d4nt

I agree, chronological feeds of people you’ve explicitly chosen to follow are fine, but AIs looking to optimize engagement have caused untold damage to society. Future generations will study it as an example of unintended consequences in AI systems. The sooner we shutdown this disastrous technology the better.

Like this: https://openid.net/certification/about-conformance-suite/

I think they’re on to something, but the solution needs more work. Sometimes it’s not just individual engineers who are playing defence, it’s whole departments or whole companies that are set up around “don’t change anything, you might break it”. Then the company creates special “labs” teams to innovate.

To borrow a football term, sometimes company structure seems like it’s playing the “long ball” game. Everyone sitting back in defence, then the occasional hail mary long pass up to the opposite end. I would love to see a more well developed understanding within companies that certain teams, and the processes that they have are defensive, others are attacking, and others are “mid field”, i.e. they’re responsible for developing the foundations on which an attacking team can operate (e.g. longer term refactors, API design, filling in gaps in features that were built to a deadline). To win a game you need a good proportion of defence, mid field and attack, and a good interface between those three groups.

Maybe it is running really slow. Maybe it’s taken 100 “real world years” for me to write this comment.

I suppose that really depends on what the subject of the simulation is. If it’s you personally, it only needs to simulate other scientists telling you about quarks, the solar system and the latest JWST images. If it’s humanity in general then other civilisations are not required. Of course it might be that this is a giant quark simulation and things like gravity, planets, life and humanity are an interesting quirk that’s arisen recently.

Soon everyone exists in a social bubble thats 1000 times worse than any present day social media bubble. It becomes impossible to interact with another actual human because there’s no common frame of reference. Micro languages emerge. AIs seamlessly translate everything into your micro language.

Any IRL interaction between two humans becomes almost impossible. It’s like travelling from the US to Japan today (assuming you don’t speak Japanese), you’re reduced to pointing and google translate.

Because you just never interact with other real people, and real people often seem like pale shadows of the hyper real AIs we talk to, the concept of fellow human beings having a “soul” or deserving rights or dignity is eroded. Policy (and life in general) becomes more ruthlessly utilitarian. Eg climate change has screwed the people in Yemen. Meh. I probably don’t even hear about it. If I do I don’t register those people as important humans because most of the “people” I intact with every day are artificial. I regularly delete the ones I don’t like and generate more. The idea that I should care about how they _feel_ is a strange and alien concept to me.

“Subspecies” feels a bit othering. Humans have been eating meat since Homo erectus (about 2M years).

One problem with productivity as a measure is that it’s measured in $ over time, and $ is an aggregate measure of “how much work are people willing to put in to get this?”. If I spend an hour and charge you $100 for the output, my productivity is 100x more than if I spend an hour and charge you $1. Even if I produced the same thing.

So you could restate the problem as “our ability to make things people want using less and less time is flattening out into a S curve rather that increasing exponentially”. And I would argue that this is to be expected.

Take cars. Ultimately, what I _want_ is to get from point A to point B as quickly and safely as possible, and that’s not getting exponentially better. My 2020 car is way more refined and comfortable than a 1970s model would be, but it’s not an exponential improvement. It has four wheels, five seats, goes at 70mph (the national speed limit) just fine. The Sat-nav is more convenient than the A3 map I used to keep in the back, but it does the same job. Self driving might be a big leap forward. So would a teleport. But so far it’s an incremental improvement on 1970.

In many areas of life, the economy is basically bumping up against the limits of physics and my ability to consume. I have more food than I could ever eat, I have a very nice artificial cave to live in, I can control the temperature, there’s hot and cold clean water, there are no armies marching over the horizon, people generally don’t steal my stuff, I can summon the worlds best musicians and entertainers to perform for me in my home 24/7 if I want to.

There are some exceptions to this. My lifestyle is not evenly distributed to everyone on the planet, and I’d like it to be. My lifestyle currently produces CO2, and I’d like it not to. The big area for improvement is biomedical technology. I’d like to not get cancer, or a cold, or age at all really.

What else do I want? Well, in some ways _that_ has become the defining problem of this era, we don’t know. I want someone to love me, I want people to respect me, I want to feel that my life has meaning and purpose. The economy can attempt to ‘solve’ those problems for me, but it’s a poor substitute. From 1870-1970 we fixed the big problems, things have gotten trickier since then.

