5
$\begingroup$

EDIT: I have now taken a bit of time to update the question, giving more detail into the origin and structure of the data. I hope it's more clear now.

It's important that the solution is generally applicable however, since not all projects have the same setup.


I have had my fair share of issues with some individuals here but I have yet not seen the kind of trigger-happy moderation like the kind shown here.

I am not sure if you guys realize that it does actually take time to formulate a good question, to make it generally relevant for a wider audience. I have mentioned what I have tried, what I think the problem is.. The question is put on hold without a single comment about what needs to be clarified, boilerplate warning and put on-hold.

What is the vision for this QA site really? Are we as a community ok with this?

| |
$\endgroup$
  • 11
    $\begingroup$ I think you have a reason to complain, but please don't characterize five years of moderating more than 100K questions by eight people on the basis of a single experience at one time with one moderator! You can easily learn about our policies, practices, and vision by looking through relevant posts on this meta site. $\endgroup$ – whuber Aug 10 '16 at 13:45
  • 6
    $\begingroup$ I voted to re-open. I'm a high rep user and I've seen some decisions that irk me as well -- it happens. Fortunately the way the site is built almost anything can be reversed. $\endgroup$ – Andy W Aug 10 '16 at 13:45
  • 2
    $\begingroup$ @whuber I didnt mean to generalize, my title was intended on that specific type of moderation, not to mean that all moderators are "trigger happy". As for the tone in the body, I might have failed in masking my frustration, not a good day at the office as days go... Sorry if I came across brash $\endgroup$ – posdef Aug 10 '16 at 14:12
  • 4
    $\begingroup$ We all have days like that. As a general proposition, no matter how you might feel about a situation, you will usually get a faster and more satisfactory response by keeping it impersonal and framing it in terms of issues rather than individuals. That avoids the distraction of defending against personal attacks (whether real or imagined) and helps keep people focused on resolving the situation. In this case you have legitimate questions about what procedure(s) might have been followed and what you might be expected to do in response to the closure of your question. $\endgroup$ – whuber Aug 10 '16 at 14:29
  • 4
    $\begingroup$ I have to agree with whuber on the matter of approach. I certainly encourage our users to raise issues with moderation, but it's a difficult task at the best of times, and it's not hard to come to different views of questions (and even when two different people might come to the same view, it's also easy to miss something or to read something the wrong way and make a choice you may not have made otherwise). As far as I can see, all the moderators try their best to help keep the site running as it should and it's generally much more productive to avoid making queries that you raise personal. $\endgroup$ – Glen_b Aug 10 '16 at 15:06
  • 5
    $\begingroup$ If you didn't mean to generalize or be "brash", perhaps you should edit the text of this Q to make "it impersonal and framing it in terms of issues rather than individuals", as @whuber suggests (& delete your personal comment on the referenced thread). It is certainly understandable that you are frustrated & had a bad day--that happens to all of us (maybe it was even happening to Peter here)--but now that everybody has had time to cool down, you have a chance to rectify it. $\endgroup$ – gung - Reinstate Monica Aug 10 '16 at 21:12
  • $\begingroup$ @gung good call... edited the question here. The comment I dont have the rep to edit $\endgroup$ – posdef Aug 11 '16 at 8:15
  • 3
    $\begingroup$ You only get five minutes to edit comments. After that your only way to alter them is either (a) delete and post a new one or (b) flag a moderator, who can edit them. $\endgroup$ – whuber Aug 11 '16 at 12:44
6
$\begingroup$

I observe that concurrent to my writing of this answer, the question has been re-opened. The question, to my reading, remains unclear, and I think the unclear elements of the question I name here still need to be addressed.


Perhaps I'm overstepping, but it would seem that the real issue here that OP would like to know why, precisely, the question was closed (the closure being the genesis of the personal attacks). I submit that I have experience working on a diversity of data science/statistics tasks, but I am having some difficulty discerning what your post is about. (Perhaps this should be a comment on OP's other post, but now it's too long to fit.)

These are the particular areas that I think would benefit from clarification.

  • You make reference to $a$, $b$, and $c$ without introducing what they are. Is that piece of information contained in the plots? Perhaps, but the plots are so small that I can't read them.

  • You discuss increasing/decreasing stimulus. Is a stimulus one of the "variables"? Or is it something else? Is it synonymous with one of your 12 observations?

