Timeline for answer to What does the public really need from us? by starball
Current License: CC BY-SA 4.0
Post Revisions
23 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 2, 2025 at 19:06 | history | edited | starballMod | CC BY-SA 4.0 |
deleted 25 characters in body
|
| Jan 27, 2025 at 22:44 | comment | added | Slate StaffMod | @NoDataDumpNoContribution Library science is a broad and complex field, and it's the one you're mentioning here. And, yes, there is absolutely a lot we can and have learned from library science. It's come up before in conversations I've had, too :) | |
| Jan 25, 2025 at 10:24 | comment | added | Joachim | Similar to what I was musing about (mostly in the comments) here: combine our current Q&A with more elaborate and more specific content, all of which is intertwined. | |
| Jan 24, 2025 at 10:09 | comment | added | starball Mod | @NoDataDumpNoContribution that's an interesting idea and good background info (I wasn't there for Documentation). maybe there could be rate limiting and a queue of proposals that people can vote up/down on, and periodically, negative score stuff or stuff stuck at the bottom gets removed, and stuff at the top gets published for people to "answer". | |
| Jan 24, 2025 at 9:45 | comment | added | NoDataDumpNoContribution | My memories about documentation were that people mostly wrote about easy stuff, duplicated efforts from elsewhere and wasted time creating silly examples. Instead for quality they want to for quantity and produced nothing substantial in the end. If we go for this it must be driven by demand, not by supply. Like you for example submit a desire to have an example for topic X and others can then vote on it. Used to be there even, but somehow didn't work, I think. | |
| Jan 24, 2025 at 9:28 | comment | added | Resistance Is Futile | It could be helpful if we could seed self answered questions where question itself would not give any reputation, but answers would. And where the question could be just a title with short description. Something similar to how Documentation had proposed topics which were then written by people who knew about the topic. Such questions would then allow everyone to contribute and fill the gaps in knowledge and cover simpler questions that might not fare well when asked otherwise. | |
| Jan 24, 2025 at 9:26 | comment | added | Resistance Is Futile | I like the Q/A organization as it gives opportunity to everyone to write their own answer and where the best can rise to the top. Articles are not competitive in that regard and this is why I don't like them. this was also the problem with Documentation where you couldn't write your own content that would be individually judged and there were bazillion useless edits because those edits would bring constant inflow of reputation to the editors. | |
| Jan 24, 2025 at 9:19 | comment | added | Resistance Is Futile | @NoDataDumpNoContribution Curated list of questions plus additional content. For instance in Documentation there was a Introductory section where we could put start up guides and tutorials. Tag Wiki could be a starting point, but now it is restricted to a single file, while for proper introduction and organization we would need more. And not all major tags have Collections which offer Articles. Also we would need Wiki Articles that don't have reputation associated, so that there is less incentive for abuse in that area. | |
| Jan 24, 2025 at 9:10 | comment | added | NoDataDumpNoContribution | @Slate "But a model for how they serve to archive and preserve knowledge, a model in agreement with how people think about knowledge, is necessary and challenging...." Isn't it always some sort of graph? We have Q&A, we have articles (Wikipedia), we have books, slides. We can potentially mix them all freely. And while I'm really no expert in this, there are experts researching about knowledge graph / information storage and retrieval systems. Maybe they could help here. | |
| Jan 24, 2025 at 9:01 | comment | added | NoDataDumpNoContribution | @ResistanceIsFutile I see. You mean curated ordered lists of questions. This would probably still not really create a book in the end, just concatenating different posts though. What I did once was sorting a tag I wanted to learn more about by question score and reading the 100 or so highest scored questions. But it wasn't so satisfying, turns out the most popular problems aren't necessarily the most interesting to read about just like this. | |
| Jan 24, 2025 at 7:38 | comment | added | Resistance Is Futile | @NoDataDumpNoContribution That is not what I mean. I meant organizing them around topics within tags. Like books. So you have startup guide, you have basic language topics, memory management, threading... and things like that. So that you can start from one page and work your way through content available on SO which allows you to learn in some structured manner, instead of just searching the sites when you have a problem. | |
| Jan 24, 2025 at 7:28 | comment | added | NoDataDumpNoContribution | @ResistanceIsFutile "organizational structure from which we would be able to reference other posts and organize them in topics." We have related questions and we have tag and we have links. We could surely add something more. Collectives are basically nothing but tag sets. | |
| Jan 24, 2025 at 7:25 | comment | added | NoDataDumpNoContribution | @Slate "Q&A pairs themselves are excellent teaching tools..." Are you sure? Maybe only in some contexts. At university we learned with scripts, books or presentations. Q&A was hardly ever used. I needed it only afterwards. I'm certainly learning from it but often rather as a byproduct. Q&A is for me rather P&S (problem and solution). Just in general: for teaching the questions are hardly needed, only the answers in a way. | |
| Jan 23, 2025 at 22:32 | comment | added | Resistance Is Futile | @Slate You are on to something. I think there are plenty of opportunities here. But only if the company puts knowledge and people into focus instead of AI. | |
| Jan 23, 2025 at 22:27 | comment | added | Slate StaffMod | So I know most of these features have caught flak for one reason or another, but I have to stay open to the possibility that the way we change the knowledge model on Stack Exchange may simply need time and a whole hell of a lot of experimentation to mature. Just like Q&A once did. | |
| Jan 23, 2025 at 22:24 | comment | added | Slate StaffMod | In a way, I think it's something we get "for free" with Q&A. A lot of work went into the knowledge model of Stack Exchange as we know it today, but the fundamental idea (ask question get answer) is already a sufficient model for how people think about knowledge. The flipside is, if we didn't have good Q&A services today, trying to figure out how to bridge the gap from "how people think about Q&A" to "how people use a Q&A service to archive knowledge" would not be nearly as obvious as it seems now. We only think it's so easy because we're sitting on a Q&A site that built it to maturity. | |
| Jan 23, 2025 at 22:19 | comment | added | Slate StaffMod | @ResistanceIsFutile Returned to this comment a few times today. You put me in mind that we need a knowledge model for good additions to the Q&A network. Honestly, Discussions, Collectives, Articles, Documentation - these are all good kernels, Documentation maybe the strongest from this perspective. I know that might be a controversial opinion. Of course these features are not perfect (well, most of them have been quite nascent). But a model for how they serve to archive and preserve knowledge, a model in agreement with how people think about knowledge, is necessary and challenging. | |
| Jan 23, 2025 at 11:31 | comment | added | Resistance Is Futile | I am always coming back to the Documentation experiment. There were various reasons for its failure, but the idea itself, having more organized knowledge and more focused starting point in particular tags (areas) of interest. Some kind of expansion of tag wikis and collectives where broader range of tags around particular technologies or vendors will have some organization starting point. But unlike what we have now, there would have to be some organizational structure from which we would be able to reference other posts and organize them in topics. | |
| Jan 22, 2025 at 22:04 | comment | added | Slate StaffMod | I'd encourage you to flesh this idea out, honestly. At minimum it helps improve the knowledge archive to draw annotated connections between posts. Turn it from being a collection of Q&A to something like a directed graph. I'd be curious to see what you come up with here in the long run | |
| Jan 22, 2025 at 22:00 | history | edited | starballMod | CC BY-SA 4.0 |
added 774 characters in body
|
| Jan 22, 2025 at 22:00 | comment | added | Slate StaffMod | Put another way, Q&A pairs themselves are excellent teaching tools - they help people who have a certain set of background knowlege take it just a little bit farther. And they often gesture at what lies beyond the question that was asked. Especially on network sites. (Also on SO, but less so for debugging or doc questions.) But taken in isolation one alone won't get a learner very far. So the natural line of thinking is, if Question A helped me learn and it gestures at Question B, what little extra bit do I need to really get the most out of understanding Question B? | |
| Jan 22, 2025 at 21:57 | comment | added | Slate StaffMod | Honestly it's something I've thought about a lot, too. We have a ton of questions which are great in isolation. But it's a bit like writing a book without any of the connecting threads. More than just being a better related questions recommendation engine, something like this could really fill a need to help new folks to a technology go from point A to B, and understand how / why questions are linked together. This is a spitball idea and this idea isn't exactly parsimonious as I've written it. But we certainly have the knowledgeable people here to do something like this. | |
| Jan 22, 2025 at 20:49 | history | answered | starballMod | CC BY-SA 4.0 |