Ask HN: Technical forums that are not dominated by pedantry?

203 points by rrwo ↗ HN
I'm finding forums like Stack Overflow to be almost unusable due to the sheer number of useless pedantic criticism that people post instead of actually answering questions.

This usually falls into one of the following:

1. Nitpickers who interpreted something in your question as incorrect. They will focus on this minor point that they think you said.

2. Dogmatists who consider what you are asking about as violating some sacred creed, like model-view-controller or following a pattern. They believe that you've violated that, and they are going to focus on that. Nevermind that you're asking about performance problems with some function, they are going to criticize you for using that function.

3. Evangelists who think you should be using newer technology Y instead of technology X. You asked a question about configuring something in X, and instead get lots of useless responses telling you to use Y. It doesn't matter your reasons for using X. X is stupid, use Y.

I suspect that if the voting/karma behaviour of these forums could be changed, there would be less of this.

But are there better forums where this sort of behaviour is discouraged?

240 comments

[ 212 ms ] story [ 3158 ms ] thread
*dominated

(But to answer your question, not a clue.)

Sorry, accidentally submitted before fixing typo. I've edited the post to be longer.
I think it was a joke :)
You would surely love going through a code review @ Google :-)
No such place exists. Let's make one. It will be you and I and nobody else.
You and me and nobody else. In the pedant-free zone.
I mean, no zone is really pedant-free. More of a low-pedancy zone.
There are a lot of subreddits that discourage all of the things you mention. They are usually full of eternal september posts, repetitive and low value.
I give you the ole reddit pedant trap:

1. Ask your question

2. Create a new account and give your question a wrong answer (or point out mistakes or whatever)

3. Watch as 90% of pedantic idiots focus fire on that guy who clearly doesn't know things

4. Profit

This is (sort of) what Cunningham's law is about.
This sounds like a great opportunity for a SaaS. The product lets you write a question and one or more wrong answers or nitpicks, then posts the answers from random accounts after some random delay.
It could be implemented by training a GPT-3 on downvoted SO answers.
Sounds similar to the "Linux is gay" technique.

http://bash.org/?152037

The flip side is that if someone asks a question like "how do I do X in Windows 10?" there will always be some dull idiot who says "uninstall it and install linux".
Not to mention mods that come in and close your question without a word because it is "duplicate" or "unhelpful".

Hopefully they got the mods under control but I think #4 should be "out of control mods". I think all 4 can be applied on almost all public forums and #4 is the most killer because of how impactful it is.

SO mods are uninvolved in the vast majority of closed questions; it’s a community vote mostly by non-mods. When something is closed as a dupe, another question has to be nominated as the answer, and this is shown in the close note.

In my experience, these are typically quite accurate. I’ve seen a small percentage get reopened upon clarification, but people often whine about the dupe closings because their question is “how do I calculate 2+2?” and the linked answer is “how do I calculate 3+3”.

Add in "I am getting an error" (without providing the actual error) and https://en.wikipedia.org/wiki/XY_problem posts, times a couple hundred thousand instances, and fairly strict community moderation becomes the only thing permitting a little signal in amongst the noise. SO's original intent was never to be a free-for-all Q&A with repeated similar questions; it was intended to also be a big reference of existing answered questions.

(See https://blog.codinghorror.com/introducing-stackoverflow-com/... "Stackoverflow is sort of like the anti-experts-exchange (minus the nausea-inducing sleaze and quasi-legal search engine gaming) meets wikipedia meets programming reddit." That "meets wikipedia" bit is key.)

But that last reason makes the behaviours described in the question even worse.

If I land on a page with the question "What does the blue button on the back of a Mark 3 defrobulator do?" at the top in large friendly letters, I am likely to be interested in that question, and I'm unlikely to be interested in any of the conversation that follows on from "Why do you think you want to press the blue button?".

That's why comments and discussions are second class citizens and comments get collapsed.

A well written answer doesn't have any of "the conversation that follows" in it.

One of the challenges with users unfamiliar to the Q&A format is that they treat SO like a forum and try to force in doing replies and back and forth into questions and answers (collectively posts) when the site's design is actively and intentionally hostile to such.

> I'm unlikely to be interested in any of the conversation that follows on from "Why do you think you want to press the blue button?".

Sometimes that conversation nets out as "you will kill someone if you press the blue button in certain circumstances", and that's an important clarifier to have access to.

Surely in cases like that you don't want to be having the "why?" conversation: the answerer should put that information in big bold letters at the top of their answer, whether or not it turns out that the person who happened to ask the question is in those certain circumstances.
Sure, but what about when it's a little less clear, and the stakes lower?

"How do I increase the PHP memory limit? I'm getting a memory error when I foreach through lots of rows." is likely best answerable not by increasing the memory limit (and running into the same problem with a few more rows), but by chunking the rows to avoid the memory leak.

I don't see a downside to "there's probably a better way of doing this you should at least consider, here's why" as an answer to a question. SO permits multiple answers; someone'll likely be along shortly and give the php.ini answer, but it's not the best approach.

I think adding helpful information like that is great.

But it's much better if it comes after answering the actual question, not instead of it, because (as you say), SO was intended to be a big reference of answered questions.

I joined an old school forum during the pandemic and have really enjoyed it. You are able to "like" posts but if you want to express disapproval you have to jump into the discussion and air your arguments. The moderators are not perfectly consistent but they tend to err on the side of leaving a discussion open even if it's a bit of a train wreck.

I've noticed that not being able to silently disagree/disapprove results in a community where people who genuinely and fundamentally disagree with each other end up interacting a lot and often become friends.

I don't think any particular format is right, just adding my experience.

I think the 'downvote' is neither useful nor healthy, indeed. Very often it is also implemented in conjunction with other UI features which make downvoted comments disappear, which ends up creating an echo-chamber and dissuades dissenting comments.
Downvote is tricky.

It is useful when used correctly but as can be seen on Reddit (and somewhat on HN but it typically balances out) it is used as "I disagree" / "I don't like this" vs being used more like a moderation tool "this has nothing to do with the conversation".

I like how HN restricts the downvote for a while but I think with Reddit being so popular there should be an opt-in where you have to acknowledge the statement "The downvote is a moderation tool and isn't used to express disagreement" (I feel like that near exact verbage is somewhere like in the guidelines but I am meaning making it required acknowledgement).

I agree, but struggle with the assumption that downvote should be for moderation. That was (and partly is) my own assumption here on HN, but after being here for a while I started wondering why we don’t treat it symmetrically. Upvote is widely used for “I agree” and “I like this” and lots of other reasons, why shouldn’t downvote do the same thing? It also makes getting downvoted feel less painful to realize people do use it like this, for a wide variety of whimsical reasons (just like with upvote), it’s not always suggesting you’re wrong or breaking the rules.

On HN before I had access to the downvote, getting one felt so strongly negative I decided I wouldn’t use it, and have stuck to that ever since. But in the mean time, I started to see real utility in it’s effects as a soft moderation tool. Downvoting someone who’s relevant but wrong or making invalid assumptions wouldn’t meet your strict moderation requirement, but downvotes do tend to sort the conversation correctly helping keep the good stuff at the top, and also being a signal (albeit imperfect) for readers. Using upvotes also helps with the sort, but doesn’t change the color of comments (maybe it should!). So I’ve questioned myself whether I should stick to not downvoting, and I only keep avoiding it out of some sort of ritual or pride or something, I dunno.

It’s interesting because upvotes here are used far more frequently than downvotes, and downvotes are strictly limited, but we still have disproportionately strong reactions to downvotes. It’s also worth noticing how because downvotes change a comment’s color, it often prevents people from piling on; many comments stay at 1 or 2 downvotes and people leave it alone after that, even when many many people notice it’s wrong or breaking guidelines.

