2
$\begingroup$

@Terdon has requested a Q&A section on a rolling thread on an investigation about gene duplication. What is exactly meant by this and how to go about doing it?

$\endgroup$
  • $\begingroup$ By the way, I realize this is asking you to do a lot of work and I entirely understand if you don't want to spend even more time on this. I just can't think of any other way of keeping the really useful content you created :/ $\endgroup$ – terdon Sep 12 at 13:57
2
$\begingroup$

Thank you for posting this, I was about to bring it to meta myself. First, a quick summary of what happened. @Michael G answered this question but in a way that didn't give a definitive answer. Instead, Michael did what most of us are very accustomed to doing, both online and in real life: he gave the OP very useful pointers based on his extensive expertise, and asked some very relevant questions that would help him provide an answer.

The problem is that these sites, the SE network, don't do discussion well. Not only because it is against the "rules" and not only because we try to have a high signal to noise ratio, with one question and everything else being answers to that question. Another issue is that, unlike traditional online forums, the SE site mechanics don't guarantee the order of the answers. So when you have two users having a back and forth discussion, referring to "the post above" or "the post below", the post they are referring to will move depending on how the person visiting the page is sorting the answers.

So, coming back to the specific example, when I saw that this had happened I converted one of Michael's answers to a comment, I incorporated the OP's answer to his answer into the original question and deleted the rest since they had, by now, moved on to a new Q&A.

However, this new Q&A was also treated as a discussion, resulting in a bit of a mess. It now has 5 "answers", but all of them are part of a back and forth discussion between Michael and the OP. The kind of thing we're all used to doing in forums, but the sort of thing that's really not a good fit for this site. Finally, they moved on to a third Q&A, which you can find here: How to make a robust amino acid phylogeny with adequate robustness support.

As a moderator, I'm in a bit of a fix now. On either of the other two sites I moderate, I would simply close these questions as too broad or delete them outright as not following the rules of the site. However, those other sites are mature, with a lot of content and this would be a drop in the ocean. We're a much newer and sparser site, and I am very reluctant to delete outright. Not only do we need content, this is really excellent content, dammit! It's just in the wrong format. Michael spent a considerable amount of time providing very valuable help, and the OP was responsive and came back with all the detail Michael asked for. This may be a bit of a "trainwreck" (by SE standards) at the moment, but I think we could actually make a truly useful Q&A out of it.

That's why I suggested that Michael and/or the OP could maybe collect all the steps into one question. Ideally, the OP would post a new question, but this time, they would include all the detail that came out of this long discussion in the question itself. That way, Michael could distill the long conversation into one comprehensive answer to the now very comprehensive question.

I am very open to other ideas though. This was just the best way I could think of to both preserve the Q&A model and preserve what seems like a very useful exchange of knowledge that could help future readers. In their current state, the threads are unlikely to help others since it's very hard to keep track of what was said in answer to what.


How this should have been handled according to the SE model

With this sort of situation, the best solution is to first try and get the extra detail needed in the comments, but when that's not enough, we have the following options:

  • Invite the OP into a chat room (one will be created automatically if you start having a back and forth in the comments). Sort out the details there, get the extra information, and then come back and edit it into the question so the question is now in a form that can be directly answered.

  • Ask the OP to break the question into smaller parts. Ask the first step in one question, then everything else in follow-up questions.


Quick links for anyone trying to understand:

$\endgroup$
  • 1
    $\begingroup$ The chat room should have been first port of call, I'm not 100% how to do this but I'll figure it out. Summarizing everything into one question will be tough but maybe doable. Let me see $\endgroup$ – Michael G. Sep 12 at 14:08
  • 1
    $\begingroup$ @MichaelG. you can create a room by visiting chat.stackexchange.com and clicking on the "create new room" button at the bottom of the page. However, new users won't be able to chat since you need to have at least 20 reputation (across all SE sites) to post in chat. In such cases, either ask one of the mods to explicitly grant the user access (any SE mod can do this) or start in the comments and wait until the system asks you if you want to create a room (happens after a few messages between the same 2 users). That room will always be accessible to the OP. $\endgroup$ – terdon Sep 12 at 14:11
  • $\begingroup$ Thanks @terdon, but I did some time ago for "phylogenetics" and someone deleted it. Next time the OP is around we'll chat. The easiest thing for now is I will review and edit the questions to be complaint to SE format and feedback here. $\endgroup$ – Michael G. Sep 26 at 14:20
  • $\begingroup$ Rooms with no activity are deleted after a while (several weeks or even a few months). That might be what happened. $\endgroup$ – terdon Sep 26 at 14:29

You must log in to answer this question.

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