Monday, August 30, 2021

I Will Not Stand For This Gross Injustice

When I was a kid I fell in love with J. R. R. Tolkien, devoured all his works, and then moved on to other fantasy authors and series, hungry for more. That's a little like how I am now with books about financial inequality. I adored the two massive Piketty books, but I need more - more! - so I must go further afield to get my fix.


But not all that far afield. The Triumph of Injustice is written by Emmanuel Saez and Gabriel Zucman, whose names are familiar to Piketty-heads: they've co-authored various research papers, in various combinations between the three of them, and are major contributors to, a massive online database tracking various financial statistics around the globe. Saez and Zucman are both professors of economics at UC Berkeley, practically in my back yard; both are originally from France, and Saez has become an American citizen.

The Piketty books have very broad scopes, looking at the entire world over centuries. The Triumph of Injustice is focused on America: every once in a while it will note where America differs from other large and wealthy countries, but otherwise it is entirely focused within our borders. It is temporally more focused as well, looking at what happened in two particular time periods: the New Deal regime from 1932-1980 and then everything that has happened from 1980 to the present.

That focus has a couple of benefits. First of all, a much shorter book, at just under 200 pages compared to the 700 of Capital in the Twenty-first Century or the 1000 of Capital and Ideology. Second, it can be much more specific: where the Piketty books look at overall trends and correlations, TToI looks at what specific laws were passed, what actions Congress or the President took. Or, more often, what actions they did not take. Choosing to not take an action is itself a choice, those choices have enormous consequences.

Saez and Zucman identify a recurring pattern that has been played over and over again to collapse tax bases. First, the administration stops vigorously enforcing tax laws. This causes evasion to proliferate and revenues to drop. Then they "solve" the problem by cutting tax rates, arguing that the evasion flourishes because of expensive and complicated taxes. The whole system resets to a lower rate, the wealthy get much wealthier due to paying fewer taxes, and move on to finding more ways to evade more taxes.

You often hear a distinction drawn between "tax evasion", which is illegal acts like not reporting income, and "tax avoidance", which is taking legal steps to reduce your tax bill like transferring money to a charity you control or to an overseas subsidiary. The authors argue that there is not a meaningful difference between evasion and avoidance. Our law includes a set of provisions called the "economic substance doctrine": in essence, this means that any action you take specifically to avoid taxation is illegal. Investing in an overseas company is of course legal; but if you are investing specifically to dodge US taxes, then that is illegal.

The economic substance doctrine underlies their argument that overall the problem with tax cheats is one of political will and not of inadequate laws. During the New Deal era, the IRS would aggressively prosecute fraudulent tax schemes, the President would publicly chastise misbehaving companies, Congress would hold hearings to shame the worst offenders. And it worked, over and over again, that the spirit of the law was upheld. For example, in the late 1960s it became popular to "donate" money to "charities" that exclusively employed the donor's family members and undertook projects to benefit the donor's lifestyle. Within a short time, the the people running these schemes got dragged into court by angry feds, and then very suddenly rich people stopped using those fake charities and resumed paying the taxes they owed.

In contrast, the Reagan team believed "taxation is theft", which meant they were not at all disposed to put pressure on rich people to pay their taxes. The IRS is nominally independent, but over the long run they are receptive to the priorities and pressures of the federal government, and they have been under enormous pressure to leave wealthy tax cheats alone. And so, when "tax shelters" began to arise in the 1980s, the administration let them flourish. Instead of fighting the fraud, they used the existence of tax shelters to argue for and eventually succeed in slashing the top income tax rates. I'm reminded of Elizabeth Warren's saying that "personnel is policy". For the most part, we have the right laws to combat cheats, and they've been on the books for a century. We just need people who are willing to enforce them.

This book includes some things that I already knew - I wasn't exactly shocked to hear that Reagan was a turning point in American inequality - but I also learned a lot of new stuff. One big thing is concretely how overseas tax shelters work; I've been hearing about them for years but have always been fuzzy on exactly what they are and what they do. Now I know! As an illustrative example, just before Google incorporated and went public, it created a subsidiary in Bermuda and "sold" that subsidiary their core Search intellectual property: the algorithms and concepts and methods behind their technology. Now, in 2018, Google earns, say, $23 billion in the US, mostly by hiring US engineers who build products used in the US to sell US advertising to US consumers. Google will then pay its Bermuda subsidiary $23 billion for the rights to use their Search IP. So Google ends up with a net profit of $0 in the US, resulting in a total US tax of $0. It does owe tax to Bermuda for the $23 billion in profit there. What's the corporate tax rate in Bermuda? 0%. So it pays $0 in tax there as well.