Some time ago I asked myself: why did gambling become so addictive? You always loose, but it’s so powerfully addictive the behaviour must be beneficial in some way. It occurred to me that you always loose when doing “artificial gambling” (casinos, horse races etc) but “natural gambling” is different. Should we explore this coastline, should we try and cultivate this plant, should we run this scientific experiment? These bets may not pay off individually, but for the overall population over time they always pay off. The natural world has ergodicity.

One of the first things that some reading on psychology gives you is the realisation that the “real self” is an illusion. You’re a bundle of competing drives and narratives. Even you don’t know why you do stuff most of the time, and you make up justifications after the fact. So if one way of looking at things makes you stay home and cry, while another leads to going out and making the world a better place, maybe training yourself to pick the latter interpretation is a good thing.

If he stopped the legislature doing something, there would be a “constitutional crisis” but whether or not he then lost the power to do that again would really depend on public opinion and how well judged his move was. I can imagine a scenario in which a corrupt and unpopular government is holding onto power by twisting the rules somehow, and the king says no. It would be chaos, but if the king had public opinion on his side it could work out well.

For me this is a strength of a constitutional monarchy over an appointed or elected but mostly ceremonial head of state. The people have a relationship with the king, they’ve grown up together and there’s a connection there that’s hard to define. In times of crisis that connection could play a key role.

Convert “$x” into “something Alice wants”. Alice doesn’t really want $x, she wants a new car, a house, a vacation, a new machine for her factory.

Now “They want to (add, prevent from decay) as many edges as possible” becomes “They want to encourage different parties to lend each other things they want”.

Until recently money was very cheap, so it wasn’t a choice between hiring an engineer OR a salesperson, just hire both.

I see so many similarities with the freelance web/mobile developer space. And having run a small dev shop (3-6 people) for 7 years I agree that the _only_ way forward is to hold your nerve, fire the shitty clients and only ever do work you can feel proud of. Yes, there will be tens of thousands of people on (the VFX equivalent of) upwork who charge 1/3 your price, and clients will put immense pressure on you because of it. But if you’re know for honesty and quality, the work will come.

As governmental systems go, having an unelected head of state is actually kind of smart. Their lack of democratic legitimacy means they don’t try to exercise any real power, leaving that to the parliamentary parties. But they represent the nation and the military swears loyalty to them, so they effectively have a one use “sorry, but they’ve gone too far, I’m dismissing this government” card they can play in case of absolute dire tyranny. There’s no way to fire them, it’s very hard to bribe them, they’re trained for the job from birth and there are backups in place in case they die. The system avoids many of the problems found with elected presidents (whether US or parliamentary style) while preserving all the benefits of democratic government. My only complaint is that the system for choosing the next one could be a bit more representative than “firstborn child of the current monarch”. But changing only that part of the system now without ruining it would be pretty hard.

A pay as you go model seems reasonable in principle, but the devil will be in the detail. In theory the information being gathered is public already. I could follow someone round the public highways if I wanted to. But having everyone’s movements in one big database obviously presents huge problems.

The UK has no written constitution, which has its advantages sometimes, but could be a problem here. What we have instead is institutions. So we must hope that one of them steps in and puts limits on how this can be used.

I’m reminded of this article: Why didn't electricity immediately change manufacturing? http://www.bbc.co.uk/news/business-40673694

In the same way that swapping steam engines for electric motors didn’t yield much benefit. Just swapping real meetings for zoom meetings feels like a backwards step. But the real benefits come when you _reorganise_ your production processes around the new technology. Sadly for many middle managers, their role is disappearing and their income is dependent on them not understanding this.

I’m a 42 year old male and have found it very hard, my whole life, to establish meaningful friendships with other men.

I have many acquaintances, I’m not shy or socially awkward. E.g. When I was running a business I would often go to business networking events alone, start conversations with people, establish a rapport and spend hours chatting, but all those interactions have essentially left me with one good friend.

I’ve often found it easier to establish friendships with women, but (being straight) they get complicated. Either I develop feelings, or they do, or there’s a suspicion from someone’s parter about the real nature of our relationship. It’s just too problematic.

I think the female “model” of friendships outlined in the abstract just makes more sense to me. “emotional support, intimacy, and useful social information” is what I want from a friendship.

I suspect there are other men in this position and that the dominant male “model” of friendship that we have (and which is outlined in this article) is more cultural than biological. But I have no proof. What do you think?

A simpler system might be to have dedicated upward and downward shafts, with change over points in the roof and basement. The cars all queue up to go up or down depending on which shaft they’re in.

