2
$\begingroup$

This discussion has the aim of convincing people that we should broaden the range of "acceptable" questions on this site, as distinct from Stack Exchange generally. It will, however, take a while to sort out the end game if it is to be done properly, by evolution, rather than by imposing a predefined set of rules.

History of the site - tech questions

Stack Exchange was founded to enable practitioners building software to ask technical questions about their work so as to get expert advice and speed up their work. Because of this focus certain questions were deemed to be less useful and a bit "out of bounds." The main guide, from the Help Center is that questions shouldn't be "overly subjective". This has been interpreted on this site to include at least three types of questions (with examples):

Too "Broad"

What does a person with an undergraduate degree in CS still need to learn to be employable as a developer? This question was closed, over my strong objections, as too broad. This despite the fact that it had many up votes (and one down) and was favorited by many people. There were several answers given, many of which also had many up votes.

I'll take it as a given that this question has no definitive answer. However, students graduating face a lot of obstacles and they need to be prepared. Their professors need to be aware of what they will face so that they can properly advise their students. As the question states, it isn't intended to change curriculum, since no curriculum can cover everything. But a student going out blind is at a disadvantage.

Update:

  • It has been suggested that I narrow the question; asking, for example, about technical issues only. However, I maintain that many of the issues are decidedly not technical and fall into many categories. The question has this breadth by design, not by accident. Such a narrowing would make the question essentially useless. It might (a) take a dozen questions to cover the possible issues, (b) I might miss some of them, (c) readers would have to, themselves, integrate over a large number of questions and answers to get a proper picture of the challenges.

What kind of game (in Scratch) would require use of a list? This was also closed as too broad. Yet it gives useful guidance to instructors using Scratch who are trying to do a good job.

Opinion Based

Which of the following programming languages will help me better to teach the basic concepts of CS and why? This question was closed as being opinion based, but the OP specifically asked for reasons. Yes, opinions would be expressed, but I would expect also reasoned comment. Where can a new teacher get this kind of comparative discussion?

No "definitive" or best answer possible.

I have no example of questions closed for this reason, but it is frequently mentioned in the classroom as a reason for down voting or closing a question.

Why is it an issue?

One of my problems with closing questions for the above reasons is that the person voting to close is making a judgment that might be different from that of educators who might answer the question. For some of them, the judgement that it is too broad can only be properly made after a range of answers are given. The same is true for opinion based. Are there, in fact, a lot of opinions, or is there reasoned comment. But closing early makes it impossible to know the truth of the situation.

I pointed out that one question was closed despite a lot of answers. I was told in the classroom that that was a bad thing. And yet our HNQs arise from a flurry of early activity. That is inconsistent at best. But see below.

Teaching is Different From Programming

This site is about education, not technology. The product of education is change in the minds of students. It isn't some program or artifact. The nature of education is that context matters more than in other fields. Every student you have brings a new context. Every group, curriculum, course, subject, book, idea, brings context to the mix. I believe that, for important questions, it is impossible in principle to provide one "best" answer. One definitive answer. Do this. Done. Success.

That isn't how it works. Only small and relative inconsequential questions have such definitive answers. "How much should I stress indentation in Java." Such small questions are not very interesting, though they help beginning teachers and those changing some part of their curriculum. But they don't matter in the long run. Only big questions do.

Update

  • It has been suggest that I provide a compelling reason for wanting a widening of the rules. The compelling reason is that teaching doesn't have an algorithm. It isn't a process that you can learn once and for all. Teaching is a relationship between two individuals, though often we try to force it into a mass process. Teaching is dominated by (a) context and (b) interpersonal contact. It is fundamentally different from the inherently simpler context of, for example, Stack Overflow.
    If Stack Exchange wanted to limit the future to the same policies as those of the past it should never have approved sites such as this one or MathEducators or Buddhism or Community Building or ... Forcing those concepts into a technical only framework does a disservice to the the network itself as well as to the newer sites.

Passing on Deep Expertise

My biggest issue concerning big v small questions is "How do we educate new teachers and help them learn the tricks that it took us old-ones so long to grok?" In the US, at least, teachers at the university level are not taught how to teach. All they have are the examples set by their own professors and the examples of those professors are similarly constrained. A professor doing something useful and really innovative has only a small audience and most of them won't carry on the new way as they won't themselves be teachers. Things like Pedagogical Patterns helps, of course, as well as workshops at SIGCSE and similar conferences. But a site like this could do a lot of good if it isn't constrained to only answer small questions. There is far too much "lecture with powerpoint" going on when teaching could be so much more effective.

Some of the questions, mine in particular, are leading questions. I hope to make younger teachers think about the process itself and the wide range of possibilities open to them.

Evolution

My goal is to have this site evolve a set of standards rather than having an imposed collection of constraints. But it takes only a few people to stifle that evolution, by closing questions before anyone has a chance to see whether they are useful - especially to younger teachers - who most need guidance. I don't believe the current constraints are helpful and note that once a question is closed it is extremely difficult to reopen as less than 2% of our members have the reputation to vote to reopen.

One of the "super beings" here (pops) has already complemented us for reopening the question of blogs, which is more useful to a site like this than to some others. I think that period of evolution - and forbearance - on the issue of the appropriateness of questions would serve us well.

$\endgroup$
  • 2
    $\begingroup$ Incidentally, some related reading: stackoverflow.blog/2010/09/29/good-subjective-bad-subjective $\endgroup$ – Michael0x2a Sep 19 '17 at 23:40
  • 1
    $\begingroup$ I don't really disagree with you here, but note that a lot of your points apply to Stack Exchange in general. Questions in many topics (including programming) don't have a single "correct" answer. Also a tiny gripe: using a bunch of quote blocks for things that aren't quotes is pretty confusing to read. $\endgroup$ – Kevin Workman Sep 20 '17 at 18:08
5
$\begingroup$

The main guide, from the Help Center is that questions shouldn't be "overly subjective". This has been interpreted on this site to include at least three types of questions (with examples):

Too "Broad"

Broad is not the same as subjective, and I'm not sure why you're conflating them. In the case of the example you give, two of the close voters left comments:

Its too vague is what it is.

I've voted to close this question because there are too many possible causes for a CS graduate feeling unprepared to enter the job market. I'd suggest that you ask a separate question about how to resolve a specific concern that graduates have (e.g. lacking in workplace skills, missing knowledge, etc)

Neither of those suggests that they were using "Too broad" as a proxy for "Too subjective".

The guideline which essentially underlies "Too broad" is

Your questions should be reasonably scoped. If you can imagine an entire book that answers your question, you’re asking too much.


As an aside on HNQ: the main point of HNQ is to drive traffic to the smaller sites. On that metric it's a success. But there are plenty of people who complain that it's traffic at the cost of quality, because HNQ selects for the wrong things and then creates a positive feedback loop whereby the traffic from HNQ pushes the metrics which HNQ measures.

$\endgroup$
4
$\begingroup$

In general, while I think while there's certainly room to expand and adjust certain policies to better fit the goals of this site, I also think flat-out rejecting Stack Exchange's policies altogether isn't a good idea.

After all, by starting this community on SE, we've basically signaled to any visitors and would-be contributors that we've accepted the core tenets of SE (and in exchange benefited from their infrastructure, mechanisms for boosting exposure such as the hot network questions, and general reputation for quality). Basically, we can't have it both ways -- we can't both use the advantages of being on SE while simultaneously ignoring their policies.

In any case, here are some of my thoughts on the topics raised on the original post (in a somewhat scrambled order):

The "no definitive or best answer" idea

Since I was one of the people in chat who originally pushed this idea, I feel obligated to state that I now believe I was wrong here after recently remembering and re-reading the Good Subjective, Bad Subjective blog post. (In my defense, it was written in 2010...)

I also think that article brings up two key ideas that are worth reiterating:

  1. Subjective questions are good when they inspire long, detailed, in-depth answers that explain "why" and "how" (as opposed to low-effort hit-and-run answers)
  2. Subjective questions are good when they insist that opinions are backed up with facts, references, and experiences.

So with that in mind...

The "What does an undergrad need to learn..." example

I wasn't initially sure how I felt about this question, but upon reflection, I think my main issue is partly with the question and mainly with the answers.

In particular, the answers weren't really useful -- they were the same old predictable "students need to learn version control/software engineering/writing tests/how to work with others/how to test" kind of answers that pretty much all instructors are already familiar with (especially the ones who are motivated enough to visit forums like this one).

To be fair, the question tried to dissuade these sorts of low-effort answers (in particular, the first bullet point asks things like "What evidence do you have for the gaps you suggest?" or "How can a person fill the gaps?"), but many of the answers unfortunately seemed to ignore these aspects of the question.

For example, several of the answers talked about how it's important for students to learn how to work on large projects. I mean sure, but how exactly do I teach students how to do that? Several other answers talked about the importance of knowing how to evaluate tradeoffs and evaluate risk. Great, also self-evident -- but how do I teach students this? A few more answers talked about how students need to learn to take responsibility for their own education -- same question.

In short, I think the answers failed to meet the "good subjective questions should invite detailed, in-depth answers" litmus test. (Sure, the answers use lots of words, but that doesn't mean they're in-depth).

I think this sort of thing is where the unease for broad questions and the preference for specific, focused, questions comes from. Broader questions tend to invite these sorts of shallower, scattered answers -- focused, targeted, and well-crafted questions let answerers focus their expertise and provide concrete advice that can help teachers actually make a difference and help their students.

(On the plus side, I'm now noticing these answers have all sorts of interesting ideas for potential questions...)

The "Scratch lesson ideas" example

I think I agree that this question should be closed primarily because the question, as it's currently phrased, doesn't meet the litmus test: it doesn't really seem to encourage deep or insightful answers.

However, I also think that this question could be easily re-opened with a little bit of tweaking in this case. In particular, I think it would help if the asker listed more constraints about what kind of assignments they're looking for.

(e.g. What concepts have the students already learned? What concepts have the students not yet learned? What concepts will the students be learning next that this one should try and pre-emptively set up? Since the teacher wants to use scratch, does there need to be a graphical component? What meta-goals does the teacher have?)

More broadly, I think I'd actually like to see more questions like this one where the community collaboratively brainstorms a variety of potential lesson ideas/homework assignment ideas. Coming up with a good homework assignment can be pretty challenging, so I think these types of questions could really benefit from the expertise this community has to offer.

There's also some precedent for relaxing the normal SE rules in this regard -- see tex.stackexchange.com's big-list tag, see some of math.stackexchange.com's more popular questions, etc...

I think the trick for us would be figuring out guidelines for phrasing questions in such a way that it encourages detailed, in-depth answers. (Perhaps we do something with community wikis? Moderate those types of posts more aggressively?)

To put it another way, I don't think this specific question is that great, but would like to see more of these types of questions, provided they're well-written.

The "Which of the following languages..." example

I'm still thinking about this question, but my instinct is to leave it closed because:

  1. It seems like the sort of question that would invite endless debate, and those sorts of questions are usually frowned about by the SE community at large. (See idea 2 up above)
  2. I question the underlying premise of the question (I like Buffy's answer best for that reason, actually). I admit this probably isn't a good reason for closing, though.

I think the "these languages could become obsolete making this answer obsolete" argument is partially a red herring on the grounds that the bulk of the languages mentioned in the post are at least a decade or two (or three...) old, and will probably be around for another decade or two. Honestly, answers with a lifespan of a decade is pretty damn good in my book.

The "Teaching is different from programming" idea

I don't have much to add here except to echo the counter-idea that our community isn't particularly unusual in this regard.

I do (now) agree that the "all questions must have a definitive answer" idea isn't correct, especially for this community, but also observe that there are many other subjective-ish communities (such as MathEducators, Buddhism, CommunityBuilding, as you point out), and they clearly figured out a way to reconcile their goals with SE's core tenets.

And if they can do it, so can we.

I also disagree with the idea that broader questions and answers are the best way to pass on expertise. Rather, as I stated above, I think it's the opposite: focused questions, targeted, and well-crafted questions give experts more opportunities to showcase their knowledge and deliver truly insightful answers.

And consider what happens when you take the idea that broader questions are better to the extreme -- at some point, the only way an answer can truly do justice to a question is to basically write the equivalent of a whole book. Nobody is going to do that, so what are people going to do instead? Write answers that don't (and can't!) fully address the question and so end up being of minimal value because of how "diluted" they are.

$\endgroup$
4
$\begingroup$

This discussion has the aim of convincing people that we should broaden the range of "acceptable" questions on this site, as distinct from Stack Exchange generally.

Unless you can point to a very compelling reason why we should have an exception, we should stick to SE norms. This is why this is a SE site, rather than any other forum on the internet. We chose to help this site because it will have the support of the Stack Exchange model behind it.

Stack Exchange was founded to enable practitioners building software to ask technical questions about their work so as to get expert advice and speed up their work. Because of this focus certain questions were deemed to be less useful and a bit "out of bounds." The main guide, from the Help Center is that questions shouldn't be "overly subjective". This has been interpreted on this site to include at least three types of questions (with examples):

I'm confused by your interpretation of the help center as saying that questions shouldn't be overly broad. The help center states "Some subjective questions are allowed, but “subjective” does not mean “anything goes”. All subjective questions are expected to be constructive" and goes on to provide details, but another section is devoted to examples of questions that do not fit the Stack Exchange format. It's far more specific than simply prohibiting questions that are "overly subjective".

I'll get to dissecting your examples in a second, but I'd like to start off with pointing out that here, the help center contains the rules of the site. That's why this is a Stack Exchange site. If you want to ask questions that are designated as off limits by the help center, rephrase them to fit Stack Exchange norms or go somewhere else. You act like this is a relic of Stack Overflow being a programmers site when the same model and rules have worked for sites with 171 different topics, not just Stack Overflow.

What does a person with an undergraduate degree in CS still need to learn to be employable as a developer? This question was closed, over my strong objections, as too broad. This despite the fact that it had many up votes (and one down) and was favorited by many people. There were several answers given, many of which also had many up votes.

A questions fitness for the site is totally unrelated to upvotes or favorites. You'll see old questions on Stack Overflow that are closed but have upvotes in the thousands. Questions that ask things like "Do you prefer mac or windows?" tend to be highly upvoted because people want to express their opinion, but should still be closed because they're not good questions for the site. Keep in mind, the only thing closure changes is the inability to add answers. Everything else stays the same.

I'll take it as a given that this question has no definitive answer. However, students graduating face a lot of obstacles and they need to be prepared. Their professors need to be aware of what they will face so that they can properly advise their student.

Then ask a specific question about a specific situation you encountered with a student. Answers will typically be generalizable to a large set of situations, and will be much more useful to the OP because they'll fix the OP's specific problem. From the help center: "You should only ask practical, answerable questions based on actual problems that you face. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page."

What kind of game (in Scratch) would require use of a list? This was also closed as too broad. Yet it gives useful guidance to instructors using Scratch who are trying to do a good job.

It may give useful guidance, but that's not what this site is for. From the tour: "This site is all about getting answers. It's not a discussion forum. There's no chit-chat." and "Ask questions, get answers, no distractions." This question is a list question, in that it is asking for a list of possible games. There is no way to evaluate answers against eachother, which is the basis of voting on this site, hence it's a bad question. From the help center: "avoid asking subjective questions where every answer is equally valid: “What’s your favorite ______?”"

Which of the following programming languages will help me better to teach the basic concepts of CS and why? This question was closed as being opinion based, but the OP specifically asked for reasons. Yes, opinions would be expressed, but I would expect also reasoned comment. Where can a new teacher get this kind of comparative discussion?

A new teacher can get that kind of comparative discussion in chat or on a different site. "If your motivation for asking the question is “I would like to participate in a discussion about ______”, then you should not be asking here." (help center) In addition, the premise of this question is flawed in that there is no "best" language, and in a few years a new language could appear that would make all the answers obsolete, and thus spreading misinformation.

One of my problems with closing questions for the above reasons is that the person voting to close is making a judgment that might be different from that of educators who might answer the question. For some of them, the judgement that it is too broad can only be properly made after a range of answers are given. The same is true for opinion based. Are there, in fact, a lot of opinions, or is there reasoned comment. But closing early makes it impossible to know the truth of the situation.

