There was a time, not too long ago, when it seemed that the best security system was a $25 padlock on the door of the computer room.
Back then, management's idea of computer security had more to do with the safety of the expensive data processing equipment than the security of the information itself.
But as the computing industry has come out of its locked closet, the focus on information security has gone through a metamorphosis. From standalone data processing machines, computers moved to distributed systems with wires stretching to terminals at strategically located spots. Then, soon enough, these terminals were on everyone's desk. The connection of PCs to the central system further expanded the realm of computing, while telecommunications obliterated the fiction of time and space. Of course, TCP/IP and the Internet have extended the realm around the globe, and Wi-Fi has severed our users from any semblance of a physical connection.
Broadband Insecurity
Today, you can drive through any suburban neighborhood and find unsecured wireless access points that--with the appropriate password--can connect to your company's mainframe information system. Such is the power of broadband technologies, and if it can work for you, it can work for any crook with the special knowledge required to hack your system.
In fact, computing is so pervasive now that the emphasis on security has flip-flopped: Hardware is now expendable, while the real gold mine is the flow of sensitive information contained in and channeled by our computing systems.
Yet in our minds' eye, the analogy of the padlocked computer still persists, 30 years beyond its relevance. Even in IT parlance, we're still "locking down" user profiles and generating complex encryption "keys" and "passwords."
We are like Ali Baba, protecting our mythical treasure troves with a simple phrase like "open sesame."
Meanwhile, the real problem of identity theft continues to grow.
Phishing Incidents up 226% in May
In June, IBM reported that phishing attacks increased 226 percent through the month of May, while viruses and worms continued to incubate and spread through email and Web applications at an increased rate of 33%. And at the same time, 90% of the exploitations of information systems were the results of Web hacking of insecure systems. Imagine that! This is the norm, despite all the concern about network security!
In late May, IBM security analysts observed active exploitation of a Microsoft Library ASN.1 vulnerability. By correlating the signatures with other security events, IBM was able to determine that several attacking sources belonged to educational institutions. They then revealed that that the attacking sources were actually compromised host computers that had been hijacked into an Rbot network.
What do these events tell us? Several things:
- Our public information systems and networks are being regularly penetrated.
- Private systems are being actively probed for security vulnerabilities.
- The focus of effort is shifting from malicious vandalization to organized exploitation by criminal elements.
- Our current security tools and security methodologies for protection are grossly inadequate to meet these challenges.
- Government law enforcement regulations and resources appear to be insufficient to combat the rising tide of computer fraud.
Clearly, the analogy of a "locked system" today is a complete fiction. The only refuge seems to be to hide and to hope your company's information system will somehow slip by unnoticed by the crooks.
Where Do We Go from Here?
Are we going to put our systems back into the closet? Can we buy a lock strong enough to hold the world at bay?
Of course not!
On the other hand, we can't sit back and hope that Microsoft or IBM or Sun Microsystems or HP or Symantec or some other company is going to deliver a product or a new technology that will secure our networks. The reliance upon so-called "standard protocols" has wed our systems to hopelessly outdated technologies that are nearing the end of their useful lives.
Take SMTP. Please! Just take it away! This protocol must have been designed in Switzerland because it has as many holes as Swiss cheese!
And how about HTTP! Or HTML! Or even XML! These are protocols and programming constructs that sit wide open, devoid of inherent security. In fact, security on the Internet as a whole is a tinker toy construction of false frontage, secret handshakes, and special nods and winks. To break through requires the tremendous skill and dedication available to any teenager with a computer connection.
Who Will Save the Internet?
The network to which we're entrusting the future fortunes of e-biz is a 30-year-old telecommunications experiment developed by ivory tower academics and funded by the U.S. government!
Of course, these are the same individuals who tell us that the open-source community is going to fix our security problems. Yeah, right!
