@LeonConrad asked about Research Portal Recommendations; several of us voted to close because we believe this is contrary to scope in the help center which discourages requests for references. @Kobunite suggested we discuss this in meta, and I agree.

I believe that requests for references are a bad fit for H:SE because - They are ephemeral - the best portal today may not be the best portal tomorrow. A new portal could come online, and what had been the best portal could be taken over by a cabal of Whig historians who flood the site with links that demonstrate that King John II was actually an immortal alien faery who fights crime.

  • They are subjective. "best" is subjectve in any case. Samuel Russell & I will probably not select the same portal as the best research portal simply because we come at any given question from radically different perspectives. (and just to be crystal clear, I pick on Mr. Russell because I respect his scholarship, but the things that fascinate him about history are not things that I find thrilling).

  • It is nearly impossible to select a "best answer" to a request for recommendations. SE is built on a "best answer model".

Any of these objections could be overcome. This question asks:

  1. Should we support requests for research portals?
  2. If we support the request, how should we handle it?
  • IF we choose not to support research portals, I wonder if there is a way to revise the question to focus on methodology/discovery of new sources. Or is that just a thin veil over the shopping problem?
    – MCW Mod
    Mar 7, 2014 at 13:17
  • I'm happy (and flattered) to be used as an example of "good but different" :). Mar 16, 2014 at 21:14

2 Answers 2


Allow me to put forward my views, as the person who put forward the original question.

I see a distinction between subject-specific research sources and research portals.

Here are a couple of examples of questions I consider to be about research sources: this (closed) thread about sources on the origins of communism, and this (accepted) question about on-line sources for maps. The latter was a popular question, by the way.

I think of a research portal as something like Google Search that Can lead me to research sources. Research portals could include:

Google Scholar
History Today magazine

I'm keen to widen the net and include more globally focused Why not draw on the collective experience of users on this SE site?

  • 1
    The problem with the general category of questions is that more often than not they attract more extremely low quality answers (including outright spam) than we can (or care to) cope with. Furthermore, most recommendation / "gimme a list" questions tend to be extremely low quality themselves, with their most common problem being a profound lack of prior research / effort.
    – yannis
    Mar 7, 2014 at 10:47
  • 1
    Nothing inherently wrong with your question, and it's certainly on the high end of the spectrum quality wise. However, if we are going to widen our scope to include recommendation questions, I think we should first be fairly certain that we can handle the low end ones. And having seen the general category fail again and again on several SE sites, I can't say I feel confident we'll ever be able to sufficiently moderate them (within the confines of the SE platform).
    – yannis
    Mar 7, 2014 at 10:49
  • 1
    @YannisRizos could you please point to or describe some of the questions for reference material that did not work out on other sites?
    – Drux
    Mar 8, 2014 at 19:30
  • @Drux As a quick example, Stack Overflow & Programmers (and I suspect other sites as well) have custom close reasons just for this general type of questions (you don't get a custom close reason unless the whole category is a serious burden on the site). There's also the canonical "real questions have answers, not items or ideas or opinions" blog post.
    – yannis
    Mar 10, 2014 at 10:34
  • 1
    (cont...) On the flip side, Math.SE is doing just fine with allowing them, and we have a brand new site (SR.SE) that's all about recommendations. I wouldn't mind if we gave them a try on History as well, and decide for our own if they work for us. I'm just not very confident we'll succeed.
    – yannis
    Mar 10, 2014 at 10:34
  • 1
    @YannisRizos I wouldn't welcome just any links to conspiracy web sites, or similar. I'm thinking of references that identify published books or otherwise web sites where it's clear who is behind (e.g. bbc.co.uk/programmes/p01dh5yg). And whoever posts links should add short summaries (e.g. similar to this format economist.com/news/books-and-arts/…) to put a little bit of her reputation (whatever that may be worth) at stake. Haven't witnessed the other attempts, so I may be naive.
    – Drux
    Mar 10, 2014 at 10:45
  • 2
    If you have a clear idea on how these questions could work, you should post an answer here @Drux. Sometimes, all it takes to convince the community is a good Meta answer...
    – yannis
    Mar 10, 2014 at 10:52

@Yannis Rizos brings up some very valid point regarding the quality of such questions if they were allowed on H:SE.

Over the year or so that I have been using this SE the majority of the reference request type questions have been of poor quality. However, there have been some very well written, and well reasoned reference request type questions that have been closed and deleted (such as the one that prompted this meta post) that I have abstained from voting to close.

Yes, they are reference requests and that is out of the scope of this SE. However, when the question is well written and well reasoned I feel that it could be in scope, and is a valid and useful question for someone to ask.

I think that reference requests could possibly be effectively policed using the existing "low quality", "opinion based" and "Off-Topic" close reasons.

The two examples posted by @Leon Conrad illustrate this nicely. The first question was closed as it is obviously off topic as currently phrased (as explained in comments by Mark C. Wallace). The question is flawed in a couple of ways. It's of limited use to other people, it's too broad (as there isn't a single book that would answer the question) and it's poorly formatted.

On the other side, the question regarding historical maps is a question for which answers will be valid of a wider range of people. As stated in a comment on that question:

While technically off topic, the answers to this question provide a valuable resource that is not readily available elsewhere on the web, accomplishing an explicitly defined goal of the site: to improve that corner of the web dealing with matters of historical interest. I suggest leaving it open, but converting all the answers to community wiki.

Pieter Geerkens

With this in mind, we could allow reference request questions that are of a high quality (as all our questions would ideally be), provide a valuable resource and are well defined and well targeted (E.g. Books that briefly explain the background to communism in Russia = poorly targeted, broad subject / Online sources for medieval maps = well targeted, well defined subject matter) and use existing close criteria to manage them.

  • 1
    The problem here is who gets to decide if a question (reference or not) is "high quality". I'm afraid if we use (our subjective opinion of) a question's quality as an indicator of its topicality, all we'll end up with will be tons (and tons) of drama. That said, well defined and well targeted are easier to define, and could work.
    – yannis
    Mar 18, 2014 at 14:09

You must log in to answer this question.

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