In the last couple of years, there's been more awareness of the leeching effect of high-tech companies, particularly the newer "sharing economy" ones: more people and economists and politicians recognize that what seems like big innovations in efficiency are mostly or entirely built around tax-dodging: the "innovation" is mostly finding ways to avoid paying benefits to your workers, or sales taxes to your state, or supporting the public infrastructure of your cities. Reading TToI makes me think that the problem is even older and deeper than these companies. "Blue-chip" tech companies like Alphabet and Apple are titans of Wall Street with massive market capitalizations, largely because they seem much more profitable than the boring stodgy old industrial companies. But, would Alphabet and Apple still look so great if they paid the taxes they owed? Would GM become a beloved company if they had a subsidiary in the Cayman Islands to which they paid $10 billion for the intellectual property of "having a motor that turns four wheels and pushes a vehicle forward"? Modern companies have the advantage of being born in an "anything goes" era of lax enforcement, making it trivially easy for them to dodge paying their fair share of the tax burden that older companies cannot so easily shrug off.

Another thing Saez and Zucman write about that I hadn't previously deeply considered is corporate profits. First of all, they make the (very reasonable) argument that every dollar can ultimately be traced back to a real flesh-and-blood person who owns it. That dollar might be on the books of some company, that's owned by a conglomerate, that's owned by a private trust, that's controlled by a family, but ultimately some real person owns that dollar. With this principle in mind, they have a different view of income from what I (and I think most people) usually think of. As they see it, if you own stocks in a company, and that company makes money but doesn't distribute the profits in dividends, then you still own the increased value in the company. Or, to put it in other words, if your net work increased by $1,000 over the last year due to an increase in the value of shares of stock that you own, then you had $1,000 of income, even if you have not yet sold the stock: that money is still yours, just sitting on the corporation's ledger instead of yours.

That might seem kind of esoteric, but it turns out to have really profound impacts on the very, very wealthiest people in the country. The vast majority of us don't really have a choice in how we realize income: we get a paycheck and a W-2, full stop. But wealthy people can pick and choose whether they hold their money in a company or receive it as pay. Many of the wealthiest people, like Mark Zuckerberg and Jeff Bezos, would not be impacted at all if we drastically raised the top marginal tax rate to, say, 90%. Why not? Because they don't earn much regular income that's subject to the income tax. It's the "$1 salary" that Steve Jobs made famous, which shifts your income from labor to capital. The ultimate effect is that you can become personally worth billions of dollars without paying tax on your billions.

Theoretically, that money will eventually get taxed: as a capital gain (at a far lower rate than regular income), or as a sales tax (if you, like, buy a corporate jet through your company), or as an estate tax for your heirs. But, again unlike regular wage earners, you get to pick when and how to pay your tax. Of course, this opens up the door for all sorts of shenanigans to obscure your wealth, to hide it from taxation, to claim paper losses or shift it overseas or otherwise break the law.

So, big-picture, the upshot of the first part of the book is basically:

  • In America, we used to have robber barons and incredibly wealthy and powerful people who ran roughshod over our democracy.
  • Starting in the 1930s, we established extremely high progressive tax rates and vigorously enforced them, while simultaneously investing in social programs, education and infrastructure. As a result, growth was more equally spread throughout the entire income spectrum, with the incomes of working class and middle class families rising at about the same rate as the wealthy.
  • Starting in 1980, a series of policy changes have removed the high tax rates on the wealthiest people while also making it easier than ever to avoid paying even their lower taxes. As a result, the incomes of the bottom 50% of the population have flatlined while the wealthiest have reclaimed their Gilded Age status.

But they don't end on this doom-and-gloom note. They remind us that it wasn't that long ago that we had a different tax policy, and different outcomes, so we know that there are possible solutions to our problems. Of course, the world of 2020 is different from the world of 1930, and some of the new challenges, like globalization, call for new solutions. Saez and Zucman close with a variety of concrete steps we can take to make life more fair here. It's thrilling to realize that at least one of their proposals is already coming to fruition: Janet Yellen, our new Treasury secretary, is working with other major economies to establish a global minimum corporate tax rate, which will be hugely instrumental in curbing the use of offshore tax havens like Ireland and Bermuda. The authors also propose a wealth tax, with some fairly stark differences from what Piketty and Warren have proposed: the Saez/Zucman wealth tax wouldn't start to kick in until much higher wealth levels, and would impose higher rates.

Why is that? Well, channeling Bernie Sanders, late in the book they start to question whether we should even have billionaires. Going back to the Laffer curve, they crunch some numbers and come up with an "optimal" tax rate to extract the most money from the wealthiest citizens: any higher and those people will choose not to work, any less and we collect less potential revenue. This works out to a top marginal income tax rate of about 75%, which results in an overall average tax rate of 60% across the top 1%. But, they offer, there might be social and democratic reasons why we want to go higher than that, with a goal not so much of maximizing government revenue as a goal of reducing the power and influence of billionaires.

Money is power. The wealthiest Americans have enormous resources that can pose huge danger to our republic: tying up important reforms in endless court litigation, funding astroturf campaigns to stoke media coverage, outright buying media companies to shift public opinion, hiring armies of lobbyists to block necessary legislation, using accountants and lawyers to devise new schemes of tax evasion, funding the campaigns of sympathetic political candidates, and on and on. We can try to beef up our laws and enforcement agencies to block these attacks. Or... we can just take away their power. We did it before! Small is beautiful, whether you're a company or a personal fortune.

Elizabeth Warren's wealth tax is kind of designed as a permanent wealth tax, with a slightly broader base and less punishing at the top. It could generate a lot of revenue over multiple generations and use that revenue to fund long-running worthy programs. That isn't really the goal of Saez and Zucman's wealth tax. It's more like the 90% marginal income tax rates of the mid-century, or a carbon tax: it's a tax to perform a specific social good by discouraging bad behavior, with the goal of eventually collecting no money at all.

Like Piketty, Saez and Zucman stress that this is only one of many ways, and there are lots of good ideas out there. But it's crucially important that it's a democratic debate, not an oligarchic one. We get to choose our leaders, choose our laws, choose what sort of society we want to live in. We now have 40 years of hard evidence from our experiment with Reaganomics. The result is dismal. It's long past time to change course.

Wednesday, August 25, 2021


I continue to get a better grasp of  Roberto Bolaño's books as I read more of them. After reading the first one, I was like, "This is what he writes like!" After reading a few more, I was like, "Wow, all of his books are so different from each other!" And now I'm like "Oh, but there's definitely some recurring stylistic choices and themes across most of these books!"

"Distant Star" is a particular case in that it's a retelling of an existing story. As he (or, well, his fictional narrator) notes in the introduction, this was previously published as the final installment of "Nazi Literature in the Americas". That story has now been extracted and retold as a stand-alone novella. It's been a while since I read NLitA, but most of this book felt familiar to me, and it's hard to remember what beats and scenes are new and which are from the original story.


Looking back over my previous post, I see that the names of all the characters have changed: Carlos Wieder aka Alberto Ruiz-Tagle was previously Ramirez Hoffman aka Emilio Stevens. The narrator of the older work was named Bolaño, this one is anonymous. And reviewing that old post helped me solve another mystery about the sisters: here they're the Garimendi sisters, there they were the Venegas sisters. I'm not pretty sure that there's another pair of identical twin poets in another of his novels (probably The Savage Detectives?), which both of those stories have reminded me of.


I'm curious about the reason for and the meaning of changing these names, especially after specifically referencing NLitA in the introduction. In both novels, the narrator seems like a novelized version of Roberto: a student socialist poet imprisoned by the Pinochet regime, who later leaves the country and wanders Latin America before settling in coastal Spain. It makes me wonder if this story is adapted from his own experiences, or resonant with his thoughts about Chile under Pinochet. If so, the changing names may oddly be a way to emphasize the real-world inspiration: the original names would have been changed before publication, and changing them again highlights that.

This book reminds me that Bolaño's superpower as a writer is the use of absence. He has a fantastic ability to write around a subject, whether that's a person or an idea or an event or whatever. He creates all this stuff that kind of silently points back at this thing that looms off the page while the narration strides forward. It builds up to this wonderfully thick sense of dread: as in a thriller movie, the monster we can't see is far scarier than the one we can.


In this story, he sometimes builds that absence before finally tapping it, as with the long wind-up before finally seeing Carlos's photographs: by the time we see them we're convinced that they're horrifying, and then the narration can just kind of skim along the surface and move us along, leaving our minds to fill in the grisly details. But other parts of the book remain unfilled holes, like the final confrontation between the detective and Carlos, or the contents of the binder he finds, or the identity of the detective's client.

Of course, this leaves us free to speculate and tentatively advance our own theories. And actually, it lets the characters in the book do that too! The narrator wonders at one point whether Bibiano, who has found some middle-class success teaching at North American universities, is bankrolling the investigation; Robero gently but firmly points out that professors don't make anywhere near that amount of money. The narrator seems to assume that the client is a leftist who wants justice for the crimes Carlos committed, but personally I think it might be someone implicated in the crimes: an accomplice or supervisor or beneficiary under the dictatorship who now wants to tie off the loose end and ensure he (less likely she) doesn't need to fear an accusation and charge in the future, to keep the ghosts of the Chilean past buried.

As with lots of Bolaño's other writings, most explicitly in NLitA but also in By Night in Chile and The Savage Detectives, a big subtext is the responsibility of art and the links between art and artist. There's a digression near the end of Distant Star about "the barbaric writers", a far-right school that impels adherents to spend a week desecrating classic literature before emerging reborn into the world, purified and prepared to write unencumbered original work. Reading about these rituals is horrifying, but shouldn't we feel equally horrified by the racism and bigotry of the writers?

Carlos was, by all accounts, actually a very talented poet: but as I read the story, it seems like he was infiltrating the Concepción literary scene to identify socialists and target them in the upcoming purge. Could any of his darkness and malevolence be seen in his poems? I don't know. It doesn't seem like the smart and cultured people around him noticed anything. (Which is ironic, since he gave off the affect of not noticing the people around him.)

Should Carlos's (early) work be celebrated, knowing what he did? I want to say "No." Personally I'm feeling more adamant on this topic in a post-GamerGate, post-alt-right, post-Trump world than I might have been back when this book was written. Art is a wonderful and beautiful thing that we should cultivate and celebrate; and, also, the evil people do and the lives they can destroy vastly outweigh any value their work might have had. It seems best for the human race if we get in the habit of shunning the work of monsters, which fortunately seems to be the tentative trend of the times. Thankfully, there is no shortage of great work being created by non-monsters, and I don't think we'll be losing much as a culture if we consign the work of Carlos to the abyss and instead pay attention to Bibiano and Fat Marta.


I do enjoy seeing linkages between books, and I don't yet have a good enough bead on Roberto to describe exactly what he's doing. I don't think that it's a David Mitchell-esque shared universe; but there are lots of wonderful little holes connecting his books, most famously the passage in The Savage Detectives that gives the best explanation of the significance of 2666. How does Distant Star fit into all this? I don't know! But its existence - its double existence, its dual life under another name - does seem significant, and I'll keep pondering it.

Monday, August 23, 2021

A Punk, But Make It Cyber

It's time for another checkin on Cyberpunk 2077! Yes, yes it is. See, it's here in your calendar! Now sit down and turn off your phone, this will take a while.


Bottom line up front: I have a lot of complaints, but I'm really enjoying the environment and atmosphere in the game, and that's keeping me going. I'm finding that I'm enjoying the game more when I play it more like a shooter and less like an RPG, even though I vastly prefer RPGs to shooters.



I initially started as a build that was focused on Intelligence and Cool (Stealth), with a little bit of Tech. Unfortunately, most of the Cool perks are based around lethal attacks, while I wanted to use it for silent and non-lethal takedowns, so the tree actually has a lot less utility than I expected. Instead I'm focusing heavily on Intelligence and recently maxed it out at level 20; next I'm leveling Tech, which gets you access to more locations, lets you craft and upgrade weapons and armor and mods and consumables, and increases the power of Tech weapons.


