Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

There’s no such thing as a tech expert anymore

wired.com

Post reply on HN
Print view View original

101–110 of 113 posts

Re: There’s no such thing as a tech expert anymore

#101

Post by _jal »

almost_usual wrote: post #85
Hippocrates wrote: post #75

This is a horribly wrong take. The premise that nobody understands these tech issues at the level needed to address them is not even close to true. In fact, this sounds like an argument one of the luddites in congress would make rather than a tech journalist.

The argument is no one knows the exact details of the whole system. That is obviously true, the only experts at these companies are the engineers who write code. There are still details they don’t understand. An expert would have to be the collective knowledge of every engineer that worked at the org and there would still be missing detail and ‘bugs’. This is literally why the security industry exists, bugs are inevi…

The implicit argument is that one must wholly comprehend the exact details of a company's systems in order to meaningfully control them.

This is obviously false reductio ad absurdum given the existence of managers doing exactly that right now.

CEOs are not testifying to congress about exploiting side channels in hypervisor memory management. On one level, they're offering industry viewpoints on policy issues; on another, they're actors in a political game.

Re: There’s no such thing as a tech expert anymore

#102

Post by grogenaut »

knorker wrote: post #73

That's a really long article to say what we already know: That "expert" means to know in detail a narrow thing. It hasn't been possible to be a true polymath in like 500 years. Surely this isn't news to the author? There's no such thing as "an expert on everything"? Yeah, no shit. It's almost tautology, due to the limitations on human brains and the advancement of science since the renaissance. Another way I think th…

Have you read any other wired articles in the last 10 years? They're all like this.

Re: There’s no such thing as a tech expert anymore

#103

Post by almost_usual »

_jal wrote: post #101
almost_usual wrote: post #85

Earlier quoted context omitted.

The argument is no one knows the exact details of the whole system. That is obviously true, the only experts at these companies are the engineers who write code. There are still details they don’t understand. An expert would have to be the collective knowledge of every engineer that worked at the org and there would still be missing detail and ‘bugs’. This is literally why the security industry exists, bugs are inevi…

The implicit argument is that one must wholly comprehend the exact details of a company's systems in order to meaningfully control them. This is obviously false reductio ad absurdum given the existence of managers doing exactly that right now. CEOs are not testifying to congress about exploiting side channels in hypervisor memory management. On one level, they're offering industry viewpoints on policy issues; on anot…

> This is obviously false reductio ad absurdum given the existence of managers doing exactly that right now.

Managers don’t know the details of any system unless they wrote it when they were an engineer (and it hasn’t been modified). They are trusting their engineers’ interpretation of the system, which is a game of telephone up the management chain.

> CEOs are not testifying to congress about exploiting side channels in hypervisor memory management. On one level, they're offering industry viewpoints on policy issues; on another, they're actors in a political game.

The largest account takeover incident recently was an unsophisticated social engineering attack executed by a 17 year old.

I’d argue most system and data vulnerabilities aren’t sophisticated or complicated yet companies like the idea of this because it makes them seem more improbable.

Vulnerabilities happen because security guarantees and understanding cost time and time is limited when shipping a product faster than a competitor.

In reality if Congress wants an expert they need to have a team of engineers at the company performing audits and code reviews.

Re: There’s no such thing as a tech expert anymore

#104

Post by almost_usual »

acheron9383 wrote: post #93
almost_usual wrote: post #88

Earlier quoted context omitted.

Your car engine doesn’t store information about everywhere you go and what your interests are. Your car engine probably doesn’t experience data leaks, social engineering, or 0 day bugs. The argument is Google and Facebook don’t 100% know their data usage and risks and I’d agree that’s true. Complex software systems are impossible to 100% guarantee expected behavior. This is why bug bounties and the security industry…

The metaphor seems apt to me, if anything, we should have and do have a much greater interest in regulation of automotive products, which themselves are absurdly complex. How is a regulator to know about various emissions, chances of explosion/fire, engine longevity and warranties? Engines can kill you with fuel leaks, poor materials engineering, or unknown manufacturing bugs. I'd argue that your car engine is much m…

In reality if Congress wants an expert they need to have a team of engineers at the company performing audits and code reviews.

No different than a company that does due diligence before an acquisition.

