16

Update May 8, 2025 - This experiment is now live.

As previously announced, we’re exploring new ways for community members to share high-quality, community-validated, and reusable content, and are interested in your feedback on the idea of long-form technical content. During an internal conversation after reviewing research that showed excitement about the potential of user-generated long-form content, it was asked:

What if we had a place that could serve as a formal, structured presentation of facts, analysis, and research, offering in-depth insights on topics that matter to developers?

We hypothesize that community members have valuable knowledge that goes beyond responding to user questions. Therefore, we believe that providing a dedicated space for long-form technical content (e.g., tutorials on coding tools, detailed breakdowns on how to approach and solve common programming challenges, and insights into development best practices) offers an opportunity to capture this knowledge while preserving our joint commitment to quality.

A version of this idea was initially introduced as a feature within Collectives, designed with a narrow scope and purpose. We are now exploring a more expansive understanding of long-form technical content and what it could be– community-contributed content that operates independently of the framework of Collectives– to shape their future direction. The outcome of this experiment and our subsequent research will help us better understand the community's interest in developing this initiative as another avenue to share knowledge outside of Q&A.

Today, we are launching a general survey to better capture interest and an idea of the value community members see in authoring and reading community-generated long-form content.

Additionally, to test our hypothesis, we’re launching an experiment as early as the week of May 5, 2025, to understand the interest in not only reading but also contributing to community-authored, long-form technical content:

  • A link to “Articles” (the name for this idea has not been finalized yet, so it may change) will be added to Stack Overflow's main side navigation.
  • The link will take users to the Stack Overflow Blog filtered to community-contributed posts.
  • There will be a form for logged-in users to submit their interest in contributing to content like “Articles”.

We will monitor click-through rates and form submissions to understand the community’s interest in long-form technical content, both in terms of consumption and contributions. The experiment will run for four weeks, after which we will assess whether to continue or discontinue it.

We look forward to hearing from you and learning more!

6
  • 6
    A comment by Prashanth hinting at this a while back: meta.stackexchange.com/questions/407025/… Commented May 1, 2025 at 20:15
  • 2
    Could you point out the difference between articles and self-answered questions? Commented May 2, 2025 at 8:26
  • 7
    @GertArnold Q&A has many rules that effectively make it impossible to setup a blog-style article. Besides the fact that Q&A has a very severe split between the question and the answer(s) while an article can weave a more structured narrative around it. Commented May 2, 2025 at 9:05
  • 1
    Nice initiative. Only thing I wish. That it would have happened 10 years ago. It looks a bit as if as long as Q&A worked, people hardly looked towards different game. Survey is a bit long. Next time maybe only 20 questions? Commented May 2, 2025 at 18:41
  • 1
    I am disappointed that the very first article I tried to look at had a glaring error in the title: stackoverflow.blog/2025/04/28/… The article is about self-supervised LEARNING and the title talks about self-supervised LANGUAGE, which is nonsense. I have a strongly negative reaction to technical articles so carelessly edited and I doubt I will spend any time looking at the rest of them. I have many other sources of information that are higher quality. Commented May 17, 2025 at 1:55
  • 1
    @ColleenV Also it's a very short article without much in-depth information and without any links to other resources. Might not be very useful and could even be just an answer to a question from the length. Commented May 18, 2025 at 10:32

5 Answers 5

Reset to default
22

There are several things to consider. Articles as such have mostly attracted low reputation users with no or low credibility and plenty of Articles have been AI generated or otherwise plagiarized. Some have also been merely reposts of the author's blog posts with links to the original which makes them look more like spam than attempts to share knowledge.

Another issue is that one of the best things about current Q/A format is the ability to have multiple answers on the same question (problem) and the best answer can rise to the top. This also concentrates knowledge around a particular problem to a single place instead of having it scattered around.

Currently the greatest issue with the Q/A part is that you need to ask a well formed question which aligns with our rules (should be focused and not opinion based) in order to write the answer and share the knowledge. Articles may offer some possibility to broaden what is acceptable without having a negative impact on Q/A as current rules for asking questions would not change.

To me the best option would be having suggested topics of interest (similarly to how users could suggest topics for the Documentation experiment) and then knowledgeable users would be able to provide answers (articles) based on those. Or a user who wants to write about something could do that immediately without having to cover any of the suggested (requested) topics.

The problem itself could be presented in a sentence or two and would not bring any reputation and then multiple users should be able to write full answers (articles) on the same page (similarly to current Q/A) based on proposed topic and reputation would be rewarded to those, unless OP makes it Community Wiki.

5
  • 4
    I think there should be a reputation threshold. I remember Shog9's answer here. Commented May 2, 2025 at 8:25
  • 4
    @GertArnold agreed. You kind of want to avoid the tutorial effect. In my experience, many tutorials are written by people who know how it works for only 5 minutes themselves and wanted to enthusiastically share it with the world while kind of forgetting that the world needs production-quality information and not enthusiasm-driven information. Commented May 2, 2025 at 9:08
  • 7
    @GertArnold Yes, reputation threshold sounds fine. Although 30K is a bit too much. I still don't have that much and I published 4 books. I can definitely write stuff ;) Commented May 2, 2025 at 12:01
  • Could this not be done with Discussions? makes me think of my two "self-answered discussions" there: stackoverflow.com/q/77979031/11107541, stackoverflow.com/q/78046082/11107541 Commented May 2, 2025 at 21:35
  • 1
    @starball good point re: reusing Discussions. We are first validating the type of content and then looking at how we would implement it. My bias is for reusing existing infrastructure where it makes sense. Commented May 7, 2025 at 19:29
12
  • The link will take users to the Stack Overflow Blog filtered to community-contributed posts.
  • There will be a form for logged-in users to submit their interest in contributing to content like “Articles”.