Any question, even with no answers, can be accurately judged to be too broad or primarily opinion based. Sometimes I see the justification used that "we should wait to see how it'll get answered" to not close early, but I disagree with that argument because closure is not the end of the road. Closure is a way to allow the question to be fixed before it gets answers that would be invalidated by the fix. Closure should be based mostly on the content of the question, except in the edge case that the answers tend misinterpret the question, in which case it is clear that the enough people are misinterpreting the question that is should be closed and be clarified.

I pointed out that one question was closed despite a lot of answers. I was told in the classroom that that was a bad thing.

Lots of answers tend to mean that the question isn't really a question, and is intended as a prompt for discussion. According to the help center: "If your motivation for asking the question is “I would like to participate in a discussion about ______”, then you should not be asking here. However, if your motivation is “I would like others to explain ______ to me”, then you are probably OK. (Discussions are of course welcome in our real time web chat.)" Lots of answers tend to mean that a question is motivated by the OP attempting to initiate a discussion rather than asking a question. Lots of answers could also mean that a question simply has lots of answers, which is totally acceptable.

This site is about education, not technology. The product of education is change in the minds of students. It isn't some program or artifact. The nature of education is that context matters more than in other fields. Every student you have brings a new context. Every group, curriculum, course, subject, book, idea, brings context to the mix.

I agree that many questions could be improved by adding more context, but sadly it seems to be challenging to get many askers to add that context. I'll support you in that quest to get askers to add context to questions.

I believe that, for important questions, it is impossible in principle to provide one "best" answer. One definitive answer. Do this. Done. Success.

Of course! That's why multiple people can answer the question! But a prompt for discussion is different from a question.

My biggest issue concerning big v small questions is "How do we educate new teachers and help them learn the tricks that it took us old-ones so long to grok?" In the US, at least, teachers at the university level are not taught how to teach. All they have are the examples set by their own professors and the examples of those professors are similarly constrained. A professor doing something useful and really innovative has only a small audience and most of them won't carry on the new way as they won't themselves be teachers. Things like Pedagogical Patterns helps, of course, as well as workshops at SIGCSE and similar conferences. But a site like this could do a lot of good if it isn't constrained to only answer small questions. There is far too much "lecture with powerpoint" going on when teaching could be so much more effective.

You seem to be constrained in your thinking to believe that small questions demand small answers. A simple, practical, answerable question can be answered with a long, general answer that gives large scale advice and talks about larger ideas and theories in teaching.

My goal is to have this site evolve a set of standards rather than having an imposed collection of constraints. But it takes only a few people to stifle that evolution, by closing questions before anyone has a chance to see whether they are useful - especially to younger teachers - who most need guidance. I don't believe the current constraints are helpful and note that once a question is closed it is extremely difficult to reopen as less than 2% of our members have the reputation to vote to reopen.

Again, utility isn't a measure of how fit a question is for our site. To avoid any confusion, I'd like to point out that the 2% who can vote to reopen is the same 2% who voted to close. I'd also like to see more users gain these privileges by participating in this site.

One of the superusers her (pops) has already complemented us for reopening the question of blogs, which is more useful to a site like this than to some others. I think that period of evolution - and forbearance - on the issue of the appropriateness of questions would serve us well.

In response to updates

I will address this set of updates, but no further ones. This site is for questions and answers, not debate, and it's unfair to me and any other readers and a waste of my time if you keep updating the question. Ask another question or move this to chat.

So, here goes:

It has been suggested that I narrow the question; asking, for example, about technical issues only.

Nope. Never suggested that. I don't think anyone has ever suggested that.

However, I maintain that many of the issues are decidedly not technical and fall into many categories.

Yup.

The question has this breadth by design, not by accident.

OK? The "purposefulness" of a question has no impact of its quality. Not sure what you're trying to say here...

Such a narrowing would make the question essentially useless.

Nope. If you have a lesson you're planning or a problem student you need to deal with or a curriculum you have to design or any other host of issues a specific question and specific answer could be very useful to you instead of a super broad stokes theoretical question that someone could write a book on.

It might (a) take a dozen questions to cover the possible issues, (b) I might miss some of them,

That's why "you should only ask practical, answerable questions based on actual problems that you face" (help center, emphasis mine).

(c) readers would have to, themselves, integrate over a large number of questions and answers to get a proper picture of the challenges.

No. This isn't a blog for teachers to read about different theories and intellectual debates. That can certainly be fit into answers, but that's not the point of the site. The point is that you can "ask questions, get answers, no distractions" and that those questions are "practical" and "answerable".

It has been suggest that I provide a compelling reason for wanting a widening of the rules. The compelling reason is that teaching doesn't have an algorithm. It isn't a process that you can learn once and for all.

Neither is worldbuilding or photography or writing or parenting or philosophy or the workplace or sound design or community building or language learning or literature or interpersonal skills or ... so many other sites. And yet they all strive to fit their subject area into the SE model. Your reason would apply equally well to those sites, and yet they still understand the importance of the model. Why are we different from those sites?

Teaching is a relationship between two individuals, though often we try to force it into a mass process. Teaching is dominated by (a) context and (b) interpersonal contact. It is fundamentally different from the inherently simpler context of, for example, Stack Overflow.

Again, we're not the only "subjective" SE site, as I listed above.

If Stack Exchange wanted to limit the future to the same policies as those of the past it should never have approved sites such as this one or MathEducators or Buddhism or Community Building or ... Forcing those concepts into a technical only framework does a disservice to the the network itself as well as to the newer sites.

No. It does a disservice to CS teachers or math teachers or buddhists or community builders or world builders or students of literature or sound designers or photographers to deny them a place where they can ask their everyday practical questions without the site getting taken over by people who want to use it as a soapbox for their ideas and who could care less about an ordinary teachers everyday issues. This site is not a blog where you can rant and share your special opinions and ideas about teaching. It's a place where people who don't know the answer to their specific teaching problem can try and find objective answers supported by experience, research (and a bit of gut instinct) given by experts in the field or experienced teachers. Its a repository of knowledge built piece by piece, authored by thousands of teachers asking their everyday questions which gradually builds, answer by answer, question by question, a resource that will be invaluable to every teacher for years and years to come.

That's the goal of this site. This isn't "forcing these concepts into a technical only framework" as so many other so called "subjective" Stack Exchange sites have proven. This is building a practical, useful, repository of knowledge. Not for any user to soapbox about their ideas without the "constraints" of facts or a real life problem to solve.

More thoughts

I was rereading the question, and this bit stuck out to me. So I wanted to respond again.

Only small and relative inconsequential questions have such definitive answers. "How much should I stress indentation in Java." Such small questions are not very interesting, though they help beginning teachers and those changing some part of their curriculum. But they don't matter in the long run. Only big questions do.

At first glance, this seems to be true. Of course big questions have more value than little ones! But if you think about this, you'll begin to realize it's incorrect. If you want to change the way people teach, you have to start small. Start with the little things, like how indentation is taught, and little by little you'll change the way people teach. Your great answer to "How much should I stress indentation in Java" could initially help just the OP. Fairly small impact. But over the years, teacher after teacher will be able to come back and read your answer and change how they teach indentation in Java. And maybe, your little piece of wisdom will allow them to apply the same techniques as you explained for indentation to other topics they teach. And, if you write a truly great answer, the ideas could permeate hundreds of teachers styles over the years and impact tens of thousands of students.

But your standard, lofty, super general answer won't help at all. Maybe you'll manage to move a couple teachers slightly closer to your view, but I doubt it'll have as significant impact over the years as new ideas and theories of teaching come into being. It'll fade into a mass of articles and opinions that a teacher is bombarded with by other teachers, by what they read, by conferences, by workshops. It'll grow old and obsolete as your indentation answer continues to be the best way to solve that problem and continues to enlighten teachers and their students. That teacher will never forget this new, amazing way of teaching indentation. But as for "that thing I read a while ago"? I think it'd be forgotten and lost in a heartbeat.

$\endgroup$
  • 1
    $\begingroup$ There are too many good points in this answer. The updates just add to that problem. $\endgroup$ – Sean Houlihane Sep 22 '17 at 9:42

You must log in to answer this question.

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