Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I agree with the premise that making is harder than commenting, and this can be frustrating. But I don't think that telling people not to comment is the right solution.

There can be a lot of value in comments, even ones that have not been thought through very deeply. In fact, casual, uninformed comments are often the most valuable as they're representative of how someone new to your product or project will perceive it.

I've found that the keys to getting value out of comments without getting discouraged are:

1. Not to take comments too personally or get defensive about it. They're not attacking you. In fact they're not attacking anything. They're just making an observation. (On rare occasions, you may get a comment from an asshole who really is attacking you personally, but it's pretty easy to tell when this is the case and ignore them).

2. Take a more statistical view on comments. If one person mentions something, maybe they're just an outlier. But if 10 people mention the same thing, they're probably on to something.



Comments are probably have positive usefulness expected value in a probabilistic sense. Unfortunately extracting that value, separating the insights from the dross, is work and requires effort. Effort that perhaps the person doing the original work being commented on shouldn't necessarily be expected to take up.

Additionally, human nature makes it hard to receive legitimate criticism dispassionately, let alone when it comes in a big pile of stuff that also contains pointless abuse.

There is value in comments, but depending on a project's circumstances it may or may not be economical to extract it.


> Effort that perhaps the person doing the original work being commented on shouldn't necessarily be expected to take up.

It doesn't apply to all situations, but sometimes you end up with less net effort by listening thoughtfully more. I expect this is especially the case when you want or need to optimize for some form of popularity. If you're just making things for the joy of creation, sure, just ignore everyone as that's certainly more work than pleasing only yourself.


The article also brings up the difference between two types of comments, the "let me know how I can be helpful" comment versus "I will connect you to five potential customers tomorrow" which is the more helpful one. Commentary can be valuable if it's been thought out more, or if it the commenter invests something themselves, both of which this example shows.

Academic advising is also extremely valuable even though it fits the definition of commentary, because they have put reputation and funding on the line, and spend long periods of time thinking about the work. It's not always just because the advising comes from an "expert" in the field, but just that they have invested in the work.

The type of commentary that I find less useful, is the vague notions of "there is a serious problem here" comments that I hear sometimes on Hacker News. Like "I find this problematic, [and then some outrage]" or "I have some concerns about the security/privacy of this" or "I am concerned about HIPAA here..." It demands a response, and usually the maker has already thought through this in more specific terms, but can't rebut because the comment is vague.


> The type of commentary that I find less useful, is the vague notions of "there is a serious problem here" comments that I hear sometimes on Hacker News.

In creative writing workshops, a common quip about the feedback of non-writers (eg, friends and family) is:

"If they don't like something, they're probably right. But when they tell you how to fix it, ignore them."

Which is to say, in this context of this thread, vague complaints usually are a reliable indicator of something wrong. But knowing exactly what is wrong and how to fix it is a much bigger ask, and even if they spent the time and effort most people wouldn't be able to help you here.

But that doesn't mean the signal they're giving you isn't useful.

> but can't rebut because the comment is vague.

Don't rebut it. Don't take it personally. Just take it as a data point and say thanks, or don't respond.


In our tech consulting onboarding guide it says (paraphrasing Jeff Patton's User Story Mapping), "Be doctors, not waiters.

Where does it hurt? is the only question you should need."


These kinds of comments are probably better worded as questions.

Many questions are simple. Others are difficult and not immediately answerable, but perhaps worth pondering.


When you get someone who’s willing to offer up a lot of information, it’s great. And when you don’t, then really honestly try to understand them and their frustration. When you get to the bottom of it, you’ll come away with something useful and they’ll know that you care enough to try to make it better.

I recently jumped on a call with some people from Miro. One of them had left a calendar booking link in a forum for something else and I tenaciously scheduled myself in. When I spoke to them I said “here are my frustrations and I want to give them to you know before I just get into the product and learn to live with them and can’t articulate them anymore”. To their credit they took the call and listened. They got something and I was happy.


+1 for the statistical view. One or two data points can be ignored, but 3+ different people who report a the same bug/problem and you need to listen up.


Your first point reminded me of the post the other day by Daniel Stenberg about the feedback the he receives for his work on curl. [0] Making things in public absolutely does open you up to personal attacks. Ignoring them is easier said than done.

[0] https://news.ycombinator.com/item?id=26192025


Also when breaking new ground, if 10 people mention the same wrong thing, you may be onto something ;)


Exactly! Lean Customer Development! We spend loads of time teaching this to new entrepreneur


I am sorry to tell you that comments have never been helpful for me to meet a deadline. They have ever been a signal that the commenter was not willing to engage intensely enough so that his contribution would have been of immediate value. Instead it remained on a meta-level, expecting me to carry half-baked and barely understandable thoughts out.

But guess what? Bosses comment, workers write. That's the fate. Don't know why. Some is to attribute to Peter principle but not all.


A simple rule for dealing with comments is « no feedback on feedback ».

Obviously works best if the person commenting doesn’t have any power in the situation. That being the case, avoiding a big discussion and extracting whatever useful information is contained in the comment is a net win.


Did he tell anyone not to comment?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: