon Musk is banging the drum about it. There's something to worry about in Mr Musk's power to do this, especially if he is putting his thumb on the scales of the Twitter amplifier. I'll write about that one day, but right now I don't give a flying fuck. I'm more interested in why Mr Musk himself is suddenly interested.

To me it is obvious that he wants to help Nigel Farage politically. This is the most perfect topic for Mr Farage and his Reform party because all their establishment rivals are complicit. Tactically the timing is right; Reform wants to make a big impact in local council elections and will be competing against many of the Labour councils who were most directly involved. By the time the election comes, some of Reform's incumbent opponents might already have resigned in disgrace.

Musk has even started calling on the King to dissolve parliament and precipitate a new general election. I think this is just shit-stirring design to amplify the isssue rather than a serious attempt to coax His Majesty into action. But in the long run, a general election will come anyway and there is plenty of disgrace to share around at higher levels of government too. For example, Prime Minister Keir Starmer is a former Director of Public Prosecutions under whom these gangs were allowed to run rampant.

Farage will want to turn all upcoming elections into referenda on whether the British state should continue to condone and collude in the mass rape of young British girls. Now there's a hell of a Brexit question:

Should the United Kingdom remain a member of the Rape Gang Union or leave the Rape Gang Union?

It won't be a difficult question for the voters to answer, and yet the British establishment has been insisting on the wrong answer for at least a generation. It has well and truly earned its predicament.

Over Christmas 2024, the Make America Great Again coalition has erupted in a splendid new Twitter spat nominally about H-1B skilled work visas category. Now an astute reader might find such topics a little too boring for a good flamewar. Indeed it looks more like that after Trump's election victory, some on the right are tired of all the winning. They want to snatch defeat from the jaws of victory by turning attention away from Venezuelan gangsters and onto Indian tech nerds.

The charitable view (the one I held before I started to think about it) is that the debate about whether American tech workers are being frozen out of the job market by Indian coolies. Deeper yet, we might dimly perceive a principled argument between nativists for whom "America First" means "Americans First" and techeads for whom it means "America in First Place".

There are some on the nativist side who really are driven by these valid considerations. But if that’s all there was to it, because both sides could easily agree on sensible changes to the actual visa. But there are darker motivations driving MAGA to choose infighting instead. More on those motivations later.

The H1-B is for temporary employment of foreign workers in "specialty occupations or as fashion models of distinguished merit and ability". This would all be much more entertaining if was about fashion models, but disappointingly, it’s about computer engineers instead. The nativists and tech-heads both hate H-1B for seemingly opposite reasons.

Although the visa is supposed to prevent foreigners from displacing native workers, the nativists see the safeguards as a sham. Is computer engineering really such a specialty Americans can't do it? The rules require workers be paid at least the "prevailing wage" in their occupation - as if that wage were not affected by an influx of foreign competition.

For its part, the tech-head camp sees only a shortage of workers and to them H-1B is a maddeningly unreliable means of mitigating it. Every country has its own visa cap, and when the cap is exceeded, the winning applicants are chosen by a literal lottery.

My own lived-experience from the junior trenches of big tech is closer to the latter camp. No matter how many capable candidates there might be in the world, very few of them ever seem to turn up at the interview table. Hiring is hard. But it is also true that many of the biggest employers of H-1B workers are Indian outsourcing firms who have replaced what used to be in-house tech staff in the US. They can also point to the fact that, despite claims of talent shortage, in the last few years, tech firms have actually been laying workers off.

These arguments have been around a long time, but the sudden Yuletide fight seems to have started with the tweet by right-wing media personality Laura Loomer complaining about the "appointment of Sriram Krishnan @sriramk as Senior Policy Advisor for AI at the Office of Science and Technology Policy". Mr Krishnan's offense was that he had once spoken out in favour of loosening the restrictions on H-1B visas.

Soon enough, Elon Musk and Vivek Ramaswamy weighed in to support skilled immigration. Musk liked America to a champion sporting team that needed to bring in "the top ~0.1% of engineering talent ... to keep winning." (The US has a different visa category O-1 for this purpose, but Mr. Musk's experience seems to be that he himself along with many people who built great American companies actually came in under the more permissive H-1B visa.

The reason I’m in America along with so many critical people who built SpaceX, Tesla and hundreds of other companies that made America strong is because of H1B. Take a big step back and FUCK YOURSELF in the face. I will go to war on this issue the likes of which you cannot possibly comprehend.

https://x.com/elonmusk/status/1872374103983759835

Mr Ramaswamy made some wonkish criticisms of the H-1B's quota system. But he also lit a firecracker by criticizing an American culture that has "venerated mediocrity over excellence for way too long". Mr Ramaswamy wants to see

More movies like Whiplash, fewer reruns of "Friends." More math tutoring fewer sleepovers. More weekend science competitions, fewer Saturday morning cartoons. More books, less TV. More creating, less "chillin." More extracurriculars, less "hanging out at the mall."

https://x.com/VivekGRamaswamy/status/1872573801416003622

These are just a few highlights from a vast and sprawling food-fight. At its best, the fight between those Americans who want to lean into skilled immigration in order to keep America in the champion's spot vs. those who want Americans to have first dibs on the opportunities their country provides.

As a matter of abstract principle of course any country should give its own citizens first dibs on its opportunities. But opportunity does not fall as manna from heaven, it is made by the skilled work of human beings. America is the land of opportunity because it has always an open order that attracts contributions from all over the world. A highly skilled immigrant doesn't remove opportunity from natives, she thickens the network of such opportunities.

But that doesn't mean every H-1B visa holder is such a creator; at least the average low-level IT worker employed by an outsourcing firm isn't. Which is why there are actually many opportunities for both sides of this debate to get what they want by reforming the visa.

1. Add a fixed minimum salary restriction. Even if this pure restriction were the only change, it would still help high-level hiring by freeing up quota slots that would otherwise go to lower-paid foreigners.

2. Simplify and debureaucratise the process. Red-tape not only makes honest hiring more difficult, it also creates loopholes for abuses. Indeed replacing the fuzzy "prevailing wage" condition with a flat minimum salary is a good example of this.

3. Unlink the visa holder from particular employers. Nativist opponents of H-1B often say the visa makes Americans compete with "indentured servants". So let visa holders to change jobs as long as they continue to meet the same requirements as the original.

It's also important to remember the bigger picture. America today has an effectively open door for illegal immigrants. And even the vast majority of legal migrants come under asylum or family reunion rules, not skilled-work visas. The coalition to restrict immigration has lots of important work to do before it can afford a fight about skilled migrants.

So why then is there a fight?

The attention economy gives certain people (call them "grifters") a financial incentive to choose division and defeat. But that's perhaps a rationalistic explanation for murkier motivations. What seems to be at work here is the psychology of what is starting to be called the "woke right".

It's helpful to look at Laura Loomer's original post. She was using Twitter to publicly rabble-rouse against an advisor on AI policy on the basis of past wrongthink on a completely different topic. This is an almost perfectly baked example of cancel-culture cake worthy of the woke left. (That she also mischaracterized the wrongthink is just the cherry on top). The only significant difference is that Ms Loomer didn't seem to have any hope of actually cancelling her target. She just wanted to complain about Mr Krishnan.

Which brings us to the towering trumpeting tusker in the room: most of the ire seems directed at people of Indian descent. One wonders if the worry is really about Indian tech workers and not about the number of brown faces and Sanskrit names appearing in the vicinity of Trump's new throne.

I doubt there are enough racists in America to really drive a political movement of any heft. But there might be enough to misdirect the energies of Americans with justified anger at mass immigration. Trump and the whole MAGA coalition would do well to keep their eye on the ball and pay no heed to the mischief-makers hanging to their coattails.

Once, not so long ago, in a place called Yarongara, there lived a yeti. He was only a ragged, floppy old thing, with shaggy off-white fur, a lolling neck and stuffing that had long lost its vigour. But Yitsun the Yeti was little Ravi's most beloved toy and closest friend.

His big sister Rosie also had a best friend, a flesh-and-blood friend called Wendy. But unlike calm old Yitsun, Wendy had been trapped in a terrible sickness for weeks now. She spent most of her days muddled and confused, drifting in and out of troubled dreams, just barely able to take food and drink in those short times when she was awake and aware. It broke Rosie's heart to see her best friend so unwell.

Rosie and Ravi were in their room after another difficult visit with Wendy. As they got ready for bed, the little boy looked to the wise old face of the yeti sitting patiently as ever on his bed. "Yitsun, the doctors don't know how to help Wendy. Do you think you can help her?" The children often asked Yitsun for wisdom; his clear blue eyes seemed like windows into a world of experience from whatever distant valley yetis came from. But they gave no answers, at least not yet.

