On Meta SE: June 2023 Data Dump is missing.
On Meta SE: Moderation Strike update: Data dumps, choosing representatives, GPT data, and where we’re holding (Restoring the data dumps is a precondition for ending the strike.)
On Meta SE: Philippe says that the data dump will be restored.
On Meta SE: Aaron Bertrand says the June data dump is complete.

Every three months, Stack Overflow published a data dump of our site's contents to the Internet Archive. This enabled projects like Kiwik, various other projects (e.g. making a better search system so I could find those other projects!!), and enabled us to migrate the site to another software platform in extremis.

This is no longer happening. This will resume happening.
The rest of this post, while true, is currently (and hopefully forevermore) outdated.

What this means?

This site is no longer a community-controlled resource.

xkcd 743: Infrastructures

Stack Overflow has positioned itself to enclose the commons. This move centralises power in the hands of the company: it allows them to hold us hostage. In my experience, this only happens when a company's management expects to need to hold the users hostage:

Here is how platforms die: first, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die.

I call this enshittification, and it is a seemingly inevitable consequence arising from the combination of the ease of changing how a platform allocates value, combined with the nature of a "two sided market," where a platform sits between buyers and sellers, holding each hostage to the other, raking off an ever-larger share of the value that passes between them.

The company has an excuse, so this may not be their intention – but it is the effect, and I don't think their justification is a good one.

Until the company resumes the data dumps and commits to that, I will no longer be contributing exclusively to the Stack Exchange network.1

  • No questions, except if they are available elsewhere.
  • No answers, except if they are available elsewhere.

In addition, I will be cooperating with with the ongoing moderation strike until its conclusion, by:2

  • No voting, except on meta.
  • No curational moderation.
  • No curational edits.

I encourage you to do the same – if not sign the strike letter directly.

If you want to continue using Stack Exchange

If you want to continue using Stack Exchange, please Publish (on your) Own Site, Syndicate Elsewhere. If you need help setting up your own website, you can post a comment or an answer (or contact me elsewhere), and I (or another) will provide it. Neocities is a good web host for these purposes: it's quick, they have tutorials, and they promise not to lock you in.

If you care about any contributions you've made since early March (2023-03-05), make sure you have another copy of them. We can no longer rely on them continuing to be available on this site.

If you've made friends here, and you haven't already, consider exchanging contact details.

1: Apart from working on accessibility tooling. Lacking up-to-date data dumps makes that a tad harder, but that's no reason to throw in the towel.

2: I reserve the right to deal with anything egregious, if I come across it. That almost never happens here, but the Smoke Detector is shut down as part of the strike, so it might become a problem.

  • 2
    And Reddit also shut down my Thunderbird's RSS feed access recently, leading me to lose interest in visiting Reddit for things like /r/rust/. Two major sites stepping forward on Cory Doctorow's enshittification cycle at the same time? This month sucks.
    – ssokolow
    Jun 10, 2023 at 18:36
  • Does the title only mention data dumps to avoid any potential blacklisting of a certain anagram of ‘matedioron sikrte’? Jun 12, 2023 at 6:40
  • @user3840170 We're a small enough site to be able to afford a 200×-fold increase in diamond mod effort, so the new AI content policy won't hugely affect us. None of the current moderators are in a jurisdiction where the binding arbitration clause is valid, so we're protected from creative interpretation of contract law. Rinse and repeat for everything else. But stopping the data dumps affects every member of our community.
    – wizzwizz4 Mod
    Jun 12, 2023 at 14:05
  • Well, with the effective prohibition on moderators removing plausible-but-wrong LLM plagiarism, moderating such content away would require constant vigilance on the part of the community casting downvotes and delete votes. (Which I would like to see, as I noted before, but am not particularly hoping for it.) So it’s not like it makes no difference. Jun 12, 2023 at 15:09
  • 1
    @user3840170 It's still possible for mods to abide by the letter of the secret AI law for sites as small as ours. It might take a few hours; some contacting of experts; some largely-manual searching through Common Crawl, WebText2, Books1, Books2 and Wikipedia for plagiarism sources; a great deal of looking at mod-only user data; and escalating to CMs… per post; but there are three of us, and only ~a dozen posts per day, most of which can be quickly ruled out by virtue of being non-bogus. It would be a full-time job, but doable, and we'd get faster with practice.
    – wizzwizz4 Mod
    Jun 12, 2023 at 15:26
  • 1
    I am not sure we have a reason to celebrate yet. After all, the wider strike demands have not been met. Sure, this is the one that affects this site the most, but without the corporation recognising the autonomy of the communities it hosts, who knows what else they are going to cook up. I see no reason to stand down until at least they reveal the secret AI policy. Jun 19, 2023 at 15:54
  • @user3840170 Seeing as I haven't had much to contribute in the way of questions and answers lately, the only thing I can do (aside from actual vandalism) is stop moderating. Currently there's one pending diamond-moderator flag (about an unwieldy comment thread): that's not a particularly effective strike. Over half the moderation is done by regular users, who are mostly not stopping. If you think we should be doing more, feel free to make a call-to-action meta post.
    – wizzwizz4 Mod
    Jun 19, 2023 at 16:52

1 Answer 1


Making a webpage that search engines treat as a Q&A site

There are other ways to do this, if you want to lay the HTML out differently (feel free to ask!), but this is easiest. This will (according to Google's documentation) make your webpage eligible for search engine Instant Answers:

<!DOCTYPE html>
<html lang='en' typeof='schema:QAPage'>
    <title>Overall page title</title>
    <!-- Disables "desktop compatibility" behaviour on mobiles,
         allowing text to be displayed at a readable size. (optional) -->
    <meta name='viewport' content="width=device-width, initial-scale=1" />
    <article typeof='schema:Question'>
      <h1 property='schema:name'>An example question?</h1>
      <div property='schema:text'>
        <!-- QUESTION BODY GOES HERE (as HTML, ideally) -->
        <p>This is an example question.</p>
        <p>It has two very short paragraphs. <i lang='fr'>Comprenez-vous?</i></p>
      <section property='schema:suggestedAnswer' typeof='schema:Answer'>
        <div property='schema:text'>
            <!-- AN ANSWER GOES HERE -->
            <p>Hello! I'm an answer.</p>
      <section property='schema:suggestedAnswer' typeof='schema:Answer'>
        <div property='schema:text'>
            <!-- ANOTHER ANSWER GOES HERE -->
            <p>I am also an answer.</p>
            <p>All the paragraphs go in the <code>&lt;div&gt;</code>.</p>
    <!-- In theory, you can have multiple questions on one page;
         I have no idea whether search engines can cope with that. -->

You can copy your posts from Stack Exchange using View Source. If you copy other people's posts, be sure to comply with the attribution requirements. If copying from your website to Stack Exchange, please review How to not be a spammer – you are allowed to link to your site, and I would recommend you to do so, but make sure what you're putting in the answer box is a sufficient, standalone answer.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .