You don't need to work on hard problems | Hacker News Reader

You don't need to work on hard problems

benkuhn.net

154 pointsgoranmoomin43 comments

I consider the main measure of quality for a problem to be how much it helps people, not the difficulty in solving it. This measure applies for entrepreneurial problems where you want to earn money, but most other fields too. No harm in solving little problems that affect no one, but for big life passions, I find that service inspires and motivates more than cleverness.

As my mentor Frances Hesselbein says, "To serve is to live." https://www.inc.com/joshua-spodek/6-lessons-lunch-best-leade...

The problem with this way of viewing the world is that in reality, this guy is judging technical difficulty from the basis of someone who can complete a college master degree in one year less than planned, then concludes he doesn't generally have problems with the mathematical side of things.

Well of course.

Unfortunately this is no more helpful than stating that Muhammad Ali usually doesn't have a problem with street crime. It does not make a generalizeable approach to the problem.

The actual problem he's describing is a lot like being a great fighter. It is not so much about utterly defeating your opponent in one narrow skill, ignoring everything else. It is about being a little better than them in one area and not fucking up in any other ("easy") area. You can be the best at giving punches on the planet, it will do zero good if you easily loose your balance. Even in already the artificially very limited area of competition fighting, you have to be decent at 10 skills, and good at one. In the real world to solve general problems you have to be decent at 100 skills, from tax accounting to sales, from algorithm analysis to team building. None of these you can be a disaster at.

Exactly. If he could already join Jane Street as an intern... Well of course somebody who already has all the privileges can easily say that "I don't want these privileges and I want to explore something else", but first you have to be in a privileged position to have the ability to choose. For many others it is totally worth it to get into this position first, then think about what their next step would be.

There are a couple of HN tropes that are very similar, but not identical, to this idea.

1. "Growth mindset"

2. "90th percentile expertise in two areas is easier to achieve and more useful than 99th percentile in one"

This post seems to suggest "you should remain open to growing into a solver of problems whose hardness stems from being fairly hard in both technical and non-technical ways while not being extremely hard in either alone, because such solvers are useful".

I think I dig your comment -- but I'm curious about your use of the word "tropes" which has a somewhat negative connotation; was that intentional?

I'm curious why you think "trope" has a negative connotation. Was that researched?

Not GP, but to me it can be 'negative'-ish in the same way as 'cliché' is, i.e. it doesn't have to be, but can be used like 'well it's a bit of a trope (resp. cliché) that [...]'

It’s not meant to be negative, just in the sense of “a commonly occurring rhetorical device”.

I do not necessarily care about hard problems, but I do care a lot about interesting work. Fighting fraud sounds interesting, learning accounting sounds interesting. I find interesting for me often correlates to things I don't know. It is hard to get hired on things I don't know and even harder to stay at the job once I've figure the things out. Happy with my current role because I'm working on stuff that I have to figure out.

So I think less so hard problems and more so learning opportunities.

Fighting fraud is only interesting if the fraudless business is interesting. If the business is stupid, fraud should happen.

> If the business is stupid, fraud should happen.

Are you saying fraud is justified if a business is boring?

Fraud detection is comprised of a lot of interesting data science and pattern recognition (example being utilization of graph databases like Neo4j and its advantages over traditional RDBMSs of index free adjacency) not to mention the leveraging of ML. If the business paying me to implement this sort of thing is 'stupid' and 'boring' (whatever your definition of that is in the context of a business), why would that make the act of working and implementing fraud prevention any less of an interesting endeavor?

http://we-yun.com/book/%E5%9B%BE%E6%95%B0%E6%8D%AE%E5%BA%93%...

It can be extremely profitable too. If you happen to identify fraud against state or federal government, and it gets successfully prosecuted, you can earn 30% of imposed penalties.

Out of curiosity does anywhere other than the US have this kind of incentive?

"The most damaging thing you learned in school wasn't something you learned in any specific class. It was learning to get good grades."

Total lie by guy who goes after what other want from them: profits.

Getting something others want from you, especially figuring what it truly is, is a huge skill and advantage.

The author doesn't have the vocabulary for it, but what they're explaining is the difference between complicated problems and complex problems. The existing education system trains software engineers to solve complicated problems, those are what the author calls "hard problems". What he misses is that the problems he's talking about end up being challenging because they're complex.

https://www.fastcompany.com/90344944/complex-vs-complicated-...

They also seem to be extrapolating onto everyone else from their own experience of feeling some kind of pressure to want to solve hard problems.

Good for you, but as someone who felt the same way in university, and hasn't really worked on any 'hard problem' since (a few years), I sorely miss academia/studying/learning/challenge, keep thinking about going back for a PhD, wish I'd done it straight at the time (I did want to even then, just thought a break first would do me good), and would relish working on hard problems.

> All I’m doing is building a CRUD app

There’s also no such thing as a CRUD app - or rather, there is, but it won’t meet anybody’s requirements as “just” a CRUD app and will require a lot of (unique, interesting, challenging) customization before it’s actually helping anybody do anything.

The majority of developers seem to make simple CRUD apps overly complex (in the code I have inherited at least).

Generally the customer doesn't want "simple" apps like just using a CLI. Then there is the "evolutionary" aspect of feature creep and requirement specs that add non-optimal constraints.

I wasn't talking about the interface, I was thinking about people over engineering code or just having poor design / architecture that makes it a lot more complex than it should be.

Plenty of 'customised' useful-to-user apps are just CRUD apps though, nothing unique/interesting/challenging, just domain words mapped on to the same multitudinous/boring/easy shell built hundreds of thousands of times before.

Deeply insightful. Follow the links.

You don't need to work on hard problems but in general, solving hard problems is usually more important than creating another TODO app or a bingo card website. The latter is more likely to make you rich (for certain definitions of "rich"), and yet the former could, if successful, be fundamental for the progress of humanity; but, you know, it's hard.

It depends on how you define "important". Fundamental shifts are great, and they advance humanity a lot all at once. But ordinary achievements that help a few people at a time, a little bit, take on a cumulative importance of their own.

Even after the big "hard" problem gets solved, there's a gold rush of a huge amount of more ordinary work to pick the low-hanging fruit. Those aren't usually hard problems, but finding relatively obvious solutions for the first time. That wheel will be invented over and over for each sub-domain and specific set of requirements, and will gradually come to look like yet-another-TODO-app as everybody craves the next big leap. But until it comes, there remains a lot of room for making somebody's life just a little better by re-applying the last one for the 999th time.

Hard problems are hard because they involve humans. For some reason, the author is confused between the two. Complicated or advanced is not the same as hard.

Anyway, impact is more important than complexity, and not quite as obvious.

A hard problem is climate change, or managing energy market, or figuring new recycling chemistry. Perhaps providing specific medical care in safe manner, or reducing health problems. Providing food or housingfor 10 billion people, that's a hard problem. Getting everyone free education of good quality is a hard problem. Electronic voting is a hard problem.

Hard problems are generally ignored. They do not have obviously actionable solutions.

Complicated problems tend to have many solutions or partial solutions. Additionally it's easy to create a hard problem with an unnecessary solution to a non-problem, see that Wave thing he mentions. (There are many other workable and cheap loan solutions.)

Or fool yourself that you're actually solving any real problem. That typically happens if you define goal without consulting the subjects. (This Wave thing smells of it.)

Reply on news.ycombinator.com