Lessons from iMeme

And in all honesty, I’m one of them. I’ve been online since 1985, when I discovered BITNET as an undergraduate, sitting at a terminal connected to the campus mainframe. Most of the people using BITNET were computer science majors and may actually have known what they were doing. I was a classicist, but fascinated by the ability to “talk” instantaneously with people at universities around the world—not to mention flattered by the attention I got as one of very few females. I did know there were other networks, because some of my friends at other schools were on ARPANET and I had to take extra steps to get e-mail messages to them.

Then came graduate school. By that time I’d already passed out of the starry-eyed “This is cool!” phase when it came to things like chat rooms. We had Macs in the Classical Studies department and a new set of protocols for communicating. It was 1990 and the Internet was just opening up to the non-university public. The text-only interface of Gopher, Fetch, Telnet, Usenet, and the amazingly primitive e-mail program the University of Michigan favored shaped my online experience. When I bought my first computer and modem, 14.4 kbps was fast.

Never at any point in time did I really understand how any of this worked. I was comfortable with computers and good at following directions. As with many tools of modern society, we don’t have to know how the Internet works to be able to use it. I can drive a car with a manual transmission, but I couldn’t build one, or repair it.

The World Wide Web and our modern browsers and e-mail clients are the equivalent of the automatic transmission. It’s easier than ever to get online, and the Internet is no longer the domain of geeks. In fact, we’ve very nearly achieved the self-driving car, and this may not be such a good thing.

Fortune invited Vint Cerf, “father of the Internet” and now Chief Internet Evangelist at Google, to iMeme to talk about the future of the Internet with David Kirkpatrick.

The first question Kirkpatrick asked was about government attempts to control the Internet. Vint Cerf is an advocate of Network Neutrality, not a fan of control, and believes (or at least hopes) that it will get harder and harder to restrict access to information the Internet. Kirkpatrick then asked why we needed to worry about Net Neutrality if it’s so easy to route around obstacles.

Cerf explained something most of us don’t think much about: that the Internet is made up of layers, with the physical layer of fiber-optic cable, copper wire, modems, and Ethernet on the bottom and the applications we use online at the top. The higher you go through the layers, the easier it is to find alternate routes. No one in Australia can get high-speed access while Telstra keeps Fibre to the Node under lock and key pending a resolution of its dispute with the government.

Despite the 150 million botnets he mentioned in Davos in January, Cerf asserted that the biggest “perturbation” in the Internet over the next few years would be the shortage of IP addresses and the need to move from IPv4 to IPv6. “IP” stands for “Internet Protocol,” and if you want to see something really scary, you can look up the Wikipedia entry.

Every computer, router, modem, etc on the Internet has an IP address, a collection of numbers which acts as a unique identifier. Although these addresses do generally indicate something about geographical location (at least to those who know how to read them), the name “address” is misleading.

The address of your home tells people where to find you, but just knowing where you live doesn’t allow someone to follow you everywhere you go. An IP address is like the “silver cord” which links the spirit to the body. However far you wander in the astral plane, all you (or anyone else who can see it) need to do is follow that cord to get back to your body.

Wherever we go and whatever we do online, we leave a trail behind us like Ariadne’s thread in the Labyrinth. The Internet offers the illusion of anonymity, because most of us lack the skill to find out who is behind a screen name or a junk message. But even people using tools like the TOR anonymity network can be tracked by skilled and determined “adversaries.” If you want to be truly anonymous and untraceable online, you have to resort to practices of questionable legality that fall into the “Kids, don’t try this at home” category of danger and difficulty.

And whatever else you know, or don’t know, about the Internet, it’s important to realize that you are neither anonymous nor invisible. The decision to post something online is irrevocable. And if Big Brother wants to watch you, it’s easier than it ever was before.

