5

We've been struggling internally the last couple weeks, I think Casebash puts it together well:

I actually think the decision to make this a research based site (like MathOverflow), as opposed to a more popularist one like StackOverflow may have been a mistake. For a research based site to succeed, it is important to start with a core group of researchers, which recent meta posts seem to indicate that we don't have. With a more popularist site, you still need experts, but they can be workers in related fields who studied psychology in college, rather than researchers

This Answer made me realize this is important enough to get it's own post. We are scaring off users which have backgrounds in psychology. Period. Even I myself have found the site less pleasant to use after the first couple weeks.

I'm not saying a research-level site is bad, I'd love for us to have one of those two. But I don't think we can start with a research-level site from the ground up. We can built a "populist", broad interest level site from the ground up. If our "populist" site is successful, we'll have an engaged community from which to attract more PhD, research level users. Doing that backwards is not working.

While I agree with many of the closures that have happened, I think we need to decide which kind of site we need to be, and we need to seriously be able to focus on attracting the appropriate audience. We can maintain SE-quality content, but we need to decide on a sane initial research requirement and tone for the site.

I think we've been courting the "researcher only" crowd for far too long and we have nothing to back that up. Anyone suggesting the site should continue in this direction needs some major back up at this point in my opinion.

Which kind of site do you want Cog Sci to be? Back it up. I want to hear your plan for how the site can survive as a quality, useful resource for a wide variety of people. How can we attract the user base you feel we need?

3

Extended comment:

I agree with the sentiment expressed by Catharsis that the site is overmoderated (e.g., long comment threads about clarification, and mod-edits instead of user-edits), though I do think most of the questions closed are completely justifiable.

I'm not in favor of a "research-only" site, but a big problem I see with this site is that it's fairly easy for anyone to come up with a question without ever having taken a cogsci class before-- which leads to lots of underinformed and poorly research questions.

Imagine if 25% of the questions on Stackoverflow were written by people who have never programmed before in their life. It would be a complete mess. On occasions, that's what I feel like this site has looked like. There's a significant amount of questions by people who have no knowledge of the field. This leads to extremely vague questions, in which it's a) difficult to understand what's being asked or b) needs an entire book to answer.

Whether we want to close questions for being "too basic", or edit questions for being too "vague", we run into the same problem. I guess what I'm saying is: changing the target audience really doesn't fix the problem.

  • 2
    "Imagine if 25% of the questions on Stackoverflow were written by people who have never programmed before in their life." ... I do often get that feeling on SO. – Steven Jeuris Apr 2 '12 at 8:59
  • 1
    "I agree with the sentiment expressed by Catharsis that the site is overmoderated" Isn't that a bit contradictory? You find the site is overmoderated, but you agree with presence of the low-quality questions. Do you feel we should just close them as quick as we do now, but refrain from starting a long comment thread helping the user out? – Steven Jeuris Apr 2 '12 at 9:01
  • @Steven you're right, that was a bit contradictory. I'm mostly just against the long comment threads-- though I'm frequently guilty of them myself. Instead of dragging out the "clarification" process, we may be better either a) closing or b) taking a stab in the dark and answering based on our assumptions of what OP meant. – Jeff Apr 2 '12 at 12:53
  • 1
    I am for closing over taking stabs in the dark at poorly made questions. I am neutral on the length of comment threads. – Artem Kaznatcheev Apr 2 '12 at 23:50
  • @ArtemKaznatcheev I agree with you on that point. I know of no SE site where poorly phrased questions are allowed. The general approach is to close, and optionally give guidance. – Steven Jeuris Apr 3 '12 at 14:04
  • I don't find "never programmed" an apt comparison; everyone uses and understands a degree of cognition, and you can't not apply it's principles. No attempt at research and no attempt at gaining a background of information are a problem, but I don't agree there's some job title you need to have or action you need to have performed to contribute to this site. – Ben Brocka Apr 3 '12 at 14:04
  • Also, I'm...really not sure what your post is actually trying to say. You say the problem is the audience and then you say the problem can't be fixed by changing the audience. You say we're over moderated and you indicate we need more moderation. – Ben Brocka Apr 3 '12 at 14:06
  • I'd also like to chime in here. I am one of the mods and I do edit questions. However I am trying to do what I think is in the best intrest of the community! If you feel my edits (or any mod's edits) are not in the best intrest of the community, please start a new Meta question about this. If we're not doing a good job, please help us do better. We're your mods, we work for you! – Josh Apr 3 '12 at 15:45
  • 1
    @BenBrocka I completely disagree with your comments on "never programmed". Having a brain and thus experience with cognition from just thinking is the same as "using something that was programmed (i.e. a computer)" would be to SO. Having actually studied (in a formal setting like university, in a professional setting, or on your own; much like how you can learn to program in many different settings) cognition would be equivalent to having learnt how to program. Thus, I think your disanalogy fails and I completely agree with Jeff's analogy. – Artem Kaznatcheev Apr 3 '12 at 18:06
  • 1
    @ArtemKaznatcheev unlike programming there are many important questions out there that can be asked by normal people, like this question. It's completely different from programming because there is no broad, vague understanding of how programming works in most people. Besides, what is your intended course of action even if the analogy is valid? If someone asks a bad question, it doesn't matter whether they've programmed or not, why does it matter whether you "know programming" if you ask a good question? – Ben Brocka Apr 3 '12 at 18:22
1

This more an extended comment and answers a more selfish question: "what kind of site do I want?". The first part explains my background with cstheory, the second part talks about what I want from cogsci, and the third part talks about how I can see us getting there. Everything here is completely from my point of view, and not that of the community.


I want this site to be a community. My favorite SE site, and personal gold-standard for quality is cstheory.SE. The reason I love that site is because it has a powerful community of experts, this allows:

Asking questions: If I ask a question there, then I know the answer will be authoritative, because I trust the person answering it. They can include absolutely no sources, and say things like "I expect this to happen" and I will value that answer above many "here is a list and summary of sources" answers I see at many other sites. Why? It is because I trust that the community knows what they are talking about.

Keeping up with the site: I usually keep up closely with the site, because I love reading the questions. Even unanswered questions often teach me something new. Further, the character of the community, lets me know that the questions being asked are one that are of interest to professionals. Thus I can get a feel for the field by reading questions without even looking at answers.


I want cogsci.SE to be the same sort of community for researchers (and professionals) in cognitive science, neuroscience, and psychology. I want to be able to read the site and get a feel of the sort of questions professionals care about. I don't want to come to a site where I learn what sort of questions somebody-who-does-not-interact-with-cognitive-sciencel cares about.

When I read an answer on cogsci, I want to trust it because I know that the person answering KNOWS the material. They didn't just do the obvious google scholar searchers (I can do them, too) and return a summary of the first few articles that popped up. I already know how to google, I want to read the views, experiences, and hunches of experts.


How can we get there? I have only hunches. My biggest hunch is that we need to focus on the community. I have no interest in reaching out to people that will ask a few obvious questions. I also have no qualms with closing poorly asked questions. Sure, sometimes a person that asks a bad question will learn from their interaction with the site and start asking good questions/providing good answers, but this seldom happens.

I want to focus on encouraging users who ask good questions, and provide good answers. I want to make them feel welcome, and provide them with a community and interesting content. The few experts (mostly grad students) that we have attracted already, have many colleagues that can be slowly lured to the site if they see a good community.

Thus, I am more than happy to build the site slowly, and focus on a forming a core user base of experts. If this makes the site frightening or unwelcoming to some of the lay audience, I am okay with that sacrifice.

It would be nice to have the best of both worlds, but if we can't then I would like to focus on the experts.

  • 2
    My problem is you've openly admitted to not asking questions on this site because you feel the audience is sub par. If you honestly have constructive questions that you aren't asking, you're actively harming the site and specifically harming your own vision for the site. The best way to shape the audience of the site is to create content that appeals to the audience you want. – Ben Brocka Apr 3 '12 at 1:11
  • 2
    @BenBrocka that isn't what I admitted. I said that I am less inclined to spend time on formulating my questions when I don't see many well formulated questions. I've already asked 10 questions related to my previous or ongoing work, and have received very good answers on 2 of them that led me to read and learn a lot of new things. I do have other ideas for questions on my "to do list", but unfortunately they sit behind all my other duties. If I am harming the site by not taking the time out of my schedule to ask these questions, then it is definitely not actively but passively. – Artem Kaznatcheev Apr 3 '12 at 1:29
  • @BenBrocka As for what I have been doing instead of asking questions: I have been editing, voting, and recruiting new users of the research or near-research level (although I have only managed to get 3 of them to participate with questions/answers and not on a regular basis). – Artem Kaznatcheev Apr 3 '12 at 1:39
  • Two comments: " I have no interest in reaching out to people that will ask a few obvious questions." You don't have to answer those questions, and if you do, I advice you to keep the answer short. "I also have no qualms with closing poorly asked questions.", agreed. I feel we are doing a good job at closing questions. Professional site or not, vague questions are always bad. – Steven Jeuris Apr 3 '12 at 14:13
  • 2
    About your "asking questions" bit about CSTheory; that's not how SE works. I've brought this up in chat; it's just not how SE works. We build up a community of experts by judging others on their activity, so the most helpful and expert users rise to the top. This is not a format where only "authoritative" users are allowed accounts and it never will be. To expect only expert users and no one else is naive and is not how the SE model operates, nor is it useful to the internet/world at large. If you want a fellowship, get one. – Ben Brocka Apr 3 '12 at 14:37
  • @BenBrocka I never suggested banning people from making accounts, I also stressed the COMMUNITY aspect of cstheory. I don't trust the users based on their external criteria (although it is nice to see an explanation of Shor's algorithm by Peter Shor), since most of them are just graduate students, and there are plenty of bad graduate students. Instead, by focusing on research the community made it extremely difficult for someone who is not a good researcher to "bubble up to the top". Thus, I can trust that anyone that did bubble up to the top provides authoritative answers. – Artem Kaznatcheev Apr 3 '12 at 18:19
  • (continued) if there are a lot of easy question around, then the rep process of SE stops measuring someone's expertise (beyond a basic threshold) and starts measuring how much time they can invest in the site. As for your fellowship comment: if it is meant as an insult, then I don't appreciate it; if it is meant as something else, then I don't understand it. – Artem Kaznatcheev Apr 3 '12 at 18:22
  • @ArtemKaznatcheev "SE stops measuring someone's expertise" SE never was good at measuring someone's expertise. – Steven Jeuris Apr 4 '12 at 9:10
  • @StevenJeuris I object to that, it's actually pretty damn good at weeding out people that don't know what they're doing, you do have to know what you're looking at though. It's the worst system for determining expertise except for all those other ones. – Ben Brocka Apr 7 '12 at 19:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .