Hacker News Reader

Surely the resulting code is better than the original code (otherwise I'd expect the author to push back on the comments instead of implementing bad suggestion).

Only if the author cares enough about the code. I've implemented things I know are bad in the past because the politics of pushing back aren't worthwhile. Especially if you work in an environment where your review is driven by the opinion of your colleagues (eg stack ranking).

"Code reviews are painful" are often a sign of bad team dynamics or a culture of politically motivated decisions.

so another story here:

code review mandatory at last place I worked, the place had a policy of any code you push in that does not match the guidelines must be changed. Guidelines was that no em allowed, must use rem. Old part of codebase assigned to me, I found some things in CSS I improved (reuse of code, a small overflow bug)

Got comment - you need to change em to be rem.

I can't do that because there is em all over the place and this can have side effects. The diff shows it as my code but if you look you can see it is actually the old code just moved to right place.

Those are the rules fix them.

since this code was in a big release of stuff and none of our stuff was allowed through unless I changed one em to rem even though I could not be sure there wouldn't be problems I went ahead and rolled back my changes.

fixed overflow bug other way. no code reuse ever in that part of codebase.

My CSS-foo is fairly weak; what problems would em cause such that they disallow it in the code base?

em is relative to the font size of the nearest parent thus you can have situations where your font size of an element node is radically different due to the parent it is placed in. This can of course be powerful if used correctly, but it takes someone really good to use it correctly and perhaps it shouldn't be done in big teams because big teams imply some people not very good in some particular discipline.

on edit: adding this link https://j.eremy.net/confused-about-rem-and-em/

Thanks! I never understood this difference between em, rem, px.

I learned something very new and I have been doing web development (backend mostly) for a long time.

Both em and rem refer to element sizing. Rem is relative to the base page font size, while em is relative to the containing element font size.

The consequence is that nested em specifications are cumulative; 1.2em of 1.2 em of 1.2em gives 1.2^3, while 1.2 rem is fixed, antwhere it is used on the page.

This is not of itself good or bad, it is behaviour. Use it to desired effect.

As examples, I prefer to specify font size within elements (header, nav, aside, article, footer) in rem, font size of contained text (h1-h6, code, blockquote, pre, generally) in em, and width of text elements in em.

This allows for less CSS (most of my styles are a few dozen selectors, often less), as relative text doesn't require respecification, and content blocks are defined in terms of sensible, reader-focused, preferable widths, padding, and margins.

Much of that is personal preference: I design around article text, mostly, not wireframes or pixel-perfect layouts. The designs also tend to be robust.

Examples:

    html, body {font-size: medium; line-height: 1.4;}
    article { font-size: 1rem, width: 40em, padding: 2em 4em }
    aside, nav { font-size: 0.9rem;}
    header, footer {font-size: 0.8rem}
    h1, h2, h3, h4, h5, h6 {margin-top: 3em; margin-bottom: 1em; font-weight: bold;}
    h1 {font-size: 2em}
    h2 {font-size: 1.6em}
    h3 {font-size: 1.4em}
    h4 {font-size: 1.2em}
    h5, h6 {font-size: 1.1em}
    .callout {font-size: 2em; width: 50%; padding: 1em; margin: 1em: margin-left:0; float: left: clear both;}

Relative sizes of h1-h6 don't need to be respecified when used in header, footer, aside, nav, or .callout elements. And in callout, they're automatically scaled to 2x 1rem * factor -- cumulative application of em sizing.

With simple DOMs, this can be useful. Complex DOMs and stylesheets may benefit from direct specification via rem.

So in essence, styling a component with em can reduce reusability of components more and more as em gets introduced in various places?

It's not that clear cut.

The 'em' sizing is inherited by child elements, which can be either a feature or a bug, depending on interactions.

If you're aware of this and design for it it can be a useful feature, and enhances reusability. If you're unaware of this, or your design is complex, it's much more likely to be a bug. "Bug or not?" is entirely dependent on that context.

Yikes! Demanding a bug fix also remediate unrelated, extant style violations is a good way to ensure nothing improves.

right. I said that to him and to the frontend lead for the whole organization (very big place) - codebases this big you can't have that requirement, do it with a small codebase/team who all sit in the same room and know what's going on sure. 10-11 teams spread around with lots of different programmers, many also remote, it can't be done. But no dice.

In fact I had people specifically recommend me at times, don't do that because when it hits review with reviewer X it will be a problem. Sometimes you had to butter them up asking their opinion before (and that wouldn't always work because sometimes they would change their mind) anyway I better shut up and get to work, I can feel the bitterness returning!

There is something to be said for continuous improvement (my company has a policy of leaving the campground neater than when we got there, which is one of the reasons I enjoy working there so much), but there is a time and place for everything...

Not that this applies to your particular situation of course but I have seen places where that duplication was actually warranted - especially in css. E.g. some code was extracted and reused when it was not meant to, just drying the code up, but those things turned out not to be the same thing and it was rather painful when we had to change them independently.

As to your particular case - I’m sure the politics of the situation were pretty bad but I’ve solved situations like this by having two separate commits/prs one that uses the old style to fix a problem so the change is obvios, the other to change all the old style to the new style of code. Helps the reviewer understand what’s happening and move the PRs along. Though this was in a much more liberal code review environment of course ...

no, it would not have worked in this situation. the problem was really the reviewer who could not be gotten around.

Oh, no! You copy-pasted code, with zero knowledge about it structure, and got caught. It's sad.

Next time use throw-away account.

Did you consider fixing "em -> rem" (in a separate commit) before you start applying your own fix to that code?

> where the author says they feel like code review is optional. If you think it's optional

I think these are two different things. The author says they don't enjoy code review, and that that probably is because it feels optional. From the post, I gathered that that's not what the author thinks - the first few paragraphs emphasises that they think it is an important part of the work. But apparently, that knowledge isn't enough to change the experience of reviewing.

> The better way to think about it is that code review simply isn't optional.

Of course it's optional. Tests are optional too. There are many projects and situations where trading code quality for development speed makes sense. Especially when you're building MVP's, experimenting, or trying to validate market demand.

It's also often not worth it on small to medium sized projects where only one person is working on the codebase (think 3-person project: designer, front-end dev, back-end dev).

Larger projects with multiple people working on the same codebase are a completely different story. In those cases code reviews are valuable and highly recommended but still optional. The only thing that isn't optional is delivering a working product.

Best practices == great, dogma == not great.

To me, code review is a combination of two things I would think we would try to avoid: 1. Testing implementation over behavior 2. Using a human being to test in a non-automated way

Proper test coverage and human QA to me is the most effective approach to not shipping bugs, because it really doesn't matter to the customer what the code does or doesn't look like if it runs. And I just don't believe that code review is an efficient means to catch that if user is in a state of X and tries to do Y to Z that the app blows up. That becomes most apparent when that code is running in the full context of the application, not when viewed as a git diff. A bug is so often a perfect storm of seemingly innocuous but conflicting/insufficient behavior than it is some block of code that is clearly broken at a glance.

There is something to be said about stylistic and architectural issues and code reviews do give an opportunity for asking questions and driving "better" code. But this is actually just another reason why I don't like code review. "Better" is subjective and there's always another level of "better" to be reached. If I held them to my own personal standards sometimes I'd be asking for a rewrite. So what do I do, suggest that they improve their code, but not to the full extent that I feel it could be improved?

I hate code review.

There are so many bugs that I've caught in code review that would not have been caught by automated or manual testing. Two big reasons for that:

1. Automated testing is still written by a human. A human who doesn't understand the edge cases in their code isn't going to understand how to write tests for those edge cases either. And "have someone else write the tests" won't work either, because frequently these edge cases are a result of the code implementation, not inherent in the problem, and so without a thorough understanding of the code you wouldn't even know the edge cases is there to be tested. Same goes for manual testing too, the manual testcases are still constructed by a human.

2. Race conditions, or rare conditions that most people won't hit. These are rather unlikely to be uncovered by testing (whether automated or manual). Race conditions in general are hard to test, and rare conditions are, well, rare.

There's also just cases like unexpected interaction paths that can produce improperly-handled states in the code, or bugs that don't result in an obvious bug to the observer but are still incorrect (the obvious example here is memory leaks, but this could also just be leaving the app in a bad state such that subsequent actions would fail, but you don't catch those because your test is done at this point).

much like 'testing', the code will get reviewed at some point - either during a crisis, or before it goes out.

> The author of this piece admits that they have to change the code they write in response to comments. This means the code review is flagging areas of improvement. Surely the resulting code is better than the original code (otherwise I'd expect the author to push back on the comments instead of implementing bad suggestion).

There's a potential problem with that idea: The assumption that, if several programmers involved offer different solutions, the best solution will prevail.

I think in many cases the accepted solution ends up being the one pushed by the person that holds more power, the one who can communicate better, or the person that 'negotiates' more aggressively.

I'm not saying that to argue against code reviews - It's just a friendly reminder that although as technical people we hope to achieve some objective level of quality, we can't escape office politics, human bias and general subjectivity.

You're right. I'd say it's an aspect of professionalism to get better at this as a reviewer. One of the things you need to try to do when you read somebody else's pull request is to put aside your ego and focus on the code.

