I'm glad it was so easy for you. For me, I've been living like a self-care saint for over a year. Meditation, mindfulness, gym three times a week, active recovery, sufficient sleep, 15k steps a day, perfect diet, low-stress job, fulfilling sex life, morning sunlight, ice baths, saunas. I'm not OCD or up tight about it, but I'm very consistent. I get constant comments about how I take such good care of myself. A lot of people say they have never met someone who take such good care of themselves. But recovery has only happened very very slowly over the course of many many months.
Despite our best efforts, we are deeply irrational. Our thinking is based on instinct, not on core principles; it's a top-down approach driven by feelings.
> At Digg, one mentor gave me the savvy advice that the fastest path to financial success was working for four years at three different just-about-to-IPO companies. A surefire way to retire by forty. (This is, for the record, pretty good advice.)
Why is it good advice? What is the context in which it operates?
The most burnout-inducing environments I’ve worked in have all had one thing in common: They had a person or class of people that had wormed their way into being in control of everything while being responsible for very little. This left the rest of us to be responsible for the consequences of their decisions while having little to no input on the things we were held accountable for.
The absolute worst company I worked for had this separation as a core philosophy. They had different managers for everything: Product managers to make all of the product decisions, UX managers to make all of the design decisions, project managers to decide when we’d do things and how to check in on our progress every few hours of every day, program managers who thought they were engineers who just didn’t write code, VPs who would choose all of your programming languages and frameworks for you, and on and on. These people would shuffle from meeting to meeting every day with a 2-hour company-paid lunch in the middle (which they ultimately got in trouble for) but wouldn’t ever do any of the work themselves. Meanwhile, if any engineer dared make a suggestion we’d get a long lecture about staying in our lanes. Then when projects were late/wrong/failed or just missed the mark about what the company needed, they would do long post mortems to assign blame to different engineers for doing it wrong. At best, they’d come up with vague statements about how “we failed as a team due to communication issues” or something.
It was the most demoralizing work environment. Every meeting was full of sad, dejected engineers. People would quit without having other jobs lined up because they just couldn’t take it any more.
it runs slightly slower on the GPU than under llama.cpp but uses much less power doing so
I would guess the slowness is due to immaturity of the PyTorch MPS backend, the asitop graphs show it doing a bunch of cpu along with the gpu, so it might be inefficiently falling back to cpu for some ops and swapping layers back and forth (I have no idea, just guessing)
I very recently purchased a MacBook Pro (M1 Max) with 64GB of ram. I haven't experimented that much, but I was able to run inference using the 65B parameter Llama model with quantized weights at a speed that was reasonably usable (maybe a touch slower than ChatGPT with GPT-4).
I haven't attempted to use the 65B model with non-quantized weights, but the smaller models work that way, if slowly. With 96GB of ram -- the upper limit of a MacBook Pro -- you might be able to use even larger models, but I think you'd hit the limits of useful performance before that point.
I should note that it can be a bit tricky getting things to work using the Mac's GPU. I couldn't get Dolly 6B to run on my work MBP, which theoretically should have enough ram, though I still want to try it on my personal laptop.
Because going through a struggle tends to turn people inward and makes them see everything, including the harm they do to others, as if they were a victim just doing what they have to do to survive. It is very hard to overcome it because you have to become a kind of martyr, who accepts the reality of their suffering but who's too holy to blame it on everyone else, or even on yourself.
Kids are often nicer than adults because their lives are so easy, but if they don't learn to bear a cross (this is the best metaphor I know) they'll become nasty as they get older, as people reject them and their health starts failing. It's one thing to be a nice young man, another one entirely to be a nice overweight balding 60 year old with joint pain and a skin disease. That's not to say it's about age. It's about how good you feel, and age is just the big conveyor belt that everyone goes down whether they're ahead or behind their demographic.
You can watch this play out in you next time someone says or does something annoying when you're still smarting from a minor injury, like a stubbed toe. You'll tend to act as if they were the ones who stubbed it because blame wants to earth itself.
Sonder: n. the realization that each random passerby is living a life as vivid and complex as your own—populated with their own ambitions, friends, routines, worries and inherited craziness—an epic story that continues invisibly around you like an anthill sprawling deep underground, with elaborate passageways to thousands of other lives that you’ll never know existed, in which you might appear only once, as an extra sipping coffee in the background, as a blur of traffic passing on the highway, as a lighted window at dusk.
