No matter what benchmark you look at, Twitter seems to be in a bad way. It’s true that revenue has been growing at the world’s favorite real-time information service, but that’s about the only thing that’s going in the right direction. The user base is flat, losses continue to mount, and the stock has lost more than 50% of its market value since October.

To fix these problems, co-founder and CEO Jack Dorsey has said that he will try to improve some of the kludgy aspects of Twitter, like the way @ mentions are handled, and the company will also focus on enhancing the live experience through apps like Periscope.

For some, however, these gestures may feel a little like getting a cheerful memo from the captain of the Titanic saying he’s planning to experiment with a new arrangement for the deckchairs. Will these tweaks be enough to turn the service’s user growth around or bring the share price back from the dead? That seems unlikely.

Some long-time Twitter watchers and even a few former Twitter insiders believe that a big part of Twitter’s current problems stem from choosing to mimic Facebook’s ad-supported, most-eyeballs-possible business model, when Twitter was never really suited for that.

This model has forced Twitter to focus on money making strategies that irritate users, these critics argue, whether it’s promoted tweets or auto-play videos or even the dreaded “dick bar,” a widely-hated feature that was eventually removed. And despite the company’s best efforts, it looks as though Twitter will never come close to Facebook’s massive reach, even if it boosts its user numbers by including those who haven’t bothered to log in.

At one point, Twitter had the chance to take a different path. Instead of focusing on the advertising model, it could have chosen to become a utility—a kind of real-time plumbing for the social web. It could have been the system of pipes that connected dozens or even hundreds of apps and services and websites, and made them all real-time capable and socially aware.

In a nutshell, Twitter could have been a little like email, or the SMS messaging system. And it probably could have become a little like what Slack is now—the messaging service that has brought social communication to hundreds of major corporations and could be worth almost $3 billion based on its last financing.

Why didn’t Twitter choose this road? In part because the ad-driven, Facebook-style, mass-market vision seemed like the best way to justify a large valuation, and the company and its venture capital investors had their sights set on going public. Many of the decisions the company wound up making were driven by that imperative. Information utilities typically don’t get the kind of gigantic market values that Twitter and its backers wanted.

Early Twitter staffer Alex Payne, the site’s former chief engineer, talked after he left in 2010 about the fork in the road that Twitter faced. “Twitter needs to decentralize or it will die,” he wrote in a letter sent to senior Twitter managers. Payne went on to say that becoming more open was also a socially valuable goal, not just an economic one.

The call for a decentralized Twitter speaks to deeper motives than profit: good engineering and social justice. Done right, a decentralized one-to-many communications mechanism could boast a resilience and efficiency that the current centralized Twitter does not. Decentralization isn’t just a better architecture, it’s an architecture that resists censorship and the corrupting influences of capital and marketing.

Instead of becoming more open, however, Twitter turned in the opposite direction and began closing down its network. After entrepreneur Bill Gross—the man who invented the AdWords business model for search—started buying up Twitter clients and showed signs of wanting to build an ad model on his own for the service, Twitter’s board had what one investor called a “holy shit moment” and started shutting things off.

Twitter rumored to increase character limit

Independent Twitter clients were either acquired or their access to the company’s API (the application programming interface, which allows other kinds of software to connect to Twitter) was restricted, and their use of the so-called Twitter “firehose” was throttled. Scores of developers who had built Twitter into their applications were ignored or actively dissuaded from continuing. Investor Chris Dixon, now at venture capital firm Andreessen Horowitz, said Twitter was like “a drunk guy with an Uzi.”

All of this was done so that Twitter could control the network, and thus the advertising and money flowing through the network, in the same way a walled garden like Facebook does. And that spelled the end of Twitter as a kind of open, federated information utility. Is Twitter making more money than it would have if it had pursued this other path? Almost certainly. But it has also forced it into doing things that have throttled other potential areas of growth.

So is there any way for Twitter to reinvent itself as an information utility, to recapture what tech analyst Ben Thompson once called “the Twitter that might have been?” No one seems very optimistic about the prospect, if only because the company has already invested so much time and resources in becoming an advertising platform, and so much structure and process is now devoted to that.

Sign up for Data Sheet, Fortune’s technology newsletter.

It would be difficult, if not impossible, to dismantle all of that or re-engineer it, especially at a time when the company is still losing money. In the meantime, revenue would likely crater, and the stock price would get even more beaten up. “They’d lose the multiple,” said one professional investor, who asked to remain anonymous. In other words, the stock wouldn’t be as highly valued. “Analysts would see an information utility as being like CNN, and that’s a low-growth business.”

About the only option, another analyst said, would be for Twitter to be acquired by someone with massive reach and deep pockets, such as Facebook. “Their best bet is probably to get acquired and start all over again,” this long-time Twitter watcher said. “I think Zuckerberg is the only one who could pull that off.”

Twitter has been trying to repair its relationship with developers, with the launch in 2014 of a platform called Fabric that lets applications use Twitter authentication for logins and other purposes. But it has been a slow road back. Many developers who were burned by the previous policy are probably leery of trusting the company too much.

For better or worse, Twitter has become the advertising-driven mass-oriented network it wanted to be, and promised Wall Street it would be. But it isn’t large enough to justify the valuation it used to have, and it isn’t growing quickly enough either. Yes, it has $3.5 billion or so on its books, but it’s going to have to boost its revenue per user dramatically if it wants to survive long-term, let alone thrive.