What are we trying to build?

Worldbuilding has a sandbox. I’m quite familiar with it.

Because of the limitations of SE, the sandbox is implemented as a meta question (“this is the sandbox”), and question drafts are answers. An answer contains the question draft, proposed tags, and any meta-commentary from the author (what you’re having trouble framing, questions for reviewers, etc). People then respond in comments. Iteration happens. Eventually the question gets either posted or abandoned, and deleted from the sandbox either way.

In a Codidact sandbox the post is a “question”, but not a question to be answered in the sandbox. It’s more of a request for critique, really, same as someone in Photography asking for help improving a photo or someone in Graphic Design asking for help with a logo or someone on Writing asking for critique of a poem. These aren’t really questions; they’re requests for help or input. I would implement that as a “discussion” post type.

Which is exactly why a “one post type per category” approach doesn’t work. Communities should have the option to have those canonical wiki posts – which are not Q&A posts but another type – right there with their Q&A if they want. And meta could host Q&A and discussions and announcements. And they might have other wiki posts that don’t belong with their Q&A, and blog posts that do or don’t.

A category defines the content that goes there. Which specific post types implement that is orthogonal.

4 Likes

“These aren’t really questions” My point is that they are questions. Sort of “baby questions, waiting to grow up and join the main Q&A”. The end result is almost, but not exactly, the same as the Discussion post type. Though even if exactly the same as the Discussion post type, but with an actual name “Sandbox” attached, that would still be a vast improvement over the “force it into Answers to a Meta Question, since we have no place else to put it” mode of SE.

But in general, I think I have come around to your vision of Categories as “content areas” where I was thinking of them as “post types but with another name”. I think that from a Navigation standpoint, there are times (for a lot of different functional reasons) that seeing each individual Post Type within a Category separately makes a lot of sense. But especially in the “looking for an answer to my as-yet-unasked question” mode, seeing all the “main” Category Post Types together (e.g., Q&A, Canonical, Wiki) in one place - and searchable by keywords & tags all at the same time - can make for a greatly enhanced user experience.

1 Like

In one sense they’re draft questions, but, strictly speaking, they are “meta-style” questions where the question is “how can I improve this proposed post?” and not, for example, “how close can the K-class secondary star be in my binary system without frying the habitable planet orbiting the G-class primary?”.

Yes! I want us to be able to support the things that are shoehorned in awkwardly on SE.

Glad to hear it! Thanks for the discussion.

3 Likes

And some sites would only have canonical questions, not wikis. Up to each community to decide for itself.

6 Likes

I just noticed that IanKemp and I both used a double negative. I’m nearly 100% certain that neither of us intended it that way.

1 Like

Yes, yes, and “vastly” is too small a word.

Incidentally I was recently surprised to find that on https://graphicdesign.stackexchange.com/ if you start to create a new topic with the critique tag then displays adds site-specific usage guidelines for that tag where one couldn’t help but see it. I didn’t know the software could do that.

image

That (little) text is probably edited by a CM after a community reaches consensus.

I’d like them to be comparable to a Meta post – including comments or “answers” where people have a chance to critique a draft.

One way it’s not like an Announcement or a FAQ topic is that the title is very short and unique, and (ok, two ways: there are two ways in which it’s different) it also has some unformatted plain text description (as well as optional rich/markdown) text because of the way the tag usage popup works when you select or hover over a potential tag when you’re wondering whch one to use when you’re editing a new post).

Something else which SE doesn’t do is support tag hierarchies – except possibly by using hyphens like C++-design versus C++-debugging – see also Tags rather than sub-sites.

Perhaps the Wiki (i.e. free-form rich text) can be used to link to related tags and to describe the relation, thus all the hierarchy anyone might want. Viewing a summary of the tag the software might display those (linked) tags as “Related” a bit like it (SE not Discourse) displays related/linked topics.

Designing important new functionality might be “not for the MVP” perhaps though, I don’t know.

Perhaps not – on SE they’re used in a space-constrained UI i.e. in the tag editor/selector e.g. like this:

There’s a fair bit of work/detail in the behaviour of that tag selector – it took me several days to clone, it was easily the most complex part of the UI.

One problem with S/O is that there is an over-powerful central group of power-users. This central group frequently uses meta- to coordinate and make changes (eg. question closing, downvoting etc) en-masse.

I think a few, well-chosen site features, widely available to users (even those with a relatively low level of trust), will negate the need for having a central clique like this.

Furthermore, culturally I see no real need for “homework questions”, “duplicates” or “poorly researched” questions to be closed or downvoted off the site.

These labels are often highly subjective.

For those cases where there are real problems with questions (abuse, rights infringement, spam etc) then crowdsourced opinion might get us most of the way there. Or maybe there needs to be a moderator option for these. I’m not dogmatic.

But for “duplicates”, for example, a mechanism to link duplicates will go a long way.

I see no real benefit in letting super-duper users swoop in and downvote or close a question they perceive as duplicate. S/O has attempted this model, and failed terribly. There are endless duplicates. And the so-called “canonicals” frequently have highly-upvoted, wrong answers.

If you can’t beat em’, join 'em.

If a newbie wants to ask what a closure is, then I say, “ask away”. Certainly, I’d, link their question to a dupe. But perhaps another more generous user will engage and answer their question there and then. A win/lose then becomes a win/win.

Similarly for the widely used (and abused) “lack of research” reason for closing questions. Live and let live. As long as people can engage with the user and ask for clarifications and/or improvements to the question, then I don’t see a real problem.

