I also believe that there is an optimal level of immigration, which is relatively high, and not limited to the most skilled. Modern, first-world societies still have jobs which are relatively unpleasant and don't require much skill, for example in agriculture. Jobs that the natives don't want to do, but an immigrant would. So even the migration of unskilled labor can create those win-win situations. While it is extremely difficult to calculate what the optimal level of immigration is, it is at least theoretically obvious that this optimum must exist. At some level of emigration from country A to country B, country A will actually hurt from the loss of labor force, while country B won't be able to accommodate any more.

The European Union is currently in a migration crisis, mostly from about 700,000 people having migrated to Europe from Middle-East and African countries, often by boat over the Mediterranean Sea. The USA is likewise having a migration crisis, in their case mostly about 1.6 million people that came on foot from South or Central America over the Mexican border. In both cases, the real problem is not that migration exists, because if it didn't exist at all we would actually all be worse off. The real problem is the legal framework under which the migration happens, with a secondary problem being the level of migration, although for me it is hard to say how close or beyond the optimal it is.

To not be too politically correct here, there are a lot of shitty countries on this Earth. In most cases, a shitty country is both politically and economically shitty. For example, over 7 million people have emigrated from Venezuela over the past years, but it is impossible to say how many of them left for political reasons, and how many of them left for economic reasons. The United Nations Universal Declaration of Human Rights of 1948 and other subsequent treaties established that "Everyone has the right to seek and to enjoy in other countries asylum from persecution", and that is a good thing. But when there is a large number of people arriving at the southern borders of Europe or the USA, it is nearly impossible to determine which of them are actually persecuted in their home countries. I am certainly no friend of Maduro and his "Bolivarian Republic of Venezuela", but I do not think that they politically persecuted over 7 million people. According to Amnesty International, "between 240 and 310 people remained arbitrarily detained on political grounds" in Venezuela. That is bad, but still suggests that a large percentage of the people who fled Venezuela did it primarily for economic reasons. And why would those reasons be any less valid?

I think that the first world is shooting itself and others in the foot by basing its right to immigration on asylum from political persecution. Yes, political persecution is bad, but so is your children starving. And if you consider the kind of person who is actually most likely to get politically persecuted, they are probably a lot better educated and a lot better off than most economic migrants. If we base our right to immigration on compassionate grounds, why would we want to limit it to an elite fraction? What is the moral basis of saving the politically persecuted journalist, while sending back the starving agricultural worker and his family?

Processing of asylum claims in first-world countries are a mess, pretty much everywhere. The EU just agreed on a compromise that allows to lock up asylum seekers for longer. In most places the asylum seekers don't have basic human rights, like the right to work, or the right to free movement. And all that because the receiving countries are trying to make the impossible distinction between political and economic migrants. It is somewhat perverse that a person who wasn't politically persecuted in his home country but left for economic reasons ends up locked up in a camp in the country he migrated to, based on a right to freedom from political persecution. Much of the current crises in Europe and the US are basically those asylum prisons being full, which certainly isn't the right way to determine how much immigration we need. Asylum rights lead to very bad "solutions", like the US expulsing Venezuelans to Mexico, not Venezuela. It "saves" the US from getting it wrong, expulsing a Venezuelan to Venezuela, only to see that person get jailed or killed there for political reasons. But neither that nor a border wall actually solves the underlying problem, and causes a lot of problems in countries that have the bad luck of being geographically between the shitty countries people are fleeing from and rich countries people are fleeing to. Everybody can see that the current system isn't working, but nothing happens, even when right-wing parties are in government, who exploit the general dissatisfaction with the migration system for political gains.

At this point in time, rich countries setting generous numbers of allowed immigrations per year and deciding who to let in and who not on a lottery system would cause less harm than the current asylum-based system. At least the people who were let in could start working and be integrated immediately. Yes, expulsion of illegal immigrants is always going to be contentious and difficult, but unless you want to establish a right to global free movement, you will always have to deal with that problem. Regional rights of free movement work quite well in the European Union, and could be a model for other continents, giving actually politically persecuted people a place to go. Much of the current migration crises is due to a legal situation that is nebulous and impossible to correctly enforce. A much simpler system wouldn't be perfect, but still a lot better than what we have.

Enshittification is clearly a product of a Silicon Valley venture capitalism tech startup culture, in which these tech startups mostly created platforms that weren't profitable. Thus users get a lot of useful stuff for free, the platform grows exponentially, and after a few years of such growth the platform is so large that it becomes basically a monopoly. Then the venture capitalists want a return on their investment, and the platforms starts squeezing their customers and business partners. A lot of those platforms are in the entertainment industry. There is the whole "creator economy", with platforms like YouTube and Twitch, but also streaming platforms like Netflix, and gaming related platforms like Steam. The recent Unity debacle was another example of enshittification. Platforms are increasingly using their monopolistic powers to extract a larger percentage of our entertainment dollars.

Now I would like to zoom out, and have a look at those entertainment dollars. From 2020 to 2022, mostly due to Covid, the overall amount of these entertainment dollars, global spending of people on entertainment, has gone up a lot. 2023 and beyond is different. General inflation and a cost of living crisis is forcing people to spend a larger part of their money on necessities, and thus less on entertainment. The global housing crisis redistributes wealth from younger renters to older home owners, but the entertainment industry is mostly focused on the younger audience. And in-person experiences, like eating out with friends or traveling, are quickly growing back to their pre-Covid levels, taking entertainment dollars away from at-home entertainment channels.

That all adds up to a pretty ugly picture, economically speaking. On the one side entertainment companies saw immense growth due to Covid, and reacted by spending more. The average cost of making a triple-A movie or game has gone up dramatically. On the other side the overall sum of entertainment dollars, and especially that of at-home entertainment dollars not only has stopped growing, but is going down. Time spent playing games or watching streams at home is going down, and people need to watch their spending more. Too many too expensive entertainment products are chasing too few entertainment dollars from customers, and enshittification means that intermediaries are trying to get a larger piece of the pie too.

As a result we had a never-ending stream of news this year about movies and games failing to be profitable. Increased monetization of games is meeting larger resistance now, with actions like the review bombing of NBA2K24, or players opting out of "games as a service" models, with Diablo IV losing players much faster than previous games. That is leading to a vicious cycle, in which entertainment companies are making less money, and react with actions that make them even less popular, like price increases, releasing unfinished products, or Netflix cracking down on password sharing.

I mentioned in my previous post that I am watching the money I spend on subscriptions more closely, and that is all entertainment dollars. As part of that, I have also unsubscribed from most content creators I supported on Patreon, Twitch, or YouTube. As sad as it is, in the end the larger content providers like Netflix and Microsoft Game Pass are giving me more entertainment value for my dollars than supporting a single Twitch or YouTube channel. I suspect that I am not the only one in this situation, which doesn't bode well for the creator economy, even if the situation is certainly not their fault. The last creator on Twitch I supported, I went out of my way to first switch from my iPad, on which I usually watch the content, to my PC, because the same subscription on iPad costs 30% more due to the cut Apple takes.

My prediction for the future would be that the entertainment industry will need to do some shrinking, because of the current oversupply of entertainment products chasing a shrinking pool of entertainment dollars. We hear a lot of companies that put out a reliable stream of successful entertainment products, like Disney or Blizzard, now increasingly failing to do so. And we hear a lot of layoffs from various participants in the entertainment industry, from game studios to large tech companies (who are in my mind often a part of the entertainment industry). The consolidation will frequently be painful, and will often be perceived as unfair, with the "nicer" market participants losing out to the more aggressive ones. This isn't going to improve the quality or value for money of our entertainment products. Unless we are lucky and enshittification actually leads to some of the shittier platforms dying, as Doctorow predicts. The world would probably be a better place without Twitter/X and TikTok.

