Good morning! I haven't blogged for (checks Archive) six months now ... sigh*. That is to say, I haven't blogged here; I just got back from competing in the 2023 505-Class World Championship, in San Francisco, and blogged about it elsewhere.
It was an unbelievable experience filled with problems and solutions and bad moment and great ones, and now that it's over I can't wait to do it again. Or something like it.
This is my fourth Worlds in three years; 2021 J/70 Worlds in Marina Del Rey, 2021 Moth Worlds in Malcensine, Italy, 2022 J/70 Worlds in Monaco, and now this one. Would be most excellent to keep this string going...
* If you're looking at my Archive you can see; I go in and out of blogging. Not sure why I stop, and not sure why I start again... Stay tuned!
Happy April not-Fools! The one day of the year on which we are not fools, because we don't believe everything we read. (Every other day, "it's on the Internet, it must be true!)
This prank was my favorite, from Secret Los Angeles: no more Dodger Dogs! The comment thread on Facebook is priceless; half the posters are horrified, the other half are earnestly correcting the first half. Personally I had a hot dog to celebrate :)
I went sailing today, gorgeous, sunny and breezy, actual heat. Nicest day in about five months. Spring has *finally* sprung.
Waay back in 2012, I posted Moved ... to Facebook. In 2011 I became a daily Facebook poster. Weird to think that was a thing. But it's interesting to think about why it's not a thing. You can do just about anything on Facebook you can do on a blog; post pictures, text, link to things; and there are Likes, and Comments, and all kinds of infrastructure. But it's not a thing, is it? In no way was that the same as posting here. So be it.
Another oldie, from 2008: the lost art of desk checking. That was fifteen years ago, so if it was dead then, it is surely dead now? Um, no. For example, for the desk programs for my series on CUDA and GPUs, I did a lot of it. I guess dinosaurs gonna dino.
This might be the reason AI models don't entirely replace human programmers. OTOH, maybe they become great at desk checking, and it might be the reason AI makes programming more efficient. We'll check in on this in another fifteen years :)
Ottmar Liebert: flow. A subject on which he is expert. "It’s difficult to know ahead of time which way the wind blows. Sometimes one recognizes what’s happening immediately, one feels the invincible flow of creativity, one feels switched ON. Sometimes one can feel the struggle."
News I can use: how to unlock the 100kph achievement badge in Zwift. You have to find the steepest downhill and ride as fast as you can in your biggest gear. Stay tuned!
Steven Wolfram: ChatGPT gets its "Wolfram Superpowers"! A great pairing. This is rather remarkable for how fast it got done as well as how powerful it is.
NotTheBee: I just asked Google's new AI chatbot "Bard" the very same question about both Biden and Trump. The difference in its answers is astounding. This is super bad. And maximally inauthentic.
Ottmar Liebert: Millions of Drops. "Water drops don't sound like rain. millions of drops do." A video editing tour de force.
One step closer to success: Reality Space's 3D-printed rocket launches, fails to reach orbit. SpaceX's early launches didn't make it either. Onward.
Interesting: Oliver Stone Releases Trailer for His Pro-Nuclear Energy Movie, ‘Nuclear Now’. Can't wait to see it.
Elon Musk links Arthur C. Clarke about the future of AI ... in 1964!
I have found this to be true: the more you want to see a video clip on any news site, the less likely it will play when you click it. Weird that this isn't 100% by now, like clicking on an HTML link.
Did the video of Arthur C Clarke play for you?
Michael Kagan, Nvidia's CTO, says other uses of processing power such as the artificial intelligence chatbot ChatGPT are more worthwhile than mining crypto. I have to agree.
Inhabitat: do you think zoos and animal parks are good or bad? A reasonable balanced analysis. They did not however consider the opportunity offered for a great afternoon with a little kid :)
And also... no pictures! I picked one for them...
If you've been around here at all, you know: I
hate dislike woke-ness. So, why?
I have asked myself this question too.
It isn't that I disagree with much of woke-ness dogma, although I do.
My fundamental challenge with many "woke" ideas is that I don't believe government action is the best way to handle them. But I recognize the contra points of view and will happily engage on them.
My dislike stems from the inauthenticity of woke-ness. It's virtue signaling; most wokies* don't understand the issues, haven't actually thought about them much, but have absorbed prevailing wisdom and eagerly parrot it in order to show that they, too, are a "good" person. Does it make them feel better about themselves? Maybe. And maybe that's the attraction.
One tell is that wokies* reject facts which don't fit their narrative. If you think X, and you encounter a fact which is ~X, what do you do? Do you inquire about the fact? Do you process it and maybe come up with X'? Or do you simply reject the fact?
* is wokie a word? No. Should it be? Yes.
I often refer back to an incredibly insightful ontology of ways to disagree, from Paul Graham:
- DH0. Name-calling.
- DH1. Ad hominem.
- DH2. Responding to tone.
- DH3. Contradiction.
- DH4. Counterargument.
- DH5. Refutation.
- DH6. Refutation of the central point.
Wokies rarely engage on the facts, even if there are legitimate arguments available; instead they go for contradiction, ad hominem, and name calling. And there is an even weaker response: silencing the source This is happening all over, and is even institutionalized.
My central gripe isn't with these people's actions, it's with their intent. I have many friends with whom I disagree, and I have no challenge with their disagreement if it is authentic.
On this day, April 1, we encounter a lot of falsehood for the sake of humor. And that's great. I've never laughed harder or enjoyed myself more on account of some of it. You might say this is authentic falsehood.
Unfortunately and irritatingly, our public discourse has moved strongly toward inauthentic falsehood. We say things we know are not true and act upon them, because others will agree, and we'll feel better about ourselves.
The other day I mentioned nut picking, "acting as if the craziest people in any group represent the group". This is not authentic.
I wish the pendulum will swing back, but I'm not optimistic. I see several parallel gradients that reinforce inauthentic behavior, locking it in. The antidote is to question everything. Not just on April 1, but every other day.
End of quarter, filter pass ...
I must tell you, this was not a great quarter for me. I am ready to turn a corner.
Today's picture: Titan! One of my favorite's, of all time; Titan silhouetted by Saturn, as taken by Cassini. This is a real photograph of a real moon of a real planet. Goosebumpy.
We keep reading about GPT and it's various chat brethren; the genesis for all was OpenAI's Codex, an AI system to translate natural language to code. As a coder myself, naturally this is of interest, and the tools now available are amazing. The big question is will computers replace humans? Or will there always have to be humans to train the computers?
Consider Wikipedia, one of the truly amazing human creations. Could it have been created by computers? I don't think so. But now that it exists, it can be used to train computers, for sure...
The New Yorker contemplates the End of the English Major. Really, the end of Humanities. Certainly Woke Politics are a part of this decline, but there's more to it than that: "During the past decade, the study of English and history at the collegiate level has fallen by a full third. Humanities enrollment in the United States has declined over all by seventeen per cent. What’s going on? The trend mirrors a global one; four-fifths of countries in the Organization for Economic Coöperation reported falling humanities enrollments in the past decade." Humanities score high on interestingness, but maybe increasingly lower on entertainment?
[Added: my daughter, a Gen Y who majored in Sociology, thinks students are more preoccupied with their future economics now than previously. How interesting, Millennials are often regarded as anti-capitalist.]
Also the New Yorker: Goodbye, my Funding. From 2017, pre-Covid. It's written as a humorous satire of course, but the underlying idea that "my funding" is somehow separate from "my activity" is an interesting observation, typical of many. Universal Basic Income is one manifestation. If we had UBI, would we have more English Majors?
NASA wants new 'deorbit tug' to bring space station down in 2030. Huh. This seems like a perfect excuse to test new weapons systems! And if I could suggest, we should launch a new satellite to relay video!
Whenever I see pictures of the ISS, I have a Titan-like "wow, this is a real photograph of a real thing" moment.
PSA: Did you know, the IOS Compass app not only gives you directions, but coordinates and altitude? Even works with airplane mode on ... I guess the GPS is still active regardless ... try it next time you're flying :)
Interconnected: My new job is AI sommelier and I detect the bouquet of progress. "I made an AI clock for my bookshelves! It composes a new poem every minute using ChatGPT and mysteriously has an enthusiastic vibe which I am totally into. Kinda."
Not real - yet - the invisible superyacht. In my experience most so-called superyachts have a "me me look at me" design, so this would be welcome.
For superyachts? The lightest pain in the world. It uses reflective properties of materials rather than pigment. Cool (literally!) but also likely expensive...
The IQ Bell Curve for AI:
I keep going back and forth on whether I think AI is dangerous...
Novel drug makes mice skinny even on sugary, fatty diet. Huh. Glenn Reynolds thinks if drugs make it easy to be skinny, skinniness will be less-valued. I'd be happy to volunteer to find out :)
Congrats to tortoise couple Mr. and Mrs. Pickles on the arrival of their three hatchlings, Dill, Gherkin, and Jalapeño. Hehe.
Note: the little Gherkin is actual size :)
Just sitting here thinking about entertainment. It's a bit related to interesting-ness; how do we chose to spend our time?
For much of the world, basic human needs like food and shelter are a given - sadly, for another much of the world, they are not, don't want to be callus about this - but for you, readers of this blog, this is largely true ... and I/you/we spend a lot of our money and time on entertainment.
In our world the most valued people are entertainers - actors, musicians, athletes, writers, etc. These are "hits" businesses - only the tip of the entertainment pyramid is so valuable - but they are broadly leveragable - a great actor, musician, athlete, or writer can entertain many many of us (indeed all of us) at the same time.
It was not always so. Before electronic communication the reach of an entertainer was limited, and hence, they were less valuable. (Writers had big reach earlier, with printing presses...) In my not-yet-written book Unnatural Selection I planned a chapter called "the piano player", in which I noted that in the recent past even a good player could be entertaining to many in a local community, while now you would have to be a great player, but could entertain everyone. Mass communication has fostered a global decanting. But that aside, entertainment was still valuable back then, maybe more so given its scarcity.
Clearly entertainment is a brain thing, so is wanting and liking entertainment a by-product of other selection, or fundamental to it?
Frequent readers know I postulate happiness comes from liking yourself; if entertainment makes you happy (it does) then does it make you like yourself (it must). So why would watching a great actor, or listening to a great musician make you like you? Maybe its by analogy, a sort of inspiration; if that human can do that thing, than I could too?
As a contra point to this: watching other species do incredible things is entertaining too. And what about humor?
It is a mystery. I will continue to ponder :)
Sitting in the United Lounge at Atlanta airport I came across this picture, of a United Boeing 377 Stratocruiser.
I didn't know what it was, had to Google (yay, visual search), but it struck a chord. All those years ago, maybe 60 or so, that plane existed, and flew, and was full of passengers; maybe on their way back to LA from a business meeting in Atlanta.
They would have been dressed differently - nicer no doubt - and would have been thinking differently about different things. Probably carrying books and newspapers. Not planning to watch a movie. Nor blogging while high :)
They would have been chauffeured to the airport and been dropped off at the entrance, porters would have taken their luggage, they would have presented paper tickets, and they would not have suffered needless security theater with long lines and luggage scanners and taking your shoes off. No taking your laptop out of its bag :)
The food would have been better, for sure! The service too. And the flight would have taken longer; Stratocruisers cruised at 350mph, vs 550mph for the jets of today. "It could carry up to 100 passengers on the main deck plus 14 in the lower deck lounge; typical seating was for 63 or 84 passengers or 28 berthed and five seated passengers." The whole experience would have felt more special.
Google tells me the average price of an airplane ticket in 1963 was $41, which equals $323 in today's money. I would gladly pay that amount for that experience. (Maybe with generative AI, soon I will be able! - I predict some of the first uses of a 'holodeck' will be time travel into the past...) Travel+Leisure tells us what flights used to cost in the 'golden age' of air travel. And they summarize: "Security risks are greater and security lines are longer. People don't wear their best suits to fly anymore. Deregulated, democratized, affordable air travel is very different from the glamorous air travel of those far-gone days, but at least more of us get the pleasure of complaining about it."
On my way home from a brief business trip to Atlanta - fun to be "on the road again"; much eating and drinking, and was treated to a nice Hawks game last night - and so I am blogging from 30,000'. You have been warned...
BTW was even colder in Atlanta than has been in LA ... brrr. Spring?
In re Global Happiness: GQ: we're thinking about Happiness all wrong. Are we though? I think they confuse happiness with contentment. True they are related, but I claim they are not the same...
To be checked out: Microsoft bring OpenAI's DALL-E image generator to the new Bing. So now when I want an illustration for a blog post which doesn't have one, I'll just ask Bing to make one :)
Related: NVidia's big AI moment is here. Yep. They are the "Cisco during dot-com era" hardware of the moment.
Gratuitous plug for my posts about CUDA...
Related: NVidia partners with Google Cloud to launch AI instances.
And so who ordered that? An aperiodic monotile! "The authors of a new preprint paper claim that they've discovered what’s called an aperiodic monotile, a single shape that can cover a two-dimensional space with a pattern that never repeats itself exactly." Penrose tiles were weird enough, this is beyond weird. And turns out they are an evolution in an entire family of such shapes...
And finally, here's a cute animal you should know ... the ribbon seal!
The indispensible Visual Capitalist today have given us:
I don't know all the details but I do know their measure of "happiness" was as perceived by the people in each country. It may mean different things in different cultures! In some, being "happy" may not be as important as others, and in some, it may be more relative than others. But anyway it is a common concept, and an interesting measure.
I've mentioned before, I think happiness comes from liking yourself. Maybe if someone was asking about your happiness for a survey about your country, you would answer based on whether you like the country as a whole, not sure. But in some sense happiness is relative to your own expectations.
Does anything surprise you here? My biggest takeaway is continents: Seems like the ranking is:
- Oceana (Australia + NZ)
- North America
- South America
I'm also surprised that India - the largest country by population - isn't happier. I think of it as a happy place, but clearly its people do not. China is the second largest country, not happy, but that's not surprising. Next in size are the US (happy), Indonesia (not happy), Brazil (kind of happy), and Nigeria (not happy). Lots of work to do.
Yay, it's Spring! And I for one am ready. This has been a cold, wet, cloudy winter, and I'm ready for some dry hot sun.
SpaceX launches two space missions in one day. And of course, landed both boosters for reuse. So cool that this has become so routine.
In other news, I flew on two airplanes today, and both landed. Yay.
Seth Godin: Shields up. "What I want is a junkblocker. A big button on my browser that says 'shields up.'" Maybe AI will give this to us, but there will always be the problem of who decides what is junk?
Shelly Palmer: run 'ChatGPT' on your computer. "Here’s something you probably won’t do today: install a large language model (LLM) chat application on your PC. That’s okay, I did it for you – in about 10 minutes."
Microsoft-owned Nuance adds GPT-4 to its medical note-taking tool. This is a perfect application for GPT, I predict this will be huge. One to watch.
Powerline: why wind and solar will never work. "They produce electricity less than one-half of the time, a fact that will never change." If you think batteries are the answer, they have some statistics which make you think otherwise.
So, what's the answer? Last Energy signs deals worth $19B for nuclear plants. Nuclear energy is the only one we have.
Kottke: The puzzling gap between how old you are and how old you think you are. The gap keeps getting bigger. My "how old you think you are" hasn't changed that much for many many years...
Russell Beattie: Quick thoughts on Mark Twain's travelogues. "Seriously, if you haven’t read anything else by Twain except Tom Sawyer, you need to do yourself a favor and read his non-fiction. It's all truly fantastic. "
Have you ever reflected upon bad product names? We are surrounded by them. Why is naming products so hard?
Especially bad are names associated with things that have variations. Like models of cars.
The product people know too much, and they assume we know more than we do. We barely recognize the name of the company, let alone the product, the model, or the latest variation on the model. I get it; you need to have differentiation. So do that, but don't hope that we are tracking every zig and zag of your development.
Take Lucid cars. (Please!) The "brand" is Lucid - and you can be forgiven if you've never heard of them. They are an electric car company founded by engineers who used to work at Tesla. Their car model is the Lucid Air. They also have a Lucid Air Pure, a Lucid Air Grand Touring, and a Lucid Air Grand Touring Performance. And another model called Sapphire, which is a Lucid Air too.
They all look more or less the same, and are distingushed from each other by various features like battery capacity, engine power, and of course price. Lest you think the latter is a detail, the most expensive of these options is approximately double the cost of the least, even though they are for many purposes the same car.
From everything I've read, the Lucids are great cars and have a real shot at being Tesla competition. But they're too young and too unknown to have so many different models and names. Confusing.
Some companies give up on names and resort to numbers. The Nimbus 2000! I get it, names are hard, and numbers are easier. Also, numbers have a sequence, so perhaps customers can figure out that a Nimbus 3000 is a better or newer model.