That’s partly related to the site’s mission statement.

I think that one of SO’s explicit aims was to become a searchable wiki of good answers to various questions – as such it makes/made sense to have/keep non-duplicate topics.

Another site’s aim might be to give personalised advice – and therefore be more reluctant to close any person’s question.

It may be beneficial, if it really is a duplicate, and if it has already been answered well.

Your question’s being closed is kind of annoying if the so-called duplicate doesn’t already answer it. Maybe it should be, at least, easy or automatic to get your question reopened if you edit it, e.g. if you clarify how the question is different from the previous one.

It may depend on the subject matter too – e.g. SO problems tend to be boolean i.e. fixed or not fixed, answered or not answered – questions and answers on other sites might be more different-shades-of-meaning (and thus less likely to be duplicate).

Even so there are FAQs which lots of beginners ask…

And I personally don’t want people to copy-and-paste existing answers into new topics! That would go against my training.

2 Likes

Maybe I can re-word this better. I think it is a balance of harms. I think a lighter-touch approach to something like a dupe would keep more people happy and lead to more return visits. The person asking a “silly dupe” about setTimeout in JavaScript today, will be answering the tough questions in two years’ time. And I think the annoyance factor for the old-timers on the site can probably be mitigated in other ways (filters?).

Furthermore, initially, often the questioner doesn’t really know what their question is. The S/O model is to quickly downvote and close. It can be a horrible experience for novices.

For someone like you or I, we might either leave the question to someone with more time, or simply paste in a link to a duplicate or two, and leave the questioner to it. A more generous, or more junior user might want to have a go at an answer. And why not? It will be a learning experience for them too.

1 Like

I don’t find it especially friendly either in that way, and I haven’t posted there regularly in a while – I believe you but I don’t really know what you’re talking about.

That’s quite a good point. I answer a lot of seemingly elementary questions – e.g. more than 1000 answers (on another SE site) – partly in the hope that other users (less junior than me) will answer other more difficult questions that I couldn’t, and partly because the research is good for me, and/or because it’s a good question to which I’d like to understand an answer myself.

Even so if I think a question is a duplicate and already answered, I’ll ask, what’s different about this question … if only to know how to answer it or what to answer.

1 Like

Question: what is actually improved in the S/O model versus marking as dupe (with a link the the dupe, of course) and leaving it open? In the former a user takes a stab at a question, stumbles into what looks like a dupe to the old-timers, and is unceremoniously shut down. In the latter, the dupe is highlighted to the questioner, and maybe they then delete the question. But perhaps they leave it open and get a tailored answer for their particular problem.

In both cases the old-timers see a dupe flash up on their screens.

Absolutely agree.

I think that theoretically the harm is that a reader trying to find an answer using Google now has to find two different SO topics, one of which might not have a good a set of answers as the other.

But one of my recent questions on SO was allowed to remain (and is being upvoted as useful sometimes), though it was duplicate and according to this comment that was only because it was phrased using a different keyword (i.e. “opaque”) which people might be using as a search term:

I don’t think any of the other branded type questions reference opaque so it might be useful to add an answer to improve search-ability

The “SO way” used to be to close the question but leave it as a pointer (a hyperlink) to the topic of which it’s a duplicate.

Also moderators can merge answers from one topic to another – but I think that’s only moderators who do that and I don’t know what moderators (as opposed to just “super-duper users”) on SO mostly do with their time.


I write the above instead of saying, “the harm is just that power users can’t be bothered to answer the same question twice” – because it’s the above that I recall seeing as an explanation/reasoning on Meta for SE’s/SO’s close-as-duplicate functionality.

If it were only the latter then your argument would make sense, i.e. “move over, old-timer, and let somebody else try to answer it.”

I can’t speak for SO since I was hardly there, but on other SE sites these kind of closures are essential to keeping the place clean.

But, it’s not clear what point you are trying to make. Clearly the software needs to have the ability to close questions. What the criteria are for closing is site specific, so not a matter for software design.

Yes that’s right.

Though specifically “close as dup” does have special software support, i.e. to display and edit in the banner a list of the topics of which this one is a duplicate – a UI feature which is enabled only by closing the question.

Yes, the software definitely needs a close as dup feature, which automatically adds a link to the duplicated question. How that may be used on different sites isn’t the point here, as long as the software supports most of the intended uses.

The dupes would be linked, just as they are currently. Furthermore, the S/O system does not prevent dupes, it merely shuts them down (sometimes). The power users are not compelled in either case to answer the question more than once. In both cases they can simply mark the dupe and link to it.

So @benaston was pondering a feature like that except without even closing the duplicate – maybe call them “similar” oslt instead of “duplicate”.

It is probably site specific. I guess it’s about the design of the “game” rather than the software, per se.

A lot of this is a matter of site culture, and communities here will have to set their own guidelines. SO tried to be a site for expert questions and users there are impatient with beginner questions, but one of my main sites, Mi Yodeya, welcomes questions from beginner to expert. We have to let communities decide what they want to do, and push them to communicate clearly about what that is.

Our “categories” feature would allow a site that’s more expert-focused to also support beginners – add a “getting started” category and encourage people writing their first program (or equivalent for your domain) to ask there. By virtue of it being a different category, users would be “primed” to encourage a different mindset. I suspect that part of what frustrates SO users is that it’s all mixed together, so those who don’t want to see basic questions have to “wade through them” to get to the stuff they want.

3 Likes