Several questions ( Tracing Elizabeth Watson of Winchelsea, Sussex after 1815?, How do I verify 2 William Watsons are the same person?, What convict records for Australia exist?, Where do I find records for artist competitions from the 1880s?, and What does "Effects under" refer to on proving a will?) have been posted by one member in a relatively short time. Some have family names in common, others appear independent.

Begin work on any one and you will quickly take a turn around the labyrinth and encounter one (and then more) of the others. My interpretation is that each questions requires, and provide context for, each of the others.

I stopped posting responses when it became clear that my frustration was beginning to show through what I was writing. I do not want to be unfair to the questioner who has obviously tried to present what appear to be manageable chunks.

What advice do (or should) we offer on how to deal with this situation? How ought a question provide the whole picture of the background and then define a single question within it?

  • When it is a 50/50 decision, I'd say divide it up. We are well short of the 15 Questions per day needed to be considered a healthy beta. – lkessler Jan 5 '13 at 5:04
  • 1
    It will not benefit anyone if there were 50 questions per day if that level is achieved by atomising the situation to the extent that essential background is not provided. The net effect will be a spike in unanswered questions. – Fortiter Jan 5 '13 at 5:56
  • I feel I may be responsible. Peter is a relative who has been investigating one particular family. He has been in contact with his cousins and they have agreed on a list of unknowns/unverified. This list has been ongoing (and growing) since about 1999, from what I understand. As part of the GE to Beta push, I asked him to commit. Since then, I have been 'hounding' him to ask some questions. The group came back and thought each question was significantly autonomous to be separate questions, so I advised him to put some up as a start. I suggested this after reading some of the comments on meta. – Those Legs Jan 5 '13 at 10:55
  • I think that any or all of Peter's questions are well worth asking. However, each of them needs to be supported by an appropriate statement about what he (and his cousins) already know. SE is about seeking assistance not a contest to see who can dig out what the questioner already has. As Duncan has mentioned elsewhere, the group protocols are not yet clear on what is expected in a question. My question was meant to advance the process of improvement not to apportion blame. – Fortiter Jan 6 '13 at 2:30

This shouldn't be any reason for concern. The OP is simply cutting the project into bite-sized chunks, as you said, which is a smart thing to do. Otherwise, the question would be way too broad and would likely be closed as NARQ. However, often simply a link to another related question will provide a bigger picture of the situation. Essential information should always be in the question, but contextual information in another linked question is fine. There is absolutely no problem with referring to another question for context as long as all information essential to the question is included (repeated if necessary) in the question.

If you believe some information from another question is important and should be included in the question, feel free to ask the OP to do so. Ultimately, how much contextual information included is up to the OP. For example, How can I locate a New England telegraph man ca1850? (Lemuel Keyser Preston seems to be hiding.) has almost four screens of information. On the other hand, When does an alternate spelling due to spelling reform warrant preservation? simply includes the essential parts of the question, but includes a link to another question with additional context, which is not entirely necessary to the question. Just remember, contextual information in another question can either be linked to or included in the question. It's entirely up to personal preference as long as all information essential to the question is in the question itself.

  • And if "all information essential to the question" is NOT there, then ... – Fortiter Jan 5 '13 at 5:53

I think this is a tough question to answer in general. I know the idea is to keep questions "precise" and answerable. However, in doing this, it sounds like the responses to this posters' topics are then overlapping in their content.

Questions can be general or specific, by which I mean that the poster may be looking for a specific answer (as in a lookup request) or a general approach/method/source to follow up on. Splitting-up a specific question is certainly going to risk this type of overlap.

I have been accused on several occasions of overloading less-specific question so that they cover a very broad topic. My response in those cases was that breaking them down would not actually answer the question/problem I have. Sometimes, trying to reduce a problem to those bite-sized questions that the site requires will risk losing the overall context (e.g. the dependencies between those finer points).

My personal belief is that a fixed policy on question structure will not cope with all circumstances, and each must be considered on its own merits. I'm sure there will be disagreements there, and it may not fit the SE policy, but I see no alternative.

Rather than the guidelines stating what style of questions are acceptable or unacceptable, I would like to see more general advice on framing the question that the poster has. For instance, thinking about the type of answer they would like to see, and the types of answers likely to be posted in response.


On one hand we don't want questions that are too general.

So then we complain when the questions get divided up so they're specific.

Go figure.

You must log in to answer this question.

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