But where do you go from there? You can add letters. The Nimbus 2000LX! The RX! The LRX! Not to mention the LRX+! It all gets very confusing very fast.
Don't be afraid to keep the same name. Just because it's new, just because it's got a new feature, if it's the same product - the same value proposition - the same name is okay. In fact it's good, because it's stable, and people get to know it.
Tesla has had a Model S since 2012 but the car you buy today is pretty far advanced from the Model S of then. Yes, there have been a few variations, but Tesla have kept them to a minimum.
Apple are another company which have kept names simple. Macintosh. iPod. iPhone. iPad. Apple Watch. Etc. They have resisted the urge to rename with every new feature and version, and we their customers thank them. (Yeah, they do do that "Pro" thing from time to time...)
Concepts like "this is a product", "this is a new version of a product", and "this is a feature of a product" are helpful. But the real thing is to keep your customers in mind. Don't assume they know anything - they probably don't - and try to help them.
I spent this afternoon with my granddaughter (7) at the delightful Santa Barbara zoo.
It's not a big zoo, which makes it perfect for an afternoon with a little kid. You can walk the whole thing and see just about every animal in a couple of hours. They don't have every kind of everything, but they do have some kind of everything: the amazing staples like giraffes, leopards, gorillas, etc., plenty of beautiful and amazing birds, snakes (!), even jumping spiders (!!). And it's pretty; you don't get that "big animal in a small cage" feeling of sympathy for the residents.
And also, lot of good signage, including detailed descriptions of each animal, where they are from, how they live, what they eat, etc. My granddaughter is a great reader and delighted in informing me of all the details of each animal, while I tried to find them. We made a great team.
Zoos are one of those things which bring up conflicting emotions; should we capture and display wild animals? Does this help us learn more about them? And does this help us educate each other about the wider world in which we live?
Those are tough questions; but an easier one is "do they make for a great afternoon"?
(sigh ... bad thing about an offline posting process is when you forget to sync ...)
Yay, it's St Patrick's Day, hope you enjoy it. I'm celebrating by recovering from a cold (cough, cough).
David Sacks comments on Bank of America's balance sheet: look what VCs did now!
Sarcasm aside, I think the key was not their balance sheet - it had problems, but so do many, many other banks - but rather the coordinated sense of panic on the part of their depositors. The vast majority of BofA depositors are blissfully unaware and will happily leave their money on deposit.
Vodkapundit: How to lose the entire Middle East with this one weird trick. Sigh.
Shelly Palmer: Web3 is really here! Woo hoo, we have a new connection type, web3://. And so it's all decentralized and secure, right? Independent of Web2 infrastructure? Um, no. It's centralized, just as insecure, and dependent upon Web2. Do not ask and so then what are the advantages...
Elon Musk: It’s exciting to see more & more public figures engaging in active dialogue on this platform! (Referring of course to Twitter.) Indeed.
I've found myself on and following Twitter way more than pre-Elon.
Also Elon: Twitter will open source all code used to recommend tweets on March 31. Wow. Super transparent but also, a great way to improve the algorithm!
Also Elon: Engineering is true magic. Yes.
Holy cow: landing an airplane on a helipad on top of a hotel. Do not try this at home.
Razib Khan considers Steven Pinker: The Blank Slate 20+ years later. A remember reading it with great excitement, but probably because I was already a student of Dawkins and Dennett, finding it a bit anticlimactic. I think their works were actually more groundbreaking, especially Darwin's Dangerous Idea.
Clive Thompson: the rise and fall of dot-com foosball. Heh. Time was, whenever I joined a new company I would buy them a new foosball table. Did this four times. Not sure I would do it again - virtual foosball is not really a thing, right? I did find it great for team-building and breaking up a day sitting at a keyboard with a little activity.
Berci Mesko on FDA regulation of AI/ML, "adaptive algorithms". This is going to be tough for everyone; the FDA regulates static systems and the processes for modifying them, in discrete increments. Regulating a dynamic process will be quite different.
Dave Winer: on reflection, it's amazing I wasn't killed in my teens. Heh. I have a great friend who was at one time a colleague, and another colleague asked him: "how have you managed to live so long?"
