Are we primarily helping the asker or building a repository (or sticking to the middle ground)?

Codidact Platform Ideas
41

Not just that. Sometimes OP knows what the “worst” or some of the worse ways of doing things are, and they can be asked to provide that criteria of quality that they seek personally. Then we can generalize and discuss the merits of that quality criteria and arrive at the version of question and answers which list some of the better ways to do X according to a publicly agreed upon criteria of quality. OP might have a starting point provided already.

The thing about “opinion-based” is that the community likes to close the questions where the criteria of quality is only beneficial to a limited group of people, like only some small subset of artists in a niche art style area, or just OP personally. If the criteria is something universally agreed upon, the close votes don’t come, upvotes come instead.

It could be useful to make this distinction clear and treat such questions as “Guide on how to do X for the largest audience” and “Please help me do this just for myself and maybe a few other people”. Both are valid questions, in my opinion and have a place on Q&A sites.

If the amount of questions that helps only a single person and the retention of users asking such questions is deemed too low, they can be prohibited in the future, such as for example identification requests like “What is this music?” with a short clip of music attached or “Who drew this art?” or “What anime is this character from?” or “What game is in this screenshot?”.

1 Like
42

My two cents: We’d be primarily interested in building a repository of knowledge, but that shouldn’t mean that someone who happens to have a highly specific (aka localized) question is not welcome to post them to one of our sites and obtain help. (I answer this with a Stack Overflow bias/mindset.)

I said this already but can’t find it, so here it goes again - as long as OP shows willingness to put in the necessary effort and the question is not, in any form, a duplicate, and happens to conform to all rules set by the community, there’s no reason for the question to be repelled.

“It can’t ever help anyone else in the future” - that’s quite an assumption to make, but even if we suppose it’s true: is the question written in good form, by someone with apparent good intentions (not a help vampire), in genuine need of help? If yes, leave it be, don’t discourage anyone to answer it - the quality of the site won’t suffer because of that. (IMO.)

TL;DR: Middle ground, biased torwards repository of knowledge.

4 Likes
43

Yes, but these are people who are learning how to swim. The site promises to teach them how to swim, and then throws them into a jellyfish tank where they get stung, and then they’re blamed for it.

No, the site (SO) never promised that. Rather, it promised to exchange advise between professional and enthusiast scuba divers. Before engaging on a scuba diving site, people already expect you to know how to swim.

Then Atwood quit and Spolsky suddenly want to turn the scuba divers into swim instructors for kids. Without the scuba divers agreeing with this. And a large portion of them aren’t the slightest interested in teaching kids how to swim, they just want to scuba dive. A swimming instruction site wasn’t what they signed up for.

And these are the people who will be most likely to adopt an alternative platform. The majority of SO contributors will not be joining us - at least not right away.

Not correct. The SE veterans are the most likely to adopt an alternative platform, because they are pretty much to a man fed up with SE in one way or another. The newbies will follow since they will go where the domain experts are.

I think our only real chance of survival is to recognize the user base we have been given, and structure our Q&A accordingly.

And that is not to build a site for inane homework questions, but a site that appeals to the domain experts. Otherwise we are just building Quora 2. SO succeeded because it appealed to the experts of Experts Exchange and stole those users away. The key to success lies in this understanding: the site cannot succeed without users who actually know what they are talking about.

7 Likes
44

Stack Overflow has been trying to strike this balance and, as far as I’m concerned, they’ve largely been failing and the part about building a repository is where we’re losing out.

This also applies to other sites on the network I’m familiar with, especially The Workplace, which has very few posts that can without doubt be used as duplicate targets, despite having a whole lot of very, very similar questions. This is because basically every answer ignores the more general question and focuses heavily on things in this specific question, even when those are only tangential to what’s primarily being asked. Also, no-one seems at all interested in actually making canonical posts about topics or making significant edits to improve the long-term value of a post and most high-rep users insist on helping the asker with their immediate problem with seemingly no concern for or interest in the long-term value of posts. But this probably goes back to rewards and such: why would you want to spend a bunch of time looking for duplicates for no reward (and likely some complaints) when you can instead just answer it for reputation and your personal visibility on the internet, which is probably also faster.

This earlier comment of mine might be relevant:

The above is also a good example of what I mean when I talk about questions primarily helping the answer.