I tend to think of forums that have a down vote mechanism (along with karma) as being like those crystal growing kits. By discussion and voting the community grows a consensus crystal and then gradually refines ways to express the consensus.

Sometimes you want that! Other times you don't.

One of the things o really liked about slashdot was seeing two comments marked “insightful” that argued opposite sides of an issue.
Part of me wonders whether splitting downvote and disagree would fix this. I know in quite a few forums there's a system where like and dislike are separate reactions from agree and disagree, with the latter ones not affecting the 'score' in the slightest. Maybe it'd be even better if neither existed, and you had to vote a reason you liked or disliked a post, like 'is spam', 'is insightful', etc.

Not letting people just say "I disagree with this" without a reason might help remove this pendantry and echo filterness.

Or mods that will reword your questions, making it sound way different than you intended it to be.

I guess it’s a strategy to increase their counters/stats or something?

Anyway, that really unacceptable.

I think we have enough experience with crowd-sourced websites at this point (Wikipedia, Reddit, SO) that we can make a general statement that crowd-sourced websites attract just the sort of people you've described precisely because the "average" person that just needs an answer doesn't have the time/will to hang out on those sites providing content for free. I'm not sure there is an easy (or possibly any) answer to that problem.
Yeah I even see this in volunteer groups offline where organisers often are abrasive and pedantic and very difficult people, who basically lure people into this trap of a free event, and then use the situation to constantly try to assert themselves and jerk people around.
My perception may be skewed by my infrequent use, but strangely math and physics stackoverflows have much less empty posturizing and generally have high quality answers that are thoughtful and helpful. Most matters of "opinon" are limited to the topics that really are matter of personal style - i.e. which textbook is better ("x is better people who prefer y but foo is better if you like bar" instead of the dogmatic "xxx is the only correct book on this topic).
If true, could this be caused by those professions being more academic, and thus full of people who both have tenure and flexible positions and who teach for a living?
If you ask a question, generally, you are soliciting willingness to receive inputs. You seem to be feeling salty that you're not getting the input from others you desired, yet could not previously articulate.

Half of your pedantic nitpickers (of which I suppose I may now be counted in the number of) are at least providing you a concrete example of what the answer isn't, and what may ultimately be direction/inspiration for where an answer may lie.

I guess I'm just saying, be thankful you even get lackluster answers in lieu of the alternative.

As far as how to divorce honests attempts to solicit input from the social mediafication of the programming space, (i.e. aggressive building of mindshare in terms of dogmatic adherence to the One True Way to Parse a Lolcat), or people arbitrarily rearchitecting your problem space (Your lolcats must be partitioned, sharded, highly available, with 9 9's uptime, with no greater than 300 ms from disk to eyes, when all you needed was a Lolcat that'd get there eventually when a friend came looking in the afternoon in less than 7kb total)...

I'm not sure. It's rather annoying. The latter I find less annoying than the former, but generally speaking, just be both polite and willing to patient game of code golf til a bored grey beard has some time to kill, or you figure it out.

I think the three groups you list match the motivations why someone would systematically (not just a one-off) put in their time to answer large quantities of technical questions without getting paid for it - it's because you really want to evangelize something, or you want to push a dogma you care about, or you're the type of person who gets off on arguing intricate details.
Don’t forget:

4. Enthusiastic X-Y-Problem solvers. You asked about X but the replier is an expert in Y, so he asks “what problem are you really trying to solve?” Hoping to steer you to Y so he can shine. No, sorry. I really do want to just solve X thankyouverymuch.

Are these people really trying to be disingenuous or does the question missing context? Often asking for the really problem reveals more information and allows the answerer to be more helpful
Agreed this is a positive thing
Very often they choice (“this project will be using X”) is beyond your control and anything related to Y usually does not help very much.

Most people don’t seem to understand this.

The problem is that the XY Problem is a classic beginner mistake and most SO questions are beginners who didn’t even consider Y, so maybe 1% of the time it doesn’t apply. And to those people, it is helpful to explain why you are stuck with X as it will likely clarify context that will help people answer your question.
That’s so condescending though.

How do you know if someone is making a beginner question?

You just can’t, you can only assume it. And that’s very often wrong.

Sure you can, beginners never research their questions and they don't give the right context.

Beginner question: "How to make a webserver using bash?".

Non-beginner question: "How to make a webserver using bash? I have this weird embedded system and it is missing SDK, http daemons or programming languages, but somehow comes with coreutils and fully-featured bash 5.0.17. And I really need to serve a single static webpage off it"

The first question will get X-Y answers like "here is how to install apache" and "you probably want to make CGI script instead, here is how you do it". The second one will get the interesting answers.

Have you ever scrolled through Stack Overflow's new-question queue? It's quite easy to separate not just beginners from advanced users but also beginners who have done their research from beginners who have put less effort into their question than you put into reading it.

Maybe it seems rude or condescending to address a poorly formed question, but I'm not sure what else you can do since you have to realize that low effort questions also have a cost.

For example, have you ever tried helping someone where you have to ask a 20 questions just to glean enough info just to help them? You can see this on SO all the time.

Stating from the get-go that the choice of X is beyond your control would help though, assuming that you’re already aware that you wouldn’t have the problem if you’d use Y instead of X.
(comment deleted)
I think it's unfortunately something which has been reinforced by the majority of poor questions. You may legitimately want to solve X, but nearly all of the questions active users see are just poorly researched questions about Y. Attempting to take your X question literally means giving that same latitude to every other question, which increases the volunteer's workload by orders of magnitude.

I'm not sure how to best get around this, other than making your X question extremely clear, and describe why Y is not what you're looking for.

There are many poorly-researched questions on Stack Overflow, though this tends to be generational (and not exclusive to StOflow).
This speaks to discoverability issues more than it speaks to generational gaps.

As a really specific example: if people don't know that certain generations of Intel microprocessors calculated sine(epsilon close to 0) incorrectly, and ask about it, it really is incumbent on the answerers to not be a-holes.

Extrapolate out and consider various IRC channels. Python has massive documentation. A common answer is to point people to the docs, as they are large, and to encourage reading the docs. This is helpful and encourages discoverability. C, on the other hand, is an almost incomprehensible mess for discoverability. So a newbie asks a volunteer channel in IRC about what may be causing a segfault simply because the newbie has no sense of the tooling to help them discover the issue. Newbie gets berated (I have both experienced this and watched this happen). C channel is therefore toxic for newbies because it pushes them down arbitrarily, Python is good because lifting newbies up.

Common thread? Make it really, really easy to use tools, have copious amount of well-described examples organized in a sane way, and crosslink definitions and concepts.

It helps to provide all the things you've already tried to fix the problem. It helps the answerers not waste their time, and tells them more about the actual problem based on what you tried and how the problem responded.
Agreed...somewhat. Sometimes there is a hard XY problem, and sometimes not. People who care about interacting with other people like actual people would, and have the time for it, would say "I think you may have this XY problem for these reasons. But I may not fully understand your situation. Thoughts?" And then be open to the possibility that the asker is allowed to discuss X even if Y would be an OK solution.

XY Problem in an Ideal world: Consider that Y, not X, may be the best route. But feel free to respond and discuss how your problem relates to X and Y.

XY Problem in actual forums: You're wrong. X is bad, you should do Y. X is forbidden knowledge. You're not even allowed to ask how X compares to Y. Y is axiomatic.

In the ideal world, the XY problem is frequently novel to the group. The "feel free to respond and discuss" is something that has time consuming overhead in asynchronous environments. If one was to ask me an XY problem in person, that's something that can be dealt with in a way that has dense communication.

In the forum, the forum has likely been there long before the person posing the question. Often the people in the forum have seen that exact XY problem many times before and it is necessary for the person asking the question to demonstrate additional knowledge to get past the "yes, I know this is an XY problem and these are the constraints I have" to get it to be a novel problem once again.