The opening panel of iMeme 2007 posited that Google, Second Life, Facebook, and Salesforce.com are (becoming) platforms rather than applications. Moderator David Kirkpatrick asked whether becoming a platform was the ultimate goal for a technology company. The responses from the panelists tend to suggest it is, at least for them—but if it weren’t, it’s unlikely they’d have been invited to present on that panel.

To become a platform, you have to let other people play with your toys. All these companies have done so, with the most recent example being the phenomenally successful release of the Facebook API. (That stands for “Application Programming Interface” and means you make enough of your source code available that anyone with the right skills can develop plugins for your software. A plugin, as my mother-in-law explained when I first met my fiancé the software developer, is like the software equivalent of an attachment for your mixer.)

So there was a lot of talk about openness and how you can bring anything into Facebook: Twitter, video, blogs, etc. and so on.

Yes, but can you take it out again? Marc Benioff, CEO of Salesforce.com, pointed out that the reason so many people are stuck with IBM mainframes is that they can’t export their metadata. You might be able to export your list of contacts, but not the connections between the different contacts and the networks within the company, not the workflow, not the customizations. And those are the things that make the data really useful.

If your computer dies, you can reinstall your operating system in a couple of hours, but it might be days before you’ve got everything customized back to the way you were using it: the fonts, the icons, the themes, the view settings, the browser plugins, the hotkeys, the e-mail filtering rules… To back all of that up, you need a complete drive image, and if you try to restore that to a computer that isn’t identical to the one you made the image from, it probably won’t work.

FIR correspondent Dan York has talked a lot about “walled gardens” on the Internet, places like Second Life and social networks where members can communicate with others inside the network, but not reach outside. But the issue isn’t really that there’s no communication between those within these new spaces and those outside of them. It’s that once you create your profile/build your avatar and build up a network, you can’t take it with you. You can’t move your Facebook friends into LinkedIn, or vice versa. (You can’t take your graphically superior World of Warcraft character into Second Life, either.)

Esther Dyson, chief asker of provocative questions, expressed this as a tension between these would-be platforms and their users over data ownership, though it might be more appropriate to describe it as a dispute over metadata ownership. And while Mark Zuckerberg blithely assured her that the solution was to give the users total control over their data, Facebook’s terms of service explicitly prevent that, by giving Facebook a license in perpetuity to use whatever you put into it.

Beyond the issues of cross-platform incompatibility of things like avatars (a trickier technical issue than matching up the categories for profile data and the type of relationship someone else has to you), beyond even the fact that many of these sites make their money from advertising and therefore sell your contact information and stated likes and dislikes to corporations, not to mention displaying ads next to your content, there’s a more serious issue.

The thing which makes social networks interesting is the thing which makes them dangerous. It’s extremely useful to type in a single query and find out who out of the people you already know has a connection to a person you want to meet. It’s fun to find out who has read the same books or seen the same movies. But boy, does it make you vulnerable. And it makes all your friends vulnerable, too.

Suppose someone I know owes money to an unscrupulous creditor. The creditor has connections to organized crime and gets their phishing/pharming specialists to hack into the social network both I and my acquaintance belong to. Now, instead of just harassing the person who owes the money, the creditor can call all the debtor’s friends and colleagues in an attempt to enlist them in getting the money. In the process of doing so, the creditor will annoy a lot of people and probably ruin the debtor’s reputation.

Or suppose you’re an author marketing your book through MySpace. You accept all the friend requests made to your book (or main character). Then one of your readers gets arrested on suspicion of terrorism. That gives the U.S. government the right to that person’s data, which means you are going to show up on the list of that person’s “friends.” So even if you’ve never met the suspect and had nothing to do with whatever s/he is accused of, you might find yourself having a chat with Homeland Security.

And that’s why we can’t just transfer our connections from one network to another without asking their permission. The people we know have a right to decide in which context they want to admit to knowing us. Even those who know and trust us might prefer to limit the amount of collateral damage they let themselves in for by staying out of certain environments and keeping their metadata to themselves.