Post History

**Make Codidact a helpful place.**

The most basic reason why people use a QA site is because they want help. Specifically, they have a question and want it answered quickly and usefully.

"Quickly" can be very dependent on the exact case, but take a classic example: You are trying to write some code, it's erroring out, you don't understand how to fix it. If someone responds to this 3 months later, it's no good - by then you have likely either found a solution yourself, or given up on that problem. As it pertains to us, it is important to have a community where it looks like a question is likely to get a response in a reasonable time frame. Currently, this site is actually doing a very impressive job in providing answers in reasonable time given its small size. But the volume of questions is low and makes you think "nobody is even asking questions here, so surely nobody will bother answering".

"Usefully" means that it's no good if the answer is itself technically correct, but does not address the question, and it's no good if it technically addresses the question but does not equip the asker to solve their problem. There's an old [joke about a man in a hot air balloon](https://startsat60.com/media/lifestyle/jokes/daily-joke-man-hot-air-balloon-engineer-management), sometimes used to poke fun at certain professions (whether warranted or not), but which happens to illustrate this very well. The moment you realize someone is going to give you unhelpful but technically correct answers, there's a big reason to avoid asking them or engaging at all even if they certainly have the knowledge you need. When you ask for help in the form of questions, a basic concern is the motivation of your interlocutors - do they want to *help you*, or do they want to just *answer the question* (whether it helps you or not).

It's not hard to see that the first kind of person is where you go when you need help. The second kind is for when you're bored and want some edutainment.

The first kind is a very broad audience. Lots of people need help. The second kind less so. For sure, lots of people are bored, but there's many other ways to amuse yourself, and even if you want edutainment specifically there are many competing options. QA sites are at a disadvantage for edutainment - supporting proper QA requires more overhead from devs and mods, meanwhile a more consumerist model like a Youtube channel or podcast can provide more with less effort.

Communities based on *helpfulness* acquire new users a bit like this: Someone asks their first question, gets helped, and learns that this is a place where they can expect to get help. After some time, they realize they've gotten a lot of help, so they start feeling gratitude and a desire to give back, and start helping others.

Communities based on *entertainment* acquire new users by providing a steady stream of interesting and entertaining content. There is not a sense of wanting to give back so much as wanting to play along with the regulars because you believe yourself to be in good company.

**Concretely,** I would translate this into:

1. **Less moderation early on.** Seeing a few good questions is enough to convince that knowledgeable people exist on the site and could theoretically help you. Seeing bad questions does not materially prevent you from asking a good question and getting help. In fact, seeing flawed questions is encouraging and welcoming to new users, because "well if even that guy isn't getting yelled at for his bad question, I should be okay with my good question". At this point, everyone has heard of StackExchange, and everyone has PTSD for how unhelpful and unwelcoming that place can be - a lot of people are now thinking "oh great, another passive aggressive waste of time" unless given evidence to the contrary. Where necessary, moderation should focus greatly on positive and constructive feedback that invites the user to do it, rather than demanding (by threatening to close, suspend, ban, etc).
2. **Commitment to not significantly ramp up moderation later on.** StackOverflow grew explosively by following what I pointed out above. It was a helpful place without overmoderation. The overmoderation crept in over time and took over helpfulness (ironically leading to some of the top questions being closed and deleted), which is why people now complain it's an unhelpful site. Just like the overmoderation PTSD, the moderation bait-and-switch PTSD is quite common and most people on the internet have experienced it at this point, so that the *a priori* assumption is that moderation will get worse later on, which sabotages efforts to execute point 1.
3. **Focus on helpfulness rather than refinement.** There's a lot of discussion here about how to squeeze the funnel really tight and prevent hapless newbies from posting content that offends the sensibilities of regulars and mods. Excluding such newbies is considered an adequate solution - "dance to our tune or get out!". I don't see enough (hardly any actually) discussion of how to actually help these people that happen to do something wrong when asking for the help - "I'd like to help you, but it would be easier if you danced to our tune". There are now even people wanting shadowbans - on a site with dozens of users. You can get away with that kind of abuse (in violation of the point 2) when you have millions of users dying to use your platform. But here, there's very little coming into the funnel in the first place, so when you start "squeezing" it, it's no wonder there's very little left. Moreover, points 1 and 2 also support this, because overmoderation leads to some regulars learning exactly where the line is so they can be passive aggressive towards new users with apparent impunity.
4. **More genuine questions.** There should be more questions that sound like they're actually someone asking for help, and not a "just-so" opener for a self-answer. The goal here is to create a safe environment for people who want help - most people are wise to [begging the question](https://en.wikipedia.org/wiki/Begging_the_question) even if they don't know the term.
5. **New users are first class citizens.** There's a often discussion here about *what new users would want*. However, it seems rare to consider the actual feedback of these new users. Instead it is handled with conjecture, and often what new users want becomes what regular users think it would be swell if new users could want. The culture of thinking that anyone asking for help is socially subordinate to the person giving the help needs to go. The idea that expertise earns you the right to be condescending absolutely needs to go. The concept of "hazing newbies" needs to go. Just because someone is a noob at one thing, doesn't mean they're not able to help at other things. If you see people asking for help being treated unkindly, it will make you ask less (see 4). And if you do stick with it and become a helper yourself, you will likely perpetuate the cycle of abuse.