I'd argue it slightly differently, but I do agree with what you're saying. The personal computer has been tried, because when Sinclair or Acorn sold you a home microcomputer in the 80s, they didn't have the technology, much less the money to maintain control over that machine once it left the warehouse. What hasn't been tried is the personal communication device. Outside of talking face to face, we have always relied on corporations and infrastructure to carry our messages greater and greater distances, with the possibility of any one operator or carrier reading the contents - sometimes it was even required as telegraph operators would have to read a personal message in order to key it into the line, and another operator would have to do the reverse at the other end. The closest we got was pen and paper, but you'd have to deliver it personally to ensure its secrecy, and even then you risk having it stolen or being seen making the delivery.
Im not sure why i would want any of this? Like i dont want to grow my own food, why would i want to handle the logistics of computing and internet usage myself.
It's irrelevant what you want, you can't have a personal computer anyway. Even Urbit is only working towards one day being able to make you one. A noble goal. Because as it stands right now, even if you wanted to embody the farmer analogy for computers your first step is buy someone else's product. That a'int farming.
@@John-rw9bv yeah I don't wanna be a farmer, I like going to the farmers market and buying their labour and products, I've worked on farms before it's full of boring tasks in glad I'm liberated from, why would I want more tasks in my life? I'm a wagnerian and that means to be taskless.
im still on a 2015 macbook pro. What you have to do is to open it up and clean it from dust and replace the thermal paster every couple of years. As long as you can open it up and change the battery they can last for a long time. And when the OS gets to archaic i can just switch to linux. I see the same thing in cars. A 10 year old car is still great and will still be great if you get a good one and take care of it properly. Will easily last 10-20 more years. Technological progress now is mostly gimmics and marginal improvments.
True. although, if you work as a software dev (or computing heavy work) though, replacing a notebook every 2-4 years is a insignificant cost of doing business. Spending $200-500 / year on my computing needs (amortized with recoup) is not a big deal. I could see getting the $2k to get into the first setup as a significant expense however. But you are right. a can of compressed air and Pentalobe driver gets you an extra 2-4 years for under $20. add $10 and an hour for thermal paste / pads and you can get 20%+ performance just from throttling being reduced.
To my mind, planned obsolescence is gluttany in practice. Consuming more than is needed. It is an actively self harming practice. (An observation, not a judgement.) I have been plenty gluttonous myself. As I move through time though, I see more of the consequences. Great video, great food for thought.
I'm not deep into this topic, but I feel like you're describing Urbit running on Arch Linux or something similar. I would expect there be some sort of symbiotic relationship between these communities but I could be missing something. Doesn't each practically require the other? What am I missing?
The basic idea of Urbit is that Arch is bloat. The "trusted core" needs to be something so concise that a human can read it in a few hours and agree that it's reasonably safe. This puts it in the league of things like Slackware, Minix, OpenBSD to a degree, and other minimalist OS's. All Linux's are basically out. Too much reparded networking protocols, e-mail, all the RFCs for printers... Do you really want to learn the ICMP protocol implemented on your daughter's iBarbie? It's supposed to be 50x simpler than Arch, but still can save and record your daughter's asinine diary updates off her barbie and into a safe space in the cloud.
@@drewbradford7608 I should have added that I also "would expect there be some sort of symbiotic relationship between these communities but I could be missing something". My response is only why they don't use Arch as-is. But your point is why isn't there a symbiotic relationship of operating-system knowledge and contributors to both projects. This is a better and broader question, and i didn't really answer it. I think it's because of a deeper and more fundamental issue than "trusted software", and that's "trusted hardware". You can't trust hardware, at least not modern hardware. Time and time again we find hardware that has features that weren't in the specification, if there even is a specification. Usually it is what you'd expect, high-level backdoors used by nation states. Arch is designed to run on Intel (x84) systems, SPARC systems, ARM systems, and i've personally even ran it on a RISC-V system. But make no mistake, the OS is always subservient to the hardware, or at least it's drivers. The hardware can always choose to lie to the OS. The OS can't lie to the hardware, it has to request access to the memory it stores itself in, i.e. there is no place the OS can write data to out of the prying eyes of the rest of the computer hardware. This is why Urbit isn't built to compile down to assembly instructions (x86/SPARC/ARM/RISC-V/etc) but to C, because essentially it isn't a real-world tool yet but an academic project into what a personal computer could look like on ideal hardware - should that hardware ever come to exist. Where as the Arch team ONLY care about writing software that does pragmatic stuff on hardware that must already exist (and in large numbers). So these design goals or rather philosophies are at odds. And i have to say, Urbit includes in it's design a lot of the best ideas about how to make private personal computers. There's a lot of convergent evolution, which is always a good sign. So the project is clearly on the right path towards making something that resonates, uncovering something that's really there and waiting to be used by people. IDK if it'll end up with Urbit's name on it but there aren't many other competing projects. The "Civ Boot" guy on RU-vid has a lot of the same thinking.
@@John-rw9bv This is helping me a lot. I'm interested because I'm considering contributing and developing myself in this direction. I guess there is an unanswered question that I have for myself which is, what forces are causing the hardware to be untrustworthy? Assuming humanity can solve whatever that force is, isn't there an academically possible future ideal state of open source trust-able hardware that runs on ideal state open source software that partitions your disks, allows you to plug in an ethernet cord, opens you up into a sovereign terminal, sovereign browser, where you can then run the already fully mature Urbit install? Let me know if you see any holes in this, Thanks.
@@drewbradford7608 The assumption that the hardware could be trusted is a big one because silicon chip manufacture and post-process verification are at odds with one another. A leather jacket for example takes a lot more effort to make than it does to QC that it's correct. A car too. A CPU however requires a microscope /more/ precise than the microscope used to burn the image into the chip in the first place, which was at the time leading edge human technology. So, in chip manufacturing they don't QC the chips optically at all, or even anticipate every circuit burned in will work. They are tested for compliance functionally, but that leaves open the door for a bunch of functionality that is "non-compliant" to say the least, take a look at Intel's Management Engine particularly the hacker conference talks at DefCon. Basically, you need a verification process (for the chip) that is intuitive that people can trust, easy and cheap to perform, and results in a chip that is fast enough to play Minecraft. No one is selling these chips. No one is making these chips, except perhaps some military chips basically. Doing what you suggest would in theory be possible, but by the time you solve the hardware problem you are so deep into this problem that Urbit is kind of moot, you'd be better off spending your time trying to just make the most secure PC practically possible (right now that's probably some combination of hardware that OpenBSD supports) and living your best life. Because no one would be surprised if in 10 years it turns out Urbit was some Fed honeypot to trap crypto/bitcoin idealists (with lots of time on their hands). The specific thing i'd also point out is that parts of Urbit are closed source.
It aggravates me that Luke has 10x the viewers, in a fair world Justin would be over a million because what he's saying is so sane, you'd really have to be insane to hear it, agree, and buy an iPhone anyway for over a grand, nuts. Everything needs to stop until the personal computer problem is solved and idk if Luke is actually the right guy to go to for that radical outlook. Luke just wants to buy more chi-fi headphones and Thinkpads, not really save the world.
@@John-rw9bv I didn't mean to cause any aggravation, I wasn't even aware who had more viewers, I'm just a fan of this podcast. Apologies if this has been brought up a lot, I usually consume via podcast feed.
@@drewbradford7608 No no i'm not upset with you haha, i just wish RU-vid's algorithm worked better than it does - i completely agree with you regarding Justin & Luke, although Mental Outlaw would be an even better fit IMHO because if Urbit worked he'd be the first person to use it.