This isn't saying it is ok to be a jerk, but it is important for the person asking the question to be aware of the history of the community and the forum.

Often this is best done with lurking for a while before asking questions (this was the norm on Usnet - https://www.24hoursupport.com/netiquette-for-newsgroups/ "Lurking is the practice of reading the postings in a group for a week or so before posting a message yourself. No one will know you are lurking, and you can get a feel for the types of personalities in the group, what types of things are discussed, and what topics or behaviors are frowned upon.").

I would contend that in many places today, the "join in and participate now" precludes lurking and so it is much easier to run into the stuff that is common knowledge (not forbidden) to the community that the new person hasn't become acquainted with.

Unfortunately it often happens that, even when people lurk for a while, and humor the experts and do the work to show why they really want to talk about X and not Y, they are still shut out. Part of this is tribal - you are an outsider by definition, and you challenged the experts and their dogma, you will now be punished for it no matter how valid your position is. And so you move on and do other things, and the culture of grumpy rigid experts perpetuates itself.
I had heard about the XY problem[0], but never of XY problem solvers. I don't remember encountering one, do you really find them that frequently?

[0]: https://xyproblem.info/

It's quite common, especially when dealing with tech support and the like, who expect that the customer has no clue what they're asking about (and, to be frank, in most cases that assumption is probably right).
SE users just love to call out 'XY problems', it's like declaring 'strawmen' on HN.

I asked a question recently along the lines of 'how can I do blah? Or is another way of achieving whatever it was'; response 'this sounds like an [XY problem](link like you can be on SE and not have heard of it)' - I told you both X and Y! Can we just solve my problem whether it's by the avenue I thought most likely or not?

This is joked about often regarding SO, but my experience is, that most of the time the users actually have issue Y, and were helped, and it’s only me who landed there via search engine and has issue X that is (wrongly) in the title, who is now not helped :/
I was just complaining about this very thing on twitter:

    One time my car wouldn't start at an Mc Donalds, and I asked inside if I could leave my car there while I went and got jumper cables.  I didn't want them to tow my car while I was gone.

   The person behind the counter didn't know if they would tow my car or not but told me the best way to get around was the bus, and here is why public transit was a better way to get around, and so on.

   My mistake was having a car at all.
https://twitter.com/adamgordonbell/status/158018073977710182...
There are batteries with jumper cables for 50 to 100 USD. You can load them up, have them in the trunk and use them whenever you need them.

This doesn't help your problem in the past, but maybe problems in the future. Helped me a lot in the winter.

I call this the "reverse XY problem." It gets so bad sometimes that you could completely frame the problem around X, defensively excluding Y from the possibilities, and still get a bunch of responses about Y.
I think this question is fundamentally misguided by being about pedants. If you look at the definition of pedant [0], it mentions people who are concerned with minor details but nitpicking [1], which you refer to in your first example, is more specifically about finding faults, so in reality you are looking for a forum _with_ pedants (who have enough knowledge about the minor details to answer your question), but _without_ the nitpickers, who focus too much on finding fault.

Hope that helps :)

/s, but like others said I think it does get to the heart of the problem where only people who get a kick out of pointing out flaws in small details in their free time hang around on technical forums.

[0] a person who is excessively concerned with minor details and rules or with displaying academic learning.

[1] fussy or pedantic fault-finding.

Problem is, nit-picking is the bread and butter of mid-level software engineers, and it becomes engrained in their psyche. If your 50,000 line code base has one wrong negative sign, or one missing brace, it might not work. So, the focus is on detailed perfection until the bugs are gone. Being pedantic about problem solving is a good thing!

Then they take this mentality to human interaction/conversation and turn into the insufferable nit-pickers that OP is complaining about. You write a 3 paragraph comment in a web forum like HN, and someone inevitably swoops in, ignoring your main point to argue: "Aha! On paragraph 2, line 5, you said 'all' but probably meant 'many' because I can clearly find counter-example A, B, and C. Therefore your entire comment is incorrect!"

