Hacker News new | past | comments | ask | show | jobs | submit | risyachka's comments login

It was expected considering during the last year everyone on Twitter was shouting left and right that everyone should use Hetzner because it is cheap.

Yeah it was cheap - only because the demand was low and they were competing on price.

Now it will only go up.


Its still a really good price though.

After calculators were invented basically no one can can do math in their head.

I’d argue the bar will be lower and lower. Yeah those who want can learn more in less time. But those who don’t - will learn much less.


I've noticed that people who rely on calculators have great difficulty recognizing when their answers are off by a factor of 10.

I know a hiring manager who asks his (engineering) candidates what is 20% of 20,000? It's amazing how many engineers are completely unable to do this without a calculator. He said they often cry. Of course, they're all "no hire".

How did they get a degree, one wonders?


I've been bashing my head against Speed Mathematics Simplified because I want to be able to do tip math without pulling out my phone.


You won't be sorry you invested the time on this.


100% Agreed. There is genuine value in occasionally performing things the "manual way", if for nothing else then to help develop a mental intuition for figures that might seem off.


This is a sort of mental math trick that isn’t incredibly useful in day to day engineering. Now if they say 16,000 or something then maybe there’s an argument against them, but being able to calculate a tip on the fly isn’t really something worth selecting for imo


It's not a "trick".

And yes, it's incredibly useful in enabling recognizing when your calculator gives a bogus result because you made a keyboarding error. When you've got zero feel for numbers, you're going to make bad engineering decisions. You'll also get screwed by car dealers every time, and contractors. You won't know how far you can go with the gas in your tank.

It goes on and on.

Calculators are great for getting an exact final answer. But you'd better already know approximately what the answer should be.


> it's incredibly useful in enabling recognizing when your calculator gives a bogus result because you made a keyboarding error.

Humans are much better at pattern matching than computation, so the safest solution is probably to just double check if you've typed in the right numbers.


> recognizing when your calculator gives a bogus result because you made a keyboarding error

It might be counterintuitive, but the cheaper (and therefore successful) solution will always be more technological integration, not less.

In this case, better speech recognition, so the user doesn't have to type the numbers anymore, and an LLM middleman that's aware of the real-world context of the question, so the user can be asked if he's sure about the number before it gets passed to the calculator.


It's not a "mental math trick", it's a straightforward calculation you should be able to do in your head.


I don't know if this is a trick but the fast way I did that problem quickly in my head is 20% = (10% X 2) i.e calc 10% of the number then double it.

To quickly calc 10% just multiply the number by 0.1 which you can do by moving the decimal point one place 20,000.00 => 2,000.000 then it is easy to double that number.

to get 4,000.

17% for example is 1.7 x 10%

in this case 1.7 x 2,000 = 3,400


For me, it's just that 20% is one fifth. One fifth of 20 is 4 and you add the remaining zeroes.

You mostly have common equivalences like this in your memory and you can be faster than computing the actual thing with arithmetic. Or have good approximations.


LLMs literally barely improved since gpt3


Be serious. Go retry GPT-3 and compare it to Claude Sonnet and come back here.


The issue is it doesn't really replace junior dev. You become one - as you have to babysit it all the time, check every line of code, and beg it to make it work.

In many cases it is counterproductive


They fail even at not really complex problems. In most cases it’s faster to do it manually then beg ai to fix everything so that the result is proper, not just “kinda works”.

For me they save a lot of time on research or general guidance. But when it comes to actual code - not really useful.


Idk, LLMs have basically stopped improving for over a year now. And in their current state no matter how many abstractions you add to them - or chain them - they are not even close capable to replace even simple jobs.


Agreed. The jump from GPT3.5 to GPT4 was truly mind blowing, from GPT-4 to Opus/Sonnet3.5 was pretty good, but if o1-preview really is GPT-5 then I feel like we're seeing the hype starting to collide with reality.


