Lawrence 'Larry' Roberts is something of an Internet legend, and for good reason: he helped build it. Way back in 1967 Roberts not only drew up the plans for the Advanced Research Projects Agency Network (ARPANET) which would eventually lead to the Internet, but he also headed up the team that designed and built it.

So when he announces that 'The Internet is Broken' and suggests that we need a radical new routing methodology to fix it, to get around the growing problem of Internet congestion, the network engineering world tends to sit up and listen.

In a hugely technical article for the IEEE Spectrum Roberts proposes a flow-management router to replace the traditional packet data network routers. The problem, as Roberts sees it, stems from the change in how we use the Internet these days. Instead of it being primarily an email and static web-based medium, for which packet data is perfectly suited to the job at hand, the Internet has become a video and voice medium for which it is not. The only reason the Internet has coped so far is, Roberts insists, that "the Internet has been grossly overprovisioned."

Roberts argues that IP packet routers simply cannot guarantee that video, for example, will steam smoothly to a user's computer because "they treat the video packets as loose data entities when they ought to treat them as flows."

Of course, and here comes the rub as it were, Roberts does also happen to be the CEO of a start-up called Anangran which specialises in, er, flow management. Indeed, I seem to recall reading that the Internet was broken or breaking a couple of years ago when the company first came onto my radar and Roberts was preaching from pretty much the same hymn sheet.

So Internet FAIL? No, I don't think so, at least not just yet.

Have something to contribute to this discussion? Please be thoughtful, detailed and courteous, and be sure to adhere to our posting rules.