"So!" They said, "Any comments before we launch?"

No one said anything.

After half a minute the presenter said "As there are no objections, we'll proceed. Silence is consent."

At that phrase, my whole body did an involuntary spasm which I'm sure was caught on camera.

I know what they meant and, it some contexts, it's an understandable shortcut. It's common in a wedding ceremony to hear some form of "speak now or forever hold your peace".

But we should, in all our endeavours, be looking for enthusiastic consent. That means a cheerful, resounding, unambiguous, delighted "yes".

Or, as some people prefer to describe it, authentic consent. We want people to agree to our proposals - even if it is only transactional. They get something, we get something. Everyone is happy.

Anyway, on to AI bots.

Last week, my site started getting hammered by a rather aggressive bot which was downloading all my images and placing considerable strain on my server. I tracked down the open source project which was responsible for developing the tool. It was designed to bulk download images - regardless of copyright status - and use them for AI training.

In its README, it said that if I wanted to opt-out of having this particular tool scrape my images, I had to add some non-standard headers to my site. There was no way to opt-in to the bot.

It seems that every few years we have to have the same argument. Someone releases a tool on the Internet and claims that, if you don't want to participate, you must explicitly opt-out.

It isn't hard to see why that's an obnoxious idea. Thousands of tools are released every day. Am I expected to play whack-a-mole and shut down every new one that appears? That is a perverse way to expect people to behave. These bots cost people time and money without offering any tangible benefit.

By analogy, I can't suddenly declare that everyone online has to pay me £5 - oh, you don't want to? Sorry, you should have opted-out last month. That'll be a fiver, please.

Informed consent is key! It is untenable and unethical to expect people to acquiesce to your demands without even giving them the courtesy of informing them of your intentions. Those of a literary persuasion will remember Adams' Parable of The Leopard.

Speaking of ethics - this gets more tangled. I told the bot-creator that I thought their tool should be opt-in. They said that was "unethical". I didn't understand that point - consent is ethical - but their response baffled me.

Letting a small minority prevent the large majority from sharing their images and from having the benefit of last gen AI tool would definitely be unethical yes. Consent is obviously not unethical. You can give your consent for anything if you wish. It seems you're trying to decide for million of other people without asking them for their consent.

Me asking for a tool to respect the consent of users is, apparently, an ethical crime against people who might benefit from whatever the tool could create in the future. That's some "For The Greater Good" shit!

Perhaps I shouldn't stop the trolley from running over the orphans because one of them might cause a traffic jam in the future...

In the UK, people are not compelled to donate their blood. Nor are we compelled to receive life-saving blood transfusions if we don't consent.

Different cultures have different ethics and one of the problems with the Internet is that we occasionally butt up against different norms of behaviour. But I think, in all cultures, you can't just start grabbing a person's stuff and then say "but you didn't explicitly say that I couldn't! Besides, it's for your own good!!"

Look - this has been a bit of a ramble - but we have to play nice with each other while we're here. Ripping off people's content because you hope that they will see the utility of what you build just isn't polite. If you think people will benefit from something, then you need to make the positive case to enable them to give their enthusiastic consent.

Consent is one of the bedrocks of ethics. And silence is most definitely not consent.

It isn't "effective altruism" if you have to force people to comply with you.

Further discussion on Vice / Motherboard and HackerNews.

This means readers of my post can see where it has been mentioned around the web. They can read your article after reading mine. Nice!

I've also set up a "bridge" service which looks for people posting comments about my work on social media. For example, if you post a link to my blog on Twitter - or reply to someone who has shared a link - I get a notification. That means if I think it is an interesting comment, I can publish it in the comment section. It usually looks something like this:

This means readers of my post can see your Twitter or Mastodon comments. They're identified by the logo. Users can go to Twitter or Mastodon to reply to you. Nice!

Everyone is one big happy family, no matter where on the web you are.

Or so I thought. There are a few drawbacks with this system.

I didn't write that!

