I don't mind paying for hardware. I do mind my privacy, and don't want that kind of information in the cloud, or even traces from encryption I haven't audited myself.
2,142 karma · joined June 28, 2022
By "I don't fully understand," I mean just that. There's a lot of marketing copy, but there's a lot I'd like to understand better before plopping down $$$ for a unit. The answers might be reasonable.
Ideally, I'd be able to experiment with a headset first, and if it works well, upgrade to the $59 unit.
I'd love to just have a README, with a getting started tutorial, play, and then upgrade if it does what I want.
Again: None of this is a complaint. I assume much of this is coming once we're past preview addition, or is perhaps there and my search skills are failing me.
The way I read this: Elon Musk wanted OpenAI to commit fraud. They refused. He went away. They decided to commit the same fraud. He sued.
They're a strong case in both directions. There's a legal principle that you can't take opposing views on the same legal issue.
If Amazon / Walmart sells something for $149, and Microcenter does as well, I might go with Amazon / Walmart for convenience of having it shipped, or Microcenter for speed of having it now.
MAP can give pretty high margins too, in some industries.
I think you're misunderstanding how these operate. Almost all of these are staffed and operated for at least the same number of hours as a typical Radio Shack location.
> to a sector you have no practical or institutional knowledge in, which doesn't exist already
No, Radio Shack had more institutional knowledge here than just about anyone else. They made electronics kits, wonderful books, etc. A better example is AoPS opening up physical tutoring center (which they did, and which has been very successful).
> close down 7,000 stores (to roughly match the number of Kumon locations), get rid of all of your inventory
If you do this, you're managing the pivot like an idiot. Here is how you pivot:
1) Start a summer camp at one location. Iterate on the program until it's efficient and turnkey.
2) Start a small number of afterschool programs in a similar fashion
3) See what it takes to be successful. Perhaps grant some kind of certificate for those, and ideally, frame it so it can go on a college application, or if you're really ambitious, so amount of college credit. I can name a dozen other places this could go, but critically, it takes a bit of financial elbowroom to have time to explore and pivot like this.
Once this is going, expand reach to more and more stores. And do this in 1990, so you have time and elbowroom, rather than in 2000.
For big institutions, pivots almost never work wholesale.
> close down 7,000 stores
Scaling down was probably necessary at some point, but bankruptcy was not.
Scaling down can be followed by scaling back up.
Let's shorten this: "if they had not pivoted, they would have gone out of business in 2000." A pivot was needed, and a long time earlier. They key question is which pivot.
Identifying a big market, with a lot of competition and no competitive advantage, was better than nothing, but worse than identifying a market with some competitive advantage.
> You literally lost Radio Shack money and hastened the transformation every time you walked past the CD players and plasma spheres towards the back of the store where the components lived.
True.
To make money, they would have needed to monetize me for a lot more than the price of a component. Have a quick look to see what Russian School of Math charges for afterschool math classes, what summer STEM camps cost, or tuition at a university. More critically, look at the derivative -- which prices are going up and which are going down.
If you earn 50 cents selling me a resistor, I'm losing them money.
If you earn $1000 teaching my kid something about electronics. All of a sudden you need far fewer of me, and you broaden the appeal to anyone educated with money and a child.
They had everything in place to do that, right down to wonderful educators like Forrest Mims.
And if they had done that well, there would have been many more of me around now too.
The ones that died were the ones who tried to lower prices to compete with online, had commercial books selling shelf space (rather than curated selections), and minimum-wage employees who didn't know what they were selling.
I don't mind paying for good books, and what I'll pay for is having a person who has read a lot of books and can recommend what I'll enjoy.
The problem is that in 2024, kids can play video games and watch stupid videos in the comfort of their own home. Perhaps screens beat animatronics, ball pits, and whatnot, but there's no competitive advantage over superior alternatives in the new space.
Examples:
- Book stores. Cut costs to compete with online sellers. Move more and more digitally. Have no upside over pure online stores. Keel over.
- Radio Shack. Stop selling electronics components, and start targeting to the bigger market of cell phones. Have no competitive advantage over Best Buy or cell phones stores. Keel over.
In those situations, it's better to either:
* Shift business models (e.g. Radio Shack could become a very competitive makerspace, host kid afterschool programs, maker camps, and refocused on Raspberry PI, 3d printers, Micro:bits and similar).
* Shrink the business to follow a declining market without taking on debt (Radio Shack couldn't support the number of stores it had, but it could very much have supported 1/5 of the stores)
Both of those approaches usually require starting to adapt before the sky starts falling.
Need something to spread on social media? Fine. Post on Twitter, Facebook, Google, go nuts.
Never use Google (or other sketchy vendors) for anything you need to rely on to be there tomorrow.
Example: Youtube is fine for promotional materials. If you run an ed-tech and need videos to be there for students (core business), pay for something which works.
Example: Running a Black Friday promotion? Adwords. Relying on Google SEO for your main business as the main way to recruit customers? Bad idea.
You get one exception to this if you're building a Google-centric business. Is your business an Android App? You obviously get to use Android in your core business. If your business a Youtube channel? You get to use Youtube.
Things like Google Workspace, Google Cloud Platform, etc. are out. If, tomorrow, Google decides to wipe all you Google Workspace data (yes, it did this to a startup I was involved with), you're DOA. Office 365 is annoying compared to Google Workspace, but still worth it, since Microsoft won't wipe out your business as a statistic.
The way I think about this is each business has risks which multiply out:
(failed to execute technically) * (odds of market fit failing) * (odds of not being defrauded by your co-founder) * ....
This should multiply out to how likely your startup is to succeed. The way exponentials work, most of those risks need to be very, very low. You get one, maybe two, big ones. Every Google product in the core critical path has perhaps 10% odds of wiping out your business. Adopting Android is totally worth it if your business is Pokemon Go, but for most aspects of your business, pick trustworthy vendors.
> Google's records retention policies were also over the top and perhaps hurt it: https://www.nytimes.com/2024/11/20/technology/google-antitru...:
If you're intentionally hiding things from government investigators, the legal presumption is there's a good reason. Judges are allowed to impute things from destroyed evidence. Otherwise, everyone would destroy evidence.
To me, it doesn't feel extreme at all, relative to former antitrust regulations or to what's needed for a functioning market. I would have felt bad about these about two decades ago, but Google has not been a good player in recent years.
Geopolitically, it feels off.
Personally, I'd love an 8k which could act as two 4k in the middle, and two vertical 1920x4320 sides.
I use vertical monitors for documents, documentation, and word-processing.
It's also not out-of-line with what credible sources observe:
https://blogs.bmj.com/bmj/2021/07/05/time-to-assume-that-hea...
I'm affiliated with MIT, and have been for the vast majority of my life, including at points in fairly senior roles. If you shut people out pointing problems, it will never get better.
There's an incredible urge to defend elite academic institutions, but it's not in the interest of those institutions. Remember your civics class (patriots criticize their government institutions).
The only way I see this fixed involves a period where MIT is viewed like a used car salesman in the public eye for at least enough years to cause enough pain to lead to reform. The endowment is big enough it'll do fine in the end. If it keeps sliding to fraud, it won't.
* I suspect that, like prohibition or DMCA, criminalizing behavior will lead to worse outcomes than accepting deepfake porn sometimes happens, at least if laws are passed targeting this in isolation.
* I suspect that before acceptance, many people will be harmed, on both sides. Humiliated teen girls. Immature hormone-driven teen boys.
* I suspect that we'll see a range of bigger problems such as real scams where, for example, a deepfake child calls their elderly parents and empties their bank account. Or, perhaps, a deepfake general calls someone in a nuclear bunker.
* Deepfakes are at the tip of technological problems which need solving. We have other things, like the cheap and easy availability of technology to make the next Covid-19. The point is we have the technology to make it in a lab, and the price point is going down by an almost Moore's Law like curve.
I don't see a good solution short of completely rethinking democracy and capitalism from scratch. The first time around took the Age of Enlightenment and the Age of Revolution.
It's perhaps time to do that.
Prove me wrong. Seriously. If there are incremental solutions which stop deepfake porn, deepfake scams, personal drone strikes, and super-viruses, please let me know what they are.
Athens had a democracy over 2500 years ago. A few Native American tribes had long-lasting democracies. Ukrainian cities were democratically self-governing 500 years ago, and Poland had elected kings.
Those were isolated examples. This was not a revolution. We also haven't regressed; isolated examples continued throughout history. If you point to a year, you can probably find some democracy somewhere. The only major regression I know in history was around 1000BC. Regressions are rare.
