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

If something like this existed for C, then we wouldn't need to learn a new language!


Plenty of stuff exists, the first one being lint, introduced in 1979.

The problem with C isn't lack of tooling trying to fix it.


Here's one C/C++ targeting tool they cite that is kind of similar: https://cseweb.ucsd.edu/~dstefan/pubs/brown:2020:sys.pdf

But both tools only catch a subset of stuff, so they're not final cures for safety bugs.


That's fantastic, but (like related problems, like "blocking advertising", "making your social media platform act the way you want", "setting up your desktop/IDE just right", and "debloating Android"), you have to consider the amount of time and effort you are dedicating to fighting the enormous amount of resource devoted to making those software tools the way they are.

For one, you will ultimately not win. For another, past a certain point, it should become clear that the tool you are using is not made for your use case.


That's exactly the problem. As someone who does not have access to 32 Eth, you need to trust a pool operator with your money.

This is a direct parallel of issues like home and car ownership. If you're not rich enough, you have to trust a 3rd party, and also get less for your money.


Vote with your money. There are plenty of other blockchains where this is not the case as delegation is part of the protocol.


You don't have to trust a third party. You can run a pool yourself and be the one who needs to be trusted.


If you don't have enough eth to stake, how will you persuade anyone to lend you their eth?


Have friends? People who trust you?


Obviously I need to have friends who already want to invest a significant amount into Eth, otherwise I'm just "that guy" shilling some scheme to their family and friends.


Of course. So what's the problem with that?


Under what possible scenario do I have a few k$ spare, but not 10k$ spare, to invest in a speculative asset, and also have a large number of friends and family already in exactly the same position?

The only one I can see would make me the "cryptobro", to use the currently-popular phrase.

It's also quite ironic that you suggest the basis of Eth consensus be based on the strength of inter-personal trust. If that's how it works, then why bother with the complexities of Eth?


I'm not saying the basis of ETH consensus is based on inter-personal trust. The conversation started by a complaint of 32 ETH required to run a validator being out of reach of normal people, so I provided multiple other options for how a normal person can take part in this.


You gave one suggestion, which was based on inter-personal trust. This removes a major benefit of taking part in the Ethereum ecosystem. If it became a widespread way for people to take part in Ethereum's staking mechanism then yes, Ethereum's consensus would effectively be based on interpersonal trust (which is something that the developers have gone to great pains to avoid).


I gave two suggestions - join a pool, and run a pool.


Both of which (essentially the same thing - "be in a pool") require trust, which destroys a key advantage of Ethereum (and cryptocurrencies in general).


Usually because <minor> has made a prototype of an already well-established phenomenon, and the really difficult part is making a production-ready design.

A bit like the "10-year-old makes a heart pump for just $10" stories. It's really easy to hook up a motor to some pipes, it's absurdly hard to make that pass medical regulations.


How could this be? They have so many fewer moving parts!

\s


It's not the mechanical parts that are the problem. It's the software. The Tesla owners are not only software beta-testers but also the crash test dummies.


But software has zero moving parts! It must be so reliable.


You realise there's an excellent chance that the "industry incumbents" have been aware of the switch to EVs and hybrids, and are just biding their time before Tesla's circus show collapses?

It's been pretty obvious for a while to a casual observer of incoming government regulations (at least, in the EU) that ICE car production will literally be banned very soon. I'm sure the companies that form the backbone of the economy of a major country in the EU have had even more notice. In fact, I suspect they will do very well out of this transition.


Maybe after enough of these stories become common knowledge, we'll hear less of the "EVs have fewer parts and are therefore more reliable" argument.

Reliability is largely a function of design and engineering choices.


EVs should by design be more reliable, since they have fewer wear&tear parts (no ICE).

If Tesla were to make an ICE car none would function the year out because their manufacturing quality is horrendous.

Reliability and simplicity goes hand in hand, but it doesn't make the simple thing immune to failure if the quality is crap.


>EVs should by design be more reliable, since they have fewer wear&tear parts (no ICE).

It's possible to make perfectly reliable software, and an unreliable screwdriver.

>Reliability and simplicity goes hand in hand

True, but modern EVs are only simple in concept. In practice, they rely on the precise operation of legions of components with nm-sized features, that are bumped and shaken continuously throughout the vehicle's life.


Also the automakers and supply chains have decades of experience making ICEs very reliable. Actually quality and reliability we now have for reasonable prices is not bad at all. On average they just work.


I recently saw a poster for a Teachers' Union campaign, saying that "unity is our strength". Obviously both cannot be true.


Surely they’re orthogonal traits. A diverse group can unify. An example might even be the military.


"United in diversity" is literally the official motto of the European Union.


To anyone who believes that in practice, I recommend Yanis Veroufakis's book "Adults in the Room"


officially: "In Varietate Concordia"


Right. You can't have both at the same time.


I'm not sure I follow. I mean orthogonal as in along different axes.

Why do you think diversity and unity are mutually exclusive?


From your example - how does the military "unify" a diverse group of people? By aligning their thought processes and decision making. That is, by removing that very diversity.

The two words also mean literally the opposite thing. You can be along a spectrum from one to the other, but you can't have both at once.


What a bizarre thought process. Unification towards a common goal isn't homogenisation. Neither is common training or experience.


Talking of bizarre thought processes - how can you unify, and/or receive common training or experience, without losing some form of diversity?

We all love to say how true diversity is diversity of experience. You are literally taking about unified, shared experiences.


I just don't follow your reasoning here. Is it your contention that a new experience deletes or diminishes previous experiences? Or that newly learnt knowledge diminishes that previously learnt?

Surely a new set of common experiences increases the diversity of thought of all the participants? Nothing's lost.


>Surely a new set of common experiences increases the diversity of thought of all the participants?

This statement is absurd on its face. The entire premise of the benefits of diversity (of thought and experience) is predicated on the subjects in question having had unique experiences.

>Is it your contention that a new experience deletes or diminishes previous experiences?

Somewhat, yes. When I went to university, the shared experiences there hugely overshadowed what I had experienced in my prior education.



>If you do > 500km riding a year it's a worthwhile upgrade IMO.

Did you miss a zero? I do 500km a year, taking my kids to nursery on my beat-up old MTB.


Probably. I did more than 20Mm (20k km) so far on my mechanical 105 in 2 years and had to replace a single rear shifter cable in between. Mechanical transmissions are not exactly high maintainance either, and maintaince is fairly trivial. Probably would have spent more time charging batteries in 2 years with DI2 than with cable fixes and adjustments on mechanical. But I'm still somewhat intrigued to try it.


I love biking because it makes me independent. Adding electronic complexity and the need to charge is the antithesis of why I own bikes.

I might have to change the cable once in a while, but I don't ever have to upgrade the firmware.


I don't understand why people are so desperate to give reddit their time and effort. It's rubbish, move on!


Imagine the job listing: Full-time (+ overtime) data entry position for faceless corporation that will treat you like human refuse and eventually ban you. Salary $0/hr (non-negotiable).

Apparently people like OP are lining up for this though.


You will never find a more wretched hive of scum and cuckoldry than reddit


Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: