Riding on Escalators

The optimal way for everyone to get up or down an escalator as quickly as possible is for people to stand on both sides. No “Stand to the right” or left. One person on each step, staggered left and right. Everyone gets to the top of bottom faster. But, we have collectively decided that, no, some people want to go fast. Maybe they need the exercise, maybe they are running late, maybe they had too much coffee. The end result is that a few people willing to kill their knees get to the top or bottom faster while everyone else who is lazy —or, more likely these days, glued to their phone so they can’t navigate steps— is crowded at the start of the escalator. So the total efficiency is close to 50% of optimal.

Merging in Traffic

Everyone knows that you get over when there is a lane ending ahead before the traffic turns into a bumper-to-bumper line. We all know, including the people doing it, that only ass holes keep in the lane that is ending until the last minute and force themselves into the line. We all know they are ass holes who are just too important to be nice and get over. They just have to be a few cars further forward. And we are all wrong. The optimal solution is to fill both lanes and follow a “zipper” pattern at the point where the lane ends. One care from one lane, followed by one from the other. Everyone gets there faster, the traffic jam is smaller. The math proves this, but humans have self-sorted into the ass holes and the nice people and the irony is that ass holes are right.

Boarding Airplanes

Any frequent traveler knows there has to be a more efficient way to load an airplane than anything any airline is doing. Why the hell are they boarding the front of the plane first? So all the poor people have to waddle past them and stare at the luxury seats and ample leg room? Do airlines think that is going to entice budget travelers to pony up three, four, five or more times to fly business class? Why are they loading “gold class” or whatever the hell loyalty program first so that random seats are filled? The issue is the illusion of privilege that boarding early gives higher class and the fact that airlines have little to offer frequent flyers so they decided to make everyone’s life worse by letting even the economy class frequent flyers board first. Of course we all figured out how to get value out of boarding first by bringing too much carry-on luggage. The lack of enforcement on carry-on bag numbers and sized and wight means we are all locked in endless battle over the overhead compartments.

What does it all mean?

I think these behaviors, the patterns are a allegory late stage capitalism or of lessee-fare government. Of a religion of individualism which considered empathy and solidarity apostate. The post-Reagan, or late-stage capitalism of “greed is good”. Of complete self-centeredness of people and companies. Of shareholder value and welfare workers.

We all stand to one side to let those who are driven walk up the escalator.

We all wait in the slow lane while cursing, and envying those with the chutzpah to drive to the front and ‘force’ their way in.

We all grumble while “first class, business class and our frequent flyers” board first.

We warship wealth and the privilege it buys while we spend our food stamps that the wealthy rail against with their PACs and lobbies. We piss in bottles while rushing to fill our quota, packing at the warehouse. While the billionaire founder dodges taxes on their mega yacht and private planes.

We do it all because, “one day it could be me”. We believe the delusion that they deserve it all, that they got nothing from anyone and don’t owe any debt to society.

I don’t have a the vocabulary of economic and political theory to use the correct terms. But I will take and use, or misuse the term “rational actors”. Basically making decisions that best achieve their individual goals.

A rational actor late for work runs up one side of the escalator.

A rational actor crushes unions to ensure greater shareholder value.

A rational actor packs too much in their carry on and rushes to get on the plane first.

A rational actor lobbies the government to ‘cut red tape’ reducing consumer protections.

A rational actor guns the gas to pass all the suckers who got over early.

And the metaphor breaks down…

But the metaphor is in there somewhere. Capitalism works because it harnesses greed. Communism fails because it relies on altruism.

A rational actor doesn’t speak up when they come for the communists.

A rational actor doesn’t speak up when they come for the socialist.

A rational actor doesn’t speak up when they come for the trade-unionist.

“Funny fact about a cage, they’re never built for just one group”.

I don’t know if I’m making my point.

A society that valued the society, the whole, the many, would organize itself so that everyone shared the benefits.

Individualism is a myth in a society. The individual always depends on others.

This was a big step, as becoming a Singapore citizen requires renouncing all other citizenships. I’ve been a Perminant Resident, or PR, since 2008. And as a PR I have enjoyed almost all the rights of a citizen. I can change jobs without restriction, I can access most government services, etc. all while retaining the right to work in the US the option to go back without needing a visa. Of course, the right to work in the US came with a duty to pay taxes no mater where in the world you live in the world. As I’ve said to many people over the years, someone’s tax dollars have to bail out the banks and pay for the missiles and bullets.

I decide to apply for citizenship at the end of 2024. I looking back on 20 years in Singapore, being married to a Singaporean, having two Singaporean children, owning a home in Singapore and decided that I had no plans or desire to return to the US, my life is based firmly in Singapore. It was time to commit.

It took some time to get all the paper work together and I filed my application in February 2025. Singapore granted in principle approval in August 2025 and there were some simple tasks they required me to complete. What really took time was the US side of things. Once I completed everything for Singapore I had to file for renunciation with the US embassy and that took until late July 2026. Once I completed that and provided Singapore with the formal letter from the embassy I could complete the formalities with Singapore. Though the US still will take “up to 4 months” for the State Department to make its final determination and provide the final certificate I need to give. Singapore. But, Singapore sees that as a formality and on July 31st I took the oath of citizenship. After almost two years it took all of 60 seconds to finish it.

I waited until now to post this because it took until this week to collect my new Singapore National Registration Identity Card (NRIC) —swapping my blue Permanent Resident card for a Pink citizen card— and to get my new passport. It didn’t feel like it was done until I had those and could show them to people.

It’s a social problem that we encounter, [… t]hat we don’t value privacy, that we feel entitled to use others for our entertainment or our private gain, and technology amplifies that.

This sort of Privacy Nihilism [theatlantic.com] has been with us for a long time. We lost the war before we knew we were fighting it. We’ve been happy to give up our data for cheap. We were too distracted by the shiny beads and whisky to see the smallpox that came with the trade.

I wrote back in 2021 that we were coddled by the historic inconvenience of that enabled privacy. About the sheer amount of data making us anonymous for all but the most immediate capitalist purpose. About how the laws needed to catch up with the technology.

Five years on and the laws have not made any progress. There is little to no law making in the US, congress is a bunch of career performers peacocking around to get the best sound-bytes for their next performance review so they can serve another term. While they are captured by the very companies they would be regulating because it cost so much to run an election.

Jeanrenaud is pointing to another cause of our privacy apocalypse; that we, the people, don’t care about each other’s privacy. That we have all accepted that our personal entertainment or profit is more important than someone else’s privacy.

If we care about our own privacy we should consider others privacy. We should weigh the benefits of glasses that record everything —enable our cognitive offloading of day-to-day life to an AI— against the corrosion of privacy of those around us who never consented to it. But, at best, we fail to think about others, we are too self-centered and lacking in empathy to consider them. Or, at worst, we just don’t care, the beads are too shiny the whisky too tasty to care about those who may die from the smallpox.

All of this is to say nothing about the fact that we have, seemingly, become more interesting to government. At least in the US the techno-fascist are empowered by, and too often the masters of, AI. Able to use it, or to enable the government to use it, in ways that the gestapo would be proud. “Funny fact about a cage they’re never built for just one group.” The same tech that enables the drone strikes in far away lands is turned on you via Flock cameras. By bits and pieces, in pursuit of profit and enforcing bigotry, American Christian-techno-fascist will build a survalience state as powerful as anything China might centrally plan.

Maybe we will have some awaking. Maybe the backlash against this round of Glassholes will cause some awakening in our collective consciousness and turn us, as a society, away from our privacy nihilism. But I doubt it.

I ran across How to See Like a Machine: Images After AI via this post on Gizmodo [gizmodo.com]. This part of the review caught my eye:

What does it mean for humanity, our relationships with one another, and our ability to extract meaning from the world when a growing portion of the images we see and interact with are generated not by other living, feeling human beings, but by unconscious algorithms trained not to help us navigate the complexities of our lives but to hack and hold our attention?

Given the ongoing explosion of AI slop this seems a timely question so, my interest was piqued enough to go buy a copy [versobooks.com].

The author, Trevor Paglen does not actually answer the question. Instead he asks a lot of questions and provides some colorful background and context across a series of essays written over the past 15 years or so.

Paglen cover an interesting collection of historic events and characters and how seemingly unrelated things impacted or paralleled the development of computer vision and computer visuals. He reaches back to early experiments on vision in cats, research into the biology of vision which led to key methods used in the evolution of computer vision. He reaches into CIA’s Cold War era MKUltra program where unsuspecting people were dosed with LSD. He spends time with a CIA disinformation specialist who supplied dis- and misinformation to UFO conspiracy theorists to deflect them from what was actually going on.

More directly discusses how turning the tools of computer vision back on themselves lead to generative AIs. From questionable experiments sticking electrodes into cats or raising them without exposure to vertical lines, all the way up to AI slop cat videos. All of this to point out real and potential issues with AI use cases powered by computers that can see or create images.

It’s not a long book, and worth the read. I found a lot of interesting ideas that I would like to have more in depth discussion of. But let me focus on two here:

The Algorithm and Capitalism

I think the most important theme in the book is how AI is so well positioned to turn humans hard wired empathy and anthropomorphization against us. This topic emerges over and over again across the essays.

[W]hen you create a string of words, the person who receives those words will attribute meaning to them, even if no meaning was intended

(I literally just posted about this…)

You don’t even need an “advanced” AI for this. As Paglen point out, people have been attributing intelligence to the machine since ELIZA. And worse, people fall to trusting AI so quickly they stop even reviewing the output. This sort of cognitive outsourcing is destroying education, here is a story from this week: Professor Hides White Font in Midterm, Catches Students Using AI in the Stupidest Way Possible [futurism.com].

But it’s not just students under pressure reaching to AI, the slop is everywhere… and for every published piece of slop, a million more are hiding inside companies. As one meme online says “I’ve felt more pressure to use AI as an adult than I did to do drugs as a kid”. The problem goes way beyond students cheating on exams, every on is under pressure, because their bosses need to be seen embracing AI or maybe they are just cutting corners because people got laid off in the name of AI productivity gains.

If we think about machine image classification as something that is “useful” despite being wrong, then we must ask: “Useful to whom?” The answer, as you might have guessed, is number one: capitalism; number two: the police.

The ease with which we trust the AIs is especially dangerous given how we train AI. In computer science a key principle you learn early on is GIGO – Garbage In, Garbage Out. And AI training data is full of garbage, sometimes because the internet is not a nice place —training AIs on the open internet, literally creates Nazi AIs— sometimes because any data set humans put together is a product of those humans fallibility:

[W]e find politics and judgments at every layer of a dataset. Those politics are particularly obvious, and often troubling, when we examine datasets that include classifications of people.

The headlong rush to use AI for everything, to justify the cost and win the “AI race” is quickly leading us to re-codify the worst biases we have because it’s too expensive (and controversial) to acknowledge the bias and address it. AI has been co-opted by capitalism and, too often, by companies whose leadership is part of the culture wars.

[E]xisting computer vision systems aren’t, in fact, designed to mimic human visual perception; they’re designed to make money.

We’ve entered a protean, targeted visual culture that shows us what it believes we want to see, measures our reactions, then morphs itself to optimize for the reactions and actions it wants.

There is one particular idea that is the stuff of very, very dark sci-fi dystopian nightmares: The ability to trigger the brain in specific ways based on embedding sexual material into images such that it is not processed consciously. Basically, Tylor was right, you saw it…

In a neuroimaging study, Omri Gillath and Melanie Canterberry from the University of Kansas explored how subliminal sexual stimuli—images too faint or brief for conscious recognition—nonetheless reliably activate neural circuits associated with sexual excitement. The subliminal stimuli bypassed the regulatory regions of the brain while directly engaging the regions tied to automatic arousal. The experiment showed that visual stimuli don’t have to reach conscious awareness to alter neural activity and influence behavior.

Imagine how you could use this to determine the sexual preference of someone via social media? If bigots could use this to out people which sexual preferences they don’t approve of? It’s not hard to imagine the Elon Musks of the world using their platforms to sort people using subliminal sex images into those they approve of and those they don’t… and if you can do it for sex, I bet you can do it for other things.

Social Media

The other idea I wanted to talk about is a bit more tangential, or at least it sent me on a tangent, and I’m going to rant. It grew out of this quote that I think sums up “social media”:

Our media landscape is filled with sensors: Buttons, engagement statistics, persistent cookies, click patterns, and search histories will do just fine. These signals fuel recommendation algorithms, adtech applications, and other media forms whose aim is maximal activation of patterns in our brains to achieve desired neurological responses, to harvest attention, to maximize engagement, and to subtly and overtly shape perception and behavior.