Re: There’s no such thing as a tech expert anymore

#106

Post by alirsgp »

knorker wrote: post #73

That's a really long article to say what we already know: That "expert" means to know in detail a narrow thing. It hasn't been possible to be a true polymath in like 500 years. Surely this isn't news to the author? There's no such thing as "an expert on everything"? Yeah, no shit. It's almost tautology, due to the limitations on human brains and the advancement of science since the renaissance. Another way I think th…

People like von Neumann were considered polymaths

Re: There’s no such thing as a tech expert anymore

#107

Post by LifeLiverTransp »

The actual elephant in the room - nobody even talks about, unlike this article, is the effectively privatization of what used to be calle psychology,neuro-sciences, you name it. If you have all the data, and you have a private database to query it- your defacto privatized a whole field of academia, which is cut off from the most valuable thing to have.

Data.

You could write simulations, displaying the outcome of each of your actions on a psychological acurate model of a society. And you would be defacto invisible with your knowledge, forever in the blindspot of us mamals.

Re: There’s no such thing as a tech expert anymore

#108

Post by echlebek »

knorker wrote: post #73

That's a really long article to say what we already know: That "expert" means to know in detail a narrow thing. It hasn't been possible to be a true polymath in like 500 years. Surely this isn't news to the author? There's no such thing as "an expert on everything"? Yeah, no shit. It's almost tautology, due to the limitations on human brains and the advancement of science since the renaissance. Another way I think th…

I also disliked the article. Systems are complex, so it's suddenly pointless to try to understand them? Or, if you don't understand every detail of a system, you don't understand the system? Give me a break. I suppose we may as well give up on trying to understand natural systems while we're at it.

Maybe there's an alternative?

https://www.goodreads.com/book/show/3828902-thinking-in-syst...

Re: There’s no such thing as a tech expert anymore

#109

Post by ekidd »

Izkata wrote: post #100
ekidd wrote: post #91

Earlier quoted context omitted.

In the 90s it was quite normal for one person to understand everything from the hardware and networks up to how to make HTML look pretty. The last time I looked, a good CS curriculum still teaches people how a half-adder circuit works, how to write code in assembly, the basics of UDP and TCP, how to build a web app, and fundamental algorithms and data structures. Additionally, students might do a couple of the follow…

That describes a CS degree heavily biased towards practical application instead of towards theory, like mine was. We only had a couple of the things in your lists. Anything that involves the physical world went in the engineering degrees - we didn't do circuits, assembly, or programming a robot. Likewise networking I think was an elective, as might have been kernels (though I don't recall that one, and didn't take ei…

I feel like I ought to defend the honor of my school's theoreticians. :-)

Their CS program had two theory-heavy courses. In the harder of the two, the homework worked out to 10 pages/week of proofs written in LaTeX. Honors students could also take a functional programming course early on, where they usually got to write a metacircular evaluator.

The same CS program also had a software engineering course that required a 5-student team to spend half a term building a group project. (Teams were picked by the professor, for that real-world feel.) This usually required a UI, which meant that we got to learn Motif or web programming or something.

This has always seemed like a sensible way to organize a CS curriculum: Everyone needs to write proofs, and everyone needs to write code. Then you can throw in kernels or networking as electives.

Re: There’s no such thing as a tech expert anymore

#110

Post by grayhatter »

MaxBarraclough wrote: post #61
grayhatter wrote: post #57

Earlier quoted context omitted.

It's still deterministic. It doesn't randomly do something different with the exact same IO. Just because it's hard to recreate the exact same state with concurrent processes on a modern CPU doesn't change it's determinism. Also, undefined behavior in C only means a compiler can do what ever it likes. It doesn't make it random. People need to stop pretending UB in C is magic. It's not going to delete your hard disk,…

Sure, but in both cases, the way to model it is with something like nondeterminism. Computers have no nondeterminism, but computer science does. Formally verified C code is guaranteed to be free of UB, as this is necessary in order for there to be guarantees about the behaviour of a program written in standard C. The underlying compiler and hardware are still deterministic. A similar thing happens with concurrency. >…

That's exactly what I mean about UB in C. I mean I agree with you If you don't know exactly what your compiler will do with the UB, (or if you ever might want to use a different compiler and or comp options) you probably want to avoid writing any UB. But it's not magic, and it can't time travel. That segment makes explicit that UB doesn't need to create a sequence point around UB. This is important because if it did need to do that (to protect the non-UB code) the compiler would need to keep a list of all UB. I'm pretty sure that if you're going to create a list of all undefined behavior, you might as well just define it. It's not magic, it's just allowing the compiler to YOLO it, using optimizations it already knows how to do.

I have a really cool example of how/why this is actually a good thing for compilers if you're interested.

Post reply on HN