Let me be clear: nothing is going to kill Twitter because no currently existing service competes with it. I don't mean "they're not as good", I mean nobody is doing what Twitter does, and the sooner people realise this and shut the fuck up the sooner I will stop screaming at my iPhone as I read Techmeme.
What does Twitter provide that nobody else does?
- SMS updates
- IM updates
- An XMPP (jabber) gateway
The first and last of these are not convenient extras, these are the absolute foundation and the bright future of Twitter and why it is so useful, so flexible and so popular.
SMS is the foundation of Twitter
The most important, far and away the most important feature of Twitter is its SMS aggregation feature. It drives me insane that people don't seem to get this. Twitter doesn't require mobile Internet access. It works on all mobile devices already. It gets better the better your phone, but this fact means that as a mobile communications platform its install base is 100% of mobile phone users, which is practically everybody, a market at least 5 times bigger than the market with even the most basic access to the mobile web, and more than a hundred times larger than the market of users of iPhone, currently the only device to provide a mobile web browsing experience that is anything but painful.
Any communications platform is powered by the long tail. For every A-list blogger with 10,000 twitter followers there are 10,000 random users who use Twitter to communicate with three or four close friends, and they are doing it on their phones because if they wanted to do it on the web there are a hundred ways to do that already. Pownce is not filling any need whatsoever. Pownce is just a really short Tumblr, and Tumblr is just a really quick Blogger, and you don't need it to be any quicker unless you get some extra utility out of it, like Twitter provides.
But it's the sheer size of the market that makes Twitter, potentially at least, absolutely staggeringly powerful. Mobile phone penetration in the developed world is 100%. Not "100% of web users": 100% of people. And the developing world has phones too. More people can use Twitter than can use the web, and even though realistically very few of them will, it underlines the fact that you have to stop thinking about Twitter as some kind of web 2.0 service. Twitter is not the web; it's global mobile communication for ad-hoc groups, and that's a whole different beast. Friendfeed isn't any kind of competition for that; Friendfeed is not even part of the same ball game. Friendfeed is an aggregator of content, not any kind of communications platform.
XMPP is the future of Twitter
SMS-based consumption is the gigantic market of potential consumers of Twitter. Add to that XMPP, aka Jabber, which is a well-defined standard for instant message passing, and you have a mature and powerful, standards-based way of producing content as well. This is what truly makes Twitter a platform, not just a website. Companies or organizations or services that want to push content via SMS no longer have to go through the gigantic mess of contractual and technical issues involved in negotiating with carriers (or even SMS aggregators like mBlox), because they can just layer on top of Twitter, and users can "follow serviceNameHere" via SMS without ever needing to know what Twitter is or how it works.
Push content via SMS is enormously valuable. Real-time information is like that. And Twitter is a disruptively easy and cheap way of doing it, a classic case of a disruptive technology: so cheap and easy it doesn't matter that it basically sucks right now, it's just barely good enough, and it will get better over time and overtake all other solutions.
Caveats et al
But wait, some will immediately cry: SMS only accounts for 5% of Twitter posts! SMS is irrelevant! But this misses the point: content production is only a fraction of total activity. People will post from desktop clients because keyboards are often quicker and easier, but they are reading posts on their phone. And the vast majority of users of any social application are lurkers who post nothing but listen plenty. The posting statistics conceal the real pattern of Twitter usage.
Others will point out that Twitter is mainly Jabber, and "all" you have to do is layer Jabber on top of an SMS gateway. Firstly, I can say from hard experience working at Boltblue that even if you're using an SMS aggregator service, managing an SMS gateway is an extremely difficult job, and the fact that Twitter has only 16 staff but manages to do this plus anything else is a source of constant amazement to me.
Finally, it also sort of amazes me that the superhumans who are managing to juggle an XMPP and an SMS gateway with anything approaching reliability are apparently finding it so difficult to keep a mere website running. Twitter's reliability is terrible, and I make no excuses for it. But they are sitting on a gigantic goldmine, and they have been sitting on it for over a year now without anybody introducing anything that even attempts to compete with their two core features. I refuse to believe it's because nobody has thought it worth trying, so it must be that they've managed to do something really difficult, and astonishingly well.
This is Twitter's game, and it's not even theirs to lose yet. So far they the only game in town.