I am open to such arguments. As soon as the telephone companies announced their “two-tier Internet” and the network neutrality debate began, I published an article that criticized the telcos but showed that the issue was much too complex to be solved by a simple regulatory “No.” Later I dissected the various legal approaches that were on the table and found them all problematic.
But all this is a joke, and you’re laughing by now if you understand the background: the local telephone companies don’t do innovation. To be more accurate, they’re good at the incremental innovation that leads to more robust and efficient voice networks, but they have few clues how to sprout the disruptive innovation made famous by Clayton Christensen.
Rather, telephone companies have historically reacted in copy-cat fashion to competition. Currently they are resurrecting the old model, shrouded in the raiment of interactivity, of shoving Hollywood entertainment at passive viewers. The things we associate with Internet innovation–search, multimedia sites, social networking, Web 2.0–all originated outside the telcos.
So let me once again put a challenge to the phone companies: be innovative. Here are three things I’d like to see them provide with their spanking new fiber networks:
- Distributed storage
Amazon.com’s S3 and the server farms being built by Google have made this a viable computing option for mainstream firms. I’m even more enamored of a new type of distributed filesystem I heard about five years ago and that is now being marketed by a firm called Cleversafe. This storage is secure and can provide complete files quickly even if large parts of the world’s Internet go down. Note that Cleversafe does not control the storage facilities themselves, but leases space from other facilities so that no one has full control over the system.
Verizon’s web site says their new fiber service offers “up to 30 Mbps downloads and 5 Mbps uploads (and even higher in some locations).” That’s plenty for high-quality videoconferencing. You could teleconference with six people at once, all with the same image quality, displaying them in different quadrants of a big-screen monitor. This kind of peer-to-peer application takes some architectural innovation though: it requires a way for users to identify, locate, and verify each other.
- Distributed anti-virus protection
Current anti-virus, anti-spam, and blackhole services update your system at regular intervals, but not fast enough to keep up with all malware in this highly interconnected age. Systems have been tested that allow sites to flag suspicious activity and send signatures to other cooperating sites; as the evidence that something is bad builds up, the sites automatically take steps to shut it out. A few people could mark something “spam” and by the time people in other timezones wake up, their computers are protected against it.
This is not necessarily a high-bandwidth activity, but high bandwidth helps to speed distribution and ensure that the updates don’t overburden networks.
Some analysts are now telling the telephone companies to just get out of the way of the application-level innovators. The telcos are better off focusing on the underlying network–their core competency–because they’ll ultimately benefit more from the increased usage that will come with new applications.
But Barbara van Schewick, using her background in law and computer science, wrote a paper last year titled Towards an Economic Framework for Network Neutrality Regulation that introduced some decidedly pessimistic analyses and placed the debate on a completely different ground.
She demonstrated some little-noted but powerful motivations for telephone companies to restrict innovative outsiders–and said they would be motivated to do so even if they weren’t in a monopoly situation. For instance, a company can increase revenue by excluding competitors because it can achieve economies of scale, and offer larger markets to advertisers.
She also demonstrated that companies in control of the underlying infrastructure possess the means to discourage competitors, such as by offering a bundled package that new entrants can’t compete with in full.
Incidentally, although the paper applies these principles only to telecom and the Internet, many of the conditions can also be found in the software industry.
Still, most of the experts I have talked to want more competition, not increased regulation on a monopoly. This competition could come from private firms, municipalities, and consortia of end-users. However we do it, it will take a long time, but it’s likely to get us innovation faster than the telephone companies will by manipulating network performance.