Rosie smiled and ruffled her kid-brother's hair. "Hang in there, mate. The doctors will figure something out. But Yitsun, if you can lend a hand, we'll take anything we can get".

Then, too sad to play up, the pair got into their beds and went to sleep. Almost as if they were well-behaved children.

That is the beginning of a children’s story that will not be called the “Yitsun Chronicle”. That title belongs to this series of blog posts about what I am learning from using an AI Chatbot to help me write fiction.

I’ve heard that (unlike it’s competitors) Anthropic’s Claude3 is very good and writing stylish prose, and adopting any style you ask for. So I’d love to report that thanks to Claude3, I was able to write superb prose with minimal effort. Sadly I can’t: although the passage above has a complicated textual history, the short version is that the text is almost completely my own. I can’t even claim my prose is superb. It’s fine. I like it. It will do. But C.S. Lewis and J.K. Rowling would have both done it better. And sadly I am talking about the real Lewis and Rowling, Claude3 pretending to be them does not cut the mustard. I know, because I tried.

I am going to keep experimenting to get Claude3 to generate good prose, and I’ll report what I find in later articles. This post, will be about the area where I really did get a lot of use out of Claude: outlining and plotting. That’s a thing I am frankly rubbish at. I often come up with ideas for characters and themes that (in my own mind) will make for a cracking good a story, but then I get plotter’s block. I am totally at a loss trying to pull these elements into anything that resembles an interesting story, or even some kind of story.

Fortunately there is an easy way to at least reach the “some kind of story” bar: just pluck random tropes out of our literary heritage and glue them together in cliched fashion. And this of course, is what Claude and other Large Language Models are born to do. So I got lots of mileage from asking Claude to outline a story and then iterating.

I don’t want to give away the Yitsun plotline, so instead I’ll show you how it’s done with another example I whipped up for this article in about five minutes in between Tadpole-care chores.

TW: Hi Claude. I want you to help me write a children's story about a talking frog. Please propose an outline using numbered bullet points.

It’s important to make clear you only want an outline. If you forget, then it will deliver the whole shit-for brains story. You don’t want that. Also I shouldn’t have asked for anything as nonsensical as “Numbered bullet points”. Claude made the best of it;

C3: Here's a potential outline for a children's story about a talking frog:

1. Introduction

   • Meet the main character, a young child (establish their name, age, and any relevant details)

   • Describe their fascination with the nearby pond and the creatures that live there

2. The Encounter

   • The child visits the pond one day

   • They hear a strange croaking sound

   • They discover a frog with the ability to talk!

3. The Talking Frog

   • The frog introduces himself (decide on a fun name)

   • Explain how the frog gained the ability to speak

   • Establish the frog's unique personality and quirks

4. An Unlikely Friendship

   • The child and the talking frog become friends

   • They share adventures and experiences around the pond

   • Highlight the lessons they learn from each other

5. The Conflict

   • Introduce a problem or challenge that the child and the frog must overcome

   • This could be a misunderstanding, a threat to the pond, or a personal struggle

6. The Resolution

   • Describe how the child and the frog work together to solve the problem

   • Emphasize the importance of their friendship and the strengths they both bring

7. The Conclusion

   • Wrap up the story with a heartwarming ending

   • Reinforce the central message or moral of the story

   • Hint at the possibility of more adventures with the talking frog

Feel free to modify or expand on this outline as you see fit to create an engaging and imaginative story for children.

Now I this is not a great story, nor is it my story and worst of all, it’s incredibly vague: “Introduce a problem or challenge that the child and the frog must overcome”! What problem?

And yet Claude had done something magnificent here: it has blown right through my plotter’s block. In one interweb round-trip, it’s pieced together the elements of something that looks like a story. Even more profoundly, Claude is educating me in the basics of storymaking: form relationships between characters, make these characters face challenges, bring those challenges to a resolution. WHOA MAN! How cool is that! I never knew.

Of course this plot is still nowhere need good enough. We still have to iterate. And the beauty is that computers are great tools for iterating . I’ll show you how this works, but for brevity, I’m going to ruthlessly editorialise Claude’s side of the chat..

TW: Let's keep the child out of it. The main character is the frog himself.

Claude flipped it to the frog’s point of view, but kept the child in. This turned out to be for the best. But Claude did come up with a really weak reason for the kickoff:

C3: …

2. The Frog's Curiosity

• The frog becomes curious about the human world beyond the pond

• He decides to venture out and explore

TW: Let's replace (2) with some troublesome event that forces the frog into action -- give me five options.

So here I introduce another technique for using the LLM’s uncanny ability to pick random tropes. If you have some vague placeholder like “challenge” or “troublesome event”, you can ask for options to fill it in with:

C3: …

1. Drought …

2. Pollution …

3. Construction …

4. Predator Invasion …

5. Natural Disaster …

How boring! Actually this is my fault, it had never occurred to me that this story would be anything but a fairy tale. The Claude beast is telling me something the world, or myself. I can’t tell which, but I can fix the story.

TW: Let's make them all more like a fairytale

C3:

1. Wicked Wizard's Curse …

2. Jealous Fairy's Revenge …

3. Goblin Construction Crew …

4. Slimy Swamp Monster …

5. Mermaid's Stolen Voice …

Ironically, I take a fancy to the least fairy-tale like of the lot.

TW: Ok let's go for a Goblin construction crew. Give me the outline again -- this time he doesn't meet with humans or children, but with a good goblin who helps him.

Now here’s the beautiful part. I a started this goblin thing by whingeing about point (2) — the frog’s motivation for the adventure. But now I have introduced a theme that weaves it’s way all through the story. And Claude will automatically do the weaving.

C3: Here's an updated outline with the goblin construction crew as the catalyst, and the frog befriending a good goblin instead of humans

1. Introduction

   • Meet the talking frog (name, description, how he learned to talk)

   • Establish his peaceful life in the pond and surroundings

2. Goblin Construction Chaos

   • A band of mischievous goblins arrive at the pond

   • They begin draining the water to build a grand mud castle

   • The frog's home and friends are threatened

3. Attempted Negotiations

   • The frog bravely confronts the goblin workers

   • He tries to reason with them to leave the pond alone

   • But the goblins don't understand and continue their work

4. Meeting the Good Goblin

   • The frog encounters a kind-hearted young goblin among the crew

   • This goblin can understand the frog's speech

   • They become unlikely friends, despite coming from different worlds

5. The Good Goblin's Help

   • The good goblin explains the workers are under the goblin king's orders

   • He agrees to help the frog by taking him to the goblin kingdom

6. Facing the Goblin King

   • The frog and good goblin sneak into the kingdom

   • They make their case to the goblin king about saving the pond

   • The king is moved by their courage and friendship

7. Resolution and Compromise

   • The goblin king allows the pond to be restored

   • In return, the frogs teach the goblins about coexisting with nature

   • The good goblin becomes an ambassador between the two communities

8. The Conclusion

   • Wrap up with the newfound understanding between frogs and goblins

   • Hint at more adventures with the frog and his goblin friend

Let me know if you'd like me to modify or expand on any part of this updated outline!

See most of the vagueness is gone, and we have a real story. But there’s still some technicalities to deal with:

TW: Propose ten names for the frog

( 10 frog names.)

TW: Let's pick Slippy, now ten for the Goblin.

(10 Goblin names.)

TW: Mudlump. Now give me the outline.

C3: Here's the updated outline with Slippy the talking frog and Mudlump the good-natured young goblin:

1. Introduction

   • Meet Slippy, the talking frog (description, how he learned to talk)

   • Establish his peaceful life in the pond and surroundings

2. Goblin Construction Chaos

   • A band of mischievous goblins arrive at the pond

   • They begin draining the water to build a grand mud castle

   • Slippy's home and friends are threatened

3. Attempted Negotiations

   • Slippy bravely confronts the goblin workers

   • He tries to reason with them to leave the pond alone

   • But the goblins don't understand and continue their work

4. Meeting Mudlump

   • Slippy encounters Mudlump, a kind-hearted young goblin among the crew

   • Mudlump can understand Slippy's speech

   • They become unlikely friends, despite coming from different worlds

5. Mudlump's Help

   • Mudlump explains the workers are under the goblin king's orders

   • He agrees to help Slippy by taking him to the goblin kingdom

6. Facing the Goblin King

   • Slippy and Mudlump sneak into the kingdom

   • They make their case to the goblin king about saving the pond

   • The king is moved by their courage and friendship

7. Resolution and Compromise

   • The goblin king allows the pond to be restored

   • In return, the frogs teach the goblins about coexisting with nature

   • Mudlump becomes an ambassador between the two communities

8. The Conclusion

   • Wrap up with the newfound understanding between frogs and goblins

   • Hint at more adventures with Slippy and his goblin friend Mudlump

Let me know if you need any clarification or have additional ideas to incorporate into this fairy tale outline!