In the process of looking at Game Pass games that I might want to play, I installed some of them, but then didn't immediately get around to playing them. Today I tried to play one of these installed games, and it turned out that it had left Game Pass since I installed it, and I wasn't able to play it without paying extra anymore. The funny thing that prompted me to make a post about it was that the name of the game that had left Game Pass is Before We Leave. Yeah, I should have played this before it left.

Now by all accounts Before We Leave is a pretty average city-builder, and if I can't play it for free, I'm not going to buy it instead. No biggie. But it made me more aware of the disadvantage of Game Pass, that there are quite a lot of games leaving the service every month. Somebody on Reddit compiled a master list for all Game Pass games, and it has 11 games coming soon, over 500 games currently active (either for XBox, PC, or both), and over 700 games that have left the platform. While Steam also has 783 delisted games, that is of course from a much, much larger library of games, and in most cases you can still install and play a delisted game on Steam if you bought it before it got removed.

Before We Leave was on Game Pass for 17 months, and it seems the median duration for a game on Game Pass is around 1 year. Not counting the games that are on the service permanently, usually because they are Microsoft games. No Age of Empires game has ever been removed from Game Pass. I think the lesson here for me is that I need to check the games on Game Pass more frequently, and then play any game I am interested in immediately. Having said that, with the 20 hours I put into Starfield I am now up to 60 hours played in 2023, and the upcoming League of Lamplighters alone is likely to get me over the 120 hours total. It was easier in previous years, where for example I played Humankind for over 70 hours alone. 2023 is a weird year for gaming, with a larger than usual number of high-profile releases. Which doesn't seem to be sustainable, but that would be the subject of another post.

The good news is that The Lamplighters League is coming out Tuesday, and it will be on Game Pass, thus available at no extra cost to me. The demo was only available on Steam, so I played that for the duration of the tutorial, which is all there is in the demo. Again the user interface could be better: In a real-time / turn-based tactical game, the decision to not include a quick save, but require a longer series of clicks to save your game is incomprehensible. And I find it annoying that I have to switch to recon mode whenever I want to move the camera. But other than that, The Lamplighters League is far more my kind of game: The tactical combat is quite interesting, I do like the art style, and the setting, which is somewhere between Lovecraftian and Indiana Jones, is also rather fresh. I'm looking forward to playing the full version starting on Tuesday.

The biggest success here was that within a day of playing, I managed to reach level 25 with xp only gained from the tutorial and from crafting, no quests, no combat. I managed to build a series of outposts, all linked together to deliver Copper, Nickel, Cobalt, and Beryllium to the same outpost. There I crafted thousands of Com Relays. Unfortunately, even with putting skill points in commerce, each Com Relay is only worth 20 credits. So I was making tens of thousands, which is okay, but short of what for example a good spaceship would cost. And vendors only have 5k cash on hand, so selling more involves sitting down somewhere and waiting.

What wasn't so good was the general game experience in this. The back and forth between outposts involved a lot of fast travel, and the constant loading screens really drove home the lack of a sense of space. Moving from one quarter of New Atlantis to another with the metro was not different than moving from one star system to another in your space ship. The outpost building and management interface is pretty horrible, and not very user friendly. I was constantly reminded of other games that involve colony / base building, which work a lot better.

In the end, that is a bit the problem with Starfield. I had thought Starfield would be hurt by the comparison with Baldur's Gate 3, but lately it is suffering a lot more by the comparison with Cyberpunk 2077 2.0, which a lot of people are playing and streaming now. You can watch two streams side by side, a Starfield stream about somebody questing in Neon, and a Cyberpunk 2077 2.0 stream, and the gameplay will be very similar. But Cyberpunk looks better, and plays better, which is probably why the Phantom Liberty DLC is getting better review scores than Starfield. And that is in spite of Starfield having space combat and outpost building, which Cyberpunk doesn't have.

Starfield is the everything game. You can do a *lot* of different activities, and it is relatively open. But for each of the activities you can do, you probably know a game that does this specific thing better. Of course there are some advantages, some synergies between the different parts of the game. My industrial empire led to me doing a bunch of research, and giving me access to crafted weapon and armor mods. But at the same time, the different parts of the game get in the way of each other, for example by competing over inventory space. So it isn't obvious that the game that does everything in a mediocre way is more fun than playing two or three games that together do all of these things much better.

It turns out that this Active Search feature also exists in the PC version of the game, but there is no keybind for it. You can't use it with mouse and keyboard. You need to attach a gamepad to your PC and then press A (for XBox controllers) or X (for PS controllers) and keep it pressed. Then a search circle will appear around your character, and the found items will appear on the left side of the screen.

Unfortunately when playing on PC, that isn't very practical. Not only do you need an additional input device, but using the controller and the active search completely switches the UI from the usual PC UI to the very different console UI. And you can't use mouse and keyboard on that UI, you would have to do all your searching of items and looting with the gamepad, and then switch back to mouse and keyboard. There are a few places in the game where there are a *lot* of small items, and if you want to loot them all, the Active Search function might be worth it. But most of the time the switch to the different UI is too much of a hassle.

Now by itself that story wasn’t very surprising. This is in parallel with the rise of online services like OnlyFans, and a growing social acceptance. A congressional candidate in the 2022 election was on OnlyFans, and other politicians at various levels have been found to have sold sexual images for money. In the UK tabloid press I have seen a shift, where stories now present examples of the big money ordinary people can earn on OnlyFans in a rather positive light. There are also growing “sex work is work” social trends, although somewhat surprisingly the resistance against that these days comes less from conservatives, and more from progressives, who have a really hard time believing that women aren’t necessarily victims in sex work. In another surprising twist of fate, porn on the internet these days isn’t regulated by governments, but by Visa and Mastercard.

What was surprising to me was the scale of sugar dating. The largest site, previously called “Seeking Arrangement”, now rebranded to just Seeking.com and promoting “luxury dating” instead of sugar dating, has 40 million users, with more women than men. As opposed to sites like Tinder where men outnumber women up to 4:1, Seeking.com at one point had a ratio of 1:3. I didn’t push my research as far as signing up for the site, but the secret appears to be that Seeking.com asks men for their net worth and there are different levels of verification. Pay $250 for a premium account, and get your net worth fully verified, so the woman looking for a millionaire can be sure that he is the real deal. According to statistics the site published before rebranding, the average sugar daddy was 38, the average sugar baby 25, and the average monthly payment from sugar daddy to sugar baby was $2,500. The sugar babies wouldn’t call themselves sex workers, but at the end of the day an older man or woman is paying their rent plus extras, and that probably isn’t just for holding hands.

For comparison, Germany with it’s legal and regulated prostitution had 40,000 prostitutes before the pandemic, which went down to under 30,000 since. The millions of sugar babies on Seeking.com and other sugar dating websites seem to be of a much higher scale. While different people will have different moral ideas about sex work, I would say that economically the spread of sugar babes is not a good sign. It shows that the sugar babies don’t make enough money to live comfortably otherwise, while the sugar daddies (or mommies) have more than enough money for two. Another sign and consequence of rising inequality.

When looking at loot boxes, many people look at the biggest possible prize you can win, and how much money it would cost to have a good chance of winning it. I am mostly looking at the other end: I assume that most of the time I’ll get a bad or average pull. So I am mostly interested in the low end rewards. What do you get if you don’t get lucky? Although I am not playing it anymore, I have to give a honorable mention to World of Tanks here, a game that at least when I played it only had loot boxes for Christmas. What was absolutely great about these is that the worst possible prize you could get was gold, and buying a loot box and “losing” would net you more gold than if you had bought the gold directly. Also, besides the big prize of some special tank, which wasn’t overly rare, there were a ton of mid-level rewards, most of which where quite useful. So you never opened a bunch of loot boxes and came away completely with nothing.

