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

Ask HN: What's the hardest problem you've ever solved?

news.ycombinator.com

Post reply on HN
Print view View original

111–120 of 441 posts

Re: Ask HN: What's the hardest problem you've ever solved?

#111

Post by rorykoehler »

I solved a performance bug that was probably going to cost my company the majority of our customers if we didn't sort it out as it made the app unusable. It had been going on for months. None of the dev team had any idea. APM was throwing out a bunch of red herrings. I haven't written a line of code in the app or a line of production Java (the codebase language) in my life. In the end I tracked it down to a single line of configuration code. Performance went from being atrocious to being one of the best performing apps I've ever seen.

This was not technically "hard" as in complicated but more so hard in the sense that it was very high stakes and the time pressure was high.

Re: Ask HN: What's the hardest problem you've ever solved?

#112

Post by jey »

amp108 wrote: post #106
yholio wrote: post #90

I somehow decided I needed to cheat to pass a certain exam because I was basically crap at memorizing stuff. So I used an analogue wireless headphone, an induction loop around my neck and a mobile phone. Since I lacked an accomplice to dictate, I read aloud the hundreds of pages and recorded myself, careful to preserve and properly serialize things like complex formulas. This was before the era of iPods and SDCard pl…

One thing universities should teach is efficient problem solving. If it was more work to cheat than to memorize, I'd say you might want to re-evaluate whether "not memorizing things" is a principle you should cling so dogmatically to.

I'm sure he learned much more useful stuff by coding up that monstrosity, and in the real world he can just google to look up those formulas he didn't memorize.

The fully elaborated version of the underlying principle is probably something like: "I have much better ways to spend my time than on memorizing formulas that I could easily look up later".

Re: Ask HN: What's the hardest problem you've ever solved?

#113

Post by hn_throwaway_99 »

nscalf wrote: post #103
yholio wrote: post #90

I somehow decided I needed to cheat to pass a certain exam because I was basically crap at memorizing stuff. So I used an analogue wireless headphone, an induction loop around my neck and a mobile phone. Since I lacked an accomplice to dictate, I read aloud the hundreds of pages and recorded myself, careful to preserve and properly serialize things like complex formulas. This was before the era of iPods and SDCard pl…

I would 100% talk about this in interviews depending on the interviewer. I can think of a couple interviews I had in SF that would probably be appreciative of this level of effort because you principally reject memorization.

Admitting to significant academic fraud during a job interview is a very, very bad idea. Furthermore, I would question the ethics and morality of any company that said "Wow, you solved an incredibly tough problem in order to commit fraud. Boy, are we the place for you!"

Re: Ask HN: What's the hardest problem you've ever solved?

#114

Post by Aeolun »

hn_throwaway_99 wrote: post #113
nscalf wrote: post #103

Earlier quoted context omitted.

I would 100% talk about this in interviews depending on the interviewer. I can think of a couple interviews I had in SF that would probably be appreciative of this level of effort because you principally reject memorization.

Admitting to significant academic fraud during a job interview is a very, very bad idea. Furthermore, I would question the ethics and morality of any company that said "Wow, you solved an incredibly tough problem in order to commit fraud. Boy, are we the place for you!"

They wouldn’t actually say that of course. They would just think it.

Re: Ask HN: What's the hardest problem you've ever solved?

#115

Post by hermitdev »

Probably one of my harder ones was finding a memory leak in one of my servers. Didn't happen often, never in development and only leaked a fraction of a byte per request. Turns out it was a bug in the logger, where if on rolling the log and it couldn't open a new file, the logger would just queue up messages. Was a real pain to track down as we normally had plenty of disk for logs.

Re: Ask HN: What's the hardest problem you've ever solved?

#116

Post by Aeolun »

rorykoehler wrote: post #111

I solved a performance bug that was probably going to cost my company the majority of our customers if we didn't sort it out as it made the app unusable. It had been going on for months. None of the dev team had any idea. APM was throwing out a bunch of red herrings. I haven't written a line of code in the app or a line of production Java (the codebase language) in my life. In the end I tracked it down to a single li…

Sometimes selectively commenting stuff out really is the best way to figure out what is wrong.

Re: Ask HN: What's the hardest problem you've ever solved?

#117

Post by blendo »

When you're younger, the problems are harder, and the success sweeter.

I earned my spurs (well, perhaps an enlisted stripe) back in the 1980's when I inherited a non-functional mess of an HP-85 Basic program to control an antenna test system. Many days in the Hanscom AFB RADC EE Propagation shack to get the RS-232 mag tape and IEEE HP-IB/GPIB/IEEE 488 custom hardware working. We kept bumping up against the HP-85's program storage capacity, so I had to remove most program comments, and actually shorten the variable names to get it all loaded into memory.

Then we got to fly that sucker on a pair of C-141B's from Wright Pat AFB, to Elmendorf AFB, then a flight over the North Pole, then to Thule AB, Greenland (9 days in November - do not recommend), then to RAF Lakenheath.

30 years on, I still make a living fixing crappy code.

Re: Ask HN: What's the hardest problem you've ever solved?

#118

Post by ttul »

I once had to fix a rendering bug in the heart of Mozilla that prevented a XUL-based code editor from working properly on Linux - this was back in 2001. The bug was a one off error about eight layers deep in the XML layout system. It took me about two months of full time digging and nearly destroyed my life. But I found it!

Re: Ask HN: What's the hardest problem you've ever solved?

#119

Post by akhilcacharya »

Maro wrote: post #3
bringking wrote: post #2

One of the hardest challenges I have had in my career is convincing our company to move to Continuous Delivery. 90% of the challenges weren't technical, but emotional. Shipping software comes with lots of feelings, fear, politics, etc. I had to personally work with various leaders across the org to help them through these feelings and perceived blockers. We aren't 100% there yet, but we are shipping numerous times pe…

Facebook transitioned from a weekly release cycle [1] to continuous delivery while I was there. It was a big project. There's a nice writeup here: https://code.fb.com/web/rapid-release-at-massive-scale/ [1] commits were pushed to employees on commit, that's how we tested, but not to the public

2017? Wow! My company shifted back in 2011!

Re: Ask HN: What's the hardest problem you've ever solved?

#120

Post by johan_larson »

I once had to design a function to open a series of nodes in a displayed tree from one given node down to another. Not so hard in general, but because of how the system was set up I had to do it using callbacks. So tell the first node to open itself and when done call the callback that tells the second node to open itself and when done call the callback that tells the third node ...
Post reply on HN