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

Code Is Never “Perfect”, Code Is Only Ever “Good Enough”

exceptionnotfound.net

Post reply on HN
Print view View original

21–30 of 130 posts

Re: Code Is Never “Perfect”, Code Is Only Ever “Good Enough”

#21

Post by mfrye0 »

jacobtr wrote: post #7
beat wrote: post #6

I have the start of a lightning talk on something similar, called "Why Software Sucks". The idea is that we only make software good enough to get by - we will always choose features over robustness if we possibly can. Every project ends when it either fails, or hits feature-complete. No one sits around polishing the pile of bugs and bad decisions that is a feature-complete app unless they have to - much more fun to s…

It seems like most consumer apps incur a crazy amount of technical debt while largely paying people in equity, then if they take off raise a bunch of money to pay other people in cash to sort it all out... Pretty good system actually.

Yeah I've learned this the hard way as being one of those people to "sort it all out". I now make a point of reviewing a company's codebase before joining. I refuse to join a startup with a shitty codebase that falls apart after to touching it once.

Re: Code Is Never “Perfect”, Code Is Only Ever “Good Enough”

#22

Post by pmiller2 »

mfrye0 wrote: post #21
jacobtr wrote: post #7

Earlier quoted context omitted.

It seems like most consumer apps incur a crazy amount of technical debt while largely paying people in equity, then if they take off raise a bunch of money to pay other people in cash to sort it all out... Pretty good system actually.

Yeah I've learned this the hard way as being one of those people to "sort it all out". I now make a point of reviewing a company's codebase before joining. I refuse to join a startup with a shitty codebase that falls apart after to touching it once.

Interesting. Are you a consultant/contractor or do you typically work as an employee? I can't imagine a company would just open up their code base to someone (even under NDA) unless they were fairly desperate.

Re: Code Is Never “Perfect”, Code Is Only Ever “Good Enough”

#26

Post by SilasX »

patrickmn wrote: post #3
unimpressive wrote: post #2

I wonder if this attitude will change once things like coq become more accessible.

I don't know if we will ever have tools that are that safe and as easy to use as the alternatives.

It depends. People were reluctant to give up a lot of bad coding practices (goto, variable scoping, writing in efficient assembly rather than a high-level language, mutable global state) until the better way had better tools or was taught better.

Edit: Maybe one day people will look back and say, "you shipped code without formal correctness proofs? Were you all high?"

Re: Code Is Never “Perfect”, Code Is Only Ever “Good Enough”

#27

Post by BerislavLopac »

drwl wrote: post #15

reminds me of http://algeri-wong.com/yishan/only-god-writes-perfect-code.h...

Right on point. Furthermore, it just lead me to realisation that achieving software perfection is like reaching the speed of light: the closer you get to it, the heavier you are, and more effort is needed to go further.

Re: Code Is Never “Perfect”, Code Is Only Ever “Good Enough”

#28

Post by softawre »

Tooling is getting better. Ever looked at what Amazon does with formal methods to "prove correctness" for some of their systems?

Of course you can argue that "good enough" for them is "damn near perfect".

http://cacm.acm.org/magazines/2015/4/184701-how-amazon-web-s...

Re: Code Is Never “Perfect”, Code Is Only Ever “Good Enough”

#30

Post by pmiller2 »

dman wrote: post #19
dwaltrip wrote: post #16

Earlier quoted context omitted.

You are missing the point. "Good enough" depends entirely on the context. In some contexts, the bar for "good enough" will necessarily be quite high, while in other context it won't matter as much. There is no perfect code -- because no two humans could agree on such a definition. Sure, people may disagree on what is good enough. But hopefully that disagreement will be easier to resolve than one about what the perfec…

My only concern is that this line of thought can be used as an excuse to stop trying. As long as we agree that one should strive for perfection given the constraints at hand (money / time / problem difficulty / experience level etc) I dont have any objection about the idea. On a related note, I feel another analogous statement that I now see used dangerously is "Premature optimization is the root of all evil". Yes op…

That's a good attitude to aspire to, but, in real software development, very little code survives contact with users unscathed. If you're writing a C compiler or something with a well-defined spec, then, sure, you can theoretically write perfect software. But, even your basic CRUD web site is unlikely to ever reach "perfect" status.

Edit: For example, the HN code was obviously not "perfect," since we just got several new features. :)

Post reply on HN