Admitting you are wrong in public is deeply unpleasant and fundamentally good for you, like a colonoscopy. So here goes: I was wrong about Twitter.

When the microblogging service launched in 2006, 16 years before it was bought by Elon Musk, I thought it sounded pointless. Co-founder Jack Dorsey described a tweet as “a short burst of inconsequential information,” and I thought, that sounds about right. I probably said this publicly somewhere—most likely Facebook or some blogging platform. What’s the difference, I remember writing, in a post lost to time and the bowels of The Internet Archive, between a tweet and a Facebook status update? Aren’t they basically the same thing? Here’s what I had for lunch: blah blah blah. And why join a platform that only allows you to write a sentence or two at a time—nothing longer than 140 characters?

What I didn’t understand then, but came to soon after, is that Twitter worked in part because of the enforced brevity (even though the character limit is now 280 characters). Yes, there are images and videos, but it’s fundamentally a writer’s platform, requiring the user to be pithy and concise.

I’d been an early adapter to blogging (and co-founded a website called Gawker that eventually died at the hands of a billionaire named Peter Thiel and my younger brother’s childhood hero, Hulk Hogan). What I’d loved about early blogging was its new mode of communication: Linking to another blog created a kind of meta conversation with other people I’d never met. You could learn new things, meet people you’d never otherwise encounter, and have conversations that would be difficult to engineer offline.

Twitter, in its ideal form, was exactly like that. Even better. It rewarded intellectually challenging arguments, articulations of shared experience, parsing of big questions, and funny jokes. I’ve made friends on Twitter who became my friends in real life. Many times, after talking through a point of disagreement with someone else on the platform, I came away understanding their point of view better, and with a more nuanced understanding of the issue at hand. It’s no wonder that conversations on Twitter drove profound societal shifts such as #MeToo and the movements for trans rights and visibility, and that it was a place to organize and consolidate support for protests such as Black Lives Matter.

Also, we all soon came to understand, Twitter is a cesspool of disinformation, gratuitous sniping, and harassment. Perhaps that shouldn’t have surprised us: The public square is not always a civil place, and plenty of people have been burned at the stake in one. But it soon became clear that the platform many began to refer to as a “hellsite” rewards provocation, and nurtured racism, homophobia, bullying, and misogyny—while amplifying it to unprecedented proportions. Twitter, like other social media platforms, has also shown that it cannot combat disinformation deeply or quickly enough to prevent it from destabilizing democracies. And the ability to be anonymous as a user is a double-edged sword: good for whistleblowers, and also good for trolls who want to harass others or create chaos.

It’s no wonder that Elon Musk wanted to buy it.

Now that the Tesla and SpaceX CEO has bought Twitter for the inflated price of $44 billion, he seems intent upon burning it to the ground. He has fired crucial personnel via email; reinstated accounts that were banned, including many with a right-wing bent (including the account of former President Donald Trump); and eliminated much of the technological infrastructure that is crucial to Twitter’s continued ability to function, including many of the microservices that power features such as two-factor authentication. Musk then realized he fired some people Twitter actually needs and frantically had managers try to recruit them back. We know that he’s done all of this in part because he has Tweeted about it, in real time. He is, in internet parlance, “posting through it.”

Billionaires generally do not need microblogging platforms to be heard—they get plenty of coverage in the traditional media. So to understand the appeal of owning Twitter, it’s worth considering how it has affected the way we process the world. Twitter’s daily active users are just a small fraction of Facebook’s, but for better or worse, a lot of public figures are on Twitter, and so are a lot of journalists and other people who shape narrative constructions of what we perceive to be reality. Whether about public health, global conflicts, or whatever Harry Styles is doing at any given moment, a lot of people get their news from Twitter, as well as their ideological worldview.

This is both good and bad. It has been used to organize and support revolutions—most famously during the Arab Spring of 2011—and also to disseminate the lies of QAnon and many other conspiracies. During the pandemic, it was useful for distributing advice about masking, social distancing, and vaccines—but also a vector for anti-vaxxer propaganda. It is such a living document of what is happening at any given moment, that if you hear a loud boom through your window, you can probably search “loud boom” along with the name of your neighborhood, and find out what caused the sound in minutes.

Twitter is, put simply, one of the most powerful and expansive mechanisms for information discovery and distribution on the internet. If you want influence, if you’re interested in reshaping how the world thinks about things, you’re probably on Twitter. If you want to dominate the discourse and squash anyone who disagrees with you or thinks your jokes are unfunny, you might want to own it.

And for those of us who have spent significant time on Twitter for 16 years, it’s hard to imagine what the world will look and feel like without it.

Will we have to find out? Right now, the site is a slow-motion train wreck. Monetization is one of Musk’s biggest challenges, especially given the billion or so a year he has to pay to service the debt he used to buy the company. Twitter is monetized primarily through advertising, though Musk seems inexplicably determined to destroy that revenue stream and has been alienating advertisers, causing Twitter to miss its ad sales targets by large margins. Company insiders told The New York Times that the company has cut its revenue projections from $1.4 billion in the last quarter of 2022 to $1.3 billion, and then to $1.1 billion. Since Musk took over, hate speech has become more viral, the site’s content moderation and other operations have been snarled by the epic staff layoffs and resignations, and Musk’s disemboweling of engineering departments are rendering the platform less usable.

As a user, you experience it in glitches: Your notification tab doesn’t work, the site loads weirdly. If you’re a liberal woman writer, as I am, you may notice an uptick in the number of men calling you unspeakable things and invoking the word “rape.” Meanwhile, half of Twitter’s advertisers have left the platform as it has been flooded with tweets that brands don’t want to be anywhere near—racy memes, overtly partisan screeds, and white supremacists whose accounts have been reinstated.

“Frog boiling in water” is the first metaphor that comes to mind, but it’s really more like looking at a Jenga tower from above as the game is being played. You know pieces are disappearing one by one and it’s making the whole thing more unstable. You understand that eventually, it’s going to all come crashing down. But it still looks whole on the surface. Musk thinks this is all going very well because there are a lot of people on the platform. And of course there are. People also gawk at car crashes.

The fact that Musk has used Twitter to amplify and agree with various right-wing figures is not particularly novel or interesting in itself: Media owners have always used their outlets for political and personal ends. Rupert Murdoch doesn’t post through it, but his ideological fingerprints are all over the way Fox, The New York Post, and the editorial page of The Wall Street Journal explain the world.

It’s a shame this is happening at Twitter, however, because that hellsite was often a great place to be, for all of its messiness.

Its mess was the mess of a certain kind of democracy. In theory, the resources available to a user who’s not paying for special features on the platform are the same whether you’re President of the United States, or @john292838374 who joined yesterday and whose avatar is still the default egg. (In practice, there is a bit of a hierarchy in that users with more followers are heard more, and users who were “verified” with blue check marks are sometimes considered more credible because you know they are who they say they are, and are not secretly three weasels in a trench coat with a broadband connection. Musk is dismantling that system too.)

But even when it isn’t entirely sincere, Twitter can have a kind of delightful, retro mischievousness to it. A favorite memory for me is the time when one of my best friends was impersonating me on Twitter under an account titled @wise_spiers. Wise Spiers was an extreme, caricatured version of me: Her bio was “tiny mysterious visionary” and she talked like a very determined and ambitious robot. The account basically existed to lovingly skewer me and advance our inside jokes. I was the editor in chief of a New York newspaper at the time, and slightly to my horror, my staffers followed the account. This was not my friend’s intention, but once he realized it was happening, it only encouraged him. Twitter’s terms of service specifically prohibit impersonation and eventually the account was banned — and I ended up with a blue check as a result. I’d give back the blue check to have Wise Spiers back, though. It was hilarious.

The question now is, will Twitter endure? I cannot emphasize this enough: Musk did this to himself. He did not bother to understand Twitter’s business before he bought the company, did not learn how its technology works despite his pretensions of being an exacting examiner of code, and did not understand how important brand safety is to advertisers. Now he has to make it up.

But I think the biggest existential threat to Elon Musk’s Twitter is simpler: We’ll all finally get bored of it. There’s a sameness to Musk’s provocations that is tiresome. And unless he fixes the things he’s broken, the jankiness of the site will drive users away. Great people leave every day, searching for new digital spaces to set up camp in — and they take with them their strong opinions, fresh ideas, hard questions, and funny jokes.

That’s a shame. I was wrong 16 years ago when I thought Twitter was pointless. It was interesting, and it served a purpose. But it’s becoming uninteresting quickly, and spending time on it is increasingly, well… actually pointless.

That messy, vibrant town square that Dorsey and his colleagues created is not completely gone yet, but I already miss it.

Elizabeth Spiers was the founding editor of Gawker, and started Dealbreaker and several other blogs. She is also a former columnist for Fortune, writing about finance and economics. She has been on Twitter for 14 years.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.