And there you go. After about no time at all, we have the plot of an exciting and heart-warming story. It’s already good enough that I could just tell it to my Toadlet, verbally ad-libbing the details.

Or I could keep going, I don’t really like the ending, and no doubt I could ask Claude for better options. This plotting business is tremendous fun. I did this frog outline in the five minutes I could steal away from dad-duty. But the Yeti story (like this blog post) was done late into the night with the help of a brandy-glass, I had no objection to spending hours lovingly honing the story and expanding some of the details.

The real challenge was to stop that game and get to the work of actually writing text. That will be for a future blog post. Not even the next post. The next, post will be more philosophical, about what process showed me about the natures of large language models, Human beings, and creativity.

That's because the livestream caught a teenager shouting "Allahu Akbar" before stabbing Australia's maverick Assyrian Bishop Mar Mari Emanuel. The victim has forgiven his attacker, but says he still wants the video to remain online (the Bishop is himself a prominent TikTokker).

But if Australia's eSafety Commissioner Julie Inman Grant has her way, it will not be available. Indeed, it would be unavailable to everyone in the world, since she has given digital platforms a global removal order for this content. The Commissioner and the rest of Australia's censoriat are very upset that Elon Musk's Twitter (aka X) is openly protesting and complying with the order only within Australia.

They are much happier with the other platforms that have meekly agreed to comply with these orders. Actual compliance does not seem to be the point: Sydney Radio station 2GB reported that the video was still readily available on Facebook, yet the government does not seem to have noticed. It looks as if the censoriat is more interested in using this attack to pursue a vendetta against Mr. Musk than it is about controlling the video, let alone actually preventing such attacks in future.

Politicians cheering the Commissioner on claim that letting other people post videos is but a narcissistic power trip of Mr. Musk who should instead be helping to "keep Australians safe online" from "violent and distressing content". As if we were children.

Now, in fact, I personally do avoid watching violent viral videos (since I am not a child, I can protect myself from content if I choose). But I did watch this video, and I'm glad. It revealed something that I could not get from any verbal description. Something about the cool ferocity of the criminal struck me in ways that can't be put into propositional sentences. I am an ineffably wiser adult for having watched it.

That's right, violent videos can be good for you! Indeed, there's something suspicious about censoring video just because it's violent and disturbing. Australia's leaders never put up much fight against films like Terminator, The Silence of the Lambs, or Natural Born Killers. Even videos of real-world violence, such as the death of George Floyd, have not only been allowed to go viral but have been celebrated.

So by what principles is the censoriat deeming this particular video to be unacceptable for public consumption? Not only have they failed to articulate any clear, viewpoint-neutral rationale for targeting this content, they also don't seem interested in having one. The exercise of arbitrary power seems to be the goal of the censoriat and the activist industry behind it. I recently wrote in Quillette:

As is usual in democratic politics, Western governments are being steered down this path less by a simple lust for power than by the efforts of an activist industry: in this case, a network of individuals and organizations that earn their keep by warning of the dangers of harmful information and taking it upon themselves to determine what information is "harmful" and what is "safe."

You should go read the whole thing in light of the Commissioner's latest rampage. It was finalised just before any of this hit the news, and doesn't mention the Commissioner at all. The gist is that the risks of dystopian private-sector censorship by Big Tech are receding, only for the old-fashioned government kind to reassert itself. You can compare my analysis there with how things are panning out in the news.

I admit that current news makes me question my suggestion that the government censoriat is not lusting for power. The Commissioner wants to “really minimise the amount of content the Australians can see”. Our Prime Minister Anthony Albanese wants social media to crack down on people who mock him in memes. No wonder such a person also finds it extraordinary that Twitter/X is “trying to argue its case”. Tasmanian Senator Jacqui Lambie thinks Mr Musk should be jailed for his cheek.

But the Commissioner is part of that activist censorship industry. According to her official webpage, she came out of Washington DC to work on "safety policy" at big tech firms including the (pre-Musk) Twitter. Then via some sort of transcontinental revolving door, she seems to have moved from government relations to actual (Australian) government.

Now, as Commissioner, she is also part of a welter of global digital safety and woke-sounding QUANGOs, including even the dreaded World Economic Forum (WEF). From her official webpage:

As Commissioner, Julie plays an important global role as Chair of the Child Dignity Alliance’s Technical Working Group and as a Board Member of the WePROTECT Global Alliance. The Commissioner also serves on the World Economic Forum’s Global Coalition for Digital Safety and on their XR Ecosystem Governance Steering Committee on Building and Defining the Metaverse. Under her leadership, eSafety has joined forces with the White House Gender Policy Council and Government of Denmark on the Global Partnership for Action on Gender-Based Harassment and Abuse.

Inman's greatest hit at the WEF was a panel comment in which she called for “... a recalibration of a whole range of human rights that are playing out online”.

I don't want to single Inman out, recalibrating human rights is what the whole digital safety lobby is all about. As I wrote in Quillette

These laws are not simply an expression of countries’ own constitutional norms. They are a dodge around those norms. Free speech traditionally flourishes in democracies because their citizens can’t be muzzled at the mere say-so of some censor appointed by a theocrat, strongman, or authoritarian. Even in democracies with nothing like America’s First Amendment, speech can traditionally only be limited by specific laws. But on the Internet, none of this applies. Posts can be quietly downranked or deleted at will. The platforms are not accountable to anyone and the censored have no legal recourse. Civil servants, government officials, security forces, and other agencies of the state can exploit this by pressuring platforms to mute their critics.

This hostility to the rule of law has been on full display recently, not merely in the insolently authoritarian language of our Prime Minister, but also in the actions of the Commissioner. The Free Speech Union of Australia points to legal gamesmanship by the Commissioner who obtained an injunction against Twitter/X over the Wakeley stabbings without giving them a chance to respond. The Union is also helping Chris Elston, a Canadian activist nicknamed “Billboard Chris” who has also had a tweet subject to a removal order. The Union claims not only that the removal order was invalid under the law, it also points out that the Commissioner failed to notify Elston of the order, as required under her enabling legislation.

Dystopian Down Under

'Spectacular backfire': Australian government's attempt to censor trans post draws heat

First it took down tweets on breastfeeding. Now, Australia’s online safety regulator, the eSafety Commissioner, has threatened social media platform X (formerly Twitter) with a hefty fine over a post characterising the trans male identification of a bearded natal female as a psychiatric condition…

Read more

2 years ago · 104 likes · 12 comments · Rebekah Barnett

The most enraging thing about the government's behaviour is that they are not interested in the actual problems. The boy in the video has been charged on terrorism-related offences. Very likely, terrorism is the right way to think about his act. Around the world, billions of people keep their mouths shut about some of Islam's darker tendencies out of fear of young men just like him. In Britain, the Mother of Parliaments is twisting her own procedural knickers in fear of Islamist violence.

Australia's leaders should be working on how to free our country from that cloud of fear, which has no place in the workings of liberal democracy. But they are uninterested in the traditions that keep liberal democracy alive. Instead, they are champing at the bit to attack Mr Musk, because he is doing the job that is rightly theirs: standing up for the ability of Australians to communicate with each other.

If you are an Australian, and want to this country get back on track so, you can start by signing the Free Speech Union’s petition to abolish the Online Safety Act. You can also join the Union.

I’m going to use AI tools to help me stand up a little web-based fitness tracking app. This app is a fairly stupid project of no real value except as a way to get some practice using AI tools.

For this episode, I’ve used ChatGPT (i.e. v3.5) to advise me about how write the app. So far I’ve done little more than stand up a simple “hello world” web-server in something like a realistic development environment. Yet this is enough for me to have some observations about the process. These observations are not particularly surprising or insightful, but they are worth recording, if only to see how things change over the next year or three.

It understands programmers well

By which I mean it is amazingly capable of unpacking and responding helpfully to the customary inarticulate grunting of the software profession. For example:

AR: What about `response_class=HTMLResponse`?

GPT3.5 understood that I was objecting it’s previous suggestion of hand-hacking a content-type header when it could decorate the calling function instead. The long-winded text that went with the code-suggestion included the actual insight:

… By adding response_class=HTMLResponse as a decorator argument for the read_root function, we specify that the response should be treated as an HTML response.

It’s long-winded

This isn’t just a programming thing, ChatGPT will never answer with a short sentence when a whole essay will do. I think that’s because it was trained by Americans.

In the programming context, that means questions don’t yield answers, they yield answers embedded in paragraphs of text and coding examples. I can’t comment on the quality of the text, because I barely read it. The code examples are pretty good, but more on that later.

It’s talented at spitting out code

I say spitting out code, not coding. I expect ChatGPT would be bad at solving some non-cargo-cultable algorithmic problem. But for the dumbshit API gluing that is most of our work, it’s a great first guesser. It pulls in all the bits that are needed and stitches them together in mostly correct ways. It even explains what its doing, though I wish it wouldn’t.

It doesn’t check it’s work

ChatGPT is not a source of correct solutions, it’s a source of very good first guesses. As such it’s much better than a human, it’s first guess will generally be syntactically correct and cover all the obvious bases. They might even work.

But it will still happily make errors at one or two levels up. For example I asked it to redirect unauthorised requests, and it happily set up a redirect loop because the user had no authority to see the target page. But then eyeball the error either.

The way to think about this is that you will always need to iteratively evaluate and fix your code, but ChatGPT makes each iteration faster and more effective. It will quickly spit out a good first guess, and than make good improvements in response on your vague off-the-cuff suggestions.

It’s not just code

Prior to writing a line of code, I had a long and fruitful discussion with the thing about the different design options, tools, and what off-the-shelf components to put in which part of the stack. It’s a great way to quickly get up to speed on the bland-common sense general opinion of what the state of the art is some boring part of the field.

The Verdict

ChatGPT is a good programming consultant. It’s helpful, and mostly pleasant. Yet I’m not sure it actually made me any faster. I could have banged this app out quickly by simply Googling and cargo-culting StackOverflow. While ChatGPT does sort-of automate that, the manual workflow was already hard to beat.

The next step is obviously to try GitHub Copilot, even though it gives me the willies.

I said above that programming with ChatGPT makes each iteration of your coding cycle faster and more effective. The limiting friction is the mismatch between your real code and the toy example code that ChatGPT is spitting out. Not only do you have to manually cut & paste, but you have to manually adapt the result to you real code (changing variable names etc).

Setting aside the willies, an interface which read your real code and offered you a side-by-side diff in the IDE would cut out nearly all this friction, and really might supercharge the edit-check-improve cycle.

Henceforth and until further notice, the name Tadpole Worrier shall refer this being and not to his older sister. The elder offsprung shall now be known as Toadlet Worrier, the new title is a promotion that she has long since earned. This is not a picture of her:

The family is well (enough) and is (mostly) happy.

End of message.

Areo Magazine has published my article "In Defence of Tinkerers: Why We Need Open AI”. It lays out the balance of risks and opportunities of artificial intelligence. Go read the whole thing, but the public-policy crux is:

This still requires sensible legal limits on AI development. But, like all the best laws, those limits should be part of a framework that secures the rights of the general public: in this case, the right to create and commercialise AI, as long as this is done responsibility. Both the freedom and the responsibility should apply universally: to individual tinkerers, governments, well-connected corporations and academics

The timing was lucky because it dropped just as chatter around AI has moved to regulation. The ever-excellent Timothy B. Lee has an excellent review of the various proposals. But the Spectacle is more about extraterritorial red tape from the EU,and the boss of Open AI asking Congress for monoporegulation. OpenAI then followed this up with a blog post.

My own post talks about a framework of laws that should both protect and limit the rights of the public to work with AI. But I am a time-traveller from a past century to be writing about laws, rights, or even the public.

The modern talk is about “regulation” and “governance”. It’s not about how people ought behave towards one another, rather it’s about how to build power structures that will then tell people how to behave. OpenAI knows that it

some degree of coordination among the leading development efforts

and that

we are likely to eventually need something like an IAEA

and that

the governance of the most powerful systems, as well as decisions regarding their deployment, must have strong public oversight

But they don’t give any hint as to what actual sorts of rules and limits they actually want. They don’t know. But they somehow do think they know that these governance structures will choose to do the right thing.

Fuck that. Let’s talk about the actual rules before appointing overlords.

Read my article to see what rules I'd recommend. You will no doubt have different ideas. That’s fine. But I urge you all to stop debating regulation and government in the abstract, and instead think seriously about actual rules and norms of behaviour you’d like to see. Let that guide your own behaviour and the guide the voice you add to the shaping of your country’s laws.

Subscribe now

This was the starting point of self-described reactionary feminist Mary Harrington's discussion of her new book Feminism Against Progress on Bret Weinstein's podcast. I don't want to pick on Ms Harrington, this essay is only a critique of the case against progress, for which she is but the most capable advocate. It is not a critique of her book or of the podcast, during which she in fact thoughtfully refined her views in response to Dr. Weinstein's objections.

This essay is not a review of that podcast; rather it is my case for progress shaped as a response to three arguments of Ms Harrington. First she admits that society evolves but denies this represents an improvement of moral value. This is a defensible motte from which she moves to a wider bailey. She reckons that history shows that there's no progress because civilisations rise and then collapse. All this underpins her view that belief in a long term arc of improvement is a teleological or religious faith.

This motte-and-bailey is no dishonest trick cooked up by Ms. Harrington. It's an earnest part of the world-view of every progress-skeptic I talk to. Yet however earnest, it's still tricksy to quietly admit that modernity has brought us electricity, obstetrics and moon rockets while loudly proclaiming, with scant context and no qualification that progress is a fiction.

That's how progress-skeptics sneak out of their narrow moral motte. Given half a chance, they reach to wider territories, beginning by branding the idea of progress as teleological or religious. But this ignores a simple a causal explanation: society progresses because useful knowledge grows.

Civilised humans took millennia to discover writing, bronze and electricity. But we have not since undiscovered them. Useful knowledge is easier to retain than win and easier to win than destroy. On the scale of history, it is quickly disseminated, replicated and used. It gets encoded redundantly in books, technologies, social practices and the genes of domesticated species. Every generation inherits a vast and waxing store of ancestral knowledge both explicit and tacit.

Next comes the view that there is no progress because civilisations rise and fall. Whenever I hear that, I wonder where archaeology has found the ruins of 20,000 year old Autobahns. Parcelling humanity up into discrete civilisations that can rise or fall is a slippery simplification. Tribes, nations, empires and networks can all collapse, but so far, there's always been others to carry on. Did Roman civilisation collapse around 410 AD or around 1453 AD? Or did it live on to conquer South and Central America where it thrives today?

This is why progress-skeptics retreat back to the safety of Mary's Motte and acknowledge the growth of knowledge, productivity social complexity and human health but deny that this is called progress. Setting aside how weird that is, we should still answer Ms Harrington's challenge to the moral value of this growth.

Her argument is relativistic: if you set out to show that things are better now than in the past, you have to define your terms, and hence assume the thing you are purporting to prove. But even she doesn't seem to take the relativism seriously, and accepts Dr. Weinstein's response that it is desirable to fulfil human potential.

In view of human potential, I place myself on the bare feet of an ancestor toiling in a some medieval paddy-field. I am grateful to him and his folk for the heritage they bequeathed me. But I do not envy them. I do not want my children born into their estate. Though I honour that bygone culture, I also know that it was a system of caste and peonage that kept people in their allotted place of material and intellectual poverty. They did not have the same opportunity as the citizens of modern liberal republics to unfold their potential.

I've said that societies advance because knowledge grows. Knowledge is power, and power can be turned to both good evil. But empirical observation shows that for whatever reason, advanced societies, on average, murder less and provide more than forebears. So far at least, the unquestionable tyrannies and disasters of modernity have been the sidelines and also-rans of history. For an exhaustive survey of the empirics, look to fat tomes Steven Pinker. But for this short essay I'll concentrate on one thing we have learned: that freedom is better than slavery.

This is clearest in the case of true slavery. Modern civilisation, almost uniquely, abhors the keeping of humans as livestock and fiercely relegates it to dark undeworlds and barbaric fringes. It also abhors the half-slavery of ancient peasants who's legal rights were subordinate to their lords. It difficult to see how these are not moral improvements. Indeed even the more modern rights revolutions fighting various quarter-, eigth- and sixteenth-slaveries have been mostly on target.

Societies don't advance morally through just learning moral facts. But the value of individual liberty is indeed a learned moral fact. Sages have for millennia tried to define justice, and that'll be an ongoing project for millennia yet. But by 1776 the American revolutionaries could contribute the non-obvious discover that governments are instituted to secure the rights of the people.

This is just one gem in the glittering treasury of liberal modernity. Knowledge is lost through disuse, so the way to keep these treasures is to take off the blinkers of their denial.