[WorkLifeBalance]

Personally I thrive in this kind of environment, it's a kind of comfortable chaos. MVC5 is positively modern compared to many of the products I've worked with which have included (within the last 5 years) VB6 winforms but mostly aspx webforms.

Understand that previous developers might not know what a DOM is have likely written ad-hoc jquery snippets, included 3 different versions of jquery-ui until they happened across the right combinations of versions to get their plugins to play and will have a mentality of "it was working".

Given that, here's some advise on how to stay sane:

1. Read and "Digest Working Effectively with Legacy Code" but accept you'll likely not be empowered to actually implement the changes.

2. Understand you'll have a grace period of ~6 months where you'll be far more empowered to make changes and get suggestions pushed through, after which you'll likely find yourself worn down by the system and more accepting of the status-quo. Leverage this time if you can.

3. Try to identify what caused the product to get to the state it is so it can be avoided for future products. If it's something you're not empowered to change such as "bad hiring policy" try to work out how you can spot that in future interviews.

4. Be pro-active in looking for bugs, particularly security flaws. Finding an IDOR which lets you get cross-account data is usually fairly straightforward in these kinds of environments. Doing so gets you noticed so you're not just another cog in the system and can provide real value. Be careful not to put others' out too much while doing this. It sounds like you only work on the front-end but even what appears to be front-end can have security implications such as template-injection.

5. Draw a line in the sand. Be entire accepting in legacy code but if someone is working on a new part of the product or is fixing a bug in legacy code be absolutely ruthless in what you'll let passed in code review. This may not do you any favours socially but a reputation for harsh code reviews isn't all bad. For one thing you'll have to do fewer code reviews.

6. Don't bite off more than you can chew. Treat it like a giant knot and don't be tempted to pull at threads unless absolutely necessary. Instead wait until given strong reasons for refactoring.

Edit: formatting, because HN doesn't do markdown.

[Floegipoky]

In regards to #5, worth pointing out that some coworkers are probably just as disappointed with the current state of affairs as you. They would probably really appreciate somebody pushing them to hone their craft.

Also worth considering that some people never learned how to handle constructive criticism and will take any suggestions as personal attacks. Stay as far away from these people as possible, you aren't going to change them but their pettiness can make your life miserable, particularly if they're closer to management than you are.

[geezerjay]

It should be noted that more often than not there are strong reasons besides ignorance and incompetence for a project to arrive at a certain state.

https://en.m.wikipedia.org/wiki/Wikipedia:Chesterton%27s_fen...

It shoud be stressed that a developer should not assume all his predecessors were wrong just because he doesn't understand why some stuff ended up being the way it is. The world is packed with developers who proud themselves of fixing things that were never broken, and in the process break stuff.

[testplzignore]

> if someone is working on a new part of the product or is fixing a bug in legacy code be absolutely ruthless in what you'll let passed in code review. This may not do you any favours socially but a reputation for harsh code reviews isn't all bad.

In my experience this is the most important thing. I particularly focus on the "what the heck does this code even do" aspect - the person who is modifying the code needs to be able to explain exactly what the code does and why it is correct (don't accept "it compiles and the one unit test we have passes"). Forcing the team to understand the code will make it very difficult for them to continue adding bad code to the pile, as they won't be able to explain what they bad code does, or will realize the code is bad in the course of trying to explain it.

[s73v3r_]

It is very, very important in this situation, however, to not be outright mean. Do not let anything through that you would not be comfortable maintaining, but don't be a dick about it. All that will end up accomplishing is that you'll be blamed for things not going through, and then you'll lose your credibility as steward of the code. People will go over your head to the boss, who will just make you pass it.

[ndh2]

Unfortunately 5. doesn't work if your company doesn't do any code reviews.