But it's not generally accepted to have such a one track mind, to apply the requirements of some part of your work, to every human interaction. There are many professions that require absolute attention to detail in many situations, and lawyers and chemists and microbiologists, even other STEM fields are perfectly capable of not being insufferable pedants in human interaction, and can choose an appropriate level of detail depending on the situation.
That was awesome! You almost had me until I saw the /s. Good job!
Problem is all the pragmatic people you want answering your questions are too pragmatic to spend time reading and leaving comments on the internet

Never say never but I don't think it's very easy to build a broad and helpful community where the factors you listed aren't at play in people's motivations.

I find better help going into very specific communities for the technologies I'm having issues with. Discord servers are a great example

> I find better help going into very specific communities for the technologies I'm having issues with.

Probably true for some sites and not others. I would definitely say the more niche you go in subreddits, the more helpful people are, but that's not really true of Stack Overflow communities.

The same effect can be accomplished on StackOverflow by more narrowly tagging questions to specific technology niches, so that it does not show up on the watch lists of generic moderators.
Technology specific IRCs are still great resources as well!
I really wish IRC was still the defecto choice for open source projects. Now days all I see is gitter/slack/discord/matrix which I find all have a significantly worse experience than IRC.
If IRC was stateful (ie, it would "remember"/deliver offline messages when I sign-on again (at least buffer them for some configurable period)), I'd be onboard with using it for almost everything

But it's not

That's what made Slack so appealing early on - IRC-like communication with history

Maybe you know already, but you can use an IRC relay to act as pseudo client, connect to that and then read back on messages. You’ll have to find a place to host it though. For example, “The Lounge”.

Edit: typo

Yeah, you can ... but that's a bolt-on/additional service that, quite frankly, IRC ought to Just Do™ at this point

It's 2022, not 1988 :)

I just keep my irssi running on a Vultr VM.
I used to do that

Too much hassle, imo

IRC has OP’s problem taken to the next level. The people clinging onto IRC in this age are increasingly insufferable.
That's the opposite of my experience even to this day. As long as you follow question asking etiquette (do research before you ask, don't ask to ask, explain what you've tried so far, etc), people are usually good at answering questions when they can.
You may both be right. Chances are this varies wildly from IRC channel to IRC channel.
Funnily enough your comment describes SO just the same. How many people complaining about SO here have really just asked poor questions, hadn't explained what they tried, didn't do enough research first to ask a good question, etc.
A lot of really smart and talented developers participate in mailing lists as well.

As someone who used to frequently help out online in the past, the problem is that the questions never stop coming and most are usually from people who refuse to do the bare minimum amount of research or reading. This would be fine were it not typically bundled with a lack of respect for your time and a strong sense of entitlement.

> I put up a free lemonade stand, and now people think they are entitled to free lemonade, and the customers never stop coming. Well don't do it then?

I don't think that's a correct or fair analogy. It's more like:

> I published a lemonade recipe, described where to get the ingredients, and how to make it yourself. Now people are complaining to me that lemons and sugar are too hard to make from scratch, and I should just grow it myself and give it to them for free.

> You published free recipes to begin with, why? This is an implication that you are offering free work and free time to other people, and give them a way to contact you directly, so why get surprised when they ask you for more favours?

This is a truly mind-boggling response to me. If someone publishes a recipe on the Internet, you're saying that they owe you even more than what they already did?

This is like saying that if a coworker gives you helpful advice, it's fine to start sleeping on their couch. Just because someone is sharing their knowledge does not mean you're entitled to something more, and in many social contexts this would be seen as (at the very least) extremely rude.

I think the point GP was making is that it's to be expected, so don't be surprised when it happens.
can you recommend some discord servers for tech help?
OP should consider listening to this answer. It’s been 17 days since this poster commented on HN last so they pass the unpragmatic smell test.
And there's the factor that communicating through writing is hard as the lack of non-verbal cues means that you have to seek clarifications to be sure that you actually understood what the other person was trying to say. More so if they are strangers.
The pragmatic people solve their problems by asking for help and part of that involves stumbling upon questions they have answers for.
Not IWETHEY or Kuro5hin I found out the hard way they turn into troll boards.