Don't get me wrong! I think open source is a great way to break the stranglehold of large monopolies (read: IBM and Microsoft) and open the way for innovative new products. But the open source movement can't resolve the basic problem of an insecure internetworking architecture.
And the same is true of the so-called "open standards" movement, which, as it's currently composed, is an alliance of corporate vested interests, each carving out niches to support their proprietary technologies. These interests can't reach deeply into the underlying code of a public network to "fix" flaws that have been consecrated into a "gospel of the Internet."
So what are we going to do?
The End Is Not Near
If you believe that the bursting of the technology bubble in 2000 represented the end of the IT industry's economic expansion, then what would be the use of improving the network? If you believe that IT is going the way of auto manufacturing--heading overseas to the lowest-paid labor pool--then what would be the point? And if you believe that the invention of the Internet was the apex of communications and networking technology, then you'd be justified in saying, "The Internet with all its warts and security flaws will be the legacy that we pass on to generations of business organizations." After all, look at the investment that business has already made.
However, what if we're just starting the real process of building a legacy communications infrastructure? What if real globalization is only just beginning? What if all this current Internet technology really was only an experiment, demonstrating the potential of an interconnected world--a kind of real-time mockup of the possibilities?
A New Internet
That's what some people working with the National Science Foundation (NSF) are starting to propose. They're saying, "To hell with the Internet! We've made a few mistakes, and we've learned a fair amount! Time to throw it out and build a new one!"
David Clark, who helped develop the Internet in the 1970s, has received a grant from the NSF to plan for a whole new infrastructure to replace today's global network. According to reports, the NSF aims to put out an RFP in the fall for plans and designs that could lead to what Clark called a "clean slate" Internet architecture. Those designs, according to Clark, could be tested on the National LambdaRail (NLR), the nationwide optical network that researchers are using to experiment with new networking technologies and applications.
Clark says he wants researchers to re-imagine the infrastructure that connects computer users around the world. According to Clark, the current Internet's 30-year-old architecture is actually preventing further growth of the network and is ripe for replacement. By comparison, Clark thinks a new architecture could allow for ubiquitous embedded wireless communications devices and sensors. It could also provide for real security for e-commerce while allowing people a world apart to collaborate inside 3-D virtual "tele--immersion" arenas.
Clark asks, "Can the research community devise a fresh, new design for an internet--a design that takes into account both the wisdom in the original design and what has been learned since, a design that takes into account the requirements the network now faces and those we can predict in the future--and demonstrate a network with sufficient appeal and merit that we might persuade the world to move to it?"
Ready or Not! Here We Come!
Are we really ready for another Internet?
Maybe so! Clark says he thinks there are two things that absolutely need to be addressed. One is a coherent architecture of security. The other is a healthy and robust economic infrastructure for network service providers, who will be funding the construction.
Why? Because the U.S. government ultimately can't be trusted with a global infrastructure.
Why not? Well, last week, citing national security and "business concerns," the U.S. Department of Commerce announced that it will keep final control of the Internet's Domain Name System (DNS), rather than hand it over to the international nonprofit organization called the Internet Corporation for Assigned Names and Numbers (ICANN), which is currently running it.
In a speech covering the agency's latest policies on broadband, wireless spectrum allocation and other national infrastructure matters, Assistant Secretary of the National Telecommunications and Information Administration Michael Gallagher said, "Given the Internet's importance to the world's economy, it is essential that the underlying DNS of the Internet remain stable and secure. As such, the United States is committed to taking no action that would have the potential to adversely impact the effective and efficient operation of the DNS, and will therefore maintain its historic role in authorizing changes or modifications to the authoritative root zone file."
In other words, the U.S. government is now reneging on its commitment to allow this essential Internet function to move to international control. If this is the kind of policy setback that the Internet is going to experience, isn't it time to reexamine the infrastructure and design something new that is robust, secure, and truly global?
Thomas M. Stockwell is Editor in Chief of MC Press Online, LP.