So what are these posts? Is this new community content? Is it just another place to see Q&A posts? There's a form to express interest in submitting that content, but what about actually submitting content? What will the content model be? Is it starting from the feature set of Articles? What will happen to community-contributed content when the experiment is over? And when you say "high-quality" and "community-validated", what does that mean practically? Can we vote on content? Can we edit it? Who can vote and who can edit?

4
  • 1
    Yes, this is new community content. Regarding the rest of your questions, these are excellent points and align closely with the questions we’ve been asking ourselves as we envision implementation. At this stage, we are primarily gauging interest and exploring possibilities. We are intentionally not prescribing specific details yet, as we plan to refine and work out how everything will function collaboratively during the building process. Commented May 2, 2025 at 18:00
  • 1
    Interjecting :) @starball, I know you have lots of good ideas on many topics, so I would see this post as an opportunity to not just ask questions (thanks for asking these BTW), but also to suggest what you think would be ideal for YOU if this were live. For example, what thoughts do you have about how voting and editing could work? What would you consider A definition for community-validated? We get to brainstorm a bit, since this is an early exploration still. Commented May 2, 2025 at 19:16
  • 2
    @EmmaBee I want to, but I can't picture what sort/genre of content this would be. I could come up with examples of content like this I've seen elsewhere, but I find myself unable to easily answer the question "if people are already creating and consuming this elsewhere, why would they suddenly want to create/consume it on SO?". maybe I'll think about that more, but I don't feel any intrinsic motivation to right now. Commented May 2, 2025 at 21:30
  • 1
    No worries! That's still helpful feedback that you can't imagine a use for it, yourself. Commented May 2, 2025 at 21:40
10

EL&U had a blog for long-form community contributed content years ago that had some really good stuff on it. There's a write-up of some of the challenges they faced.

Are articles going to be another feature that only Stack Overflow gets?

How are y'all planning on addressing the participation issues that caused Blog Overflow to be shut down?

How are the articles going to be curated to ensure they aren't just AI/engagement slop?

1
  • 3
    As of right now, we have only plans for Stack Overflow and have not considered rolling this out to the rest of the network. However, if this initiative proves to be popular, we will consider looking into it. As for curation, we are talking internally about ways to ensure quality that reflects what is expected of the current Q&A content. Commented May 2, 2025 at 17:59
4

One possible way to provide content for this could be to take a couple of related Q&A and merge them together mixing question texts and texts from all relevant answers to give a more coherent experience. (And I don't mean with AI in case someone thinks about that.)

People may find it easier to learn from that. By removing clutter it might present knowledge more efficiently. It might give a better sense of overview and connectedness of the content, increasing the value of the existing content.

One would then link to the chosen Q&A (giving credit and attribution) and maybe also backlink from the Q&A to these articles, so both formats can profit from each other.

It might require different skills to effectively summarize content instead of producing a particular bit. Maybe more like a Wikipedia like mindset, which we don't really have here. Think about separating multiple answers into smaller bits, remove redundant bits and then re-arranging these and connect them into a single comprehensive text. Consider not displaying a single name below an article but a list because there would probably be multiple authors.

The charm of this: it's a natural extension of Q&A and plays to the strengths of the knowledge base. The backside: it faces stiff competition from AI and maybe people simply do not approve of it. But the goal would only be to have a high concentration of useful knowledge (pearls) that cover more than a single Q&A topic, but still more focused than an average Wikipedia article.

For example, covering a range of git operations in a single article comes to mind. There is a couple of highly scored git Q&A, but the knowledge about git on SO feels not well connected.

Additionally, I could also imagine a place for extended workflows/recipes/best practices guides, which can be even opinionated. Emphasis should nevertheless be on lots of links to existing Q&A, not only for attribution but also for context. This is what others maybe cannot do that easily.

1
  • 2
    I think this is a great idea, connected to another answer's idea of "suggested topics". Commented May 3, 2025 at 13:51
1

I'd love if long-form content, which is already existing on the internet, would have the same quality as Stack Overflow.

With Google and other search sites increasingly being flooded by AI slop, I think voting (including down voting!), editing, and commenting on existing articles could help a lot. Also you'd need less of the hours and hours of volunteer effort this way.

If there's already a good article on a topic: I want to share it, if I find it. I want to know about it, if someone else already did find it. Hackernews, Reddit etc. are only slightly useful for me there.

External content goes away, but the Internet archive has more or less solved this problem.

2
  • medium.com has quite high quality long-form content. However, you could not easily share it here without violating some licenses (unless you also created the content yourself). And replicating that level of quality will require laser sharp focus on quality. Nobody knows if this experiment will succeed. Commented May 5, 2025 at 12:34
  • Would you be allowed to vote on the link and back it up on archive.org? Commented May 5, 2025 at 12:55

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.