Hacking is awesome. Early in the game, my normal MO would be to Breach Protocol and get cheaper access to a network, then disable any cameras and alarms in the area. After that I would either Distract enemies by hacking objects in the environment (lights, radios, soda machines, etc.) and grab them while their backs are turned; or use Reboot Optics to blind them, then sneak up and take them down. I was maybe 20 hours into the game before I encountered my first robot and learned that you can't grab or subdue them, but fortunately the Short Circuit quickhack can usually take down a robot instantly, before they get a chance to squawk off a warning.


Now that my Intelligence is at level 20, I have access to insane programs on my deck. Cyberpsychosis is a standout: it makes an enemy hostile towards everyone, including their allies, and will usually cause them to hunt down their squad, then eliminate themselves. I can turn turrets into allies with Friendly Mode and direct them against my foes. I have passives that instantly blind foes as they spot me. Overall it's a lot more lethal than hacking was earlier in the game, but also more fun.

Hacking is also a great way to earn money. A single Access Point will usually hold between 1k-2k eurodollars. When upgrading my Cyberdeck, I would lean towards the ones with higher Buffer ratings, which increases the odds you'll be able to extract full value from an access point. I'm personally fond of the Stephenson line. 


That said, it is good to also invest in Tech and/or Body. I'd read in the tooltips that those skills can open doors, and had thought "Oh, I bet I can find a way to hack them open or find a key." That's true in most other games, but not here! Those doors just stay closed if you don't have the required skill level. For story missions, you can find another route in. For all other missions, though, it means leaving behind whatever loot was locked in that room. Oddly enough, the requirements for Tech and Body are much higher than the requirements for Intelligence; for much of the last 10 hours or so of play, all of the Access Points I encountered required an INT of 4 or 5, while the doors required between 7 and 9 Body or Tech. I'm curious if that's because they assume that all players will dip a little into INT or what.

Attributes max out at 20. From what I've read online, an end-game character can bring 3 attributes up to 20 and have a few points left over for the other two. Besides Attributes, there are also Perk points. Those are more plentiful than Attribute points: you get one while leveling your character, and sometimes get one when leveling a skill, and also can get points through specific item pickups. It's still a limited commodity, though. From what I've read, it sounds like there's eventually an expensive option to respec your Perks, but no such option for Attributes.

There are a lot of progression systems in this game. Besides Attributes and Perks and Skills, there's also Street Cred, which is kind of a parallel XP system. You seem to get most XP from completing quests and most SC from defeating enemies, although you can get both from both. Street Cred is mostly used to unlock better equipment from vendors: they won't sell you the really good stuff until you've impressed them and built a reputation. It also opens up higher-level Fixer gigs and options to buy fancier and more expensive vehicles. For cyberdecks in particular, new decks usually unlock on the 5s or the 10s, so those are good times to go shopping (e.g., Street Cred level 25 or 40). Each vendor has their own stock of items, and some items are only sold by one specific vendor. Each item has the same requirement, but there are a couple of times that you can negotiate or threaten a permanent discount from a specific vendor.


Speaking of vendors: so far the overall money curve seems pretty decent for an RPG. On the downside, it's the worst possible style of sourcing economy, where you collect hundreds of pieces of junk and crummy weapons and armor and stuff to sell off for money, leading to insanely tedious inventory management shenanigans. There are lots of vendors in the world who sell clothing and consumables and stuff, but pretty much anything you buy would be eclipsed by the pickups you get from defeating enemies, so there isn't much reason to buy stuff, except for the aforementioned cyberdecks and cyberware in general.

By far the single best thing to buy is a pair of cyberlegs: they run for about $16k but are absolutely worth it. You can choose between one that lets you double-jump and another that lets you charge up a super-jump. There are about a dozen slots to put cyberware into (eyes, skeleton, musculature, arms, legs, cardiovascular system, etc.), and each item has an associated cost, a Street Cred requirement, and often a stat unlock, like 12 Body to get a Supra-Dermal Weave that makes you immune to Bleeding. As in Shadowrun, you can't sell back old cyberware where you upgrade to more powerful models; but once you've purchased a particular piece, you can swap back to it for free at any ripperdoc. At least so far, there doesn't seem to be any downside to maxing out with as much cyberware as possible, which is a little surprising to me: from my understanding of the pen-and-paper Cyberpunk rules, installing too much 'ware causes a loss in Humanity and Empathy which can eventually turn you psychotic. In Cyberpunk 2077, that's part of the lore but doesn't seem to be part of the gameplay.


