-2

The perennial concern about low quality questions and low quality answers is resurfacing again. It is my impression (not backed by data) that many VLQ answers are posted by rep 1 unregistered people. This seems to be true also of some low quality questions.

Is there data or could data be obtained to back up or refute this impression?

Could ELU run an experiment (for say, 3 months) requiring new users to register before posting?

This seems like an effort which might: (a) reduce the VLQ posts; (b) be unlikely to scare off people we want to attract; and (c) not require a lot of work to implement and analyze.

Addendum in response to comments so far: Is it likely that ELU will be permitted by SE to require registration, either as a test or permanently? If not, we should move on.

  • This query might help: data.stackexchange.com/english/query/352554/… It guesses whether they're unregistered by how long they hung around and sorts posts by score. To be information and not just data, we'd have to limit the query to a time period and compare it with a similar query for registered users over the same time period. There might be a better way to select registered/unregistered - that was just the first query I came across. – ColleenV Jul 23 '16 at 0:27
  • @ColleenV Thanks for the link to a query. However, I am an ignoramus on the mechanics of writing computer queries. For VLQ Answers, how about counting the As in the VLQ Queue, and the total number of As per unit time (to be defined) before, during and after the experiment. If there was a significant drop in N(VLQA)/N(A) during the experiment and no significant difference in the period before compared to after, we might have done something worthwhile. Have to think about this. There must be a flaw; it sounds too simple. PS almost everything in the VLQQ is VLQ. – ab2 Jul 23 '16 at 1:47
  • I don't have the reputation to access the queue for EL&U, but if I have some time I'll see if I can figure something out using ELL. – ColleenV Jul 23 '16 at 2:15
  • 2
    When I joined SE, I wondered why the hell these guys allow unregistered users to ask, answer and edit. After 10 months, I think I got the answer. All they need is attracting new users. They don't care whether they register or not. The task of sorting them out or chasing them away belongs to us. What an honor!! This issue should be decided by SE and I don't think ELU can perform this experiment alone unless SE helps it and I doubt SE will help. – user140086 Jul 23 '16 at 5:20
  • 2
    @Rathony I agree that SE is unlikely to support us forcing registration, but I think the motivation is more about getting content than users. I've seen a lot of Wikis fail for lack of content, and it seems that some low quality content is the price of fixing that problem. I have to wonder how many registered users would have signed up if the site was less active. – ColleenV Jul 23 '16 at 13:31
  • 1
    @ColleenV We might have some difference in understanding what's happening on ELU. I don't think asking new users to register will contribute that much to solving the LQ post problem as registration process is not that complicated and doesn't take much time. Spammers and one-time users will not hesitate to register when they decide to post here. The real Q is "Are we doing enough to deal with LQ posts? If not, what do we have to do?" Considering ELU is the fifth (sixth) largest site in terms of traffic, would you allow it if you were the CEO of SE? I wouldn't. I will just say, "Deal with it". – user140086 Jul 23 '16 at 15:52
  • Re addendum: "Permitted"? What you propose is a technological change, something the SE developers would have to implement. Even if this change were overwhelmingly popular across all sites (and form the current voting, not really strong in any direction here at ELU), it would have to go onto the end of the "We'd love to have this new feature" queue. That is, it is unlikely to be implemented. – Mitch Jul 24 '16 at 21:47
  • 7
    The fact that I didn't have to register before testing the waters with a sample answer or two (or five) was a significant factor in my beginning to participate at this site. I'm suspicious in general of the often intrusive and aggressively mercantile aspects of registration at many sites, and the fact that Stack Exchange promises not to behave badly with registration information is less impressive to a newcomer than you might think. Stack Exchange sites give you time to see how well the site suits you and how receptive the community is to your contributions before you have to commit anything. – Sven Yargs Jul 25 '16 at 7:08
2

I'm against the idea of making it a requirement to sign up before posting on ELU.

However, I am curious what good it will bring to ELU and I'm okay with conducting a trial run, but for just one month.

2

I don't think the founders of Stack Exchange (SE) were imprudent enough not to think about this issue. The whole purpose of SE is to let any user, regardless of whether they are registered or unregistered, ask and answer a question free of charge. If registration is required, it could block some of potential users, who could contribute a lot in the future, from participating in SE.

Furthermore, SE has various mechanisms such as automatically flagging low-quality posts, spam filters, close-votes, delete-votes and flags, to name a few, necessary to maintain the quality of the SE sites. As long as we use them more actively, we don't have to worry about unregistered users. The key issue is how to deal with unregistered users, not how to block them.

I have just found a related Meta SE question, Why does Stack Exchange allow fake email addresses for registration? and @Shadow Wizard's answer seems to hit the nail.

... This is just an extra step that will make it harder for ordinary people to post, while not making the spammers' life even a bit harder. I'm 100% sure they can use, and are using, real email addresses and can verify them just as any other (real) user.

All in all, just a pointless barrier that should not exist in the first place, and I'm glad it really doesn't exist. (emphasis mine)

You must log in to answer this question.

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