Hacker Newsnew | past | comments | ask | show | jobs | submit | paulgerhardt's commentslogin

Impressive work for a solo effort.

Would love to see less front loading on the registration side - I fell off onboarding because I couldn’t get through the 12(!) page questionnaire.

The value proposition is clear, just let me use the app. Notes (my current solution for this) doesn’t make me read summaries of other people’s research every time I open the app :-)


I agree. And I dont understand how this isn't a universal rule for all of software: let the user use the thing ASAP. Just let me fucking play with it instead of forcing me to read intros, watch videos, click through a tutorial, etc. Just let me explore and interact! And only then also offer me some guidance that I can jump in and out of.

I couldn't agree more, for any apps I make I try to make the on boarding zero

Yeah, maybe it would be nice to have a fast-track onboarding option for common goals: lose weight, finish homework or chores, etc.

Thanks for the feedback, I will implement something to skip onboarding maybe for users who clearly understand the value proposition

I’m sure I’m missing something here but isn’t this common knowledge in sailing and climbing?

Specifically tying a knot with opposite chirality to one existing on a line can cause both knots to capsize and roll out.

One would not take it as given that three knots plus three knots would yield six knots in this scenario.


It does seem the obvious place to start looking, and the only surprise is that it took so long.

Given they’re near Hangzhou they can source a lot of stock for 1/4-1/40th the price as the US. I would guess $1.5k-$2k for the hot mess they show at 6:06 and $8k for the polished setup plus $0-$850,000 in billable hours to some PLC engineers.

The old school way to learn this field is an Allen Bradley or Siemens certification[1] but it’s pretty dry and tedious [2] with good money [3] because once you have a network you can show up at any factory or bottling line as they all use Siemens or Allen Bradley and they’re always breaking down.

The new school way is AI, raspberry pi’s, and 3D printer tool chains.

It’s worth watching the Valve Steam Controller Factory video to appreciate the scale of these lights off factories: https://youtu.be/uCgnWqoP4MM

[1] https://www.youtube.com/c/TimWilborne

[2] https://youtu.be/zDmGSHGH_is

[3] https://youtu.be/aLd2Y7pQ79o


Are there any places in the US where you could get the stock as easily? Ignoring the cost differences for now, I'm more thinking about where you might want to be if you wanted to build machines like this as a small business in the US. Of course, maybe the right idea is not to do it in the US.

[2019]

(In 2025, using Tailscale simplifies a lot of the configuration and reachability parts. This guide omits a lot of the hurdles one will run into with NAT traversal and the macOS section is a little dated.)


I am a huge proponent of Tailscale, and I've moved my entire stack into my tailnet. Even my steam deck.

With that said, as another user (th0ma5) pointed out, "The only downside is that the Tailscale organization will be privy to your actions online as well."

While Tailscale and Wireguard serve the same purpose for most, it is not a direct replacements. To most users, this doesn't matter. To a few, it may be a breach of OpSec

With all this said, Tailscale reportedly cannot see anything. Per https://tailscale.com/blog/opensource

"By making the Tailscale clients open, you can see that we don’t collect your private keys. And by making Tailscale’s DERP servers open, you can see that we can’t capture your encrypted traffic. We don’t see your data and we don’t want to. We hope that keeping this code open increases trust and transparency in Tailscale because anyone can review the code and see that Tailscale really works the way we claim."


That's a total red herring. They see the metadata (data about data) of every connection you make. They don't need to capture any encrypted data to tell a whole heck of a lot about you: https://kieranhealy.org/blog/archives/2013/06/09/using-metad...


Exactly, also, as long as there is no way of verifying that the source they opened up is the one they are running, it’s still a trust thing.


> Cash is also—I regret to say—disgusting.

Not true!

For a while I was manufacturing cash and did some research on the hygiene component.

Surprisingly, I learned tap to pay was the most hygienic iff one didn’t use a common terminal for tips or signing.

Coins were the second most hygienic with a germ score of 168RLU . Cash third. Credit fourth, and dead last was shared payment terminals one physically touched (this includes tap to pay if one “answers a couple of questions.”) A separate study found smart watches were dirtier than terminals.

[1] https://www.electronicpaymentsinternational.com/news/what-is...

[2] https://www.self.inc/info/dirty-money/

[3] https://www.nmi.com/blog/how-contactless-payments-impact-glo...

And amusingly [4] https://nocash.ro/oxford-university-european-bank-notes-on-a...

Reports in this space are full of confirmation bias so take the RLU scores with a grain of salt. Likewise some cultures have cleaner methods of dealing with cash than others.


Well it was radar. The first Raytheon microwaves were really pushing for 3GHz not 2.4GHz. If you like to play Connections, the reason for that is the first mass produced magnetrons were made by gun manufacturers like Colt and Smith & Wesson and the tooling for gun bore holes and magnetron cavities lined up at 3GHz.

The official FCC minutes from 1945 [1] indicate that publicly they were marketed for heat therapy massages not food, with a weird wink, wink that if they could get a carve out for using for medical reasons they could also sell it to the Navy for reheating food as well.

The ISM carve out came after by a couple of years in 1947 because Raytheon had got an exception for this machine, not the other way around.

The whole origin story of why this particular slot of spectrum is full of carts before horses. That water oscillation thing is a common misconception - water oscillates at much higher frequencies [2].

[1] https://www.worldradiohistory.com/Archive-FCC/FCC-Annual-Rep...

[2] https://www.sciencedirect.com/science/article/abs/pii/S01691...


> Most crypto coins (not Monero) are pseudonymous.

It's worth reading the OSPEAD report from the Monero Community: https://www.getmonero.org/2025/04/05/ospead-optimal-ring-sig...