The big money sink is vehicles. These are also unlocked by Street Cred. You buy them directly from Fixers, and will only be contacted by a Fixer with the option when you're in their territory, so you have to travel a bit to see what all is for sale. They are very expensive; it might be feasible to get them all by the end of the game, but certainly not where I'm at. I don't think it's strictly necessary to buy any vehicles, since you get one or two for free during story missions, but there is a pretty nice variety, and definitely some solid specializations: some are particularly fast, others sturdy, some designed for off-roading.


Personally I pretty much always buy motorcycles, which are a lot more maneuverable when weaving through traffic, navigating alleys and moving through drainage tunnels. One complaint I have is that you can't try out a vehicle before purchasing it, and this game has an awful habit of delivering vehicles with miniscule fields of view. Because it's first-person, you aren't seeing your surroundings, you're seeing the inside of the car you're driving, which hopefully includes a slice of the world through your front windshield. On lots of cars, that windshield is narrow as hell and doesn't offer much of a view at all. Here too, I thought I was lucking out by favoring motorcycles, which don't surround you in a chasis and let you actually see what's around you: only then did I buy my second motorcycle and make the unpleasant discovery that it has a ridiculous bug-spattered windshield that blocks out the center of my screen, i.e. the part with all the most important stuff for me to see like oncoming traffic. This is a recurring theme of Cyberpunk 2077: for better and for worse, style always wins out over fun and usability.


And ha, just after writing that last paragraph I learned that you can switch between first-person and third-person, but only while driving. (I learned this because the game put me into a race with a predefined course, switched camera modes so I could see the path, and didn't switch back at the end.) So, yeah, you should definitely always drive in third-person mode because third-person games are better than first-person. BUT, this does then bring up another flaw: the clothes in this game are, for the most part, butt-ugly. Well, the individual pieces aren't all bad, but the combination is terrible, and it's pretty much impossible to look good after the first time you upgrade a piece of gear. That's honestly a big part of why I'm investing so much into the Tech tree: I'm hoping that I'll eventually be able to craft a selection of gear that both looks decent and has acceptable stats. So far, though, no luck at all. The "stats vs. looks" dilemma is definitely not unique to Cyberpunk 2077, most RPGs have it in one form or another, with a few like DA:I having innovative solutions while SR:HK has a dead-simple one. But I think there's a big difference between, like, a fantasy RPG where my complaint is about the shape of a pair of metal greaves, and a cyberpunk RPG where my leg slot can be filled with jorts, or jodhpurs, or camo pants, or track pants, or suit pants, or a pencil skirt, or hiking shorts; and a shoe slot can be filled by combat boots, or strappy sandals, or low-rise heels, or high-rise heels, or sneakers, or hi-tops, or dress shoes. All with radically different Armor values. Ugh. I'll still take looking at that ugly mess over playing in first person, though.


I seem to be getting cranky, so let's continue the airing of grievances!

I'd alluded to this before, but stealth gameplay is really annoying in practice. Enemies can be Unaware, Alert or Aggressive. They're pretty easy to deal with while Unaware. Once they suspect that something is wrong (they spot a body, an alarm goes off or they hear a call for help), they become Alert: they will look around more frequently, and they will spot you more quickly. Once they actually spot you or take damage, they become Aggressive, and once they are Aggressive, they will never go back down. So, if you want to do stealth takedowns, and you slip up for a second, the only way to do stealth is to reload your last save. Like I mentioned above, I've lately come to find it vastly more satisfying to just murder everyone, or make them murder one another.


Some technical issues: the bass sounds terrible, particularly in the early missions in the game. It might be my speakers, but I haven't had this problem with any other game.


A lot of items have pickup symbols but can't actually be picked up. Sometimes if you move your cursor a couple of inches above or below or to the left or to the right you will be able to pick it up. More often you can't.