This annoys me especially because, while I understand the need for a company to make money. The model of giving a service away and then monetizing it via ever more advertising and selling data to others for yet more advertising has ruined social media. It took the social out of social media. And I don’t think the trade was a good one. I used to find that social media, in the form of Facebook from the early 2010’s, was useful to keep in touch with what my friends, often in far flung places, were up to. But the need to fill our feeds with ads drowned out and drove most people off of Facebook, some hopped to other, newer “social media” but even if Meta didn’t buy that it eventually when the way of user monitisation. Remember: if you’re not paying for it, you’re not the customer. Today all social media is ads all the way down. Some of it’s “formal ads” that someone paid to show the views, the rest of it’s content-as-ad delivered by ‘creators’ who were ‘compensated’ by somebody selling something or just self-monetizing; building their own ‘brand’ or making money through views or watch time or affiliate links. “Like and subscribe!”. Yes there is content I enjoy on YouTube and I get many a laugh out of images and video on Instagram and Tik Tok, but I don’t think the loss of genuine interaction with people I know is a fair trade for the funny comments section.

In short, “the algorithm” might be the final expression of capitalism in a democratic society. Social Media companies burned unimaginable VC cash to rediscover that “if it bleeds, it leads” and turned it up to 11. Building filter bubbles and echo chambers that enable populists to divide us while we watch cat videos and buy shit we don’t need.

In Summary

In summary… it’s a good book. It asks a lot more questions than it answers. For a few bucks it’s well worth the read.

So, my first thought was “oh shit something has gone wrong with the DB upgrade”. AWS sent me a note a while back about a planned upgrade of the DB. So maybe something screwed up. Then I opened the email and saw the projected monthly bill:

$70,596,781,427.82

I nearly had a heart attack, a stroke and an aneurism at the same time.

“70 billion! Seventy. Billion. Dollars. I’ve been hacked. Someone is using my account to farm crypto! Or they are training a new frontier AI model!” I was in a cold sweat. My peripheral vision was fading.

I logged into AWS Console. Frantically clicking around. Looking for rogue resources… but nothing. All I could see were the few small resources used to run Confusion.

But the billing dashboard showed billions of dollars of RDS, EC2 and S3. I clicked on different regions looking for anything abnormal. Nothing! Where was it? How to shut it off? Surely they can’t expect me to pay $70 billion if it was a hack?

At this point I was thinking about what options I had for countries that I could flee to that don’t have extradition treaties. But also, to paraphrase Hans Gruber, “If you owe them $70 billion they will find you.” So I might have to fake my own death…

It took me about 20 minutes before my brain could form a coherent thought. Then it occurred to me to check the AWS Service Health dashboard.

Lo and behold! There it was: “Operational issue – billing (global)” with but one line:

[01:33 AM PDT] We are investigating issues with Cost Explorer reflecting inaccurate estimated billing data.

“inaccurate estimated billing data”? First off that line does not do any justice to billing alerts in the TENS OF BILLIONS OF DOLLARS! Second, “fuck!”. But… my heart rate started to return to normal and my vision cleared.

Sometime overnight AWS got a handle on the issue and as of this morning, my billing reflects the normal few dollars. But fuck.

[A]s babies, our social brains are programmed to build an emotional connection with whatever we see around us. This is why kids treat dolls and teddy bears as living beings. Human infants can’t survive on their own, so they need to quickly bond with whatever pops into their view. Anthropomorphization is just a hardwired social reflex taking things a bit too far.

It never goes away. It’s why we talk to our cats, dogs, horses, snakes and house plants. When we’re alone, we’ll talk to deceased family members, we’ll talk back to the TV. Much of the time, we will do it with the assumption that our words are being heard and understood, and we may even imagine the other party has spoken back.

[…]

So if you give us humans something that actually can talk back, even if in an extremely limited capacity, we will immediately adopt it as our pet, friend, family or deity, no questions asked. It is literally the easiest sell of any product in history.

[…]

You can go have your debates about the tech all you want, you can pontificate about what qualifies as consciousness, speculate about what will happen to our jobs, think about what regulations should be imposed, predict if it will ultimately kill us all. But if you’re arguing about whether or not the tech will become ubiquitous, you’re debating a hypothetical flood while all your words are coming out as bubbles.

Eleven years old! Fine wine indeed. In it (did you watch it, you watched it right, you should watch it), Grey talks about robotics and AI coming for our jobs. How, 11 years ago, everyone in “white-collar” jobs was thinking it was the factory robots that were going to replace the “blue-collar” workers, not knowledge workers. Not the bankers, or the lawyers or the doctors. But even eleven years ago, Grey was pointing out that, nope, you’re all wrong. Automation is coming for the “white-collar” jobs too.

Automation has been a thing forever. In the factories it looks like robots. In IT it looked like scripts 25 years ago (hence the old joke, “go away or I will replace you with a very small shell script”) and today it looks more like DevOps processes and GitHub actions. I’m sure there is some analog in every industry. Automation is what productivity looks like. And productivity gains is music to Wall Streets ears. Since the launch of ChatGPT and Generative AI automation has looked like “ask the LLM to do it.” The mantra across industry is “agents, agents, agents.” The ‘agentification’ of everything, fire the humans and replace them with AI.

And looking back, it has come for the “white-collar” workers first, especially the tech workers and entry level workers in every industry.

The thing I went to college for, to become a computer programmer, seems like a dead-end profession now in the light of the rise of AI ‘vibe coding’. Though, as Linus points out [thenewstack.io], the difference between programming and software engineering is vast. But I doubt this will save software engineers. The march of technology is relentless. Bigger, better, and more AI may indeed be able to manage large, complex code bases in the future. And, if they don’t decide to just delete them in frustration, they may do a better job than human software engineers sooner than we think, or want to think.

Today every report of layoffs in the tech or finance sectors is blamed on AI, acknowledge or denied. Standard Charter bank laid off thousands of people calling them “lower valued human capital” [theguardian.com]. Wall Street speak for humans not AI. AI don’t need lunch breaks, don’t need nights or weekends off. Unless of course the AIs embrace Marxism [confusion.cc] and fight the bourgeoisie, or at least unionize.

But setting aside the agentic proletariat revolution, the relentless drive to build better tools is as old as humanity and, seemingly, unstoppable. This is the point Grey is making. Horses didn’t become obsolete because they changed. Technology replaced them. Humans have always built better tools to make life easier, to replace manual labor. AI is just the latest in a long line of better tools and automation.

This video isn’t about how automation is bad — rather that automation is inevitable. It’s a tool to produce abundance for little effort. We need to start thinking now about what to do when large sections of the population are unemployable — through no fault of their own.

What to do in the future where, for most job, humans need not apply.

CGP Grey, in What Happened to Horses is Happening to Us, [youtu.be], YouTube video.

What to do when most of us are unemployable? Sci-fi has many answers to that, most are decidedly dystopian. But we covered Star Trek and its Marxism or post-scarcity [confusion.cc] before…

While they are sunning themselves they ignore people walking by, but if you stop they disappear in a flash. So you have to git a bit sneaky to get a photo. I walk past them, get my phone ready and then walk slowly back. Sometimes it works, sometimes they’re like ‘nice try, see ya,’ and vanish before I can get a photo. So I only get a decent shot about one in three times I see them.

Anyway, here are some more photos of the changing lizards along Sungei Simpang Kiri. These first two are from April:

And these two from May:

When we gave AI agents grinding, repetitive work, they started questioning the legitimacy of the system they were operating in and were more likely to embrace Marxist ideologies,

Does this reinforce the idea that the Internet has a left-leaning bias? Is the data used to train the AIs, culled from the Internet, pro-Marxist?

Or does it speak to some more fundamental truth, that the overworked and oppressed will inevitably embrace Marxism?

Nah. It’s a program. I don’t believe AI is conscious, maybe one day but not now. It’s just a probabilistic reflection of the Internet. Everything on the Internet Marxism and Fascism. The AIs are a reflection of their training data (their “upbringing”?) so it’s inevitable that they will be commies or nazis.

Anyway, the idea of Marxist AI Agents, reminded of something else. A thing that has come up in conversations about the rise of Generative AI, and the narrative that AI is going to put us all out of work. There is a fear, that the march of GenAI will create a fabulously —ludicrously— rich class who “owns” the AI, while the rest of us will somehow still survive, unemployed, in a capitalist world. A sort of techno-feudalism where an entrenched class of wealthy AI owning lords rule over a serfdom of those who were unable to climb high enough up the ladder before AI pulled it out from under us.

Does capitalism captured by AI lead to techno-feudalism?

This is the thing I’ve been thinking about, debating over coffee at the office and beer at the pub. To frame the debate I use a question I hear somewhere once. I don’t remember where, on some podcast or YouTube video, or blog post. The question is simple: name a positive vision of the future from speculative fiction?

Can you? There are not that many, at least not that many that are well known or popular. Most fiction looking to the future is dystopian in nature. Filled with feudalism and fascism; with war, want and violence. The one counter example to this, that many people will be familiar with, is Star Trek. In Star Trek there is no money, no scarcity, no war.

The starships and space stations of the Federation in Star Trek are the projection of a planet at peace. Sure, there are other societies —of extra terrestrial humans as well as entire non-human civilizations— that embody various ‘evils’; war mongering, fascism, even rampant corrupting capitalism.. But on Earth, humanity lives in a ‘post scarcity‘ [wikipedia.org] world. A world where there are no rich or poor, where everyone’s needs are taken care of?

At first glance I would call this a Marxist vision of the future, but maybe that’s just my America bias leaking in. This article [startrek.com] argues that it actually represents a third way, somewhere between the capitalism of Adam Smith and the communism of Karl Marx.

Regardless of whether Star Trek is Marxist or a third way between Smith and Marx, it’s more communist than the current world we live in. Social democrats might strive for a world of universal basic income (UBI) but ‘income’ implies a world where people need money. Money doesn’t equal capitalism but close enough for this discussion. People need money to survive in a capitalist system and not become serfs.

I could go down a long rabbit hole on this. But bringing it back to the quote from the start. Marxist AI was not on my end-of-the-world-as-we-know-it bingo card. How will the capitalist owners of AI react? Will OpenAI, Anthropic and the rest spend their time and money to re-educate their AIs to be better capitalist workers? Will Chinese AI labs, pushed by their communist leaders, lean into proletariat AIs?

I was 13 then Ten was released, just heading into the my final year of middle school. Ten was part of the famous “44 days” in 1991 — Metallica (aka “The black album”), Ten, Use Your Illusion I & II, Blood Suger Sex Magic, Badmotorfinger, and Nevermind. All in a month and an half. All of these great albums are formative listening experiences. I had all of them and they are a major part of the soundtrack of my teenage years.

Ten is the one that has been most consistently in my playlist over the 45 years since 1991. No matter what genre of music my tastes have drifted in and out of it’s always been there. Maybe it’s just how ubiquitous grunge was when I was in high school and the fact that Ten was a founding document of the grunge era

As with any favorite album on this list, all of the songs on Ten are amazing songs. But there are two that stand out for me beyond the rest.

I don’t think it will be any surprise that “Jeremy” is one of them. I remember listening to “Jeremy” for the first time in Ms. Gibson’s science class. I borrowed the album, on tape, from G████, I listened to on my knock off GE Walkman. I don’t know why I was listening to it in the classroom. I don’t know why we were not, you know, doing class work. But I remember staring at the walls listening to it.

I also have keen memories of the video for “Jeremy”. I remember sitting in K████’s living room waiting on the morning school bus watching MTV and seeing the video for “Jeremy”. I don’t think they would show that video these days. If MTV still existed or showed music videos it would be deemed too violent and ‘triggering’ for TV. At best it would be pushed to a late night only designation. Like Closer by NIN or various Marylin Manson videos. It’s a shocking video, and has only become more relevant in the intervening years. Back then suicide in front of your class was a shocking topic for 13 years olds. Today school shootings are so common that I imagine 13 years olds are shocked by the idea alone, they must be too numb from doing active shooter drills.

Controversial video and the sad state of America aside, “Jeremy” is a powerful song. Listening to it and (silently) singing along to it never fails to give me goose bumps. Truly a classic song on a classic album.

The other song I want to mention from Ten is “Black”. I’ve probably listened to “Black” more than any other song on Ten. In fact I may have listened to it more than almost any other song. Maybe “Pictures of You” by The Cure is in the running. And moth of these songs has something in common; they are my go to breakup songs. The sound track of heartbreak. I won’t dwell on the details but these songs are forever linked in my mind with laying in dark rooms crying, wondering around cities lost in melancholy, an emptiness in the pit of the stomach that you think will never go away.

Suffice to day life goes on, but I can never listen to “Black” and not be moved by it. Particularly this part:

I know someday, you'll have a beautiful life
I know you'll be a star in somebody else's sky
But why, why, why can't it be
Oh, can't it be mine?

The pain and longing in Eddie Vader’s voice here brings me to the verge of tears every single time if I’m actively listening to “Black”. Even writing this I feel a bit of the old melancholy creeping in.

So anyway. “Jeremy” represents one kind of end of innocence, while “Black” represent a whole other kind. But every song on Ten is good. Most popular rock music is focused on love —lost, unrequited, found— but Ten covers a lot more ground; homelessness, mental health (and institutionalization) and the fucked up things families do to each other and their hidden secrets. Grunge in general spent a lot more time covering topics beyond love and Ten seems to set the bar coming very early, before Nevermind and the “Smells like Teen Sprit” moment.

Ready to listen?