3

The Making of an Underclass AOL

Why are AOLers so clueless?

—Technology correspondent of a major daily newspaper, in conversation, 1995

Here’s one of the secrets they don’t tell you when you first whip that modem out of its plastic wrapper and fight your way through arcane commands to log on: cyberspace is full of cliques.

One of the more famous examples of this was the 1994 invasion of the newsgroup rec.pets.cats by a disruptive gang from alt.tasteless, a perfect clash between a group to whom nothing is sacred and one to whom cats, in their ineffable fluffiness, are. The way the story got told in Wired, the alt.tasteless crew had a fine old time posting messages about nailing cats to breadboards, cooking them, electrocuting them, and spraying them with acid while the rec.pets.cats regulars writhed in agony. Eventually, the rec.pets.cats people were taught how to use killfiles so they’d never see the invaders’ messages, and alt.tasteless gave up after complaints to their system administrators nearly cost them their Net accounts.1

Other examples abound. On systems that allow such things, small groups will set up their own closed conferences where they can snigger at other, less with-it users in private. On systems that don’t, the same kind of behind-the-scenes, backbiting discussions go on by email or live chat; if you’re very clever about such things you might be able to pick up hints of hidden alliances by watching which users regularly back each other up in arguments or fights. Closer to the rec.pets.cats invasion is the kind of trolling and baiting that goes on when a group of, essentially, playground bullies hound some other user for offenses real or imagined—he might be a fundamentalist Christian, say, or have no sense of humor, or just be generally annoying. Or he might simply come from the wrong domain—the Internet word for a system’s name.

This last seems to be the situation of 8 million (and counting fast) America Online (AOL) users, many of whom may even believe that AOL is the greatest thing since television. That the system had problems became known at the end of 1996, when AOL switched from hourly to flatrate pricing and immediately found its system swamped with users who got on and wouldn’t get off. In early 1997, several state district attorneys began studying the company’s new pricing scheme, and AOL announced a $350 million upgrade over six months to its network to handle the volume. Although the service was still adding users, the effort to acquire them was expensive; as a result of a change in how AOL amortized those users, at the end of 1996 it declared a loss bigger than all the profits it had ever declared put together. Nonetheless, if you have 8 million users you’ll have no trouble finding business partners.

AOL was a lot smaller—only a million users—and far from the market leader in March 1994, when it set up its “Usenet feature,” which allowed a seemingly endless stream of people to tap nervously on their newsreaders, type out, one after another, “Hey, is this working?” and then hit the SEND button to relay this world-shaking message to all of Usenet.

The problem was where they said it. There is a newsgroup called test, and its purpose in life is to provide a place for people to take their newsreaders once around the dealership parking lot to make sure they understand the controls. Various things, some of them people, monitor the test newsgroup and send replies to people who post messages there.2 And there’s no doubt: it is a thrill the first time you see the message you wrote on your computer come back from Deep Cyberspace with replies attached to it. But that’s not where they said it. A lot of them picked alt.best.of.internet. However anarchic Usenet seems, particularly the alt groups, there is often a kind of internal logic to the way newsgroups are named. This particular newsgroup was intended to counteract the generally low signal-to-noise ratio of Usenet postings and serve as a place where people could repost their favorite messages from other newsgroups so everyone could see the gems without having to do their own strip mining. So the rule was and still is: no comments, no original messages, repostings only. Very strict. And this setup worked remarkably well as long as the number of new users popping up with comments stayed at a manageable level. Unfortunately, the arrival of AOL changed all that.

Within a couple of months alt.best.of.internet had turned into a battleground. AOLers would post hello messages, old-timers would follow up with vituperative diatribes about reading the FAQ without telling them how to get it, and other old-timers would pile in and take up more bandwidth and create worse useless noise than the AOLers’ messages did in the first place. Tempers got short. Repostings got lost.

If that had been AOLers’ only sin, they might eventually have been forgiven, especially by the many who do not read alt.best.of.internet, “due to the collective memory of the Net being about one week, maximum,” as David DeLaney observes in the “Net.Legends FAQ.”3 But several factors ensured that AOLers’ transgressions would not be forgotten. First was the sheer volume of new users; if only a small percentage of a million people causes trouble, that’s still a lot of people. Second was the fact that, unlike each year’s arriving freshman class, all AOLers came from a single domain: aol.com. Every message, every crude sexual come-on, every misplaced question reinforced the initial impression of that particular domain as populated with willfully stupid people—or, as the Net would put it, clueless. In the collaborative effort of one newsgroup, AOLers “couldn’t get a clue if they stood in a clue field in clue mating season, dressed as a clue, and drenched with clue pheromones.”

The final factor was one of instinctive resentment of any hint of commercializing the Internet. Where traditionally, Internet users shared their resources for the public good, the perception was that AOL neither knew nor cared about net.traditions but was only interested in sticking a meter on a free resource and billing its users extortionately.

“AOL’s philosophy borders on net-abuse,” wrote David Cassel, the maintainer of the alt.aol-sucks FAQ,4 saying that the earliest version of AOL’s Usenet newsreader was buggy and wasted resources by reposting articles multiple, unnecessary times, and complaining that AOL had failed to consider adequately the impact of its users’ demands on FTP sites5 and made no such facilities available on its own servers for the rest of the Net. “This gets into an ideological war,” noted Cassel. “Technology now allows people to freely exchange information at an amazing rate. AOL attaches a meter to that process. In addition, aggressively pursuing new users, AOL exploits the lack of awareness of existing technological capabilities, and establishes a model that follows the traditional role of pre-packaged entertainment designed for a mass audience.”

At the time, a thoughtful and intelligent user named Edward Reid did some research and came to an interesting conclusion: AOLers weren’t (necessarily) stupid; they were software-disadvantaged. In a carefully written and thoughtful article (reposted to alt.best.of.internet by Ron Newman, an old-time user with widely respected technical knowledge), Reid analyzed the interface AOL had given its users and concluded that it was the source of much of their disruptive behavior.

One problem was that the software AOLers used to access Usenet offered no offline reading or editing facilities. Therefore, AOLers, who were then paying $3.50 an hour for their access, were under pressure to read and write as quickly as possible, encouraging them to skimp on what Reid called “think time.” Reid noted that many AOLers were complaining about this in the system’s internal newsgroups. Reid couldn’t figure out why AOL, which even then provided offline facilities for email, didn’t provide similar facilities for Usenet.6 (The answer may be that AOL doesn’t supply offline facilities for its own rather rudimentary message boards.) AOL, others have commented since, is geared toward instant messages, online chat, and real-time interaction, creating a culture where a hasty “Me, too!” is acceptable comment—another culture clash, since Usenet norms consider such messages a waste of bandwidth.

A second problem, in Reid’s opinion, was that AOL’s software interface confused mailed replies (private) with posted follow-ups (public), encouraging AOLers to post publicly messages which to old-timers seemed more appropriate for mail. Quoting, a staple on Usenet because follow-up messages may arrive before the originals, was not available. Reid also complained that AOL’s threading—the facility that shows how a series of messages on the same topic relate to one another—was weak, and that features built into Usenet to allow newer postings to supersede old ones (used with regularly updated messages such as FAQs) were not enabled. There was no search facility (common in Usenet newsreaders), and limits on the number of articles in a single newsgroup the AOL software could show further restricted users’ ability to find, and therefore read, FAQs.

The biggest problem for the embattled alt.best.of.internet specifically had to do with AOL’s suggested list of newsgroups for its members to try out to get acquainted. It’s understandable that the service would want to put alt.best.of.internet at the top—it was a great showcase for Usenet. But the result was that as all those AOLers trooped to the edge of their world and stepped off (I imagine this as one of those long parades of goofy, green-haired aliens dropping through a trapdoor in the game Lemmings), they exhibited normal, human behavior—that is, they hit the first newsgroup they came to and said, Hello, world. And they got flamed.

As one AOLer complained to alt.best.of.internet in May 1994, in response to the presumption that all AOLers were bozos, “Is this any kind of behavior for people in an electronic community? Just because AOL subscribers PAY for their Internet access, not have it provided free through a university, some other users are assuming that they are uneducated morons.”

Some of these things have been fixed since 1994. But the deeper problem had to do with AOL’s decision to make the “Usenet feature” look as much as possible like the colorful graphical world of the rest of AOL. Consequently, many AOLers may not have understood that they were not on just another part of AOL, and so they couldn’t possibly have registered that the standards of behavior were different. In any case, there is a natural tendency to assume that whatever service you first use is the way online should be, and that anything that deviates from that is wrong. Many people started with online services like AOL, CompuServe, or Prodigy partly because these came bundled with new computers, but also because for a novice these services were substantially easier to set up than a direct Internet access account. This balance began to shift in about 1995, when Internet service providers like Netcom and Pipeline began marketing their services nationwide, including software packages that were designed to be easy to set up and use. The advent of the World-Wide Web as the most important unifying interface to the Internet helped a great deal.

Further resentment was created on the Net side by AOL’s habit of advertising itself as “the Internet, and a whole lot more,” further confusing where the boundary, if any, might lie. AOL also took it upon itself to improve upon certain Usenet conventions: some newsgroups are listed on AOL by descriptions supplied by the service rather than their actual names. For example, alt.aol-sucks is listed as “Flames and complaints about AOL.”

It’s fair to say that AOL as a company can’t have understood how many problems its interface was going to cause for its members and for the Net at large. Although it was slow to change, it did correct most of the mistakes Reid listed over the next two years. However, it made errors again when it launched its Web browser, which irritated Webmasters (the people who maintain Web sites), who were left to field AOLers’ complaints when the company’s browser didn’t support some common Web features correctly or failed to update pages regularly enough on its proxy server.7 (Ironically, the company that benefited most from AOL’s Usenet debut was probably its nearest competitor, CompuServe, then nearly double AOL’s size, which observed the situation and determined to construct a gateway that would cause less trouble. CompuServe has never suffered from anything like the same image problem on the Net, although it, too, is resented as too expensive. However, in early 1997 WebTV users were joining AOLers in the “clueless” ranks.)

Whatever the Net thinks about it, within its own modem ports AOL has been a roaring success. Between 1994 and 1997 AOL’s claimed user base grew from 1 million to 8 million; the company launched a public stock offering; and it became the number one domestic U.S. service, estimating its daily contribution to Usenet at 300,000 postings. Plastering the world with free disks and free trial accounts helped create for AOL a throwaway accounts culture whose flame-and-run tactics were in general more destructive to the Net in encouraging spamming and other types of abuse than the far more controversial anonymous remailers that allow users to interact on the Net over a long period of time without revealing their real-world identity. But the strategy netted AOL a ton of subscribers (and supplied a generation of computer users with free backup disks).

It only added to the Net’s contempt that there were several significant Internet services that AOL didn’t offer, notably outbound Telnet, the function that allows you to log on to remote computers as if you were directly connected to them. The buzz may be all about the Web, but Telnet is a vital service and one the other major providers were supplying by 1995. More than that, it seemed that no matter what you did on AOL you ended up twiddling your thumbs while AOL downloaded “artwork”—all those colored graphics that give the service a large part of its character. And on top of that, the ability that old Netheads take for granted to multitask—like being able to download a file in the background while browsing the Web in one bit of foreground and hanging out on multiple channels on Internet Relay Chat in another—just couldn’t be had. AOL, like all dial-up services of the era before the widespread use of Internet standards, only let you do one thing at a time. AOL did have the capability of running Internet sessions like those offered by flat-rate ISPs for those with the knowledge to seek out their own software, but it was an expensive—and, people complained, slow—way to get your Internet service.

AOL’s chat rooms8 were another sore point. Chat is one of those functions that most systems offer to let groups of users type messages to each other in real time, emulating a live meeting or conference. Internet Relay Chat (IRC) is far more flexible than AOL’s setup, which limits users to one chat room at a time, with a maximum of twenty-three users. That relatively small objection was easily trumped by the activities of AOL’s Guides, volunteers detailed to keep the service “clean.”