The game is pretty glitchy when reloading. The game blocks saving when in combat, but if you're being stealthy you can be doing stuff to enemies while technically not being in combat, and that can make things wonky if you reload that save. In particular, if you're in the middle of uploading a quickhack when you save, then when you reload you won't be able to do any hacks against that target ever again.


It seems like all the game-breaking bugs that plagued the game at launch have been patched, and after 50+ hours I haven't yet had a single crash or forced reload. That said, in non-essentials it remains pretty glitchy. It reminds me a bit of Mass Effect Andromeda. Nothing terrible, but every hour or so you run across something goofy and unpolished.

I'm not a huge fan of the immersive conversation system. I'm used to having a separate interface for dialogue, but in Cyberpunk it's all delivered in-game. This can be cool and feel natural, but the dialog controls overlap with game controls in some annoying ways. "C" fast-forwards the dialogue, and also crouches, so sometimes I'll try to crouch and end up missing something important. "F" is the all-purpose control for almost everything in the game, including looting items, opening doors, and picking dialogue responses. Sometimes I'll be busily mashing F to pick up loot, and accidentally pick the first response without knowing what it is. On the other hand, I was once trapped in a medical shop because I couldn't dismiss the dialogue options, and so couldn't press "F" to open the door and exit. And this isn't a big thing, but most conversations don't have a "Bye" choice to dismiss, so you just walk away when you're done. For some stuff that's fine, but it's weird if, like, you have a heart-to-heart with Vektor and then just silently leave.


My single biggest gripe so far is with the enormous volume and the low quality of the side-quests. What I loved most about The Witcher 3 was how there weren't any boilerplate sidequests. Pretty much every quest felt bespoke, with its own cut-scenes, unique motivations, mechanics and strategy. In Cyberpunk, though, 95% of quests are boilerplate. "Assault in Progress", Fixer Gigs, etc., all boil down to the same thing: go to a place, defeat all the enemies, collect the loot. That's it. Usually there isn't even any talking or "press the button" element. Quests in this game feel more like the quests in Oblivion or Fallout 3, which is not a good thing. As in those games, I feel actual dread when I open my map and see it filled with quest symbols, and I have dozens of quests listed in my journal, and I know that all but three of them will be tedious and boring and just like every other quest. 


The one bright side is that at least there's not the town-cave-town cycle: you usually get the quest over the phone as you approach the destination, and call back your fixer after you're done, so at least in-progress ones don't clutter up your log too much.

Of course, you don't have to do any of the side-quest stuff. The main quests are all great and interesting and varied, and I think you could just do those, or only do a handful of sidequests to level up for the main quest. It's a sign of my OCD that I feel like I need to do all these things and get annoyed at myself for not having fun doing these things that I know I'll hate.



That sounds harsh! So, why am I still playing this game? The single biggest reason is Night City itself. The world looks awesome, and I'm still finding entire new districts with their own vibes and culture and style.



 While there are a lot of gripes I have with the game, no particular thing has been unpleasant enough for me to bounce off of: combat is fine, the difficulty feels okay, advancement is fine, driving was acceptable once I learned there was a third-person view, the voice acting is pretty good. Oh, and I'm pretty sure there are romances in the game, so I'm definitely sticking around to verify that.

So, yeah! That's where I'm at now. No story stuff in this post, which honestly is mostly because I've done 75 terrible side-missions about "kill all these people because [REASONS]" and like five main story missions. I'll probably recap and react to said story once I better know what it is!

Sunday, August 08, 2021


As I've previously mentioned, once I decide that I really like an author, I take pains to learn as little as possible about any of their books before I start reading them. That's now the case for China Mieville as well. I picked up Embassytown with literally no idea what it was about, and about one or two pages in I went, "Oh, huh, I guess this is a science fiction novel!"



I tend to associate China more with the "weird fiction" of his Bas-Lag books, but I shouldn't have been surprised that he'd do something different, after the awesome detective fiction of The City & The City and the history writing of October. I guess that the Bas-Lag books do sort of nod at steampunk a bit, but Embassytown is a much more classic science fiction setting, complete with interstellar travel, alien species, androids and more.