I heard Soylant News is good.

Usually the most practical programmers are too busy focusing on coding than internet arguments. They _occasionally_ appear and drop nuggets of gold.
I remember getting flamed for making some comment about C on FidoNet at 2400 baud. It's an old problem.

There are things that would have to be done to produce a solution:

- understand the behavior: what drives people to post answers in the first place?

- define moderation rules that would shape the desired behavior

- build a community that focuses on this

- reward the best answers by the new metric

Community building is hard. We should remember to thank dang for doing a tremendous job of it here.

Every forum, no matter how its reputation/karma/power system works, is dominated by those who are willing and able to invest the most time into participating in the forum.

As you and many others have noticed, it appears that the type of people who can and do spend dozens of hours every month answering questions or writing instructions tends to overlap with "insufferable pedants" to a very large extent.

Thus I don't believe a forum that works the way you want can exist. The pattern you're observing is simply a reflection of the people that use the Internet today. By discouraging pedantry and by reducing opportunities to show off perceived intellectual superiority, you are effectively excluding the exact group of people who make tech forums work in the first place.

I dont think you can. And the reason is your problems all come from common human biases.

1) Observer expectancy bias, when they expect you to be wrong they work from there to prove their expectation

2) Anchoring bias, or prefering the solution you already know. Dogmatists will always exist, the upvote/downvote system helps them gain prominience and egg each other on.

3) Pro-innovation bias. People being exited by new shiny toys is another basic bias that is hard/impossible to overcome. same if the opposite of status quo bias and some people being terrified of changing anything.

I think the issues you find are made worse by the upvote downvote system. But the reality is that forums without that system still exist but have other problems. Such as discoevrability. Secondly by having no input from lurkers into the best solution, you will have to shift through the same haystack of domatics, evangelists and nitpickers (and hopefully those actively trying to help) without external validation.

You are arguing that the external validation (the unhelpful replies in stack overflow) is currently misaligned with helping your problems, and I agree. I have sometimes found the same problems in those websites. But they have some clever solutions, like the discussion below proposed solutions allows for a back and forth related to a specific proposed solution.

So if you ask about a performance issue and 3 guys mention MVC stuff and 1 guy brings up moving it to Electron and finally someone mentions a particular way to multithread your problem, you can comment below that guys comment and start a discussion there completely side stepping the ones you don't care for.

I think perhaps the person who asked could have a bigger sway on which answers rise to the top instead of people outside upvoting. But I believe there is a way to pin a reply as having worked so thats another way to dissuade the nitpickers and evangelists from showing up

Dev.to is pretty good at promoting a constructive community, but you still have trolls who will double down that their opinion must become your opinion
Came here to suggest this. I find the Dev.to community to be generally beginner-friendly, helpful, constructive, and kind.
It's like product/service reviews, which are dominated by unhappy customers since few happy customers feel sufficiently motivated to take the time to review. The people most likely to answer technical questions are insufferable pedants, who relish the opportunity to demonstrate their vast technical knowledge.

That being said, it's worth looking around on Reddit's more focused subreddits for what you're asking about. No guarantees, though.

I think there's a kind of CAP theorem for that kind of forum. It's hard to combine:

- open and free

- well-known

- respected

- has knowledgeable people

- willing to understand and discuss the unique context of every post in good faith

- also able to identify when you truly DO have an XY problem (because they do exist)

- scalable

I think asking questions is hard; it's rare to combine:

- knows the domain well enough to ask a well formed question about what they want done, with precise terminology.

- doesn't know the domain well enough to be able to find an answer.

That, combined with the flipside: if you're responding to questions, and most questions aren't very good, your default is to assume that the question isn't a very good question.

True. There will be some tradeoff between experts having to answer bad questions and deal with bad questioners, and unfairly treating good questioners as bad ones.

