Live data from Hacker News

OptiFi Program Incident Report

medium.com

21–30 of 158 posts

Re: OptiFi Program Incident Report

#22
post #4

Earlier quoted context omitted.

I think there are some lessons that can learned from this. For example, the command line: solana program close didn't seem to sufficiently represent the destructive nature of the action. The program's designers should likely have chosen the verb "shutdown", instead of "close".

A simple confirmation step with a relevant warning could possibly make developers think twice before proceeding.

From personal experience - no it won't. I have made mistakes with overriding warnings - when under pressure, hyper focused and already in the mental mode of - of course I am sure I want to do it - your slow brain is just shut down so you are moving a bit like on muscle memory.

Re: OptiFi Program Incident Report

#24
post #16
post #7

Earlier quoted context omitted.

When the price tag is “everything you own”, how is this an improvement? Who wants to give over their financial life to a computer program? This aspect of crypto is a 1960s paranoid computer fear come to life, and somehow it’s often presented as an improvement over the existing system of human checks and balances.

> Who wants to give over their financial life to a computer program? Where do you think your income and bank balances are tracked and stored? On pieces of paper?

This is missing the point to a degree that makes me think you're being intentionally obtuse, but maybe you're just ignorant so I'll bite. Banking computer errors can easily be rectified by humans, banks are regulated, your funds are at least partially guaranteed by the government (depending on where you live). The degree to which you're trusting computer programs with your finances is orders of magnitude less than with cryptocurrency where it's possible to lose any amount $ of asset value in an instant with absolutely no means of recourse.

Your bank analogy is silly and nowhere near analogous.

Re: OptiFi Program Incident Report

#25

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?

Re: OptiFi Program Incident Report

#26
post #5

For all its flaws (and there are countless) the one thing about the show cryptocurrency space that stands out to me as a programmer is that programming errors can be suddenly very costly (granted, in this case it was more of a DevOps blunder). Being able to very easily put a price tag on sloppy programming is intriguing to me.

You (and everyone really) should probably read some more comp.risks http://catless.ncl.ac.uk/Risks/

Certain fields have always had a high cost of programming error, including a cost in human life. It's just that cryptocurrency combines this with a first-to-market rush that's somehow still going on, encouraging a rush to error.

Re: OptiFi Program Incident Report

#27
post #4

Earlier quoted context omitted.

I think there are some lessons that can learned from this. For example, the command line: solana program close didn't seem to sufficiently represent the destructive nature of the action. The program's designers should likely have chosen the verb "shutdown", instead of "close".

While it may not accurately represent what it does, "destroy" may be a better fit.

Even "irreversible_destroy"

Re: OptiFi Program Incident Report

#29

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.

An example of such a command came up on Linus Tech Tips (LTT), from apt (IIRC) on Pop_OS!. Basically (paraphrasing) "this will destroy your system and render it useless; type 'I'm an idiot and wish to proceed'". Our eponymous youtuber went right on, typed the phrase and blitzed the system, and then moaned about how the system shouldn't have let him break it ... I suspect such prompts only work for cautious people, who possibly don't need the prompts as they'll actually read the scary warnings.
Post reply on HN