  • You discuss categorization/clustering of your data. Categorization suggests that you are performing a supervised learning task, e.g. logistic regression, while clustering suggests an unsupervised task. Which is it? If you know where increases or decreases occur with respect to stimulus, you can label and classify those examples. But it seems you hew towards the "unsupervised" meaning by reference to DBSCAN and $k$-means. Remember that the essential part of clustering (at least as I understand it) is that you're constructing groups of "nearby" things, where "nearby" is characterized by the choice of a distance metric. So perhaps deliberate selection of a distance metric particular to your task will give traction.

My recommendation would be to sit down with a friend (who doesn't do work in your field and who is unfamiliar with this problem) and have him or her read your question. If they're not quite sure what you're after, work with your friend on how to refine and clarify the question.

In general, users of CV are often working with confidential or proprietary data, which causes them to be circumspect. I appreciate the difficulty, as I once worked on data that was very sensitive. However, without some concrete notion for readers to latch on to and fix in their minds, it is very difficult to formulate an answer. If this is the case, I suggest attaching your problem to some analogous or toy scenario.

| |
$\endgroup$
  • 4
    $\begingroup$ On reading the original question I agree with the tone and the content of this. I'd add that (at least in my browser/machine/OS combination) the graphs are too faint for me to read at all. It's too difficult for me to estimate whether I would have voted to close the original. $\endgroup$ – Nick Cox Aug 10 '16 at 22:39
  • $\begingroup$ But -- my main point -- note that this kind of critical discussion is much more than any high-reputation user (meaning those with enough reputation to vote to close) or moderator is obliged to give even when agreeing with a vote to close. While no comment at all is inevitably disappointing to any OP, the tough love of SE votes to close means "Sorry, we can't understand what this is about. You'll need to rewrite it if you want to persist". A question has to come up to scratch to be accepted. $\endgroup$ – Nick Cox Aug 10 '16 at 22:40
  • 4
    $\begingroup$ I think that meta is precisely the place to ask "why was my question closed?" and that is plausibly the underlying concern of OP's meta question (though that's meeting OP more than halfway). I didn't have to write this, and no one had to provide an explanation for why the post was unclear, but since the question has (implicitly) been raised, i thought it was worth taking the time to address $\endgroup$ – Sycorax Aug 10 '16 at 22:47
  • 1
    $\begingroup$ Yes indeed; my second comment was perhaps more aimed at @posdef. $\endgroup$ – Nick Cox Aug 10 '16 at 22:52
  • $\begingroup$ Let me start by saying that I appreciate the time you took to write these comments, especially the bits about potential clarifications, I'll come back to them separately. I strictly disagree with the "Tough Love of SE" rhetoric you seem to be pushing however. I have been a member on SE sites for almost twice as long as you guys, and my experience is that the sites strive to be constructive rather than destructive. In practical terms, if you think something isn't right, you tell the person how to do it right rather than tearing down what they have done... $\endgroup$ – posdef Aug 11 '16 at 8:24
  • $\begingroup$ ... If you don't have the time or patience to do that, you refrain from acting. This is especially true for mods who are supposed to be a role model for the community. That's been the code I was thought on SO and that's the rhetoric I went by, which explains why I don't have several hundred thousand rep or a thousand edit badges; i.e. I try to contribute where I can, as much as I can. $\endgroup$ – posdef Aug 11 '16 at 8:28
  • 5
    $\begingroup$ The principle you expound is admirable. The practice is that if anyone does not understand a question, they often can't even say how it should be rewritten. And if X's question is quite unclear to Y, it's not Y's job to rewrite it. BTW I have ~20K reputation both here and on SO and in my small experience moderators here behave less capriciously than on SO, and lean over backwards in dealing with difficult members, so candour about personal experience is something we can all exercise. $\endgroup$ – Nick Cox Aug 11 '16 at 8:54
  • 2
    $\begingroup$ stats.stackexchange.com/questions/228242/… could be cited as an counter-example to your thread, which builds entirely on one example in making a general case. Here several people tried in various ways first to indicate that a question was unclear and then to get the OP to improve it. In the end people gave up one by one but there was no lack of effort in trying. In short, it is unfortunate that you got no specific guidance on improving your post, but that example does not indict the forum generally or moderation in particular. $\endgroup$ – Nick Cox Aug 11 '16 at 9:03
  • $\begingroup$ Update to my first comment above: The graphs in the original post are now better. The lines still look faint, but thanks for editing. $\endgroup$ – Nick Cox Aug 11 '16 at 9:20
3
$\begingroup$

Let's try another tack -- one that I think is a better way to think about how this problem relates to CV in general.

People can reasonably have disagreements about where the line is to close or keep open questions. I agree with General Abrial's specific critiques about your question - but I disagree that these are reasons why it should have been closed initially or that those critiques are beyond what can be reasonably expected for people to hash out in the comments. I don't think there is much point in debating that though -- questions are too idiosyncratic for me to formulate more general guidelines beyond this particular instance.