Honkai: Star Rail and Genshin Impact have the absolute opposite system: There is not much variety, and you either get a 4- or 5-star character or weapon that you probably wanted, or you get a 3-star weapon (called Light Cone in Honkai: Star Rail) which you absolutely didn’t want. You can use these 3-star weapons to boost the level of the good weapons, but it’s by not much. And although the game is full of other interesting materials and rewards you might be interested in, you simply never get any of those. Either you luck out, or you get something that is nearly completely worthless. And while there is a “pity system” that guarantees a better reward every X pulls, even that system gets worse over time, because pulling the same character a second time just gives you a minor boost. The more characters you have, the less interesting loot boxes become.

Now of course even back in the days of early Magic: The Gathering the opening of boosters had diminishing returns. But at least with every new set you could open up a box or more of boosters and get common and uncommon cards you could well use for deckbuilding, the rares weren’t the only useful thing in these packs. And that is what good loot boxes do: As long as you open only a reasonable number of them, you always get something useful. In Honkai: Star Rail I did spend some money on a battle pass and supply pass. But already with the free tickets for loot boxes the game gives you, you end up with mostly worthless 3-star weapons. That doesn’t encourage me to spend money on loot boxes at all, because I don’t believe in luck. And in these games you either get extremely lucky, or you just completely lose out. There are no consolation prizes.

The biggest examples were X/Twitter and Reddit making API access not free anymore. In the context of the rise of large language model AI, giving all your content away for free in a format easily readable by a computer wasn’t a good idea. And handling millions of API requests isn’t free. With the currently still unresolved Unity runtime fee story, I wonder if there isn’t a bit of this here as well. If Unity wants to charge per install, I at least guess that the installation of Unity Runtime is using servers of Unity Technologies, and that somebody in the company figured out that in the millions these installs cost Unity something. That doesn’t excuse the horrible execution of the move, the excessive pricing, and the bad communication. But I am not excluding the possibility that these installs aren’t really free, and that the company is unwilling to subsidize them in the future.

Imagine you had to pay 1 cent per e-mail you. That wouldn’t really impact any regular user. But it would make the business model of spam a lot less attractive. This is a typical example of possible negative examples of giving something out for free. It creates wrong incentives, especially at large scale. A million times 1 cent is serious money, but a million times free is still free. The whole capitalist model and free market mechanisms stop working at “free”. This hit home hard this year when people discovered how “free” data were feeding the AI boom. It is one thing to give content for free to a single user, but something completely different to give all of your data for free to an artificial intelligence that might use them to compete with you.

I do think the idea to hand out stuff for free only in limited quantities, but charging for the same stuff in large quantities has merit. The transition from “free” to “only free in small quantities” will have to be better managed and communicated. But the general argument of there being costs like server costs that add up in large quantities isn’t unreasonable.

Video game technology advances. Maybe not as fast anymore as it used to, but playing a game that is over 10 years old and didn’t get some major engine overhaul since then is likely to feel outdated. I’m not saying that Baldur’s Gate 1 and 2 weren’t great games, but you can’t play them followed by 3 without noticing the technological gap. And that is if the old game is still available: With live service games and MMOs there is a significant risk that instead of getting better over time, they get worse due to lack of other players, or downright get the servers shut down for not being profitable anymore. Another aspect is social media: Some 10-year old games still have a strong following, e.g. Rimworld. But others have completely disappeared from the public conscience, and you can’t find a Twitch streamer still covering them, and if you talk about the game on your social media there is no echo. My Steam library has 70% unplayed games, and that is actually not an unusual number; lots of people buy Steam games, decide for various reasons to wait before playing, and then never get around to it. And not every game actually gets good developer support over the years, some games get released half-finished and never get more than a few hotfixes.

The cult of the now has advantages when it comes to playing new games that are relevant to others, and thus it is easier to find some conversation about them on social media. But even highly praised games like Baldur’s Gate 3 have bugs and missing content in Act III, and will probably be a better experience once the definitive edition is released. I played Pathfinder: Kingmaker for 180 hours, then bought Pathfinder: Wrath of the Righteous and stopped playing after 37 hours. A combination of too much of the same sort of gameplay in series, and the newly released game being buggy, killed it for me. Wrath of the Righteous is now at patch 2.1.5, and presumably a lot better. But of course I wasted money by buying it on release. And 2023 is a year filled with great RPGs, which doesn’t help dealing with a backlog. A game like Starfield I want to play soon, when I come back from my holidays, because with subscription services like Gamepass I never know how long a game will be included in the subscription, games are leaving Gamepass every month. But I would also be interested in finally trying out Cyberpunk 2077, with the new 2.0 update.