Of course, he makes all of these things his own. I think China's single best ability in his fiction is his skill at inventing something indescribable and then attempting to describe it. In Embassytown, the biggest example of this is probably the "immer", the sort of underlying reality through which people travel between stars. The narrator Avice tries to explain the immer to her husband and to us, trying out various metaphors and realizing that they all come woefully short. There are certain things that she can say with confidence, like that the immer is older than our universe (and in fact has supported at least two universes before ours), and that it's dangerous, and that the time to travel between two points via the immer does not correspond to how distant those points are in regular state. But when it comes to the feeling of the immer, the way it works on you, the way you work at it, she can gesture at it but never really explain it. It reminds me a bit of HP Lovecraft, but I think that China is better: he actually attempts to describe the indescribable instead of merely stating that they're indescribable. 

He's also really good at slowly revealing information. He'll introduce a thing, and let it be a mystery for many chapters. The narrator and the characters will wonder about it, find evidence, draw closer, and eventually reveal the mystery. From then on he doesn't belabor the mystery, since it's over, and becomes accepted reality. But the journey is great, and China gets a lot of high-stakes mileage out of it instead of just introducing it and immediately pushing the story forwards. (I'm thinking in particular about the slow reveal about exactly what effect EzRa is having on the Ariekei.)

And the worldbuilding itself is also really great, maybe not quite as dense as the Bas-Lag books but very well-done and intriguing. Over the course of the novel we hear about the Terran diaspora, the Brennan empire of established planets and far-flung planets, relations with the indigene. Most of the focus is on Embassytown and the planet it resides on. "Biorigging" is a fun concept, a novelty even within the novel, of engineering items out of living, often sentient matter (like portable "batteries" that follow the Ariekei to help provide more energy when needed; even weapons that can detect threats and fire projectiles).

The most interesting stuff, though, is with the Ariekei themselves. Due to an evolutionary quirk, they have two mouths (one apparently a feeding apparatus, the other originally a specialized organ for sounding alarm). They speak simultaneously out of both mouths, which is rendered in the book with two pieces of dialog vertically placed on top of each other with a line separating them. As the early settlers eventually realized, the language (or rather the Language) was even more unique, in that it wasn't really a language at all. Language gives symbols and markers by which we denote things and concepts in the world. The Language, though expressed audibly, is really just a manifestation of thinking, almost a form of vocal telepathy, with each individual automatically expressing the true thoughts inside their mind. It's a very odd and cool concept, and much of the novel deals with how the humans are responding to that dynamic, for their own advantage and/or for greater cooperation.


Looking back over Embassytown, it seems to be a novel about language, metaphor and meaning, which is a really fun and ballsy theme for a novel. It's really twisty, with some things repeated nonsensically throughout the whole book ("The girl who was hurt in a dark place and ate what was given to her"), and a great progression of revelations about how things work. Some are things we learn early and the Ariekei learn much later, some are things the narrator learns early and tells us about later, some are things we all learn together.

Some early stuff ties together really satisfyingly later in the book, like Bren and lying and Scile. A few other things seem to be red herrings. Ehrsul feels like a potentially significant character in the first half of the book, but never really does anything over the course of the novel. And while the immer stuff is some of the most fun writing, it doesn't turn out to have much of anything to do with the main events. But, I suppose you could draw some thematic connections back to the central language/Language concern of the novel. Ehrsul's existence poses an existential question about thinking, which in turn connects to the possibility of relating between fundamentally different beings. (Perhaps Ehrsul is communicating with humans in a way similar to how humans communicated with the Hosts, taking advantage of a fundamental misunderstanding. It's interesting how some people, like Scile and Ez, instantly dismiss Ehrsul, which seems similar to the Ariekei's dismissal of non-ambassador Terrans. And I wonder if and when the New Language people will speak with Ehrsul. And I wonder why she isn't interested in speaking with them.)


Embassytown is pretty different from the other China Mieville novels I've read, but one similarity is that I'm left with a sense of awe at all of the incredibly creative stuff that he invented for this book, and then presumably abandoned. A lot of authors would happily make entire careers out of exploring Bas-Lag or the Immer, while China does the work to create this vast and rich universe, and then gets one (great!) book out of it and moves on to the next thing. I do really like how these worlds continue to linger in my mind after the book is over, and it is exciting to think that I may encounter yet another new world from him in the future.