Ask HN: What is the biggest challenge you face as a developer?

Not being permitted to do my job.

Yes, I'd love to help you, if only you'd let me. Won't you let me?

But you won't, because you can't seem to control yourself--it's like a compulsion to self-sabotage. (Similar to what happens to creative works in the entertainment industries. Too many cooks...)

And it isn't the CEO, who is typically a practical business guy, enjoyable to speak and deal with. He's never the problem (and it's best to deal with him directly when possible, no matter your role). Instead it's the VPs and CTOs and others in lead or similar roles, who are inherently political creatures. Witness there the worst sort of decisions in everything from hiring to "culture" to technical and business strategies.

Personified, the typical company is a capricious and self-defeating entity.

What was the question again? Oh yes, here's me trying to write an app, up against all that...

Dealing with other people's perceptions. When I start a new job, everyone sizes me up and forms impressions about my skill or 'level' or whatever. What impressions? How do they form them? Do they actually have anything to do with me? Typically, at first everyone assumes I'm an idiot. Then, after I've tackled something hard that impresses people, they upgrade to "not a total idiot, but probably green/ignorant". Then slowly over time, as I solve a large variety of problems with decent, well-performing, and not-so-buggy code, they upgrade me to "pretty smart, actually!" Eventually I'm the senior guy. Now, begin a new job and start at the beginning again. Even with 20 years experience and a long list of successful shipped products, I still start every new job with the presumption, "probably an idiot."

The “software is a team sport” mentality.

At the start of a project, people saying “this needs a team”, even if you can already see a clear path to getting the thing working.

In the middle, “stakeholders” who want to set schedules and define how things are done without adding value.

Once something’s working, others presenting it as incomplete or untrustworthy purely because it’s the work of a lone developer and doesn’t tick some process boxes (and completely regardless of demonstrated capability/stability/etc.)

(This isn’t to say collaboration isn’t the right answer some of the time. But in my experience when it works out, it’s usually arises pretty organically and often has clear boundaries rather than “working on a shared backlog”).

1. Not being able to cover all corner cases. Didn't implement auth on Mongo, some evil bot deleted the database. Forgot to set the timeout on HTTP client, ending up with resource leak (goroutine) and downtime. Integrating a lot of services, often I miss so obvious thing which results in service going down.

2. People with expertise in some key areas not willing to help with your seemingly easy questions.

Figuring out how to show and talk about what I’m working on with my friends and family. Other mediums such as ceramics and painting are much more accessible to a general audience. It would be nice to have a way of showcasing software in a similar way.

Every project I’ve been on goes through periods of crunchtime, at which point process goes out the window and it all becomes word of mouth. The team chat app becomes a firehose of information and if I look away for a minute, I miss something important and end up working on something that someone else is already working on. Developers become an extension of the team lead who effectively becomes the sole thinker and commander of his little army of code monkeys. It’s like being a unit in Command & Conquer. And all because of an arbitrary deadline to build a demo for internal stakeholders, which sets up the project for ultimate failure.

Marketing. I think for most that's number one.

The variance in productivity.

1. How do I estimate how long something takes?

2. How do I deal with the fact that some developers are an order of magnitude more productive than other devs? For some reason it's not acceptable to pay some devs 10x more, unlike sports stars.

3. When it's normal for an individual to do huge things, it becomes their expected of them. And so people often work a little slower than they can.

4. When you're not getting paid accordingly to your output, is it fair to work as fast as you can?

Trying not to sell myself short to potential employers.

It's a far more challenging problem to solve than anything I've encountered at any job.

The hardest challenge as developers IMO is understanding and selecting the best technology that suits the current era, your use case / goal, and your current situation.

Most people that I know made technology stack's decision making based on hype.

FOMO on the latest and greatest framework and the feeling of being left out. But deep down I know it's a distracrion; albeit a mentally taxing one.