The elevator cars couldn’t have cables of course, but they could be driven by electric motors on the cars themselves (powered by induction), driving a gear wheel that runs on a fixed track on the side of the shaft.

I've also had the occasional issue with serverless platforms. Not to do with Google indexing, but I had a similar sensation: "well now we're just completely stuck". When something doesn't work the way it should, the whole premise of "you don't have to think about servers" quickly becomes "no, you're not allowed to see the servers" and its horrible.

These days I try to make the case at work for targeting containers and using the cloud platforms' container hosting tools instead.

The awe and wonder I get from an old building comes from 3 things:

1. This building is beautiful. 2. How did they build this all that time ago? For example, the Great Pyramids, Stonehenge, or a large stone cathedral built 800 years ago. 3. The sense of unbroken continuous activity in this building from the time it was built until now. Which connects me to history and a sense that by being in this place I am a part of something bigger than myself. This is stronger if the type of activity in the building has stayed the same for all that time. E.g. Attending a service in an 800 year old Cathedral makes you feel awe.

#1 Is still relevant if the building is restored, but #2 isn't.

#3 is interesting. If a building was damaged but restored immediately then the sense of continuity might remain intact. But if a building was left unoccupied, or it's use has changed then this effect is lessened. For example, visiting a 500 year old pub is better than visiting a 500 year old building that was recently turned into a pub. Visiting a royal palace in England has some quality to it that visiting Versailles doesn't have because there is no longer a French monarchy. Stonehenge is very impressive (because of #2) and even though some groups have started using it for rituals, there is no real connection to the original builders so it lacks awe on the #3 dimension.

I think the point is that laws/regulations should always be followed. Not followed _unless it's cost effective to break them_. If they're simply setting aside money and then carrying on, or objecting to the fact that a fine is unusually expensive. Then they're not framing the regulations in the right way.

Having worked with a few large IT services companies, I’d question their definition of “productivity”. I suspect it’s related to Jira tickets completed or short term revenue (eg billable hours). I doubt it’s measured in terms of end user outcomes.

My experience of these large IT services companies is that their middle management is very focused on maximising profitability for the work their engineers are doing. To the point that they hold engineers back from doing the “right” thing. Eg “don’t add logging to this new feature - it wasn’t explicitly mentioned in the change request”. You could argue this correct from a purist capitalist perspective of course. But that just underlines my belief that outsourcing IT to outside services providers is always more expensive in the long run.

For me, the fact that remote working lessens the influence of middle management is a feature not a bug, even if it hurts the large IT Services provider business model.

This article is everything I dislike about the postmodern perspective. It seems to start from a position of: humanity is bad. Then adds: any promotion of space exploration must be an covert attempt by the people involved to "normalize" their agenda. It can't just be an overt promotion of their idea to spread humanity to other planets, because surely everyone agrees with me that humanity is evil. See! I used the word "colonize" - that was a clue. Because what we're doing is exactly like what the imperialist Europeans did.

I find it disturbing the way a trait that seems to be very common among humans is being framed as a problem.

External stimulus means you’re with other humans that you can cooperate with. That you’re sharing stories and learning. That you’re hunting for food. That you’re building a shelter. All these things are necessary for survival. It seems possible that being sociable and industrious might well have been bread into us at some point. And it’s probably the reason we aren’t like big cats in Africa, just hunting and then resting to conserve energy.

We should not be suggesting to people that their instincts are a moral failure to be cured.

Aside from other factors like better treatments, it seems logical that mutations of the virus that make it less severe and therefore less detectable would spread more easily. Social distancing could inadvertently be breeding the virus towards banality

Companies have processes and changing those processes is very difficult and risky for the managers involved. A new software provider really needs to fit round the existing processes if it wants to make a sale.

Buying one person, or one team, a tool like Adobe CS doesn’t really change any processes so it’s easy. But dropping a new system in that changes a process and everything changes.

“Helen from risk wants a new fraud detection system” is effectively the same as “Helen wants to review all our fraud detection policies and if anything changes and subsequently goes wrong it won’t just be Helen that gets fired”. There’s a huge pressure on the software provider to make their system support whatever unique, crazy or arcane processes the risk team were previously using, and support all existing interfaces into and out of the risk team, because nobody wants to sign off on any changes. So enterprise software has to be crazy flexible. And flexibility makes for bad UX.

Why is Rome sinking?