Huh, checking the archive I see I recently passed 3,500 posts. That's kind of cool. It's not exactly a "round" number, 5 x 7 x 10 x 10, but it's round-ish. For those of us with 10 digits it's sufficiently round to mark a moment and trigger some reflection.
Back in January I marked the new year by checking in after 20 years; at that time, I had made 3,406 posts containing 10,771 images, and 24,965 links, of which 912 were back to blog content.
All through that time, as I've enjoyed blogging, I've reflected on why it is fun, and come up with various answers, including introspections on what it means for something to be "fun". It's definitely fun to look back and see what the world was doing and what I thought about it at previous points in time. For example, three years ago the world was shutting down! A time of big change, and who knew what was going to happen.
(This infographic from Visual Capitalist remains interesting!)
So what comes next? We have AI coming to the forefront, a big political divide which seems to be getting bigger, concerning financial news, and big social changes partially caused and mostly exacerbated by the pandemic, like hybrid work. It does feel like things are happening faster; we live in exponential time, for sure.
Anyway I'm going to keep blogging for a bit longer, and hopefully you'll keep reading. Cheers and onward!
So, the Ides are upon us. Did you know:
The Romans did not number each day of a month from the first to the last day. Instead, they counted back from three fixed points of the month: the Nones (the 5th or 7th, 8 days before the Ides), the Ides (the 13th for most months, but the 15th in March, May, July, and October), and the Kalends (1st of the following month). Originally the Ides were supposed to be determined by the full moon, reflecting the lunar origin of the Roman calendar. In the earliest calendar, the Ides of March would have been the first full moon of the new year.
We of course know of the Ides via Shakespeare, who had soothsayer Spurinna warn Julius Caesar about them.
I've already had a terrible, horrible, not so good month, so the warning is late. Onward.
When you read about Chat GPT 4 and things like this, do you feel things are happening too fast? Jason Kottke notes Ezra Klein's use of the phrase "exponential time". There are more and more things happening faster and faster, it's hard to keep track. Best way I've found is to follow blogs as filters :)
Reid Hoffman: Last summer, I got access to GPT-4. It felt like I had a new kind of passport. Huh, sounds interesting.
Powerline note: Biggest victim of SVB collapse? The Climate, of course. "What hasn't received as much attention is that Silicon Valley Bank was particularly important to the climate-tech sector. More than 60 percent of community solar financing nationwide involved Silicon Valley Bank." Hmm.
Scoble: Google should win everything. Are they victim's of the Innovator's Dilemma? Time will tell.
xkcd: flatten the planets. "I don't know why NASA keeps rejecting my proposals to improve the Solar System." hehe :)
Blackberry, the trailer. "Picture a cellphone and an email device, all in one thing." Supercool. Can't wait!
BTW, could YouTube be any more annoying? Just when you think they've reached peak cruft, they prove you wrong.
Onward, hope you have a good Ides!
Big trends in healthcare are wellness, consumerism, and wearables. These things come together in various kinds of monitors which you wear, which tell you, as a consumer, how well your doing. These include smartwatches, various bracelets, and rings, and a leading ring monitor is the Oura. I've had one for about eight months, and after eight months ... meh, not sure what to think.
Yes, I did register a 98 sleep score once, but I actually did not sleep for 10h 26m and honestly didn't even feel like it was a great nights' sleep after...
The Oura "works": insofar as it measures various vital signs and tracks them, and let's you view the result. Heart rate, respiratory rate, blood oxygen, etc. And it's comfortable, has reasonable battery life - charge about once a week, for an hour or so - and it's attractive (I have a black one). It sort of makes a statement: I'm wearing a smart ring, I care about my wellness, I'm hip, etc.
I do wonder about the value though, on an ongoing basis. I can't say that anything the ring tells me has changed my behavior. It's more like I check the ring scores to see if they're right :) If you took it away from me I wouldn't fight hard to get it back.
There is a preventative medicine aspect to these devices; if something changes or goes wrong, you will know sooner and can alert medical professionals to take a closer look. That might be valuable. And if a physician wanted to see my heart rate history or something like this, then I would have it, and that would be valuable.
I'd say this is a transitional technology. Someday we'll have implanted monitors and they'll be more comfortable, more accurate, and more useful. For now, they provide a reasonable excuse for a blog post.
Fairly recent posts (well last handful, anyway):
For older posts please visit the archive.