Who said rail industry should be profitable in the first place?

I would fully support if part of my taxes would subsidize rail industry and I get cheaper prices. Imo good use of tax dollars.


Yep, thats how they work, there's a ton of data to back it up.

Unless you are from few *US* unis, you have zero to none chance of getting in, unless you have significant traction already, in which case you may as well go to any investor and they will fund you.

And if you are among those lucky few, they will fund you with the most shitty idea you have, or even without any at all.


>unless you have significant traction already, in which case you may as well go to any investor and they will fund you

This is the part that really bothers me unfortunately. I honestly grew up thinking that YC invested in people and ideas more than traction. If I had traction, I could raise money from anywhere. YC does invest in people without traction, but you have to be a certain type of person, from a certain background, and from certain institutions.


I went to a state school (and not a prestigious one) and I got in ~a year after graduating, I don't think this is super accurate


I'm a solo founder and during the application process, I got 3-4 emails telling me to strongly consider finding a co-founder using https://www.ycombinator.com/cofounder-matching.

That's great, but just look at the profiles advertised on the front-page there: Harvard, Berkeley, MIT, etc. Nothing against them personally of course, but I really think this is the trend YC is going towards these days. It's the old school institutional prestige back, but this time it's blended together with SV startups.


> I got 3-4 emails telling me to strongly consider finding a co-founder

A lot of YC's advice is terrible, such as their advice on finding a cofounder.

You can easily tell when they give bad advice: ask them for their dataset. When they don't offer it, consider doing the opposite of what they tell you.

Now, some of their advice is great, when it is data backed an easily testable like Jessica Livingston's "I don't know of a single case of a startup that felt they spent too much time talking to users."

https://breckyunits.com/advice.html

Now, what's a version of their "find a co-founder" advice that doesn't suck: "I don't know of a single case of a successful founder that wasn't building things all the time with many different people."

Don't find a co-founder or two, find dozens of "building buddies" who you just build shit with. Every now and then you can form a legal entity and be "co-founders", but the more important thing is you just create stuff together.


>>> have pretty limited context windows and limited ability to come up with brand new ideas, right? But they do write very confidently

so LLMs are basically senior management? :P


Did they mention anywhere why they decided to write their own proxy instead of using Traefik or something else battle tested?


They were actually using Traefik until this "v2.0.0" (pre-release right now) version.

There are some context about why they switched and decided to roll their own, from the PR.

https://github.com/basecamp/kamal/pull/940


As SRE, that PR scares me. There is no long explanation of why we are throwing out third party, extremely battle tested HTTP Proxy software for our own homegrown except "Traefik didn't do what we wanted 100%".

Man, I've been there where you wish third party software had some feature but writing your own is WORST thing you can do for a company 9/10 times. My current company is dealing with massive tech debt because of all this homegrown software.


> Man, I've been there where you wish third party software had some feature (...)

It looks to me that they were already using Traefik, and Traefik does indeed support the feature they wished to have.

Apparently they also got rid of support for healthchecks, which all by itself would be expected to require an extensive justification. But no.

I would love to hear why the people behind this decision felt that Traefik's support for dynamic route configuration wasn't suited to configure routes dynamically.

I would also love to hear the rationale behind the decision to invest their time rolling their own reverse proxy instead of just doing docker run on one of the many production-grade reverse proxies readily available.

I don't know. This has vibes of spending months in a development project to avoid spending hours reading a manual.


> Apparently they also got rid of support for healthchecks

The did not remove support for healthchecks. They're directly mentioned in the README.

> It will immediately begin running HTTP health checks to ensure it's reachable and working and, as soon as those health checks succeed, will start routing traffic to it.

I don't know. This has vibes of doing little to no research and then shit-talking a product.


I tried to put some reasoning here https://nts.strzibny.name/kamal-proxy/

But I don't think they ever published comparisons to anything else than Traefik


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

Search: