9
$\begingroup$

The current interest in the "homework" vote-to-close (VTC) option and related matters leads me to wonder how the internals of this site work. Compare Facebook, which despite a huge user base edits policies rather frequently, and often in dramatic ways. By contrast, at Chemistry Stack Exchange (Chem SE) changing the name of one VTC option (or just adding another) elicits a great deal of opinion. Is it because this is technically complicated (involving major alterations to the code base)? I can understand that the moderators don't have access to the code, but presumably they have a line to the developers, or? Or is it because it is difficult to agree on policy? Presumably things have to be brought to a vote etc.

So my question in summary is:

How difficult is it to alter the choice of VTCs? If it is difficult, why, and can this somehow be changed?

As a bonus question:

What can be changed on the site, how, and how fast? Specifically as regards aspects such as close menus and other global features (not user-specific) of Chem SE, what is easy to change on this site, and what is not?

$\endgroup$
10
$\begingroup$

Chemistry Stack Exchange is essentially a grassroots democracy. Every user gets a say in the development and the change of 'rules' (our policies) or guidelines. Additionally we are embedded in a large (if not huge) framework of sites, that follow the same rules (e.g. the Code of Conduct), same User Interface (with all the limitations and advantages), similar usage patterns (good stuff = up vote, not so good stuff = down vote), similar policies (the homework policy used to be an example for that), and probably other stuff. However, some of these communities have also differences in their usages or policies, as some simply cannot be adapted to other sites.

The policies we have in place now were discussed starting in the private beta phase and growing/extending/cutting back since then. Certain usage patterns have established, and some users interpret these policies ever slightly so differently that it may cause disruption. Those usage patterns are usually the reason why things around here change ever so gradually. Let's get back to all the questions that were asked one-by-one and see what may be missing.

Compare Facebook, which despite a huge user base edits policies rather frequently, and often in dramatic ways. By contrast, at Chemistry Stack Exchange (Chem SE) changing the name of one VTC option (or just adding another) elicits a great deal of opinion.

Facebook is a company, if they decide to change their policies they just do. Stack Overflow, as our service provider (and also a company), can do exactly the same thing. That happened when they implemented changes because of changes in laws, or applying the Code of Conduct, etc. SO is usually much better in communicating these things and discussing such changes beforehand, than Facebook would ever be inclined. However, that this doesn't always work well was seen, when they introduced the new UI.
If they make these changes you have a choice: accept these changes, or leave.

A VTC option is something that we as a community agreed upon, we make this 'rule' and we enact it. If we cannot find a way to change this into something we all agree on, we will enact it differently, which will lead to disagreement along the line.

With the homework reason that is exactly one of these points. Some people se it this way, others see it another way. But as it is now, we kind of all agree on the fact that something needs to change, but we can't exactly grope what, and how to write it in a way that a new user will immediately understand it. And this is basically where we got stuck a couple of years ago.
Meanwhile the current system still works, albeit sub-optimal. We better have this in place than leaving us vulnerable to unnecessary discussion once it completely breaks away.

Is it because this is technically complicated (involving major alterations to the code base)?

No, it is easy to switch this on and off, and modify the text. We have three custom ones available, two are currently in use (and the medical one is not going anywhere). So this is already implemented, and we experimented about with this some time ago.

I can understand that the moderators don't have access to the code, but presumably they have a line to the developers, or?

We have, but changes, which come from our community might have an effect on other communities, and they also might not be highest priority. (I wouldn't count on ever having mhchem help in the edit side bar for that matter.)

Or is it because it is difficult to agree on policy? Presumably things have to be brought to a vote etc.

Exactly. Establishing a policy which everybody (or at least the large majority of the active users) agree on it the most important part. It has to be easy to understand and follow for new users, and easy to enact for our current users. It's not simple to make major changes once we have something working in place. Radically changing a policy (on meta no less) doesn't automatically mean you change everyone's behaviour, it doesn't even mean everybody knows about this change.

How difficult is it to alter the choice of VTCs? If it is difficult, why, and can this somehow be changed?

Difficult enough. (Let me know if anything remained unclear from the above.)

What can be changed on the site, how, and how fast? Specifically as regards aspects such as close menus and other global features (not user-specific) of Chem SE, what is easy to change on this site, and what is not?

Tags, descriptions, some parts of the help pages can technically be changed quite fast. These basically require a meta post with a proposition, and then we fix a deadline and regard it as part of our policy. Same applies to anything withing . Reaching a consensus (or sometimes only compromise) is the hardest.

Functional changes should be brought up on mother meta, and could take ages.

I currently cannot think about anything more or specific right now, so if you thinks theres something missing, comment and we'll get to it.