> Monero users with extreme threat models should be aware that anti-privacy adversaries can leverage timing information to increase the probability of guessing the real spend in a ring signature to approximately 1-in-4.2 instead of 1-in-16.

So slightly worse odds than Russian Roulette. Cool. Cool.


It's not Russian roulette. It's a guess you still don't know if you are right. There is still plausible deniability.

Hopefully FMCP will get implement soon to mitigate these issues.


How’s your 13 mini holding up? I have a recently refurbished one (6 months old) and I can’t make it to 2pm without recharging.

Additional my mail search and photo search broke with Apple Intelligence/iOS18 integration.

Debating jumping ship to a epaper phone or holding out for the rumored iPhone Air.


My 13 mini is also not great on battery. I've been debating ordering an iFixit battery and doing the swap, but in the past I've felt it was kind of mixed results from that. I don't think those batteries are newly manufactured units, but rather leftovers from the original production line that have been sitting on the shelf for 2-3 years. So although they'll be an improvement over one that's been through 1-2k cycles, they won't be like it was when it was brand new.

For now I'm just making do with having a power bank in my bag when I'm out and about.


I don't think the iPhone Air will actually be smaller in the dimensions I care about, just thinner which I assume will compromise battery life.

My mini is holding up ok. Battery needs replacing but I haven't done it. Like mikepurvis, I carry power banks around if I'm doing anything where I'm not going to be able to recharge easily. I use one like this https://www.amazon.com/Anker-PowerCore-Magnetic-Slim-B2C/dp/...


I prefer this form factor, but yeah

https://a.co/d/aFqI38o


I was able to get a new battery and screen for my 13 mini via AppleCare, but even the new battery won't get me through the day. Recent OS updates also make my camera shooting experience really slow for some reason.

Even with all that, I'm keeping the mini as long as possible because every year brings bigger and heavier iPhones.


> Recent OS updates also make my camera shooting experience really slow for some reason.

I’m on a 16 pro and it’s bad. It’s worse if I use the side button or do it from lock Screen, quicker from actual camera app. However it’s by far the slowest camera I’ve had on an iPhone, and I find the speed and quality a disappointing.


I have a 13 mini for about 3 years now - still holding up for most of the working day (about 15 hours). The trick is to reduce the number of apps you have on the phone, reduce the number of apps which like running on the background, and not watch a lot of videos.

I figured it out that treating it as a communication device + payments device + maps + very occasional content viewer, ie mostly as a utility will make the phone last much longer.


I’m on a 12 mini and it lasts all day easily.

I generally like the phone except it’s a little too big.


Years later we put a newer version of this chip with a secure element into Bitcoin bills.[1]

Hand waiving a lot of the details, each note basically becomes a hardware wallet with some additional features to prevent double spending.

[1] https://offline.cash


This is cool, I thought of something similar, didnt know someone implemented it and the swizz like bank note design is neat.

but its still has flaws imo and just makes it a novelty rather than something practical or useful, for example unless the vendor has change, you would have to spend your btc cash denomination as a whole or exchange it for fiat. what if both the vendor & customer didn't have internet access, how would they check if the cash hasn't been used? idk if you can write back to an rfid that may solve some problems


Are you involved in the project? I assume this is essentially a paper-thin HSM/Yubikey, right?

Is there any way I can use these as general-purpose HSMs outside of bitcoin/cryptocurrency contexts? Just as a novelty.

Let's say I just want to store a private key on it and use it for SSH/etc authentication - would that be possible?


With respect, this gives me serious ick. Black mirror vibes.


It's a hardware wallet that just looks cool. Where's the "black mirror vibes". There's not much nefarious tech here.


The ick and black mirror vibes comes from the technical fact that bitcoin is non fungible and publicly traceable by design making it a surveillance states dream.

Luckily there are superior alt coins that don't have that flaw.


Err, actually the ick factor for me was the crypto bro mentality and the general seedy nature of the state of crypto and those who use it.


Turns out if you only look at people trying to turn a profit you will find bad people. It's like complaining about investment being seedy because you only watch wallstreetbets.

That should be the point where the intellectual curious hacker should try to broaden there horizon beyond the superficial surface.

I don't see the seedy nature you try to apply to a whole industry if anything i experience the opposite.


Dosage makes the poison.


looks at the crypto community and gestures broadly


I would wager the benefits of this model come mostly from the 2 sigma boost one gets from one on one instruction and not from any sort of optimal skill tree progression a master would impart on a student in a pedagogical environment engineered for optimal knowledge and skill acquisition.


> a pedagogical environment engineered for optimal knowledge and skill acquisition

I'm not sure how many of those we have available to us. Many are compromised by politics, funding, or the need to act as a daycare.

I learned a lot at the various schools I went to, but the amount I learned seemed to correlate more directly with how invested I was in learning than how well the school was funded. Plenty of schools with better per-pupil funding had significantly worse student achievement rates than where I was.

The only real exception to that is not all schools offer the same curriculum. Back in my day, not every secondary / high school had someone who could teach calculus, though now there's districts that are getting rid of calculus entirely to promote anti-racism. Honestly, I think learning calculus in high school was good for me, even if I've really only needed to calculate integrals once in my programming career.

At University, things were much the same. Undergrad courses focused a bit more on synthesizing than memorizing compared to high school, but not really by much.

All of this is to say that I'm not really sure it's fair to knock the apprentice program since we don't directly experience optimal pedagogy elsewhere.


> Undergrad courses focused a bit more on synthesizing than memorizing compared to high school, but not really by much.

Sorry about that. At Caltech, we were never given formulas. Everything was derived from scratch. I never memorized anything (but I found after a while I simply knew all the trig identifies!).


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: