![](/_next/image?url=https%3A%2F%2Ffirebasestorage.googleapis.com%2Fv0%2Fb%2Fmantic-markets.appspot.com%2Fo%2Fdream%252FAGLCIJoQuI.png%3Falt%3Dmedia%26token%3D5ec98944-8a28-4e1e-9d77-604e0c72dc26&w=3840&q=75)
There's been concerns: Elon Musk has said Twitter is at risk of declaring bankruptcy, many engineering staff have been laid off, rumors of lurking technical debt that could bring down the site irrecoverably, landlords not being paid on-time, and likely more-to-come.
Resolves YES if Twitter is accessible by web browser as of market close, and any Tweet is able to be posted on the site. Otherwise NO. This market does not resolve NA.
Temporary outages do not count, and if it's unclear whether an outage is temporary or permanent the market can be extended by up to 1 week before it will just resolve NO.
Twitter is any website that refers to itself as Twitter located at either the domain twitter.com or any subdomain of twitter.com redirected to past a login step. The locator is a web browser that follows redirects, not purely a DNS mapping.
A Tweet is either a message referred to as a Tweet or a publicly-visible reply to anything currently(as of 2023-04-04) named a tweet.
A Website is a visual rendering produced in my web browser from HTML, CSS, and Javascript. So DNS, comments in the source code, outside comments by any owners of X, would not count as references to Twitter.
Description changes:
2023-04-04: Defined Twitter and Tweet.
2023-07-24: A web browser is used to convert "twitter.com" to a website, not a DNS lookup.
2023-08-02: Defined "website"
Related questions
This market can't resolve yet because it's based on the state at the end of the year. Elon could always change it back. It probably will be NO though.
To resolve YES, we need Twitter to be accessible and for a Tweet to be postable.
X currently allows posting of Tweets because replies to things that once were tweets are publicly-visible. Example: This is a reply to something that was a tweet of mine on February 19, 2023 that is publicly-visible(tested in a private window). Strictly speaking I would have to post another reply, but nobody's questioning new posts behaving differently from old tweets. Also, posts are still called "Tweets" in the Ad analytics, so new posts would count regardless:
So Tweets are both postable and exist by name. But we also need "Twitter is accessible by web browser".
A website at twitter.com can be Twitter even if it redirects to x.com as long as a web browser produces a rendering of a site that calls itself "Twitter". The recent domain change does not affect resolution.
So every requirement is met for YES except for "website that refers to itself as Twitter". As far as I can see, it does not call itself "Twitter". There's some scattered references in developer documentation:
Case 1: https://developer.x.com/en/docs/twitter-for-websites/cards/guides/getting-started
Case 2: https://developer.x.com/en
Case 3: https://x.com/en/privacy
But every instance of "Twitter" either refers to a company or an old instance of the website. Developer APIs are more of an interface: Manifold used to have "groups", now they're called "Topics", but they're still "groups" in the source code and API, and the current name is still "Topics".
Case 4: The only questionable instance is this one: https://ads.x.com/user/_Mira___Mira_/home
It's not a static page, but the language "new version of account analytics beta here" implies it's an "old version". And help documents already don't count because they're holdovers. I did previously count this mention in a comment here, but that was because the "new version of account analytics" wasn't working so this was still the latest. Since it works now, I don't think I count this. Hopefully it gets changed so it isn't a point of contention.
I don't clearly see a "website that refers to itself as (currently being) Twitter" at x.com . Elon's being thorough. So it's looking like a NO. But we do have to wait until the date that it's being measured at.
You've said if it's "unclear whether an outage is temporary or permanent the market can be extended by up to 1 week before it will just resolve NO." Thus the measurement date should at latest be one week after June 18, the day Twitter.com stopped being available.
The DNS is still irrelevant: The June 18 date is not special.
But you might be right anyways: There's currently no website referred to as "Twitter" located at that domain.
There's "tweets" at that domain, but no Twitter.
So whenever the last qualifying update was deployed, you could argue we've had a "Twitter outage" ever since.
I think I agree with that.
There's no evidence of when the "outage" began, and nobody was thoroughly looking at the site until my comment. So the 1 week timer would be as of 16 hours ago when I went looking, and maybe I should resolve NO 6 days from now.
However that clause only permits "extension"(pushing the close date back) and the resolution is "as of end of 2024".
So by words that doesn't count.
By intent it also doesn't count, because that was intended to avoid "the site is down on New Year's so it's NO" when otherwise Twitter is alive and well.
So I think resolution still has to be "at end of 2024". But the chance of it being reversed seems pretty small. If Manifold moderators or staff say they're willing to change the resolution 6 months from now in the unlikely event Elon renames Twitter back, I wouldn't care about advancing you guys the mana.
Also Manifold changed the loan system. I would've written early-resolution clauses into my market if the rules for the money supply were different. Betting markets combine "facts about the world" with "value of money", which means changing the money without my consent is equivalent to changing my market rules without my consent.
@Gen Is resolving NO and reversing if Elon renames Twitter back something I can offer them? It looks the volume is $65 worth mana risk to Manifold at most. And realistically more like $5. You should consider an official early-resolution advance-mana policy as a replacement for loans: It should have a similar effect on horizons as loans.
It is hard to claim that the recent redirect from twitter.com
to x.com
still counts as being located at twitter.com
, especially since all official references to Twitter are being purged except where it would break APIs. The spirit of OP’s distinction between DNS and the browser URL is about what’s user-facing. Typing in the word Twitter or using an old bookmark is user-generated, not user-facing.
@Mira_ I recommend this resolve NO based on the first criterion (location).
If people can only post on x.com but it is still a derivative of twitter, having posts from before change to x.com, how does this resolve?
@Mira Are you trying to restrict this market to techno experts who understand the language you have used?
Does x.com count as twitter or not?
Answer to that from @Mira @Mira_ seems to be No: https://manifold.markets/Mira/will-twitter-still-exist-by-end-of#rfXosVQy7Tdvn1FSB0Sq
Just waiting for resolve now.
@jacksonpolack https://twitter.com/home still works and doesn't redirect, and the mentions of "Twitter" that Mira pointed out previously still exist, e.g. https://analytics.twitter.com/about says "X analytics" but also "Measure and boost your impact on Twitter."
@jack regarding the redirect - if you're not seeing it, you likely need to clear cookies. it is officially redirecting.
https://x.com/elonmusk/status/1791351500217754008
So... if at market close I can post a message saying "This is a Tweet" and it appears on a Website reachable by a browser by typing twitter.com into the address bar, then this market will resolve YES?
@ErwinRossen Hmm, no response after three months doesn't instill confidence in this markets proper resolution. :-(