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

OptiFi Program Incident Report

medium.com

Post reply on HN
Print view View original

81–90 of 158 posts

Re: OptiFi Program Incident Report

#81

Post by progrus »

FabHK wrote: post #76
bhaak wrote: post #48

Earlier quoted context omitted.

I don't disagree with that there have always been niche areas where there is a high cost to programming errors (like space or medical). But with crypto, the effect is much more direct. The programmer is handling money much more directly and if something goes wrong, they are much more directly affected and not being insulated from the effects.

Yes. In space, aviation, and medical, it is usually someone else that is affected, not the programmer.

And in space, aviation, and medical, it takes a long time to go to market.

Re: OptiFi Program Incident Report

#82

Post by synu »

soco wrote: post #25
RedShift1 wrote: post #14

Commands like that really need a confirmation prompt and a command line switch to override like --do-as-i-say (long form only). Good example of developers being put in an end user's shoes, I hope they learn from that mistake and update their programs.

I don't know about you folks but I can do stupid stuff also with confirmation. Maybe splitting the damaging actions in some substeps, so instead of a big bang you have a slow death?

Also, for me when I make big mistakes like this I think I’m in a different environment or something. It’s not that I don’t realise what command I’m running, it’s that I miss the context.

So I would just confirm but not realise what I was confirming.

Re: OptiFi Program Incident Report

#83

Post by andrewaylett »

Grimburger wrote: post #69
renewiltord wrote: post #64

Earlier quoted context omitted.

That guy is a savvy YouTuber, man. He did that because the resulting content has high memetic replication properties. I find the prompt for dangerous ops useful. GitHub will ask you to repeat the name of the repo before you destructive actions, Terraform will ask you to say yes to the prompt. These are all good things.

It's a solid practice in UX design. Physically having to type or copy/paste it in really highlights the action for the end user. There's only so much you can do to stop people setting their own house on fire but something like that puts the onus on them and fairly places the blame where it belongs.

The problems start when you know you want to light a house on fire, but you pick the wrong house.

It's almost never the case that blaming the user is actually going to help nor that adding more eyeballs will prevent people from making mistakes. If it's routine, we'll apply it to the wrong entity. If it's not routine, we'll not understand all the implications of our actions.

Re: OptiFi Program Incident Report

#84

Post by radicalbyte »

bhaak wrote: post #50
SkyMarshal wrote: post #35

Earlier quoted context omitted.

Sounds like they only lost $661k worth, which is peanuts by comparison to most other crypto hacks/losses, probably didn't destroy the entire company since they're compensating their users, and not really worthy of the sensational title.

Apparently 95% of that money was by the company itself or employees of the company. So the loss for normal users is very limited.

I can take a box containing 1001 conkers, sell one to you for $1000, then claim that I have $1m in my bucket.

What was lost was not money, but electronic conkers of which a small number people paid real money for.

Re: OptiFi Program Incident Report

#85

Post by tinco »

benreesman wrote: post #75
tinco wrote: post #66

Earlier quoted context omitted.

Was it Cambridge Analytica that was going to pay 50MM for that data? ;)

No, it was features for a P(click|impression) model, which are surprisingly mundane: it is basically impossible to care much either way, let alone be outraged, if you look at the actual features used in these kinds of things and it’s not explicitly your field. It’s not a microphone listening on your smartphone! You can even read about it in this paper which publicly described the system contemporary to that event: ht…

Thanks that's interesting! Definitely not outraged, was just poking a bit for fun. 50MM is an insane amount of money/value that would bankrupt most companies and have a noticable impact on many industries. Funny that for Facebook it was probably just a small hiccup on the ad revenue gravy train.

Re: OptiFi Program Incident Report

#86

Post by mpeg »

blibble wrote: post #73
mpeg wrote: post #56

Earlier quoted context omitted.

It's not exactly unique to crypto... in adtech, a configuration mistake on a web server can cost 6-7 figures as ads are being paid for, but take people nowhere.

and in a live trading system a small mistake can cost you 6-7 figures per second see: Knight Capital

indeed! the fintech world is already full of these footguns, but HN is very anti-crypto, so everything wrong that happens there is proof the whole industry is doomed

Re: OptiFi Program Incident Report

#87

Post by googlryas »

benreesman wrote: post #34

I once deleted like, maybe a few petabytes of data valuable enough to store on FusionIO cards with a bad config change. I think it destroyed $50-100MM once the dust settled in all the various ways that it cost money were added up. I’m quite sure the hacker who pressed the button feels plenty bad enough already, and I hope the people around them were as kind as the people around me were.

As someone who just heard about FusionIO cards...how did the config change eat all the data in a manner that wasn't recoverable?

Re: OptiFi Program Incident Report

#88

Post by programmer_dude »

benreesman wrote: post #34

I once deleted like, maybe a few petabytes of data valuable enough to store on FusionIO cards with a bad config change. I think it destroyed $50-100MM once the dust settled in all the various ways that it cost money were added up. I’m quite sure the hacker who pressed the button feels plenty bad enough already, and I hope the people around them were as kind as the people around me were.

If MM is million-million then that is a lot of money.

Edit: thanks for replies! :)

Re: OptiFi Program Incident Report

#89

Post by darrenf »

selfhoster11 wrote: post #58
darrenf wrote: post #38

Earlier quoted context omitted.

Command line switches like that won't, IMO and IME, make developers think "wow, I'm so glad this exists to guard against footguns". Rather they'll just set an alias to get around it. Likewise confirmation prompts albeit with a little more resitance. Once a repeat user is frustrated enough they'll bust out `expect` or similar, angry at the original command for mollycoddling them because they know what they're doing an…

If you alias a —do-as-i-say switch to a command without wrapping into a script to make sanity checks of your own, IMO that’s negligence. The name of the flag should make you think thrice about invoking it without safety checks.

I don't disagree. I think negligence happens.

Re: OptiFi Program Incident Report

#90

Post by mpeg »

LetMeInNowMate wrote: post #80
mpeg wrote: post #56

Earlier quoted context omitted.

It's not exactly unique to crypto... in adtech, a configuration mistake on a web server can cost 6-7 figures as ads are being paid for, but take people nowhere.

Quoted post unavailable.

That's your opinion. Who do you think loses out in a situation like that? The ad network is getting paid regardless, the adtech company (DSP, etc) will pretend nothing is wrong unless their client has hard proof and the client (probably a company you buy things every day from) was simply trying to promote a new product and will assume performance was just off.

Ultimately it is their customer that pays for all this by continuing to buy their products and use ad-powered platforms.

Post reply on HN