I had one reader complain that someone else was impersonating them in the comments of my posts. It wasn't immediately clear to them that I'd syndicated their comment and reposted it. After that, I added the Twitter logo to make it a bit more obvious. But many people still find it unintuitive that content can be replicated outside of its original publication.

I deleted that!

I had another reader who automatically deletes their Tweets every month. My blog retains a copy of their Tweets because it doesn't check for deletions. This might be against the user's wishes - especially if they had posted something inappropriate. Twitter doesn't send a "deleted" notification to services which have stored Tweets - and it would be impractical to periodically check every single Tweet I have stored. The reader was ambivalent about whether they should be kept on my blog.

I'm not that person any more!

One of the comments was a person who had changed their name & Twitter avatar. Understandably, they weren't happy about still being referred to by their deadname on my site. Again, neither Twitter nor the bridging service notifies me when a user changes their name or avatar. Naturally I deleted the comments when they contacted me.

I have the copyright to that!

An overly aggressive person was furious that I'd copied their © content onto my blog without permission. Personally, I thought that adding their 7 word reply was covered by fair-dealing, but I didn't fancy pissing them off. So I deleted it. If I'd embedded directly from Twitter it would have been fair game - but some people feel there's a material difference between embedding and copying.

Now what?

This is a complicated problem. I want to see what people are writing in public about my posts. I also want to direct people to the conversations which are happening elsewhere on the web. But people - quite rightly - might not want their content permanently stored by my site.

So I think I have a few options.

0. Do nothing. My site; my rules. If you don't want me to grab your hot takes, don't post them in public. (Feels a bit rude, TBQH.)
1. Be reactive. If someone asks me to remove their content, do so. (But, of course, how will they know I've made a copy?)
2. Stop syndicating comments. (I don't wanna!)
3. Replace the verbatim comments with a link saying "Fred mentioned this article on Twitter" . (A bit of a disruptive experience for readers.)
4. Use oEmbed to capture the user's comment and dynamically load it from the 3rd party site. That would update automatically if the user changes their name or deleted the comment. (A massive faff to set up.)

What do you think I should do?

‘Understanding how the zeros and ones increasingly influence and control our lives is critical to understanding how we can reciprocate influence and control back onto those zeros and ones.’ This vital book is a call to action: to reduce online harm, to protect the integrity of our digital lives and to uphold democratic participation and inclusion. A diverse group of contributors reveal the hidden impacts of technology on society and on individuals, exploring policy change and personal action to keep the internet a force for good. These voices arrive at a crucial juncture in our relationship to fast-evolving technologies.

Possibly the most important and timely book I've read about the state of Digital Policy. Written in the aftermath of the horrific fascist/terrorist attacks on New Zealand - and in the early stages of the COVID-19 crisis - this book examines some of the knotty problems facing Aotearoa.

This is not a parochial, inward facing text. All the problems it identifies are applicable worldwide. It presents a compelling case on how we should construct digital policy as we emerge from the shadow of COVID.

The section on Indigenous People's data rights, and the colonialism present in the current infrastructure, has lots of Te Reo (Māori language) phrases. That can be a bit intimidating for a newcomer - but it's nothing that a few minutes with a decent dictionary can't fix.

There were two particularly interesting questions that I was left with.

1. What's the carbon footprint of, say, GOV.UK? Do we make enough use of renewable energy? Can we schedule computationally heavy tasks for times when green energy is at its peak?
2. How do we make the language we use more inclusive? We need to take people on a journey, and if they can't understand our acronyms and jargon, we lose them.

The book is available DRM free from https://www.bwb.co.nz/

Is that sensible? 13 years later and I'm still holding a grudge. Is that healthy? It it useful?

I was reading a discussion on Microsoft aquiring GitHub - one of the commentors didn't understand why so many people were upset by the news, saying:

The idea that some companies must be forever tainted by their misdeads in the past seems odd to me.

I'm sure there's a company who screwed you over personally in the past - whether they messed up an important order, or overcharged you for fries - you've sworn never to do business with them again.

Let's move this to a less rational, more human domain. Would you invite your childhood bully to your wedding? You see a job application from the boss who fired you, do you toss it? The kid who mugged you for your wallet wants to stay in your AirBnB, do you let them?

Humans find forgiveness hard. We have rules on punishment (prison or fines) and rules on forgiveness (parole, spent convictions) - but it doesn't stop us from behaving in a human fashion.

How do you, personally, punish a company? How do you materialise your desire for justice?

In capitalism, the only effective protest you have is to withraw your capital.

Let me be clear; I like capitalism. I'm a landlord, investor, shareholder, and tax payer. But I'm under no illusion that it is a perfect system. It massively favours those with huge capital resources - especially when it comes to changing behaviour.

I get to choose where I spend my money. That is the essense of capitalism. I can choose the cheapest product, or I can choose the more expensive one with the longer lifespan, or I can decide based on the logo, or I can buy the thing which makes me happiest.

Giving money to bad people makes me sad. I don't want to feel sad. So I don't spend my money on people who test their products on animals. I withold capital from organisations which have given me poor service in the past.

Perhaps you don't give money to an organisation which disrespects your religion. Or one which donated to a political cause you find abhorent. That's OK too.

Perhaps Sony has learned its lesson. Perhaps Microsoft really is a friend to Open Source. Perhaps Nestle has stopped promoting its baby formula to vulnerable parents. Perhaps the leopard has changed its spots.

Perhaps not. How long would you wait to be sure?

In the early half of the twentieth century, certain physicists made breakthroughs in relativity, quantum mechanics, and nuclear energy. Many of these scientists were Jewish. The Nazis called these heretical ideas "Jewish Science" and suppressed their teaching.

Jewish physicists based in Germany fled the oncoming war. Many ended up in the USA where they worked on the Manhattan Project to develop nuclear weapons. The Nazis had caused such a "brain-drain" of expertise that it critically hampered their ability to wage atomic warfare.

It has long fascinated me that a culture expelled the set of people which could have saved it.

I'm going to tell you an anecdote which is a gross oversimplification, and is an unfair comparison.

In the early part of the twenty-first century, many people working in the fledgeling Internet industry started making noise about privacy, security, and ethics. The mainstream technologists called them fearmongers, idealists, and anti-business. Their ideas were unwelcome and they were thrown out of both the cathedral and the bazaar.

Many retreated to academia, some stayed and tried to cultivate a sense of responsibility in the industry, a few started lobbying governments around the world. By the time trust in the existing structures had begun to collapse, there were too few privacy-focused employees left to reverse the damage.

By expelling the boring and pessimistic doomsayers, the Internet behemoths had sowed the seeds of their own destruction.

All analogies break down eventually, and all simplifications obscure the truth. But there is an undeniable fact - Internet companies could have prevented their current difficulties if they had baked in privacy from the start. If they cared about their users' security. If they acted in an ethical manner.

But programmers want to concentrate on fun and exciting things, they don't want to be depressed by "experts" telling them they are acting irresponsibly.

In 1907, seventy-five people died when the Quebec Bridge collapsed. Ever since that day, Canadian engineers have worn an iron ring on their finger. Forged from the remnants of that bridge, it serves as a constant reminder that an engineer holds life in their hands. Mistakes can be deadly.

The computer industry has nothing like that. We have voluntary codes, which are mostly ignored. Programmers who commit blunders can shrug off responsibility. They face no professional sanctions and are sometimes lauded for their recklessness.

Indeed, one of our most sacred text proudly disclaims the very notion of a programmer being responsible for anything their code does:

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE

Everyone who loudly and publicly complained about the lack of privacy on the modern web was eventually proven right. Those who were initially dismissed as tinfoil-hat-wearing paranoid freaks, now have the grim satisfaction of being able to say "I told you so!"

The security experts who screamed their heads off about the gaping holes in consumer devices are modern day Cassandras.

I doubt that "Web 2.0" is facing irreversible collapse. But I also doubt that people who raise issues of ethics will be dismissed quite so casually in the future.