AOL Guides act as monitors. If someone starts abusing other people in a public chat room, using sexually explicit words, or trying to trade illegally copied commercial software, any of the users can complain to a Guide, who will join the conversation and monitor the situation, warning the miscreant if it seems appropriate. If unacceptable behavior persists, the Guide can eject the person from the chat room and even the service—alt.aol-sucks posters call this getting “TOSsed,” a word derived from “Terms of Service.” AOL has also faced complaints about censorship from other users, such as the Creative Coalition, a group formed to protest the disappearance of members’ poetry from the AOL message boards.9

If you think of AOL as a privately owned commercial service aimed at the family market, these policies make some sense even if they fail. And they have failed on a few occasions: some of the most frightening stories about the Internet and pornography or contacts between children and pedophiles did not happen on the Internet but within the supposedly safe confines of AOL. Their being reported as Internet stories is yet another source of resentment on the Net at large.

In a long article on the service in Rolling Stone, writer Jeff Goodell called sex AOL’s “bedrock,” estimating that sexually explicit real-time chat was contributing at least $7 million a month to AOL’s free-disk fund.10 Goodell’s story of one AOL user—a schoolteacher who discovered sexual freedom online and then incorporated it into her real life—would be enough to horrify many in the religious right even though it combines the best qualities of experimentation with those of safe sex. Her experiences were possible because, besides the closely monitored public chat rooms, members may set up unmonitored private ones at will. These are used for anything from private conversation between real-life friends to the jointly created one-handed online typing fantasy sessions known as cybersex.

In the Net world, this monitoring puts AOL on the wrong side of one of the Net’s major continuing flame wars: censorship. A significant portion of Net users hold freedom of speech to be sacred. The answer there, of course, is simple: if you don’t like having your speech controlled, don’t subscribe to AOL. There are plenty of ISPs out there to choose from that have adopted no-censorship policies. But on the Net, you don’t just disagree with somebody and go away quietly. (Well, you may, but if you do, no one will know you did it unless you post a large, public announcement). It’s much more satisfying to make fun of them as publicly as possible in alt.aol-sucks, the newsgroup for people who love to hate AOL. Many of the inhabitants are themselves former AOLers, and their relationship to the service is not unlike the attitude of zealously reformed smokers. Others just hate corporate America on principle.

It may have been this kind of thinking that inspired the writing of AOLHell, a free program that adds a slew of functions deemed to be missing from AOL’s client software (besides a few facilities that are illegal); a Web site with a test to take to determine if you’re ready to leave AOL for the wider Net; a site designed to show up the failings of the AOL Web browser; and a site listing what are claimed to be the words that will get you TOSsed.11 Those assembled on alt.aol-sucks therefore cheered when, in 1995, several AOLers brought a class action suit against the service challenging some of its billing practices, specifically alleging that various built-in connection delays inflated users’ bills. AOL denied the claims but settled the case, which included all AOL customers between July 14, 1991, and March 31, 1996, by giving the affected customers free time according to their service use.12 The group cheered again at the end of 1996, when AOL declared its overall corporate loss.

I know what you’re thinking. You’re thinking these people need to get a life—to share, if they can’t afford one each. You’re thinking they have way too much time on their hands. You may even be wondering why Steve Case, the CEO of AOL, doesn’t sue the pants off of all of them instead of continuing to be their chief provider of (free) floppy disks. And if it weren’t for bisk poetry I might agree with you.

“Bisk” is the alt.aol-sucks subcultural name for one of those free trial disks that show up everywhere from magazine inserts to airline lunches. Anything that ubiquitous has to be a source of jokes, so posters have come up with all sorts of imaginative uses for these: props for wobbly tables, toys for the cat, even bathroom tiles. You figure this out after reading the newsgroup for a few days or by reading the FAQ (or by posting to ask, if your address isn’t on AOL or your computer happens to be coated with asbestos). Bisk poetry is doggerel written in the same deliberately semi-literate style that produced the word “bisk,” and late at night, when the peanut butter sticks to the roof of your mouth, it can be hilariously funny. Here is the official, earliest known sample:

Subject: Aol is sucks!!!!!what you can do with their cd rom bisk

From: xxxxxxxx@xxx.com (xxxxxxx)

Date: 1996/04/27 alt.aol.sucks

cost to mutch

it suck

no good

send to many disk.