Lacking a crystal ball, I don’t have a good answer for any given new game whether it would be better to play now, or wait at least a year or two until the game is actually in a finished state, and can probably be bought a lot cheaper. Usually I decide based on emotions, my current interest, but that isn’t always a good guide.

An NPC doesn’t have to have a full ChatGPT 4 large language model behind him to be believable. In fact, if the NPC is supposed to be a peasant in a medieval village, you probably don’t want him to be able to give his opinion about mobile phones, television shows, or modern day politicians. But currently, if you are looking for the village smith, and nobody thought to give the peasant a dialogue option to ask where the smith might be, the NPC is useless to you, and not very believable. It wouldn’t be too hard to give that NPC limited AI, enabling him to answer simple questions about the village and its inhabitants.

While there are some issues about the compensation of voice actors used to train text to speech software, that technology also has much advanced. Thus the NPC peasant could possibly even use a voice over generator to reply with audible speech, not just written dialogue. And for triple-A games with lots of NPCs, the advantage of simple chatbots and voice over generators is that they scale up better, and thus could end up being cheaper than getting the traditional and far more limited dialogue voiced over by voice actors.

The current state of the art in games is NPCs which are just dolls: Most of the are stationary or have very limited movement patterns, and most of them have just one, or at best a few, canned responses to being interacted with. That also leads to stupid strategies for gamers, who simply talk to every NPC in a settlement until their dialogue runs out, to be sure not to miss anything. How much cooler would it be if you couldn’t find out about the murdered traveller unless you actually asked about him?

While this example seems far fetched, something very similar just happened to a lot of game developers. Unity Technologies, makers of the Unity engine, which according to Google has a 29.4% market share among game engines, announced a new “runtime fee” of 20 cents every time a user downloaded and installed a game. X, formerly known as Twitter, exploded with devs pointing out specific cases, like Humble Bundle or Microsoft Xbox Game Pass or free-to-play games. There was some backpeddaling from Unity, and some proposed solutions that for a Game Pass game the fee would apply to Microsoft, not the game studio. At this point it became clear to me that this would end up with the lawyers. You try to send an unsolicited bill to Microsoft for a fee you just made up, and a Microsoft corporate lawyer is going to legally blast you to smithereens.

In other, related news, the share price of Unity Technologies dropped by 5% yesterday after the announcement of the runtime fee. Some angry people pointed out the publicly available information that the CEO and senior management of Unity Technologies had recently sold their shares, which might be considered insider trading. Meanwhile game studio Massive Monster, makers of the highly successful game Cult of the Lamb, made with Unity, announced they would delete their game from Steam and all other shops on January 1st. Epic Games, makers of the competing Unreal engine, pointed out that their royalties haven’t changed, and are just 5% of any revenues above $1 million, with revenues from the Epic Games Store not being counted. The proposed Unity runtime fee would already kick in at $200,000 revenue.

A point that caused little outrage yet, but seemed curious to me, is the idea that Unity would base their fee on their own tracking data, which is impossible to verify by the game developer who has to pay the fee. And the idea that because with every installation of the game the current Unity Runtime software is installed too, Unity is technically capable of preventing an installation in case of non-payment of their runtime fee bills. Imagine Unity sends that runtime fee bill to Microsoft, Microsoft refuses to pay, and suddenly nobody can install games running on Unity from Game Pass anymore. The resulting legal fallout would make Epic Games vs. Apple look like a minor skirmish. The more I think about it, the clearer it becomes that Unity Technologies will either be forced to scrap their “runtime fee” charge, or this is going to end up in court.