Moderators and users have discretion to shape the site. This is a good thing. That same discretion can result in either poor decisions or decisions you disagree with. (I'm reminded where the originator of CV had a question migrated.) The site is set up so that these decisions can be reversed. Besides asking a question on meta, you could pop into the chat room - I would have nominated to re-open same as I did after seeing this meta thread. I've used meta to protest decisions as well.

If you think there is a more general trend in how the site is run, you can bring that up in meta. That is an argument beyond one act by one person though.

| |
$\endgroup$
0
$\begingroup$

Since this is a meta discussion, I'll provide an answer as well, specifically addressing the issue of clarity.

It seems to me that there is a significant issue here on CV with respect to general <=> specific questions. My understanding of SE sites is that we want to accumulate useful information for a general audience. Thus it makes more sense to formulate questions in ways that make them as relevant for people as possible.

A good but possibly silly analogy would be as follows; say I want to know how to throw a baseball, because I don't know how to throw a ball. I could ask the question as "how can I throw a baseball on a grass patch behind my house?" or I can ask something like "how to throw a hand sized ball outdoors?". The second alternative makes it immediately more relevant for wider audience, those who might want to throw a tennis ball, or a lacrosse ball, or a softball... I think putting the effort to generalize and make the question relevant for many people should not be penalized here.

The second issue is how much details one has the liberty to give. Now in this case the details are neither government nor corporate secrets. However they are research results and it is rather tricky what I am allowed to say and not say, especially considering the figures. There are novelty concerns for both academic and IP settings.

However, not all the details are relevant to the question, for example the exact definitions of $a_i$, $b_i$ and $c_i$, are not relevant to what I was trying to ask, IMHO. They can be increasing doses of a medication each, they can be measurements with increasing amount of time, amount of food or dollars received for a task in a behavioral experiment... The point is that $a_4$ is more $a$ than $a_3$. The underlying assumptions is that $a_1$ to $a_4$ are increasing level of stimulus of a specific kind, analogous with $b_1$ to $b_4$ and $c_1$ to $c_3$.

The idea is to find variables that respond to $a$ or $b$ or $c$ and not to combinations of them. That's what I meant with classification or categorization. So yes it is supervised, if you in advance specify which clusters to find. As an alternative I tried to automate the process by adopting an unsupervised method to see if it would pick up the same clusters. They did not, as I have explained in the question body.

The issue here is that people seem to get stuck on details that are not going to be relevant for them. I am fairly certain that practically nobody here would be able to relate to the kind of processes $a$, $b$ and $c$ refer to. That's because I barely grasp how they work myself. :)

I tried to explain the data as best as I can, and I am still willing to explain bits if someone is interested in the specifics of the problem at hand. At the end of the day I am looking for leads to solve a problem I have, I don't want to outsource my job to someone else. I think it's the more respectable thing to do.

| |
$\endgroup$
  • 2
    $\begingroup$ Thanks for the expansion of your comments. Abstraction versus detail is always a difficult trade-off. On a second reading of your question I tripped up on DBSCAN, which is not at all familiar to me. But that is just why I (still) can't answer your question and is not a personal reason for closing it. $\endgroup$ – Nick Cox Aug 11 '16 at 9:32
  • 7
    $\begingroup$ The more abstract you make a question, the harder it becomes to define your terms with sufficient precision; even if you succeed it's still not necessarily an easy read. I find this question hard to follow because I struggle to keep track of the observations, variables, values, stimuli (at different levels & of different types), responses, samples, & copy numbers. When people "get stuck on details", what they want is something concrete to latch on to, as @GeneralAbrial said, to help them understand what's going on, & to fill in any gaps in the information you've supplied. $\endgroup$ – Scortchi - Reinstate Monica Aug 11 '16 at 12:20
  • $\begingroup$ @Scortchi fair point... I'll try to reformulate the question when I have a bit of time on my hands, with more details and hopefully more consistent terminology $\endgroup$ – posdef Aug 11 '16 at 12:22
  • $\begingroup$ Making up a context is one option, not without its drawbacks, but perhaps worth considering. $\endgroup$ – Scortchi - Reinstate Monica Aug 11 '16 at 12:25
  • 7
    $\begingroup$ "In mathematics context obscures structure.... In data analysis, context provides meaning." (George W. Cobb) amazon.com/Introduction-Design-Analysis-Experiments-George/dp/… $\endgroup$ – Nick Cox Aug 11 '16 at 14:26

You must log in to answer this question.

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