What changed was a revolution. From just before 1800 to just a little after 1900, virtually every country had a revolution which led to either being some form of democracy, or pretending to be one. Democracy was no longer isolated. We had the creation of a free world covering much of the world's population, and the creation of what was pretending to be a democracy (today, even the Democratic People's Republic of Korea pretends to be a democracy).
The number of countries which claim to not be a democracy, you can count on your fingers. Iran. Vatican City. Saudi Arabia. UAE. Oman. Eswanti. Did I miss any?
https://en.wikipedia.org/wiki/List_of_countries_by_system_of...
I'd like my critical supply chains to be in North America. Ideally, I'd have a free trade zone with Mexico and Canada. Tariffs, two-sided, reduce trade, and bring me closer to that reality. I'm okay paying slightly more for that security.
I'd like the same thing in the EU too, and in other major world blocks.
There's a slight efficiency gain from specialization, where each country brings a skillset, but it's totally not worth it to me versus the brittleness which comes with it.
Remember how the Covid supply chain disruptions cascaded? That's a tiny fraction of a really bad natural disaster.
https://www.nytimes.com/interactive/2024/09/09/technology/ai...
https://www.nytimes.com/interactive/2024/01/19/technology/ar...
These are a little bit unfair, in that we're comparing handpicked examples, but I don't think many experts will pass a test like this. Technology only moves forward (and seemingly, at an accelerating pace).
What's a little shocking to me is the speed of progress. Humanity is almost 3 million years old. Homosapiens are around 300,000 years old. Cities, agriculture, and civilization is around 10,000. Metal is around 4000. Industrial revolution is 500. Democracy? 200. Computation? 50-100.
The revolutions shorten in time, seemingly exponentially.
Comparing the world of today to that of my childhood....
One revolution I'm still coming to grips with is automated manufacturing. Going on aliexpress, so much stuff is basically free. I bought a 5-port 120W (total) charger for less than 2 minutes of my time. It literally took less time to find it than to earn the money to buy it.
I'm not quite sure where this is all headed.
The heat-affected zone is caused by the weld. Ergo, welds to weaken the base metal.
In most cases, this also doesn't matter. I think all but one of the things I've welded, even a bad weld would have been way more than strong enough, and for many, even the tack weld would have held fine. Welds are very, very strong, and it's usually cheap and easy to use sufficiently strong materials that all of this is moot.
But for something like an ultralight bike frame, racing car, or airplane, it is something you do need to worry about.
* The equivalent is insurance. Insuring a $2B ship carries pretty good due diligence. A ship simply failing is rare. Of course, ship insurance doesn't care about employee rights, safe food, medical care, or other things one might expect to keep people safe. It's about protecting the capital expense. If everyone on the ship dies, but the ship survives, that's okay!
* Welds are quite strong -- it just extends the metal. This is especially true when the baseline quality of the metal is not high.
On something heat-treated SAE AISI 4130 steel (what e.g. fancy steel bicycles are made of), you see significant weakening. There is a heat-affected zone where the normal tempering is taken off, and the joining material isn't the fancy CrMo of the baseline material.
I'm not a nautical engineer, but I doubt cruise ships are made of overly fancy steel. When you're making a 180,000 ton ship, your best bet is to use cheap steel, and if you need more strength, to simply use more of it. A good weld should be every bit as strong as the cheap steel around it, and the heat-affected zone is a lot less important if the steel isn't heat-treated or tempered in any way in the first place. It will harden the steel a bit, of course, but it shouldn't be the same level of impact.
It's also worth noting you already have welds, and things need to be engineered for welds. It's not hard to reinforce the welds. Indeed, on a bike frame, the welds are where the stresses are highest, and you get around that by making the tubes a bit thicker (or, for fancier bikes, thicker just near the welds -- that's what a butted bike tube does).
I think cruises are horrible, horrible things for a whole slew of reasons, but none having to do with the ship sinking Titanic-style.
> In the real world a lot of work was done to minimize that over x86’s lifetime. In the early days RISC did do a lot of what was promised, but clever (some would say hacky in some situations) updates to x86 CPUs and compilers made these advantages less (although one could argue the x86 microarchitectures that showed up in the mid to late 1990s were more RISC like).
They were identical to RISC architectures, with the exception of a more complex decode unit, which led to extra transistors burning extra power and costing speed in the early Pentium days.