Me and my friends took a bisk and lit it on fire and froze it slamed it angaisnt the boor.13

At this point, prejudice against AOL and all those who click in her is probably not going to go away, even though it did join the coalition against the Communications Decency Act (see chapter 4). It’s sort of appropriate, though, that evidence to support this comes from the WELL, the system whose users arguably believe they run cyberspace in the same unrealistic way some tiny secret conferences I’m in believe they run the systems they’re on.

The WELL is sort of the other end of the coolth spectrum from AOL, even down to its austere, text-based interface, which is about as far from AOL’s whizzy graphics and cute trivia quizzes as you can get and still be on the end of the same modem. The WELL’s cachet comes from the fact that most of the Netizens of any fame as net.activists have at one time or another hung out there: Electronic Frontier Foundation founders, Wired editors, and technology wizards jostle with journalists from the major national media and the organizers of the annual Computers, Freedom, and Privacy Conference to argue about the most vital issues affecting cyberspace. The result is that the WELL, with 10,000 users, is the most written about online system and probably the most influential, at least in its own estimation.

In late 1995, a user on the WELL decided to test her perception that AOLers were unfairly discriminated against on the Net. She posted a blank message to an unfamiliar newsgroup from an address on a “plain vanilla” ISP. She got mailed offers of help and advice, plus a couple of jokes about her “profound” message. A week later, she posted another blank message to the same newsgroup from an AOL address. She got flames and abuse—from the same people. Reporting on this afterwards on the WELL, she said, “Seeing aol.com in the domain and making assumptions about them, reading their posts with a filter that says they are all jerks, is really not far removed from your basic garden-variety bigotries.”

Viewed from a distance, these petty prejudices must seem only amusing. Many AOL users are completely unaware that their address is on the wrong side of the telephone lines and will never find out. It’s more serious in terms of the sharing of resources the Net was designed to facilitate if valuable sources of information decide that AOLers are just too stupid to talk to (or they’d choose a better service provider), or if, conversely, vital information is discounted simply because it comes from AOL. Unlike real-world identifying factors such as gender, skin color, and accent, AOLishness can’t be hidden—although it can be changed at will. However much we would like to believe that humans are universally good-hearted, kindly creatures, we have a built-in tendency to divide ourselves into “them” and “us” and to create and maintain prejudices against classes of people, presumably to convince ourselves that we are OK folks. This is the dark side of the network of trust that will come up in later chapters, but it is not limited to the Net itself.

There are two other important lessons. First, as more and more of our communications are mediated by computer, AOL’s online hazing experience shows how vital it is that the influence of system design on human behavior be examined and understood. Different cultures develop in cyberspace in part because of the technology that supports them. The WELL has a system design that fosters highly structured discourse by allowing no threading within a topic, forcing a would-be participant to read through to the end of the discussion before adding his or her thoughts. Repetition is therefore rare. On Usenet or CIX, with built-in threading, the interface encourages responses to specific points; while this allows discussions to branch into other topics without confusion, repetition abounds because many posters answer without reading to the end of the thread to find out that their point has already been made. This is especially a problem with the widespread use of offline readers.

Second, it’s easy to lose perspective on the Net. The embedded sorting of the Net into topics is an efficient way to sort computerized discussions. But a consequence of that structure is that people tend to focus only on topics that interest them, and because those topics fill their computer screens they tend to imagine that those topics are the most important ones, not only on the Net, but possibly in the world. This kind of intensity is vital in some professions (including writing, researching, and programming, some of the earliest classes of Net user). But it leads to skewed fantasies in which people whose interests are different from yours cease to exist or may be discounted: out of sight, out of mind. Any group that thinks it runs cyberspace must remember that the real in-group with the real power are the bastard operators from hell (BOFHs).14 And they don’t let any of us read their newsgroups.