This word was popularized by the site Dictionary of Obscure Sorrows, and I believe it, as the opposite, hearkens back to solipsism, the philosophical idea of you being the only one to exist. It is interesting to see this word "sonder" not referenced at all in the article, I would have imagined it a meme within the Internet at this point.
My limit3d understanding of the world showed me a certain close friend as an extremely anxious, borderline narcissist person.
This person has over the years, progressed to a highly sucessful career, broadly admired by peers. But having known the person for years I know whata under the hood and the constant calls let me know about all the deep insecurities that are unfathomable to me.
This person has an attention to detail and focus that I didnt think humanly possible. He will replay every conversation from many angles, looking for clues or things to attach insecurities too.
This person would be a formidable spymaster for any king. Its incredible how much information iteration a person can do with the right incentive (anxiety)
Here is the rub. Beauty is in the eye of the beholder. What I used to see as a flawed, broken individual, I've come to see as a wonderful example of how nature has a purpose.
You may need to keep your distance from these human steamrollerss, dont get in their path! but dont forget to appreciate the machinery at work. And of course, if there's a path that needs carving..
For any activity in which I do not feel safe and threatened for my Identity, I utilize the middle relay as a full on end-to-end Wireguard VPN itself to route all traffic through a VM I've got specifically built for this.
I run a Tor middle relay on one of the 8 IP addresses I have purchased as a block from a certain ISP that allows you to, I have been for around a year. The amount of traffic passing through it is heavy. Obviously, this comes with certain caveats (the middle relay's, or any TOR relay IPs are publicly available and published weekly on GitHub and as you can imagine, some places like to instant ban anything to do with TOR).
Since it is only 1 of the 8 IP addresses; the other 7 remain free from blockages of any kind and the one running the TOR middle relay is setup in a manner in which I can use it normally (for the most part) and my traffic would just "blend in" with the normal tor traffic passing through it.
You might ask, what is the purpose of this? Well, if it is normal for a lot of TOR middle relay traffic to be passing through one of my IP's on a daily basis, plausible deniability becomes a real defense as checking DNS logs becomes a moot point as there are requests being routed 24/7/365.
This is an excellent, detailed, and in-depth guide of the process of going through running a TOR middle relay. The statistics provided and data presented are simply superb, Great read!
Quoting myself: I very seldom feel inclined to refuse. That said, interviewing for my current position (fully remote team), I was given the option of video or voice call, and I chose voice. I chose that cautiously because I knew it might be offputting but I also knew it might be a good indicator of my prospective team’s culture if it had been offputting. Fortunately that wasn’t an issue!
And I now do a video chat about once a week, which isn’t particularly demanding, and it’s generally a positive experience because we work well together… but I do take an hour or two to decompress afterwards.
Tor -> buy mullvad for xmr -> use it for clearnet ip after Tor
Best for privacy, best for abuse.
Arent there any problems like captchas everywhere because the ip was overused?
Or CP distribution lawsuits towards mullvad?
I actually got into software/systems engineering working on internal software for "minor" player groups in EVE. The level of integration[0] those orgs had 10+ years ago outranks 80% of my current real-world clients in terms of discoverability, documentation and depth.
I mean that's what you get from a bunch of eve-playing nerds committing to a labor of love. Best projects I've ever had.
[0] Real time updating mining & trading boards by location, 100bn+ ISK inventory tracking down to the cent across many corps, characters, inventories & contracts, ... Killboard & market data feeds, risk evaluation, resource allocation optimisers... Absolutely impeccable archives & backups going back 5+ years, on top of postgres & flask iirc.
edit: And audit logs EVERYWHERE. Seriously, I've seen better CYA/KYC/chain of responsabilities from eve recruitment corps than some actual for profit entities.
Tesla and SpaceX are technology/engineering problems. This is what Musk is good at and can take huge risks to go after it. Twitter is NOT engineering or technology problem. The politics and policies are not Musk's forte, at least not yet. However, the more concerning part is his continued defragmentation of attention. Tesla is now genuinly falling behind with massive new competitions rising up. New models haven't been on market for a while. Previously announced models aren't getting delievered. Pace of growth of superchargers is not keeping up. There is a lot at stack where Musk can make huge difference.
It's probably worth noting the ongoing LinkedIn vs HiQ case [1], where at least initially the legality of web scraping, and generation of databases via that scraping, was upheld.
To paraphrase my brilliant cofounder, mentor and tough sparring partner:
"The biggest lie you techies tell to yourselves is 'Build it and they will come'. Nope. Nobody ever comes by. I've beaten competitors with the better product again and again. Know why? Because marketing builds growth, growth builds revenue and revenue enables hiring brilliant people to fix your product, all the while the brilliant product people are still waiting for customers."
Let's say I've have been having kicked my ass more than I can count on some beliefs I've had, but I yet have to regret working with him.
The essay "Consider the Lobster" by David Foster Wallace came to my mind. In that essay he describes how extraordinary the nervous system of a lobster is and how mindlessly and cruelly we boil them live. If David were still alive, he would definitely welcome this development. I really miss David. I am awestruck by his other essay "This is Water". I am amazed at his deep insights into human nature and how well he describes it, kindly.
I find the idea that pain is something profound and not just the reception of a signal strange and based in non-scientific beliefs.
Your CPU has a temperature sensor, does that mean if you hold a flame to it and it can detect its own impending destruction, it deserves protection under the law?
Just because something has sensors somehow makes it our contemporary?
Did somebody in the room say, "We've crunched the numbers, and our best estimation is that if we orient all our resources toward achieving this goal and put forth our best possible effort, we'll be able reach 5x."
And then did somebody else in the room say, "I agree. So let's be extremely audacious and shoot for 10x"?
Either the first person was miscalculating or the second person was being unreasonable.
One of my issues with OKRs is the oft-repeated line that the goals should be so audacious that you should only expect to achieve about 70% of them.
If you're following this advice, you are either purposefully choosing goals that you do not know how to achieve, which is a problem, or purposefully choosing goals whose achievement relies on way too many externalities over which you have little control, which is a problem.
I'm not saying external factors shouldn't be considered, or that we should only set goals that are within comfortable reach.
Rather, I'm saying the uncertainty should be accounted for differently. It should be baked into our goals, not treated as if it only matters at the OKR level.
Imagine if you were to plan a sprint, and your story points did not factor in uncertainty at all. Then, you committed to a sprint size that also did not take into account uncertainty. But your stakeholders agreed that a 30% sprint failure rate was acceptable.
Now imagine baking in uncertainty into your estimates and then committing to a sprint size you can realistically hit. Add stretch goals if you want. Doesn't that seem more sensible?
Jamie's channel is one of my all time favorites. I love watching him work exactly because of his general attitude toward problem solving and how different it is from mine.
The video they put out is extremely cringe. The whole “metaverse” thing is cringe.
It makes me believe Zuckerberg is completely surrounded by yes-men who never challenge him and only tell him what he likes to hear. He is truly detached from us normal people and our human experience.
As a freelancer, every time a client decides for a cheaper alternative, I make very clear I would be delighted to work with them in the future anyway. It rarely fails, one or more years later, the clients calls me back because their cheap alternative turned out to suck and be expensive eventually. Last month, a client from Luxembourg called after 6 years of total silence. They still had me in their listing. 3 years ago, one called me because 2 years prior, the 50k quote they rejected from me turned into a 400K bill from my competitor, and still no release yet.
My rates have been steadily increasing for years thanks to this. Before, geeks were at a disadvantage because people didn't know better, and teams with a good marketing would destroy us. But now, they have been burned so many times. And it pays because more and more devs coming to the market are becoming dependent on their tooling. Now, more often than not, I work with teams that have been copy/pasting git commands not knowing what they do, that have never, ever looked the source code of their framework or don't know how to use a debugger. The HN bubbles tends to blind us to the reality of the corporate world.
Yesterday I did a deployment, but was not allowed to touch the machine. Instead, they made me call a guy sharing a screen of a Vista machine, while he was sshing prod using cmd.exe, and I had to dictate him the instructions to debug the deployment on their custom linux setup. A near retired sysadmin that couldn't type with 10 fingers, pressing 30 times the up arrow to find a bash command in the history every time. He could click on WinSCP very well though.
This 20 minutes job turned into an afternoon of billing.
Though I suppose that's what I look like as a Python expert to an old timer from the 80s that can code in assembly, debug using strace and understand L1 cache :)
People are scared we are going to get automatized by AI.
I am preparing for the most lucrative decade I ever worked in.
