Twitter acknowledged Tuesday that "from a site stability and service outage perspective, it's been Twitter's worst month since last October." It's a big embarrassment for a company that, over the past year or two, has managed to clean up its reputation for technical instability and that this spring one-upped critics by unveiling a business model that looks like it might actually work.
"Last Friday, we detailed on our engineering blog that this is going to be a rocky few weeks. We're working through tweaks to our system in order to provide greater stability at a time when we're facing record traffic," the post by Twitter representative Sean Garrett read. "As we go through this process, we have uncovered unexpected deeper issues and have even caused inadvertent downtime as a result of our attempts to make changes."
These "deeper issues" tap into something that's, well, even deeper than that. Twitter may soon be faced with a choice: become a long-lasting, crucial part of the Internet's fiber, or continue down a path toward corporate profitability that could seal its fate as more lucrative but less legendary.
In his blog post, Garrett insisted to concerned Twitter users that ultimately, these changes will make the service more stable. For its 190 million users, that's a small reassurance. Many of them weren't yet using the service in its much less reliable days and hence won't be quite as forgiving about how far it's come. But this is the smallest part of the issue at hand: Twitter probably won't fall victim to a Friendster-style exodus of fed-up members leaving for greener pastures. It's bigger and has achieved cultural resonance far more significant than Friendster had when it "lost out" to the likes of MySpace and Facebook. Meanwhile, Facebook's attempts to make members' "status messages" more public and more searchable has not curbed Twitter's growth.
The real problem--why Twitter's ability to stay afloat is still under question--is highlighted by its occurrence as two massive global news stories have taken root--the growing disaster following the explosion of the Deepwater Horizon oil rig in the Gulf of Mexico, and on a less dire note, the quadrennial FIFA World Cup soccer tournament. When there are big headlines of international interest, it's understandable that more news junkies would gravitate toward Twitter more frequently, and more media outlets would want to harness it as an easy way to tap into conversational zeitgeist.
Short-form, rapid-fire digital communication, particularly in a form easily translated to rudimentary mobile devices, is here to stay, and Twitter will invariably be remembered as the company that birthed this phenomenon. But we've seen before that the company has been thrown into situations for which it was neither technologically nor logistically prepared. Last summer, when Twitter was such a key medium during the shaky aftermath of the Iranian presidential elections that the U.S. Department of State requested that the company postpone server maintenance, it became clear that Twitter was going to have to forsake some of its laid-back attitude and get into business mode.
Maybe I'm reading into the situation too much, but I think Twitter's at a crucial point once again. Instead of being on the cusp of deciding whether it wanted to be a quirky San Francisco social-media experiment or a global communications tool, right now Twitter seems to be at a point where its management team needs to decide whether it's a communications tool, or the communications tool. This quandary could get in the way of its plans to finally turn a profit: is Twitter going to be the company that makes money when Starbucks shows you "promoted tweets" if you complain about needing caffeine, or is it going to be truly universal, a technology as lasting and ubiquitous as e-mail?
If it chooses the latter, as Mathew Ingram at GigaOM noted on Tuesday evening following remarks from Web pioneer Dave Winer two months ago, this could mean that Twitter could distribute its technology globally and put the burden of access and stability on companies that would be accessing it to run their own consumer and enterprise services. "If it's as important a service as it seems to have become (or is becoming), should it be looked at as part of a larger infrastructure, the way Ethernet or TCP/IP was in the early days of the Internet, or like IMAP and POP for e-mail?" Ingram posited.
As Winer wrote, nobody in the history of the Web has really been able to have it both ways. The closest example he could pick out was Netscape's open-sourcing of Mozilla, "I don't think the founders (of Twitter) are such daring thinkers as (Netscape founder) Marc Andreessen was back then."
The catch here is that in the most simplistic sense, Twitter shouldn't need to counter its problems of instability by potentially forsaking some of its financial goals. There are only so many people on the planet, and its growth won't be exponential forever. With the right kind of infrastructure, yes, Twitter will be able to handle even the biggest sports tournament's worth of tweets.
Instead, it's the philosophy behind it that comes into question: whether it's logical or right for such a vital communication protocol (whether it's disseminating calls for help from earthquake victims or cries of frustration from soccer fans) to be controlled by only one company. Qualms about Facebook's security led to support for an open-source alternative that has since raised about $200,000 from members of the public. Granted, the hysteria over Facebook's privacy has calmed down in the wake of changes on behalf of the social network as well as the realization that this privacy "scandal" may not have affected the majority of its nearly 500 million members.
But the young developers behind Diaspora, the conceptual Facebook alternative, say that it's not just about a single privacy scandal, it's a belief that "the social graph" should not be owned by a single company. The same issue could apply to Twitter: should short-form global text communication be owned by corporation? Or, like e-mail and telephone technology, should it be an interoperable protocol that can be a worldwide product that might be trading in a short-term billion-dollar valuation, but could be getting a decades-long shelf life in the process?
Winer might have put it best back in April: "I'd like to see Twitter trust the universe."