It’s not that we shouldn’t help users with their on-topic questions because we don’t see the future value, it’s that both the question and answer should be general (and specific) enough to potentially help others with similar problems, even if this means losing out on the specificity the asker might prefer or even “need”. But if the question is too far from salvageable, closing it may also be appropriate.

And it also comes down to saying “yes, that is a duplicate”, even if the asker isn’t happy about it and that means they don’t get a 100% direct answer to their question.

In some cases we may also decide some specific subset of questions are off topic because they just fundamentally can’t help others with their similar problems (but I’m not sure such questions even exist).

The problem with allowing questions that aren’t quite ideal is some combination of:

  • a low signal-to-noise ratio: there are just too many such questions and a user seeking to build a repository won’t want to stay here. Also, some such questions may end up ranking higher on Google than a “proper” related question.
  • the broken windows theory: people see many bad questions so they think their bad question would also be fine and others try to get rid of fewer bad questions and improve fewer questions that could’ve been great.
5 Likes
45

Here’s my two cents as a long-time contributor to the Photography site.

First, there are large number of perfectly wonderful discussion forums out there, many of them with help sections. It’s not useful to duplicate those. Having a curated source of expert knowledge is of value.

I’m not at all above helping newbies — this doesn’t need to be a rarefied expert-to-expert site. Photography as an art is growing like never before; as an expert-domain profession, it’s almost collapsed. Let’s gather knowledge and help make more experts (and capable intermediate photographers on a learning path).

At the same time, I’m not super interested in one thousand “what filter was this” questions (where the answer is usually “the thing you think is a filter is some combination of skill, setup, and standard post-processing”), nor by a bunch of “I don’t want to learn anything, I just want to buy something that makes my pictures better” questions.

But, to flip again: the SE site has been pretty stringent on turning away equipment recommendations. There’s good sense in this and it’s help us avoid the brand flamewars that plague other sites. But it’s also problematic because it leaves out a key “feeder population” of users.

And finally, we’ve always wanted more questions around the art and skill of actually taking photographs, and it’s been hard to attract those questions. The engine isn’t really well-suited for it, and often these don’t have a simple, straightforward answer.

I’m kind of thinking the problem is best served by breaking it down: different areas with different expectations.

  1. A knowledge-base focused area for learning about aperture and ISO and focal length and all that, plus standard lighting setups, techniques for posing a model, and so on — topics with definitive answers. This is different from Wikipedia because Wikipedia makes a good reference for many such things, but is often a terrible learning tool.

  2. An immediate-problem area for things like “what camera is right for me” and “what lens should I get to go on safari?” There may be some general advice which scales to everyone, but mostly the answers are personal (and not much use six months later).

  3. An area specifically for learning how to take photographs in a certain style, or techniques for covering a type of subject, and so on. This is sort of between the two: answers should be interesting to many people and educational on their own, but there’s usually not a “right” answer. This might even include a photo-critique section. On Stack Exchange, one problem we struggle with is unique titles for these questions; ideally, this could be organized visually by example images, with any title secondary.

4 Likes
46

I (and others) have suggested something of this sort. A “blog” type of posting in that it isn’t Q + A but simply “organized information on a very specific topic”, but edited/updated by the community (sort of like Community Wiki on SE). Your examples about aperture, IOS, focal length, lighting setups, etc. sound similar to the DIY things that we’ve tried. @Harper and I on DIY tried to set up one as a Q + A but it didn’t work well because there was no way to tell newbies “look to this first”.

2 & 3 seem like just 2 sections of Q&A. One option is 2 separate Communities. I know I saw something about a few communities that already have that on SE (Math? Physics?) but I don’t know how many communities are large enough to be split that way, and the line between types of questions may be a bit fuzzy.

3 Likes
47

Physics Overflow has “reviews” built into their site; I gather that a review is something between a critique and a peer-reviewed submission, but my point is that it’s an additional type of content that they’ve decided to support on-site, as opposed to saying “go to this other community over there for that”. I could imagine Codidact supporting more post types beyond the ones we’ve identified for MVP, with communities deciding what to enable. Each post type would have its own “rules” for how that type works; for example, blog posts don’t accept answers, perhaps critiques support anchored comments but not answers, perhaps “personal advice” (like the “what camera for safari” example) posts accept answers but don’t support voting, etc. (I’m just making these examples up as I write; blog posts are the only case I’ve thought about before.)

5 Likes
48

I do feel / agree these sections are largely distinct and trying to welcome all of them (in one place) is probably not going to work that well.

I would say we focus on #1 first and foremost and the others should be secondary (and optional).

I would, however, stick to the Q&A format, since that was the starting point and the direction this whole thing seem to be going towards (and, of course, the format has its advantages, like very directly answering actual real-world questions, which is something the millions of knowledge bases scattered across the internet tend to be missing). A different format may work, but that should probably be another site.

#2 can often be turned into canonical posts (#1) if users (askers or editors) are inclined to do so. If someone wants to know what camera to buy, of course one can say “because you want X, you should look for a camera with Y”. But surely one would also be able to discuss the various factors that goes into picking a camera without being specific to any one asker (or, if that’s too broad, discuss individual factors or features in individual posts, like how much zoom or how many megapixels or whatever).

Of course there are cases where a more personal touch would help, but it’s arguable whether this needs to be supported at all, or at least supported in the beginning (there are always other places to go if such help is really needed). Chat on Stack Exchange seems like it could’ve been a good place to support this, but it wasn’t never all that well integrated into the platform, people tend to chat, as opposed to using it as a place to deal with questions a bit beyond that scope of the main site, and it suffers from the typical chat problem: threads tend to get lost within minutes (if it’s moderately active), or less, and things can be quite hard to follow (kind of like what we have here).

Posts under #3 seems like they would fall into one of the follow:

  • Basically #2. If you can’t make something searchable, it’s essentially useless after a few months or even days. Although maybe it’s a bit different because of the greater short-term value or the potential to have some ordered lists of questions for people to look through.
  • Can be turned into a canonical post, i.e. #1.
  • Would probably be fine under #1 as is. Even if there are many possible answers, or it’s pretty subjective, it can still make for a good question that can be found by and help others. Stack Exchange does support “good subjective” questions, which is kind of what I’m talking about. But some of the restrictions on Stack Exchange regarding what’s allowed (or at least how it’s enforced) has often seemed fairly arbitrary and not all that conducive to building a repository.

Code Review is probably a good example of #3. I don’t have a problem with the site, but it does seem fundamentally at odds with the goal to “build a library” (which is written on the tour page of most Stack Exchange sites). It’s built on a platform intended to be searchable and referenceable, despite the questions themselves being neither. It does make me question a bit what the actual goal of Stack Exchange is, which brings into question the scope of other sites and whether the goal is even building a library or if it’s more just focused on the short-term needs of users. Users questioning these things (or, worse, simply making some assumptions) probably isn’t good for the site.

The next question would be: if these separate areas exist, would users be able to figure out what belongs where? There’ll probably always be some ambiguity about that. I sometimes ask things in chat which I’m sure doesn’t belong on a Stack Exchange site I’m quite experience with, yet other experienced users will recommend asking there. Also, having areas for #2 and #3 may encourage users to ask their specific question there instead of create something more general.

49

This is an important question, and goes back to the general issue of “how do we get users to do what we want, instead of what they want?”

The catch is that “we” is not “a group of motivated experts building a new system” (i.e., the current “we” discussing this right now). “We” is whoever shows up and wants to participate in a meaningful way.

This is absolutely key. And is something that SE has never, IMHO, done right. We need a separate place (call it “canonical answers” or “blog” or “FAQ” or whatever, possibly community dependent) where the experts (and/or motivated ordinary users who want to help) can build this knowledge base of key Q&A. Might not even be Q, just A. I see this over & over in DIY - the duplicates often aren’t so much exact duplicates as “questions that could/should have a master document that explains the ins & outs of ‘x’”.

2 & 3 - the various questions - are a bigger problem. I contend that the typical user comes to SE (and will come to Codidact) to get a clear answer to a specific question. They don’t know how to search (effectively) for duplicates. They don’t know why questions have certain criteria to make them “good”. They just want to ask. And a small number of those people will get their answer and search more and start answering others and eventually become part of the “we” that really runs the site. Education of new users at the very first step is critical. But without being too overwhelming or treating them like idiots.

Along the way, gamification is a factor. I honestly doubt that I would have done as much with some communities in SE without the gamification. It may be that just keeping a “information only” tally of Upvotes front & center for users to see would be enough even without badges & privileges & hats… But I don’t know. I worry that “no gamification” will result in a less “involved” experience for new users and then they don’t stick around to become part of the core.

4 Likes
50

Why can’t the whole place be that place?

Every question that’s worth having on a public site should be a “key” Q&A that’s reference-worthy. Otherwise what’s the point of it being public?

If you try to section off the “key” Q&A from the rest, you’re creating two separate things. Two things that may both be useful, but two things nonetheless. It’s more complex (potentially confusing) and less focused than one thing. One of those already exists in a number of other places (with varying levels of quality and different formats - SE, Reddit, Quora, Yahoo Answers, etc.) - what would be our unique selling point, if not well-curated reference posts? Yes, we’ll have those too, but if it’s not the “core” to build everything around, it will just be some icing on what’s basically the same as everything else (I’ll leave it up to individual opinion as to whether this is a cake or a turd). And if those are the core, why do we need the other part? What why do we want another one? If our goal is just “Stack Exchange with better tools, intentions, management and/or icing”, I can’t imagine we’d end up with a much better site, if we even manage to gain critical mass. The tools aren’t that bad, everyone starts with good intentions, management is objectionable but basically competent and icing is just that. Although maybe, for some experienced users, quality is secondary to not being managed by the same people.

I have in the past advocated for something related: having a separate dumping ground for lazy homework questions on SO, so we can have high quality without having to try to stop people from asking their garbage, but this leaves a number of problems, some already mentioned: people wouldn’t know where to ask (if given the option), the “high quality” section wouldn’t get questions because they’ve already been answered “fine” on the other side and low quality would seep across.

The better solution is really to “just” position the site as a place of extremely high quality from the get go. How? Well, that is the million-dollar question.

Gamification is certainly one thing. SE rewards answering above editing or closing, so it’s no surprise that no-one bothers writing proper canonical questions. Wikipedia seems like a decent example of a community-curated site with much less, if any, gamification (and much, much higher quality). And yes, it can probably get a lot of more contributors with more gamification, but it still works as is (and we can always find some middle ground and award the behaviour we actually want).

It’s probably also way too easy for a (new) user to ask a question on SE and it’s especially easy to do so without reading any of the rules. So yes, education, but also maybe some gatekeeping. Some users don’t want to be educated, and we don’t need to welcome them to be successful, and some sort of review process before a question ever goes live would seem especially promising.

51

I think there’s a difference between a “canonical” or “reference” post about, say, how Judaism interprets the creation story (stories) in Genesis, on the one hand, and a specific question trying to understand why Rashi says what he does about the emergence of plants. It’s all valuable reference, but some of it is way more specialized. That has to be ok. In particular, if somebody asks the latter question, it’s not ok to force everyone involved to instead produce the former – if it happens great, it’ll be well-received, but sometimes we just need an answer on a specific point. That has to be ok.

4 Likes
52

I think my point is that there is a difference between “a highly ranked Question/Answer that answers key things asked by many users” and “a separate reference area with that same information”. The problem is that people have trouble finding a question that closely-enough matches what they are trying to get answered right away. A separate small (might be anywhere from 2 to 50 posts, depending on the community - probably not hundreds and definitely not thousands) set of “reference” or “blog” or “canonical questions” or whatever is, IMHO, an extremely useful idea. It fits into the idea of the Q&A being used to build a knowledgebase. It is in some ways similar to Wikipedia (collaborative editing of “articles” rather than “questions”) but I think it could really benefit the site a lot. For example:

  • DIY: One so far: First time changing switches and outlets (receptacles). Anything special I should know? and Harper (who started it) and I occasionally reference it, but since it is “just another Question”, it is hard for users (e.g., one who just asked Replacing a power outlet safely to find it, unless they know how to search well - and a typical search ends up finding a whole bunch of other questions first. We could easily have more - e.g., explaining basics of GFCI & AFCI, explaining main panels/subpanels, grounding rules, etc.
  • Mi Yodeya - your example of creation stories is a good one. There could be even more basic (less philosophical) explaining (just making this up as I go along) traditional historic timeline of Tanach or some basics about the Jewish calendar

and I can come up with plenty of ideas for Photography, EE and others. The basic idea not to create an entire encyclopedia but to collaboratively create articles that can answer a significant number of new (or old!) users’ questions in an easy to find way. Then if a user says “OK, I tried to replace my outlets with your little guide but I have this complication” then they can ask a question and hopefully be asking from a more advanced level because they already took care of the basics.

I don’t like the idea of simply saying “top 20 questions by votes” get a separate area - that will not work because popularity can be skewed by HNQ (a lot!) or by current events. Maybe these special posts could only be started by very-high-rep users (to make sure they are appropriate for this special status), but everyone (except the very lowest level) should be able to edit & improve the posts.

1 Like
53

Oh, I see now. Does it need to be a separate area, or is it sufficient if there is an easy way to get to the collection? In addition to those references being good canonical references they also provide answers to questions, so I’d like them to be there alongside the other questions. But that doesn’t mean all posts are of the same type; if posts like these are of the “canonical” type (just to choose a word) and we had a link for “canonical posts” that showed you just those, you’d get the effect of a separate area. Meanwhile, if one of them was updated today, it would still show on the front page alongside the questions and blog posts that are also there, so these posts are more visible to casual observers.

While I like that some sites build reference-quality tag wikis for selected tags, you have to know they’re there to make use of them. I’d like all the usual ways of seeing content – main page, search, links from profile pages, whatever else – to work for all types of posts.

2 Likes
54

To make an analogy with academia, in academic journals we have two different types of articles (usually in separate sections of a journal, or even separate journals): There are research articles, and there are review articles. Research articles are about new results in your research: It’s about answers to scientific questions that you just have found. Review articles are collecting results in a certain area (and, of course, referencing those research articles; being cited by a review article is a major boost for a research article).

I would say this corresponds to the situation we are discussing here with Q&A: The normal questions correspond to the research articles: They are about specific problems and their solution, they arise from the everyday work (for articles: current research). On the other hand, the canonical questions/FAQs correspond to the review articles: They collect and sort the knowledge found in the individual questions on a specific subject, and bring them into a form that serves as many people as possible.

Now from the analogy, we can learn several things: First, we cannot expect to have canonical questions right from the start. Just as a review article is only written if there are enough research articles about a subject, a canonical question can only be written after we’ve seen enough related questions so that we know what needs a canonical question, and what should it contain.

Second, the review article doesn’t obsolete the research articles; it references them. If you want to see the details, you go to the research articles. Besides giving an overview, the review article also helps finding the relevant research article. Probably we should do the same with canonical questions.

Third, a review article is a completely different beast from a research article; similarly, a canonical question would be a wholly different beast from a regular question. In articular, the regular questions it is based on would not be closed as duplicates of it (there may be a prominent link to it, though).

Note that this doesn’t mean that duplicate questions would become acceptable; also if you write a research article that doesn’t add anything new beyond the already existing articles, that research article will get rejected (unless the referee doesn’t know about the prior publication, of course — same problem as with duplicate questions!).

4 Likes
55

What I would absolutely support and encourage is some functionality to make common questions much more visible and easier to find.

This would work great on Workplace too, where there are a bunch of really common questions about interviewing and employment and whatnot that everyone in the workforce might’ve once wondered and, especially if they hadn’t, be given the answer to. [Which are, of course, endlessly duplicated, because there’s no proper reference for most of them and everyone just answers instead of going in search of it]

On specific tags (i.e. languages) on Stack Overflow, this may also come in handy to reference particularly common errors or misunderstandings.

There are remnants of this happening in tags wikis, Meta and some main posts on some sites, but nothing really works all that well because it isn’t functionally supported.

I don’t think such FAQ’s should be separated, but rather come about in the natural progression of things.

In some cases, yes, users might need to create a canonical post from scratch that’s a bit broader and more complete than the typical question on the site. Even these should ideally be phrased as actual questions though, and don’t much benefit from being separated either. Although, in an ideal world, they too would come about in the natural progression of things, as users realise they need to keep adding more details for a post to fully make sense and cover the more significant variations of the question.

Perhaps these FAQ’s or canonical posts could be tagged in some way after the fact and/or linked in manually curated pages or lists somewhere. Manual work is best avoided, but it might be necessary because an unordered, unformatted list might not be that useful, even if the posts themselves are extremely useful. The tag wiki is probably a good comparison (despite its weaknesses), as it allows a detailed article to be written about some topic closely related to what’s on the site, but, at the same time, slightly outside it.

I don’t have a problem with either of those.

I should probably stop using those terms interchangeably and as if everyone agrees on the meaning.