I think people get to a point in their life or their expertise where they lose sight of the fact that there are always new people. Literally people who didn't even exist when they were first learning their craft. Children can only learn by making mistakes. You have to be generous and sacrifice for them. These kind of forums do a bad job of handling the "children" of the subject matter who would be willing to learn. And I don't necessarily mean the forums have to change, just that they may be bad places for channeling that kind of generosity and sacrifice.

> I suspect that if the voting/karma behaviour of these forums could be changed, there would be less of this.

And less people answering your questions.

Frankly, few people are so selfless that they spend time answering other people's questions simply to help them. Many of them do so precisely because they can judge other people and feel superior. You take that away, and you'd have to pay people to waste their time doing your work instead of theirs.

hot take: I suspect a similar effect goes on in staffing police departments. I still believe that a majority of police officers (especially the one that don't make the news or go viral) are good cops who have secure egos and really want to uphold the law, but a fraction of officers really are motivated by cracking skulls and ego tripping (i.e. those things actually feel good to them, and are a significant part of why they are taking the job). The reason why these behaviors haven't been ruthlessly stamped out is because if it were completely not tolerated, then departments would have significant shortages, so there's an implicit "perk" of being able to ego trip and push people around and exert dominance (of course, up to a limit) which is a necessary cost of hiring.
Use your "block" button early and often. At least that's what I do on Reddit at the slightest sign of pedantry or personal attacks. Those people really are beneath you, they live in a bubble of their own reality, and it's best to not respond, just block.
At times I wonder whether this stuff is a sort of “I’ve been bullied by this tactic in the past so I’m going to bully with this tactic now” psychological pathology.

Society’s suffering from a lot of whataboutism, gaslighting, and authoritarianism. The internets made it so there aren’t many real islands of information isolation. Iran comes to mind. (Granted, North Korea kind of doesn’t.) The propaganda is clearly crossing national borders more easily than ever before now.

So maybe we can’t have nice forums because the internet is a reflection of the world.

> At times I wonder whether this stuff is a sort of “I’ve been bullied by this tactic in the past so I’m going to bully with this tactic now” psychological pathology.

There's a flip side to that, though. I've been guilty of XY problems in the past, and the experience of having them pointed out was illuminating; I learned both why not to do something and what the better way of accomplishing the goal was. That new knowledge is often widely applicable.

It seems a bit goofy to go to a site, ask for help, and reject the assistance offered out of hand just because someone suggests you might be on the wrong track.

All of these are essentially forms of bike shedding, and it's become a massive problem here on HN too. I don't know if they do this because of the classic bike shed where they don't have any actual insight but feel like they need to say something for some reason, or something else. I don't know of any place that isn't so dominated by that anymore.
I'm not sure you're going to find anything better. These websites are already pretty incredible at getting fairly competent people to write technical content for free. If you want something even better, still for free, that's a big ask.

Are you asking this question because you want to contribute hundreds of answers for free to this better forum?

Is there any paid solution to this?
Sure; hire an expert, or get a support contract in the tool you're working with.
That's not exactly akin to technical forums or any generalistic Q&A solution, is it? I guess a better question would be: how much money would experts need before they share bits of their expertise in a broader platform while also following guidelines such as non-pedantry, stay on topic, etc.
It's about time. How long does it take to ascertain someone's level of knowledge, the specific aspects of the problem they need solved, etc. What if their approach seems fundamentally flawed, and while it could be jury-rigged, they'd really be better off coming at it from a different angle? How many hours are they willing to be billed for?
> I guess a better question would be: how much money would experts need before they share bits of their expertise in a broader platform while also following guidelines such as non-pedantry, stay on topic, etc.

However much it would take to distract them from their day jobs where they're adding tons of value.

That's my question as well, I just want to pay for it to get actual good quality. Had enough of freemium crap.
Stack overflow isn't freemium. It's actually free. As you don't pay the volunteers they don't put up with your questions being bad.