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

Ask HN: I got duped into working on legacy code, should I leave?

news.ycombinator.com

Post reply on HN
Print view View original

41–50 of 126 posts

Re: Ask HN: I got duped into working on legacy code, should I leave?

#41

Post by ekarulf »

I find it impossible to replace a legacy system without first understanding the legacy system. It's easy to say "X is old" or "X sucks at Y" but keep in mind that applications have a strong survivor bias. Ask yourself, "what does X do that is awesome" or "how has X lived this long" and you'll find a lot of hidden scars and requirements.

Start documenting those requirements and create a plan to map those to technologies that you prefer. Understanding how to successfully migrate platforms is a learned skill, and it's going to become useful again when React is no longer en vogue.

Re: Ask HN: I got duped into working on legacy code, should I leave?

#43

Post by masor »

db48x wrote: post #32
throwawaysbdi wrote: post #13

Earlier quoted context omitted.

I should clarify that's over 20 deep of our code , not counting all the framework stuff.

The depth of the stack is not a very good metric for code quality.

Why is everyone quibbling with him about that. He just wants to express his opinion that the codebase is very bad in a more colorful way than by simply calling it doubleplusbad. Let's just take his word for it and move on.

Re: Ask HN: I got duped into working on legacy code, should I leave?

#44

Post by buzzybee »

I think the basic truth to work off of here is: "Shit got real."

Yes, you can step away, in the figurative and literal sense. In many possible worlds this may even be the best plan, if you have no real agency over the code. And you got lied to. It'd hardly be the first time someone has been convinced that the work would be cooler than it is.

But -- you also have to decide if this is an opportunity in disguise. Stories of folks who gradually build ownership over the creaky "legacy" codebase and polish it up with Sisyphean refactoring, until it shames the "rewrite", are out there, and they're real.

On the other hand, fixing the legacy code may be one of the hardest things you can do in software architecture. There's no sense of joyous freedom to it, just correcting of old mistakes. But if you did that, and published documentation to evidence what you did and how, lots of folks would want you - and not necessarily for the task of fixing their own crusty code.

So there is an opportunity here, if you can find a way to sneak it through the management, which has surely decided on the balance sheet that the code is a "deprecated asset" and will not want to hear about new ideas. In fact, to maintain it you don't want to present anything as a "new" - you have the much harder task of justifying every small change as an opportunity to improve the debuggability - to get the callstack down from 20 deep to 19 deep, to remove a superfluous parameter, to improve turnaround times, to reduce the code size by a few lines, or something similarly miniscule, but which might add up over time into real understanding.

Re: Ask HN: I got duped into working on legacy code, should I leave?

#45

Post by wpietri »

throwawaysbdi wrote: post #7
bigiain wrote: post #4

Put in an amount of effort that you consider professional at work (by which I mean, put in enough effort that you're proud of yourself, even if you're less productive than you'd consider ideal for someone specialising in what they need). Use React on side projects. Look for better opportunities, and be more picky and curious about the actual work you'll be doing. You owe them no more loyalty or accommodation than the…

It's difficult to be proud of anything I'm doing :). Whenever I propose a design for something, I'm told to do it the fastest way which is usually a bad hack. I'm amazed the code works at all. The stack traces are more than twenty deep sometimes. I can't stand to look at it for more than 20 minutes without a break. I need to be more careful next time I'm looking for a job

> Whenever I propose a design for something, I'm told to do it the fastest way which is usually a bad hack.

And why do you agree?

One of the differences between professionals and other workers is that professionals are experts in how to do their work. If I tell my doctor to just write me a prescription because I said so, she'll just scoff. I can make requests, but she decides what will happen and how.

Especially if you're already likely to quit this job, then you don't have much to lose by insisting on using best practices. Sure, you have to honor their desire to do things reasonably quickly. But you don't have to keep digging the hole deeper if you feel it is professionally negligent to do so.

Re: Ask HN: I got duped into working on legacy code, should I leave?

#46

Post by dkarapetyan »

sauronlord wrote: post #40

"..hope I don't get fired for underperforming?..." What kind of passive language and thinking is that? Everything is legacy once it is written. Imagine you, the trailblazer, only having to write new code and not having to maintain the crap you wrote. The actual problem here is your attitude and way of seeing things. Legacy? Sucks? Fucking write a better version and show some initiative instead of just hoping of being…

Was with you until you got a little too aggressive. There is a better way to say what you're saying.

Re: Ask HN: I got duped into working on legacy code, should I leave?

#47

Post by realPubkey »

ekarulf wrote: post #41

I find it impossible to replace a legacy system without first understanding the legacy system. It's easy to say "X is old" or "X sucks at Y" but keep in mind that applications have a strong survivor bias. Ask yourself, "what does X do that is awesome" or "how has X lived this long" and you'll find a lot of hidden scars and requirements. Start documenting those requirements and create a plan to map those to technologi…

Yes. Thats what the college-lecturer tells you. But in the real world you should not stay longer than necessary in a company which lied so hard into your face.

Re: Ask HN: I got duped into working on legacy code, should I leave?

#48

Post by dkarapetyan »

realPubkey wrote: post #47
ekarulf wrote: post #41

I find it impossible to replace a legacy system without first understanding the legacy system. It's easy to say "X is old" or "X sucks at Y" but keep in mind that applications have a strong survivor bias. Ask yourself, "what does X do that is awesome" or "how has X lived this long" and you'll find a lot of hidden scars and requirements. Start documenting those requirements and create a plan to map those to technologi…

Yes. Thats what the college-lecturer tells you. But in the real world you should not stay longer than necessary in a company which lied so hard into your face.

Always several sides to a story and during interviews everyone is on their best behavior. I have yet to be in an interview where I get the full and complete picture without any embellishments. It's only after you have been on the ground floor for a few months that you see things for what they are.

Also, OP's story is currently colored by how upset he is so I'm sure the interviewers made the job sound more appealing than necessary but I'm sure there is also some revisionist history happening.

Post reply on HN