That mattered a lot when a 1993 Pentium had 3.1 million transistors.
That matters a lot less when a modern Intel CPU has ≈5 billion transistors.
Fundamentally, that's what allowed x86 to pass all the various RISC architectures in speed. Those kinds of differences in instruction set just don't matter anymore. At that point, it was just R&D dollars, which Intel had more of due to volume.
As a footnote: It's worth remembering StrongARM. The SA-110 was just as fast as the fastest Intel had to offer, when introduced in 1995, at a much lower price and power budget. It was crazy I could get a little embedded board for $100-$300 which was just as fast as $5000 computers. Intel acquired it from DEC. Oddly enough, it barely improved from there on. A decade later, clock speed went up 4x on the XScale (and only on the very top-end), and more than 15x on the x86.
NVidia's focused R&D efforts to improve Intel platforms.
The Intel platforms are still very, very good, and very, very hard to replicate. It took many decades of concerted optimization to get here. AMD is a little bit ahead right now, but it took many decades of R&D to get there.
Other efforts to do so -- even when x86 was much simpler -- went nowhere. Remember WinChip, MediaGX, Geode, C3, C7, Nano, 5x86, 6x86, and that whole debacle of Intel-competing products?
I'll also mention:
- ChatGPT was a lot smarter for the types of things I'd like to use it for before the "safety" work. The efforts to make it never say anything offensive also made it a lot dumber. I saw a paper to that effect too. Perhaps some of that was a result of optimizations to shrink the model and make it more efficient (but then I'd expect GPT-4 to be smarter).
- GPT3, which was an autocomplete, could do some things well the modern chatbot-ified models can't.
- I have a lot of practice making the models smarter by having them check their own work, vote, etc.
I have no idea how high the ceiling is. I might be a few inches away, or it might be the sky.
That's a good time to hedge.
Probably "legacy" was a better word than "obsolete." My language was sloppy.
* AOL was very, very overvalued.
* They bought a company with real worth.
* When the .com bubble popped, they were left with Time Warner.
It'd be a dead company without it. On the other hand, it did very badly by Time Warner and its investors.
There are two possibilities here:
1. AI is gold, and NVidia is leading AI.
2. AI is a bubble, and NVidia is overvalued by a factor of 10-100x.
In the case of #2, buying Intel converts funny money into real value, gives Intel engineering capability it so obviously now lacks to continue to push x86 forward, and gives NVidia engineers many, many decades of x86 R&D. In that sense, it can be thought of as a hedge on AI not succeeding (or someone overtaking NVidia in AI).
In the case of #1, it either gives NVidia huge marketing and distribution channels, an end-to-end platform play, and a bunch of supporting IP, while chaining it to an obsolete corporation with 5x the employees. It also allows NVidia to do some antitrusty things by connecting with the dominant CPU platform (e.g. making iGPU chips, and various sorts of hybrids where Intel + Nvidia work best together).
The $1800 version is exactly what I'd love to have as an extra monitor on my main computer, if not for the price tag. 3200x1800. 25.3". It looks really good. Unfortunately, as is, it's a little outside of my wallet range.
There's a smaller $800 version too. 1650x2200. 13.3". That looks more usable.
To me, a major upside of the OkMonitor is the wireless connectivity. That's very, very nice. Especially if the computer dongle could be made small enough for use with my laptop while traveling, it would be a very nice solution.
If money is no object -- or developer time is expensive -- or failure is expensive -- it'd be my go to source.
For what I'm doing now (education), pricing is quite a bit too high.
I don't think there's a business model there (I suspect the sales would be in the thousands or tens of thousands of dollars -- so no matter where you set your profit margin, you're not covering your time or much beyond costs), but I've sometimes seen things done like that on hobby projects just for impact.
I'd love to have a wireless eInk display for my computer, ideally from a USB-C dongle (so it gets video and power), but HDMI + USB power would be fine too.
If investors think it's a good idea, market cap goes up by more than 24B. If not, by less.
I am looking for honest thoughts. I have not been spending any time on wallstreetbets, and I had hoped for a good discussion of pros and cons. I have no clue if it's a good idea or bad.
AOL / Time Warner comes to mind.
FWIW, I was pointing out aspects of such an acquisition, and looking for honest thoughts. I have no idea if it's a good idea or bad idea. I'm really quite divided.