I'm fortunate enough at my company to not have to submit to the code review bureaucracy - and I can justify it by having metrics which indicate that my code ships with fewer bugs than the code of other teams which do go through code reviews.

I think everyone should have some level of freedom to work in a manner which works best for them, rather than forcing everyone into the same process. Success should be measured by achievement of business goals and revenue.

For every anecdote you have about code reviews finding bugs, I have my own anecdote about people writing code going through a thorough code review process and the code crashes the first time an end user tries to use the functionality.

So yes, code review IS optional if you can justify it.

I like how you emphasise the "IS" instead of the "if".

For every anecdote you have about code reviews finding bugs, I have my own anecdote about people writing code going through a thorough code review process and the code crashes the first time an end user tries to use the functionality.

Sounds like you have bigger problems than code review!

What's the justification here though?

Saying "my code is better than other teams' even without review" does not prove that reviews are not helpful - maybe if you get your code reviewed you'll ship even fewer bugs, and if other teams don't review they'll ship even more bugs.

Further, eliminating bugs caused by any individual code commit is not really the goal behind code reviews in the first place. Finding bugs that the user notices on immediate use is an indictment of the QA process (whether manual or automatic).

Code reviews help maintain sanity in the code base, and are useful for sharing knowledge (both in terms of quality of code, and ensuring that there are at least 2 people who have some idea behind the code written). It provides an opportunity to have it read by someone who doesn’t share the same priors as you and helps you identify sections of code that may be confusing for another developer (including future you).

If Code review does lead to the identification of bugs that’s just a bonus.

There’s a reason Code review does not have a “run the code and test it” component to it and is entirely focused on code, and not whether it runs correctly.

>Code reviews help maintain sanity in the code base

But why delay the shipment of features/fixes to the customer for any of these reasons?

If code quality and human understanding is the actual rationale, it could make more sense to review code after it has shipped. In that way, it would cease to be a bottleneck for the developers or customers. Additionally, any bugs that appear in production could be accounted for during the refactoring process.

Saying that there is value to a practice is the easy part. We're forgetting to ask when to code review and whether or not code review should ever be a blocker to other processes.

Because in most environments it's unusual that a developer would go back and change the code after shipping it -they'll have e.g. another high priority bug to fix. Reviewing and refactoring code prior to shipping helps to keep development going sustainably, reducing build up of technical debt.

> people writing code going through a thorough code review process and the code crashes the first time an end user tries to use the functionality.

This is not an argument against code review. This means that the change wasn't even ready for code review. Whoever put up those patches skipped the step before it, where they verified the functionality of their own code.

Code review shouldn't -- can't -- be a time for someone else to mentally (or actually) execute the code and sign-off on it being bug-free. It's a sanity check for read-/maintain-ability. You can generally validate that the patch isn't doing something stupid or dangerous, but it's 100% on the submitter to ensure that their code works. If you have people putting code up for review without testing it first, then you have a bigger (people) problem, that code review will never be able to help with.

You are on a team though. If you are better than everybody else then your team is missing out on your code reviews.

> Surely the resulting code is better than the original code (otherwise I'd expect the author to push back on the comments instead of implementing bad suggestion).

Sometimes the suggestions add very little / nothing. Push-backing (i.e. adding a back and forth) would be a waste of time so you just do whatever the code reviewer says.

> So code review seems to be doing its job here. It's producing better code.

As with a lot of things, balance is needed. Better code comes with a cost (time), and that time may have been better spent somewhere else.

Wouldn’t bugs be caught during dev testing, or QA testing or automated testing or while running on a staging server? It seems nuts to rely on a developer looking at a code diff to catch bugs in an app, it happens sometimes but I generally think review is more about quality.

Code review doesn’t catch bugs. Sure, sometimes an experienced dev can spot a bounds error or race condition, but CRS are for knowledge sharing and style adherence.

Completely disagree. I catch bugs, and have my bugs caught, at least a few times a month.

Beyond the "this will likely deadlock without a timeout", "this does not clean up after itself on exception", "if the cache is not hot, this operation you assume is 1 sec will take 1 min" type thing — one of the strongest points of code review is having a conversation about "this code is hard to understand" and turning code that's correct if you think about it really hard into obviously correct code.

It takes a reasonably strong reviewer to get there though, and that's not always possible.

> Code review doesn’t catch bugs.

It can but it's definitely not a 100% surety (as some people seem to think.)

I review code everyday and it's my main way of finding bugs (besides automated tests). It saves us a lot of time that I can detect issues before asking testers to run the application and go through test scenarios.