What I basically mean is that we should avoid having another over nine THOUSAND questions (meme partially intended) asking for help with a NullPointerException. One is good. Two is too much. And no, your question is not different because yours contains 200 lines of code that isn’t in the other question or because you were wearing a silly hat at the time. [But yes, there may technically be some non-duplicates if this exception happens in the depths of some specific library and is in no way caused by an error in your code, although this is more the exception than the rule]

The important part, to start off with, is to agree on the core content: the questions, not to figure out the specifics of how to support FAQ’s or what sort of articles we should have on top of that. That can come later.

3 Likes
56

That difference was evident on the Buddhist site too.

  • Experts wanting Q&A about expert topics
  • Novices wanting about more basic topics

Two other constraints on the problem:

  • The community was keen to accept all types of question (out of compassion perhaps)
  • The experts didn’t want to answer “seeded” questions – e.g. “How does Judaism interpret the creation story (stories) in Genesis” – saying that feels like editing Wikipedia and a chore … I think they too wanted to answer sincere question from someone who actually wanted to know.

In result we accept novice questions, and though there are experts their expertise isn’t often called for.

The novice questions tend to be quite repetitive. Even though one tries to close questions as duplicate, there get to be so many that it’s hard to be sure.

Back in the day, did people perhaps split into some two communities? Like I think there was a comp.lang.c++.moderated and perhaps another which was more open to novice-level questions.

So that’s a thought.

Furthermore, I wonder if it might be fun, on the novice-level site, to build a wiki – so the aim is the site is to build a structured repository – like a book with chapters and editors, and maybe a “table of contents” as well as or instead of a “tag cloud” or “index” – and when a novice asks a question then the ambition isn’t to answer it directly, but rather

  • Point to them to the relevant section of the wiki which already answers their question
  • Say, “if that doesn’t already answer your question then ask again”
  • If they ask again, then try to edit the wiki to answer their question

Do you think that might be possible/desirable?

Whether the answer is “yes” or “no” might depend on the people and the subject.

I’d dearly like the software to support that, though, as a first-class use case. IOW the ability to collaborate on, edit, and discuss, an evolving large/structured document.

2 Likes
57

Search, main page list of recently changed posts, etc. should show the Canonical posts too.

But in addition, there should be some obvious way to find them. For example, if the right-side has “related posts” then these should be a separate section on the right, probably above the “general” related posts (Q&A), because the idea is that as a user is looking at something they may be inspired to find out more and these posts are a great way for them to do that.

58

Yes to all of this! Review articles, or in our parlance canonical posts, are emergent, and they build on other material with copious cross-linking. I think this capability would be a valuable addition to Codidact.

Implementation-wise, I see it as a different post type that (a) is not “owned” in the usual sense (intended to be group-edited) and (b) does not accept answers, like blog posts wouldn’t. (Assume that all posts accept comments.)

Because it’s emergent I don’t think it needs to be MVP, but it’s something we should be planning for from the beginning, and I think it wouldn’t be hard to add soon after MVP.

Yes, agreed – I’m thinking of some UI affordance, perhaps a button or at least a search configuration, that shows you all the canonical posts. And, as @celtschk pointed out, there should be a way for questions to link to a related canonical post if one exists.

3 Likes
59

a) Like “community wiki” in SE (which I don’t totally understand, but seems to be something like that)

b) Instead of Q + A(n), just one “thing”.

This was actually my motivation for “Blog” posts in Codidact, but I can see reasons for Blog posts as well - they can serve very different functions.

I would also make them both be “per Community”. For example, it may be that Code Golf has no need for Canonical posts or blogs and that Worldbuilding needs Blogs but not Canonical posts and that DIY needs both. Or whatever.

2 Likes
60

I was with you up to here, but this is going to be a problem. Someone spends a lot of time writing something, then suddenly they don’t own it anymore, and worse, others put words in their mouths.

Here is an example: https://electronics.stackexchange.com/questions/28251/rules-and-guidelines-for-drawing-good-schematics/28255#28255. We were getting a lot of crappy schematics, and were getting tired of telling people twice a day to use component designators, draw junction dots, etc. I took the trouble to write a “question”, then spent a decent amount of time writing what I was hoping would be a good canonical answer. Then the whole mess got converted to community wiki, and all of a sudden a bunch of other people started messing with what I said instead of writing their own answers.

I had planned to expand on that post over time, but after that experience I developed a screw this attitude and never added more points to it. Once the zombie hordes were let loose, I quit. Note that the zombies didn’t ever add any points to that answer either. If it had stayed “my” post, it would have significantly more content by now.

Don’t pull the rug out from under people who spent a lot of time on something.

1 Like