What are we trying to build?

Codidact Platform Ideas
61

It seems to me that gold tag badges seem like the closest that SE has to point to ‘topic domain experts’. That being said, @gilles’ excellent point about rep/voting having to do with popularity (versus usefulness) still stands there.

Pluralsight and GH might make good proxies or they might not. They also add an external dependency, and relying on outside sources to mark VIPs on our service seems like it might be problematic.

It’s also programming-centric. SO is the ‘800lb gorilla’ of the SE network, but we’re starting from scratch and I think it would also be problematic to focus on solutions for identifying programming topic experts at the expense of other sites- especially as we might try to attract smaller communities in the early iterations.

Special marks are an area to tread lightly in too. Twitter had issues with ‘verified users’, where some mistook the blue check mark as some kind of official endorsement. if we do identify VIPs, it should be very obvious and transparent as to what their knowledge is and why they are so designated.

Despite what I’ve said, I would also dearly like to see a system that works to reward useful content, which encourages and values contributions from topic experts.

6 Likes
62

Great feedback everyone. I forgot about the gold tag badges - how are those "earned’ or otherwise assigned? Could we use something similar here?

Yes, agreed that SO and SE are different animals. Expertise there is very different and seems to be based on more of something like experience or life experiences or exposure. The SE sites I believe rely on more experience or citation I think? Not all that different actually from code-related SO - if you establish someone as an expert, they’re an expert and provide the best answers.

1 Like
63

From the description of Gold badge (copied from the math.SE help site):

You must have a total score of 1000 in at least 200 non-community wiki answers to achieve this badge.

So it’s based on the number of answers and the number of answer upvotes/downvotes in those tags.

I think a reasonable deviation would be to only count the upvoted answers.

Another reasonable deviation would be to demand a certain median score on your answers in that tag. That way you can’t get the gold tag badge when you have many bad answers, but a single one that got voted over the top.

3 Likes
64

Please don’t lock this thread. It serves a fundamentally different purpose from MVP threads. MVP threads are about designing a specific aspect of the system. But we can’t design specific aspects if we don’t know what our goal is. We need a holistic view. And we’re still far from a consensus on that.

Let’s keep discussing the general goal here. Specifics (about voting, rewards, presentation, etc.) should go in their own thread.

3 Likes
65

At this point, the open question is whether our MVP should be a strict subset of Stack Exchange, or whether we’ll deviate in some ways.

I don’t want to build another collaborative encyclopedia. I know where to find Wikipedia.

I want to build a platform for sharing knowledge about topics where the Wikipedia model fails because citations are not available. Science where nobody has published this particular case in a peer-reviewed journal. Coding questions where nobody has written code that does exactly this. Language questions where a native explains nuances that dictionaries don’t capture. Literature questions where someone writes their own analysis of a book.

Reputation has always been there and it’s too entrenched to remove. That’s why it’s important not to put it in. If we start with it we won’t be able to get rid of it.

Jeff has a very high opinion of gamification. He’s clearly driven by it. I’m not and I don’t. Gamification has both good and bad aspects. Rep whores are a thing. I want to take care that they don’t cause too much harm.

I think Codidact will have answer voting, because no one has proposed any better way of rating content. This means that there will be the technical possibility of adding up votes. This does not mean that we have to make all participation scores figure into this total, show it all over the place next to the user’s name, and make it practically the only way to be recognized as a contributor and unlock additional ways to contribute. We can have gamification without shoving it into everyone’s face and making it inescapable.

I have a very different point of view, maybe because I’ve participated on many Stack Exchange sites and not just Stack Overflow, and I’ve been a moderator. The SE model has been demonstrated to work, but it’s also demonstrated many shortcomings. We are in a position to fix some of these shortcomings. Let’s fix what we can.

I fail to see how the fact that people who wrote high-hit questions got high reputation scores has made SE work well. Changing the reputation gained from question upvotes didn’t have any measurable impact on question quality.

6 Likes
66

@gilles - Please list the many shortcomings or pain points you’ve encountered. We can consider these anti-patterns for the design and can serve as a list of things to avoid. You’ve covered some of the issues with voting. Maybe that list goes in the other MVP threads?

I’m only an occasional SO user, but have lurked a long time looking for answers. And I have not read the SO posting rules/faq because I think the site is broken if I have to read a manual in order to use it. I feel that SO is not at all welcoming to new users and crucifies people when they don’t read the rules of engagement with a barrage of downvotes and on-hold questions. It should do a much better job of guiding the user. I more or less refused to play their game after I learned it was so painful to even use. That - IMO - is anti-pattern #1. Don’t force ridiculous requirements or pain on the user. That may be hard to fix, but if we can fix it, that’s huge.

1 Like
67

@sfors Sorry, I’m a bit lost here. Shortcomings of what?

You do need to read some of the site’s rules before you start participating. But you should definitely not need to read everything. Even more than reading rules, you should lurk a bit before participating, to get a feel of the place.

Indeed the site should provide advice when it’s needed. Stack Exchange doesn’t do a very good job of this. In their defense, doing a good job is hard.

2 Likes
68

Sorry, was talking about shortcomings and pain points of SO.

69

Oooh! I love the mention of Trello. (I should get back into using that for my own projects). Habitica RPG uses Trello to inform users about progress on new quests or other design changes, take suggestions and polls, inform them if something’s rejected, etc.

I also definitely love the suggestion of a more formalized “sandbox” for all areas to workshop questions. Instead of quick Votes-To-Close, perhaps a Vote-To-Sandbox?

Related: Instead of “Close” as Duplicate (which makes the asker feel dumb), perhaps more of a way to fold it into the “accepted” question – right now questions can only refer to prior ones, but if newer ones get added to the initial Good Question, then they can be connected as another way people may encounter this problem. (I wrote this paragraph after the Memex one below.)

One of the things you mentioned that grabbed my eye was mention of Meta-hidden… that reminds me of something I told my students: Use Wikipedia (it’s better than the children’s books I’d have to use to get a basic understanding of a topic back in the day!), and of course the citations, but also check out the TALK page for each Wikipedia page you’re interested in. (And the history can be enlightening too). But the TALK is a great place to see what experts in an area are debating. What areas are contested, and what’s settled? What are the divisions within a topic?

Part of why that’s useful is it HUMANIZES the research. Research and Information is always growing and changing. (I also encouraged them to try to find professors on our own campus researching topics similar to their papers, and to ask R.A.s or other upper classmen about “their research”). Partially that’s to invite college freshmen into the “community of scholars,” partially it’s so they wouldn’t feel like they had to quote giant blocks of text – it was written by “just people” and maybe the researcher wanted to emphasize one finding, or had to spell out details that our little ENGL100 papers didn’t need.

To me, as someone who misses teaching, the human element IS part of making the Q&A stronger. (Quality Assurance for Question & Answer?)

Part of it is diversity of viewpoints (a k-12 educator, a research psychologist, and a linguist can all tell you useful but different things about dyslexia). The fullest answer comes from combining those viewpoints (and more), and building from them all. Each answer suggests a branch in a slightly different direction, perhaps.

I am hoping for something that CAN be collapsed into a very linear Q&A site when people need a direct answer. But I’d also hope that there would be elements that could allow some depth and subtlety. I would like it to be “sticky” like the way starting with one TVTropes link ends up leading to dozens of tabs. (But one hopes that many of the tabs produce more than mere lists.)

Since we’re starting from scratch, I suggest we revisit first principles: The Memex, as proposed by Vannevar Bush: https://www.theatlantic.com/magazine/archive/1945/07/as-we-may-think/303881/

Sorry this is chaotic. :dizzy: I’m just sneaking this in at work, and wanted to throw in my half-nickel.

9 Likes
70

I was thinking about making a Stack Exchange clone the past week. Even though I think it fundamentally goes against many of the points raised in this thread I’ll explain it just so you can have an additional POV.

I was thinking about making a site that would be a live mirror of Stack Exchange. Any content posted on SE is available on the mirror, all content on the mirror is available on SE.
This would enable SE users like me that want a better SE but don’t want something foreign and new to pick up the site. And would mean that lulling users, and sites, away would be easier.

At first in an MVP the mirror wouldn’t lull anyone, as it would be strictly inferior to SE. However it means that we would be forced to write something as good as SE and could fall back to SE for any missing features. After getting the site to a similar level, we could focus on features that set us apart.

There’s many things to dislike about Stack Exchange. One thing I want changed is the ability to get notifications to new review queues. With an open source SE I could post a Meta/Discourse/GitHub/Discord issue, argue my point for why it’s good. And if I get upvotes I can then go out and actually get the change to be made. Rather than waiting on Stack Exchange to never add the feature. (Please note this is merely an example.)

Whilst yes this would limit the amount of possible features, when joint at the hip. It would mean that once the split happens then you can start focusing on where to go as a community.

However, understandably, it looks like this isn’t a desired path to go.

So why is this important to me? Well the site I main at is fairly independent of SE. Most changes by SE don’t affect the site. It’s also been fairly immune to the recent problems. This means the community’s a bit in a bubble, and not many are really disgruntled.

There’s a fairly real sense of community. Whilst that’s died down a bit of late. We have a lot of regulars that have known each other for years and have had friendly banter for many of them. Many know each other outside of the site, and talk about their ‘real’ lives with each other.

The site is either; in a small state of decay, that the community would like to fix, or it’s starting to recover from this decay. People have expressed feeling on how to improve the site and generally, it seems, everyone’s been doing their part to address this for over a year. I also feel HNQ exposure has helped bring in new users that have helped with the levels of zombies.

Finally I doubt their would be any content if we moved to Codidact. We very rarely get users that post more than a handful of questions. And those that do, normally are help-vampires. This would require the 99% to move over too.

And so I feel the sense of community would prevent me and others from really migrating away.

I feel I’m part of the target market for why you’re building this site. But I don’t think I’ll be a contributor to the site because I won’t be able to contribute (code reviews).

How are you going to lull question askers over? How are you lulling answers and curators?

Furthermore, from some of the things said in this thread I’m thinking the outcome of Codidact will be more like a mix of Reddit and Quora rather than Stack Exchange. At which point why would I choose Codidact over Reddit or Quora? I can review Python on Reddit, and I probably can do the same on Quora.

Whilst I’ve talked about myself and my site, I’m sure there are other people and sites in my position.

71

@Peilonrayz You are actually raising a number of issues that have been discussed already - so I don’t know that you are so far away from what we are currently stating as goals of Codidact. In particular:

There has been extensive discussion about doing something very similar. Mirroring both large chunks of existing Q&A and also feeding in new stuff from SE “live”. Nothing is definite yet, but there is some real possibility of that, pending license and other issues (which have gotten complicated at SE recently).

There has also been quite a bit of discussion of the possibility of hosting specific communities from SE that have been considering a move anyway, in some cases for quite some time. The catch is that while we might easily get most of the top participants, in most (not all) communities the new questions largely come from outside - new users who find SE via Google. It will be a long-term challenge to figure out how to get new users to an effectively unknown site.

So there is much to think about - don’t give up on us yet!

4 Likes
72

If we do it that way, I don’t think we’ll go anywhere. It will take years to reach feature parity. Will the people involved in the project stay on? Will potential users still be interested by then?

I favor building something that differs from SE in some aspects partly for technical reasons and partly for marketing reasons. The technical reason is that there are things that have proven not to work well on SE and this is a chance to do better. The marketing reason is that this gives us a chance of having advantages over SE early on. Advantages beyond governance, which a vast majority of users don’t care about.

4 Likes
73

I was reading this this morning and combined that in my mind with the above input regarding noise from a few folks above. In particular, I liked the answer with one heading “Moderation, down voting and close voting and feelings”.

It would be nice if we could distill the Q&A into a very focused, noise-free and to the point repository of knowledge, maybe like a wiki. Yep, I know, that sounds like a lot of work for moderators. I’m not sure if any of you were thinking about something like this, but noise and ability to search through the fluff is not easy. That’s one thing that SO did better than forums, but it still has major issues here, especially when there’s millions of questions. The core issue that SO solves is ultimately quality answers. Just brainstorming about how codidact could provide that service better than SO. I’d also love to see a distilled site/wiki that includes references maybe to the best/most valuable questions/answers.

Another idea - is it possible for us to distill some of the top Q&A from SO - or other sites, maybe even from github issues - into a sort of wiki that cut through the noise?

5 Likes
74

Similar ideas have been discussed a few times. A Wiki and a Q&A are fundamentally different types of products, yet there is overlap in the “repository of knowledge” concept.

Another example where some variant of simple Q&A would help:
On SE DIY, a couple of users (I’m one of them) wrote up essentially an FAQ for replacing switches & receptacles, a topic that generates a ton of duplicate and almost-duplicate questions https://diy.stackexchange.com/questions/168564/first-time-changing-switches-and-outlets-receptacles-anything-special-i-shoul However, a day after the last change it has scrolled out of view, new users don’t know what to look for and effectively it becomes a wasted effort. Having a way to keep certain knowledge readily available & up-to-date and obvious to new users would be great. The same could apply to the top Javascript (or any other language) questions or the most common Travel questions (typical passport & visa issues), etc.

7 Likes
75

Yeah, almost like a sticky. But, dude, you forgot to mention the catchy phrase “black to brass.” : ) I’ve done a ton of building, some electrical and a lot of DIY and that’s a cool site, but I just haven’t contributed much. Too busy DIYing it…

I agree that we need to prevent duplicates or provide users with a way to easily find like-minded questions. It’s a problem, and that’s why I was trying to think of ways to distill answers. I still think the Q&A can grow into a monster and it’s less useful as a result because there’s so much noise.

Yeah, I have read the references to wikis, because I think we’re all thinking of better ways to do this. Someone else mentioned discussion too, and I think that’s part of the SE culture - with users that like to discuss things - and that’s very different than creating a knowledge repository. We need to get back to the core goals of what we’re trying to accomplish. Because those two things are in stark contrast to one another.

