Also relevant:
https://www.quora.com/How-many-questions-have-been-asked-on-Quora-1
Try moderating that.
Lol and I imagine the id would include deleted questions, such as those used to get the question count… So it’s not even accurate
That’s just it. The con isn’t just fewer questions, it’s fewer people. Everyone starts out as a n00b and learns along the way… unless the system prevents them from doing so. Then you don’t end up getting the high-level community members asking the high-quality questions at all.
I can honestly say that I did not join for that reason. My first answer is 3 months before my first question, and I wrote about 30-40 answers before posting my first question.
Wow, really? That’s actually kind of awesome.
However, I’m aware that I’m the exception in that regard.
That’s good to be aware of.
Yes, I do see the potential problems. But just to compare, I’m discussing a lot on another forum that has THREE days before you can make your first post, and that actually works well there. Granted, it’s not a Q/A, it’s a discussion forum.
Right, and that’s a vital difference. The bulk of newcomers ask a question because they need an answer, often in a professional context, which means they need an answer now, not 3 days from now (or even 1).
One drawback on having it so easy to ask as it is on SO is that some punishments they have is virtually completely pointless. For instance, if you get too many downvotes, you are prevented from asking questions until you have improved your reputation. I don’t remember the exact details on how it works, but my argument boils down to that if that happens you can simply just create a new account.
I’m not sure that’s completely pointless, though. Mods have tools to detect and deal with problematic users who circumvent the rules. They shouldn’t be too difficult to set up for our site.
And I can admit that some of my comments are influenced by my frustration of SO. Thank you for challenging my thoughts.
Sure! I’m frustrated with them too. Have been, to one degree or another, for about 8 years now. But my frustration has largely been with them being too hostile to newcomers rather than being too welcoming to them. I’ve been around various forums and communities for a looooong time now, and IME that’s always been the surest way to kill a community: make it hostile to “new blood.” Sooner or later, people get the message and new users stop coming in. Meanwhile, RL continues happening, and people end up leaving for one reason or another, and in a few years you have a ghost town.
We shouldn’t setup more barriers to posting, we’ll just be seen as the exclusive StackExchange site.
Take a look at the smaller sites on SE, noob questions are a lot more welcome and it’d be a graveyard without them.
That’s a drawback yes. But more people is not necessarily good either. Especially if the quality of the people (forgive me for how that sounds) is low. You do have valid points, but I’m not convinced that they necessarily leads to the things your claiming.
Thank you.
While I’m aware that I’m not in the majority, I don’t think that I’m so rare that my kind is almost extinct.
True but some (the exact ratio is extremely hard to estimate) will reason like this: “Oh, I cannot ask a question now? Very well, but since I have seen this site coming up at google searches and because I can see that other people have received answers I suppose I save my account and use it for the next problem I have.”
Also, you’re saying that the bulk will do like you say. That’s true, but the bulk is not everybody. How many are they? Maybe 10%? Maybe 1%? Even if it’s that’s low, it can still be enough. Let’s say that one million people gets the message that they need to wait 24 hours. Even if that prevents 99% from asking a question, it’s still 10k questions.
Ok, almost pointless then. Besides, those tools can only catch really persistent help vampires or trolls. They are virtually completely useless against people who asks 1-2 questions a year, starting a new account every time. Especially if you want to avoid false positives.
Which is kind of my point. I argue (with not much more than a gut feeling) that the reason for the hostility towards new users ultimately comes from letting them in without preparing them.
In an earlier post, you mentioned a bit that we have to take into account how humans work naturally. While that’s true, it applies to ALL users. The frustration you see from experienced users on Stackoverflow IS a natural reaction.
Also, I’m a strong believer of “a community for everyone is a community for none”.
You definitely have a point there. And I don’t think this would be a good thing in the beginning either. And afterall, in the beginning I don’t think those newcomers will be a problem anyway. But it can be worth considering if this site grows.
I suspect that number is very low. Personally, I only “join” a site for 3 reasons:
- I need something now that I can’t get without joining and that I actually need:
- Ask a question, need to register first, I’ll register.
- Read an answer and see a link to related stuff that looks interesting (but is not why I actually found the site in the first place), I won’t bother. In fact, almost out of spite, I will go back to Google and search again for whatever I saw that I thought was interesting. This is a key reason why I have never joined Quora. (OK, poor quality answers to many questions is another reason.)
- I want to actually participate in a project - that can be specific like Codidact (forum, Discord, etc.) or more general (Github). This would generally be something more than “just to read information” (which is why I treat the registration on Quora and similar sites as almost as bad as a paywall- don’t get yourself indexed on Google and then tell me I can’t actually read the stuff without registering, that’s evil)
- I have a personal connection (e.g., someone I know IRL has asked me to do something, whether personal, volunteer, friends/family, work-related).
“This site has a lot of information and if I join then I’ll be able to ask a question or upvote or comment” doesn’t work for me, until I am actually ready to ask a question or write a comment. And just to upvote, not going to happen. The tiniest risk of “what if they spam my email” is enough of a reason not to bother to register for something as insignificant that.
I joined SO to answer a question, and was answering for about 3 years before I asked my first Question. I’d probably been using it as an anonymous user read-only resource for a long while before that though, just to benefit from questions that others had asked.
Even so, I found the process a little alien from the “normal” forum etiquette that I was used to. I think that’s where the “unfriendly” reputation would align with my experience of SO. I had expectations of how an online forum worked and I had brought those with me to something that, at first glance, looked like a forum, but wasn’t.
The guidance given to users, at the time, wasn’t very rich unless you specifically went looking for it… which obviously many people don’t do.
UI and UX would play a large part in guiding the new user (of any level) through the process of asking a good question, and if it can’t lead them to ask a good question, it might at least prepare them for the experience of the feedback they’ll get for asking a bad one. A part of being able to endure the slings and arrows is knowing that they are coming, and that they’re aimed at the “enduring Question”, rather than at you.
Having some well-worded pro-forma responses was sorely missing at the time I was processing review queues (I think), and many of the commenters that I saw were seemingly incapable of wording responses to address solely the content. However benign the comment is, it’s easier to accept if it doesn’t feel like an indirect personal attack - “Don’t do the bad thing in your question” vs “A question may get a better answer if it does the good thing”.
I also seem to recall having encountered many instances on SO where the ability to give feedback (either through flagging or reviews) did not align with the feedback I was trying to convey. I often felt like I was using “reasons” that didn’t truly match.
Yes. I just checked my questions and answers on SO (the first site I joined) by date, and I had given more than 30 answers before asking my first question.
BTW, after now having seen that I’m obviously not the only one here who did this, does anyone know how to write a SEDE query to determine the percentage of users whose first post was an answer?
Would you consider joining if someone you know told you that “There is this new cool SO-replacement. I’ve tried it, and it’s awesome”? Maybe if you read some article about it?
Also, the system could work in the way that you get the notification that your question will be published in 24h after you have posted it. Maybe a bit evil, and it could certainly annoy some people. But at the same time, then you have registered. I’m not sure about things here. Just spawning ideas at the moment.
But one thing we NEVER should adopt is to require registration just to read. That’s one thing SO did right.
That’s why I often use temporary email. But yes you do have a valid point here. Same as you said before. It does not matter if I can show solutions to the problems user will experience if they don’t use them.
But another thing. We have both made assumptions on the ration of people doing this and that based on how we personally reason and behave. You sounded surprised that I joined SO to answer questions primarily. I assume you’re even more surprised that two more, @celtschk and @klors said the same thing in this very thread. I’m not claiming that there are enough people out there that do this to make it sustainable to rely on them, but it seems like it’s more common than you believe. (And probably less common than I believe)
One thing that I mentioned above in this thread that’s relevant is that it’s about how we sell ourselves - if we sell ourselves as a place for ‘answers to your questions’, then proceed to not answer all the questions (including closing them), we’re being unfriendly because we’re implicitly lying by omission of ‘answers to some of your questions’.
So, I think we need to take a different path - more like ‘we are a community of people helping each other learn and solve problems and questions’. It changes the emphasis and states that we’ll help, not necessarily answer your question. The phrasing could likely be better but it’s a step up from lying by omission anyway.
This. No community can set up the right rules for what they want to be until they’ve agreed on what the want to be.
The software can (and should) support policies that lead to very different sites.
We should also be aware that the good people in these discussion are here with differences in opinion. We can’t plan on launching a lot of different sites, but we may not be able to agree on just one either. We need to discipline the tendency toward fragmentation, but we don’t want to set up a single site that is destined for failure because it doesn’t know what it wants to be either.
We need to be willing to compromise enough to get down to a very small set of initial sites without giving up on having the right sites. Which may mean giving up having the site you are really here for in the initial launch and instead supporting the ones we can launch enough to prove their viability.
Yes, exactly. I think that’s the point that a lot of the people advocating for “high standards” don’t understand: their vision may very well work if this site existed in a vacuum, but it doesn’t. We’re very specifically building it in competition with another site, with the intent of drawing people away from it to ours. This will never work if our barriers to entry are higher than the other site, especially with the other site already being well-established.
Anyone who hasn’t should read Joel’s article “Let Me Go Back”. It’s about backwards compatibility, but the broader point about barriers to entry – and how significant even a single one can be – ought to be required reading for anyone contemplating a project like this.
OK, I now created a SEDE query for the first post type. If I didn’t make a mistake in my SQL, on SO there are far more people whose first post was an answer than people whose first post was a question.
Edit: I’ve now tried on a few other sites, and while the exact relation is of course different, the answers seem to win out everywhere.
Very interesting. That’s not how I guessed it was.
So now that we know the majority of users come to a site first to answer something, what does that mean to us and what do we do about it?
Actually, I suspect it is a bit more complex than “come to a site first to answer something”. My guess is there is quite a mix, some of which is impossible to figure out!
- First activity: Question
- First activity: Answer
- Lurk for a long time and eventually ask a Question (after finding answers to all the easy questions by searching well!)
- Lurk for a long time and eventually post Answers (after finding answers to all of their own questions and decide to help others)
- Ask a Question on one site, find plenty of answers to other questions and eventually through HNQ find other sites of interest and start Answering on those sites where they have more knowledge. (That’s me, if I remember correctly. I started with finding answers for many questions, then eventually joined SO to post a question, then started to look around via HNQ and started getting more involved, Answering but not asking, on DIY & Retrocomputing).
A large part of the problem is that we just don’t know how long people lurk before joining. I suspect (I don’t know if this can be discovered via SEDE or not) that the time between “registration” and “first post of any type” is relatively short. But “time before registration” is impossible to know with any certainty. That is one feature of SO/SE that I absolutely want to keep! I do not want to require registration for read-only access to normal information (to see deleted questions or other more sensitive stuff is a different story, but basic Q&A, Blog, etc. should be totally open to everyone to read).
I joined SE to upvote helpful questions and “star” them, although I had been lurking for about a year before that. During that lurking time, I noticed how well the whole system worked, with my key (somewhat fuzzy) metric being accuracy of accepted answers. I still remember finding SE after trudging through the wastelands of Experts Exchange and Quora, and to me, it felt like an oasis.
I was talking with some other developers last week, and all three of us mentioned that although we had joined SE, we had NEVER posted a question, ever. We each had posted a number of answers and comments out of a sense of duty to “give back” to the community that was likely responsible for about 1/3 of our salary.
SE was not built on “cud u hjelp w/ my homwrk pls?” content.
I think they first come following a link from a search engine and get an answer passively (that is from an existing question, because the site is a repository of knowledge).
Eventually, if they ever become a active user, they either have a question that isn’t already answered or they think they can contribute. If the latter is dominant we have a measure of just how complete the repository is.
Despite the experience of active user the most common experience of a good, functioning Q&A site is just looking up existing content (often without an account or any kind of on-going relationship with the site).
So we need reliable evaluation and working search (even if that is just making it easy for the big search engines to index the material).
Well, one thing is a sort of indirect conclusion if you continue the search for the “root” of users. If we take SO as an example:
It’s safe to assume that nobody gets internet access and their first thought is “OK, where is a place where I can ask/answer questions about programming?”. We can safely assume that people lurk first - until find a question they can answer (or, to a lesser degree, until they want to ask their own question).
From that, I also feel confident in assuming that there’s a big ratio, something like “1 in 100 lurkers transitions to a user eventually”. All the discussions about rep and duplicates and whatnot, they all focus on moderators and users (as in, signed in and asking/answering questions). Lurkers have been mentioned, but they haven’t been given the importance that they are owed.
And in that light, I’d say this gives more weight to the “quality over feelings” side of things.
Lurkers will be happy if they to find an answer to their question.
Even if another person feels bad because their bad question got rejected.
Lurkers will be unhappy if they find 3 bad posts (and won’t even get to the 4th one which is finally good).
Even if another person is happy because their low-quality question got accepted.
We just won’t know without hard numbers, but I can easily see this working out something like this:
High quality: 10 000 lurkers => 100 first time posters => 20 long time users
High tolerance: 2 000 lurkers => 20 first time posters => 10 long time users
I’ve been almost exclusively a lurker on SE for several years. I’ve lurked pretty heavily on Meta in recent months (and on the other sites much less), keeping up with the various issues without chiming in. Here’s my individual perspective:
This lurker gets turned off by some of the attitudes witnessed in comments under some unanswered or unpopular (sometimes poor) questions. It hasn’t stopped me coming back yet, but is contributing to my drift away - and it has definitely discouraged me from taking the leap to ask anything myself.
My usual mode of lurking is to browse Hot Network Questions for interesting-looking topics, which I then read to waste time. Low-quality Qs tend not to appear there. On occasions where I land on a question because I actually want to know the answer, it’s after a Google search - so I usually (within 1 or 2 tries) arrive straight at the relevant question. Sometimes it’s a low-score post but it meets my query nonetheless, and I am happy. I don’t remember ever having to trawl through a lot of junk questions on the SE site, perhaps because I never use SE internal search function.