$\endgroup$
  • $\begingroup$ You're 15 minutes and 20 seconds late, Slowest Gun In The West. The two answers eerily look like election questionnaire answers. $\endgroup$ – M.A.R. Jun 5 at 17:24
  • $\begingroup$ Thanks very much to you and @M.A.R.ಠ_ಠ for your answers. Yes, I can see how this might feel like an interview, but remember I have little info on this matter. Probably the community at large appreciates understanding how this stuff works. The details you provide are very useful to understanding what can be done. Arguing in a vacuum without this sort of background info is pointless. It also clears up what makes this homework dispute complicated. $\endgroup$ – Buck Thorn Jun 5 at 17:50
  • $\begingroup$ @Buck It is one problem that information is often scattered around meta, and often impossible to find (especially for newer users). It's good you have asked the question, I think it'll feed positively into the whole process of redefining our close policies. Even though, as you might have realised now, the process can be quite tedious, it is worth the undertaking and we appreciate your input. $\endgroup$ – Martin - マーチン Jun 6 at 9:23
7
$\begingroup$

Well, I feel like I'm in an interview.

Compare [F]acebook, which despite a huge user base edits policies rather frequently, and often in dramatic ways. By contrast, at chem SE changing the name of one VTC option (or just adding another) elicits a great deal of opinion. Is it because this is complicated?

Off the top of my head, I can count two factors: Different opinions, and that we care about them, and the lack of a benevolent dictator.

Changing a close reason can be as easy as one of the mods just . . . doing it. Mods are indeed the 'leaders' in some sense, but no ones sees themselves in a position to act unilaterally unless most of the community has already agreed with how things should run. Furthermore, being the cautious people that we are, we experiment things time and time again, and we're wary of failure. Facebook is too big to fail overnight, and there is ultimately one "boss" who will decide how things should go, which might or might not regard other opinions.

I can understand that the moderators don't have access to the code, but presumably they have a line to the developers, or?

Yes, they can contact Community Managers and probably other developers, and have means of communication that we don't, but we don't need anything that requires that level of access, I imagine.

How difficult is it to alter the choice of VTCs? If it is difficult, why, and can this somehow be changed?

As I said, it's really not a big deal per se, but reducing the inevitable friction that comes after every change to a minimum makes us wary of failure. It's going to be even more soul-crushing to clean up the mess if we mess up. Hence, we should make sure we have a thorough analysis of what needs closing and what doesn't, and get most people to agree on something projected to have positive results before going further.

Another thing worth pointing out is there is a certain intricacy to tweaking anything related to 'homework'. It turned out to be an established norm by the time chem.SE came to be, as Physics.SE and math.SE were already around. The more established it is, the harder it will be to get rid of. As far as I recall, Physics ended up disallowing such questions altogether, but we decided there is still some merit in the traffic they bring if we do get rid of the bad apples effectively.

What can be changed on the site, how, and how fast? Specifically as regards aspects such as close menus and other global features (not user-specific) of Chem SE, what is easy to change on this site, and what is not?

Things moderators cannot tweak are almost guaranteed to be there. We can't have a more efficient First Posts queue, for example. Moderators can change close reasons, so discussing it sounds worthwhile because once a consensus is reached, it can boil down to me asking one of them in chat. Regardless of the fact that Stack Overflow is "Godzilla, king of sites, and center of attention", there's a load of sometimes crazy but usually good ideas floating around and just not enough developer time.

We usually wait for some sort of consensus before acting on anything, reflected, usually, in community voting the related meta posts. So that's the rate limiter right there. (As you can see, I'm still giving my previous post some extra time to be read)

$\endgroup$
  • $\begingroup$ "It's going to be even more soul-crushing to clean up the mess if we mess up. " This is very dramatic and makes me wonder what that scenario might be. But between your and Martins answer I can imagine what it might be. I suppose building up the site as a Q&A magnet has not been trivial. $\endgroup$ – Buck Thorn Jun 5 at 17:54
  • $\begingroup$ Should I assume that Martins answer is "canonical" and accept it, or do you think there are there merits in your answer vis-a-vis his that I should take into account first? :-) PS You know I'd hire you but we are working on our budget... $\endgroup$ – Buck Thorn Jun 5 at 17:56
  • 1
    $\begingroup$ @Buck your choice, I think we just said the same things more or less but he's the one who can exactly tell what moderators can and cannot do, and he's got the more recent version of your questions in his quotes. (P.S. hiring a 20-year-old is probably a bad idea for any project ;) $\endgroup$ – M.A.R. Jun 5 at 18:05
  • $\begingroup$ You should know that discriminating on the basis of age is against the law in some countries :-) $\endgroup$ – Buck Thorn Jun 5 at 18:39

You must log in to answer this question.

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