3 Likes
76

I saw a really good, detailed wiki-style compilation of information on a particular topic on SO once. It was in a tag wiki. That seemed like a good place for it; if you’re interested in that specific topic you can find it via the tag, regardless of how recently the post was edited or how many others like it there are. (Approaches that rely on bumping or pinning only work if the number of such posts is very small.)

On distilling Q&A into focused wikis, I’d like to point out that Mi Yodeya has produced four publications drawn from our Q&A, and while I’m delighted by the results, the projects required a lot of work. When targeting people surfing the web (our publications were meant to be separately distributed), I think it’s better for a site to have one destination and make it the best it can be. Why send people to a hand-tuned wiki page derived from a Q&A page instead of sending people to the Q&A page itself? If the page is noisy, clean up the noise instead. If the topic is complicated and it really does take 10 different answers to elaborate all the issues, you’d need to have that complexity in the wiki too – so think about sorting.

6 Likes
77

In addition to the above thoughts on wiki-like repository of knowledge, there’s also the concept of ‘canonical posts’; as in “topic X keeps coming up again and again in similar guises, so here’s all the possible As to that situation’s Q”.

Posts like the a wiki-style one, FAQs, or canonical posts might be excellent candidates for ‘featured posts’, a feature idea I think I mentioned somwhere…

5 Likes
78

I read from several comments that the discrepancy between wiki encyclopedia and wiki q&a seems too large.

But possibly we can make a tiny change in organization of the q&a to make it look more organised?

(I’d say that an important difference -which we may wish to copy - is that wikipedia has a more clear quality control on what is considered a good/bad article; also the connection between articles is much more tight)?

Currently SE/SO is a bit like a wild landscape, and biodiversity is decreasing. Too much relies on the automatic organisation (which is not happening and it is all questioning answering without caring about maintaining the database). We do not need to turn it into a botanic garden, but maybe we can make it a bit tidier at least?

(This all depends, on whether you want to go into direction of do-my-homework-q&a or high quality-q&a. Maybe we need a bit of both and just separate better. The homework is a good activity for newcomers, or at least, for me SE was a site where I learned from doing other people’s homework. It is not a bad activity, but just not good for a database of clear Q&A that can be generalized. Maybe, all those specific questions could be grouped in clusters, with a more general canonical question as leader and the homework type questions become examples of the canonical question, e.g. on CrossValidated we have many questions about computing a maximum likelihood, but everytime with a slightly different function)

3 Likes
79

Do you have some specific ideas?

Grouping related Qs on a topic together in a closer coupling than tags give is something to consider, down the line. Of course SE does this automatically for intra-site links in Q and A bodies (and comments (?)).

In my view there is a potential scope for those who want to help curate (edit, link, group, generally refine and increase quality) and may be something we want to think about early on.

There is an obvious advantage to having a culture that promotes improving quality, in that quality brings people, references, participation.

As you rightly point out, the former is useful for newcomers. An elitist mindset discourages a lot of people; though I am mindful that those from a high-traffic site (like SO) will have a different perspective on average post quality than those with less traffic (everywhere else ;-P).

For my money, it’s fine to aim to strive for high-quality. But there are a lot more learners than there are experts; we should welcome useful content from anyone.

6 Likes
80

This is problematic because it (different levels of content) may not work together.

A possible solution (I guess there might be many and a brainstorm might give dozens) is to provide hierarchical levels of questions by which they are linked (canonical question - standard question - specific question). Then all the homework stuff or other highly specialized questions can be allowed when they can be ‘attached’ to a more general canonical question.

  • The specific/homework stuff will have non-specialist but handy contributers helping out in a simple way. (I belong in this category. I regularly answer low hanging fruit where I am not even an expert but can help out anyway after some little research effort. I like these questions because it makes me learn look-up experiment with things)
    (e.g. debugging a specific code)

  • The standard stuff will be questions that are neither canonical nor specific. These will be often more theoretic general or deeper questions and require interested or experienced contributers ( these are not in a hierarchical relation ). They have some broader extent and other users may stumble on the same problem
    (e.g. a typical bug or programming task)

  • The canonical questions will have answers with much broader applicability. But, more specifically they function as an umbrella to capture the essence of a theme/topic that is reoccurring a lot in many other (homework) questions.
    (e.g. a typical error code)

Currently about half the questions on SE should be ‘specific’ and might need to be required to be placed as being linked to one or more ‘canonical’ questions. The canonical questions will be usefull to explain content from the specific questions in a more abstract way. The specific questions will be usefull in providing practical examples for application of the topic in canonical questions.

This is definitely not MVP but it might be useful to think ahead of time to organise everything to keep the options open and make it such that it can be easily reorganised at a later time.

1 Like