Personal Background

When I first got an account at soda.CSUA, a computer system for Berkeley undergraduates interested in computer science, many of my friends who already had accounts there warned me about what I should expect. They warned me not to believe what people suggested I do to my account, but then added "maybe people will be nicer to you, though, since you're female." They encouraged me to ignore people who were mean and "flamed" me, but then told me "but you're female. Maybe people will be nice." They told me that because my login name was somewhat female, people would probably be more tolerant of my general lack of computer clue. Along with all of this, though, was communicated the connotation that men who use computers are often sex-starved and desperate. Unfortunately, much as I wish I could argue with this, I've found much of the above true. I was treated more nicely than male users are, generally; I also have been the prey of many "desperate" computer men, and have received many unsolicited talk requests, mail, and write messages. Because of my experiences online, and what I have heard from other women, I decided to concentrate on this aspect of the Internet; I wanted to see how women's presence online is viewed by both men and women.

Technical Background

In this paper, then, I am focusing on gender interaction within the social phenomenon known as the Internet, which current jargon terms the "Information Superhighway". The vast majority of Americans have at least heard the term, with Vice-President Gore's focus on its use in communication, massive amounts of media hype (including Newsweek Magazine's "Cyberscope" page), and the fact that large numbers of people are connected to the net, either directly or indirectly. The worldwide network of over forty million computers includes a number of separate networks (commercial networks like CompuServe and America On-Line, as well as private business networks and public educational and research institutions), melding them together into a conglomeration of people and ideas and communication heretofore unseen on the face of the earth.

Although the Internet started out as a forum for scientific research, it has since become a very social place. From its inception, where users were almost completely male researchers, to the present day, with an ever-increasing number of casual users, including greater numbers of women and students on the net, its emphasis has changed rather dramatically. Users today focus more on the social aspects, ranging from email to irc (Internet relay chat), from MUDs (multi-user dungeons) to usenet groups. Indeed, the common phrase when people ask for accounts these days is "may I have an email account?", which implies that the accounts are useful primarily for their access to email. Significantly, many, if not most, of the college students who have accounts are largely unaware of the many options available to them on the Internet. However, there is a specialized subset of people, often those interested in computer science, who are more likely to use more of the Internet's resources. Many of these students are members of the Computer Science Undergraduate Association, and as such have computer accounts at soda.CSUA. The CSUA organizes social activities and also provides a forum for on-line socialization which is, I believe, rather different from the vast majority of systems on the net. In addition to providing the usual amenities for its users, soda has a thriving social aspect including nwriting, "talking", a newsgroup, and the "wall", a sort of electronic bulletin board or party-line phone call in which everyone "listening" to walls receives messages from other people who choose to send them. It is important to realize, then, that the soda community and the soda user is rather unique when compared to other systems and their users.

Because of the specialized aspects of the soda community, I focused primarily on this subset of the Internet community (although I did stray outside of it for some of the material). Looking at the power computer user, or as is sometimes called in the vernacular, the geek, I focused on soda for several reasons: I know the system rather well, many people there trust me and would be willing to take the time to answer my questions thoughtfully, and, logically, soda is the system where I spend most of my time so I am most familiar with the environment there. Although the CSUA has more than 1600 members with accounts, only around 1000 of these people log in within any given 2-week period. A much smaller percentage of those people compose what I would consider the active user community, made up of 40-60 members who attend meetings, wall (discussed later), and generally make themselves known. These people tend to have a fairly high level of knowledge about both computers and the Internet, and soda is fairly often used as a means of transmitting this information. Additionally, active users tend to know each other, if not in person then definitely online, and very often socialize within this group. Many have similar jobs or fields of study, and soda serves to provide a forum within which they can interact with people of like mindsets. (As an example of this, one survey respondent said he liked soda because "it is a social-type hangout that doesn't require physical presence (and is therefore convenient)") As a sort of indicator of this, one of the more common uses of soda's walls is to find people who want to go out to eat with you, because it is almost guaranteed that someone will be hungry at any given time. Thus soda is much more than simply a machine; for some "sodalites" it provides their only social interaction, and as such studying it can be quite valuable. Although it is possible to have an account on the Internet and never actually interact with other people (which you can do by not accepting messages, writes, or talk requests and not attempting to make yourself known within the community), I am focusing on people who do not choose to do that, people who interact, to some extent, with other members of the community.

Several idiosyncrasies set soda apart from the rest of the computing community, and prove very important when considering the gender-related issues that this paper treats. One significant difference is the availability and inordinate use on soda of a program called wall, or wallall, which broadcasts a message to all users who are currently logged in to the machine, interrupting what they are doing so that the message pops up on their screen. This can be considered invasive by some, since it is an interruption, mangling the screen and intruding on the privacy of the users. However, recently (within two years or so) the program was modified so that people who do not wish to see the messages flashing by can configure their system to turn them off, or become anti-social, as it is called. Entire conversations are often conducted between various parties over the wall, sometimes by people who have never even met in person (it bears to mention here that the average population of soda -- that is, the average amount of users logged into the machine at any one time, probably lies around 70-80 during the normal hours of the day, and 40-50 in the middle of the night. The average amount of people listening to walls at any given time fluctuates between 15-20 during normal hours down to 2 or 3 in the middle of the night.). Arguments,friendships, and even hatreds can flare completely in this virtual world.

Though walling was originally used primarily as a way to transmit computer knowledge -- asking questions and being answered very rapidly -- and find people to eat with, its focus has broadened over the past few years. Walls today serve many functions, including the traditional "foodP" (This is the syntax used when requesting company for a meal; the expected response is either "#t"signifying "I would be interested in joining you," or "#f" meaning "sorry, not interested." These symbols have their origin in computer programming, where the "P" stands for Predicate signifying a question that must be answered with either true or false. It's a typically "geeky" joke.) and technical questions, but also incorporating gossip, random trivia, discussions about current news and scandals, flaming, and other types of communication.

Another aspect of social life on soda that exists on other machines, though to a much lesser degree, is "nwrite," a hacked version of the "write" command present on so many machines. The write program sends a message directly to the screen of another user on the same system, preceded by a line indicating who the message is from and the time. Write in its normal form is usually used for a short, one-line comment to someone; as it just appears on the screen with nothing differentiating it from the plain text after the initial line, it can be confusing but is nonetheless a useful tool. Differentiating soda's version from that on most other machines is the fact that each line after the original "from" message is preceded by the person's login name. This makes it possible to differentiate an nwrite from other things that may be happening on your screen; an example of an nwrite conversation follows:

Message from norby on ttyrj at 11:55:39 ...
 norby> Morning Tara ... how's your paper going ???
11:55pm soda% nwrite norby
hi peter...
papers are evil.
how's work going?
 norby> Yes, they are.
 norby> Work's hectic ... people are having lots of network problems
oh dear.  so it's not only soda, eh? 
norby> My boss is doing several things at once, as am I.
hmmmm.  then i'll leave you alone.
talk to you later?
 norby> I think today just isn't a good day for networking.
how evil.
 norby> Yup, I'll see you later ...
 norby> nice talking with you ...
2:57pm soda% EOF
In that conversation, every line written by the person who wrote me, Peter Norby (login name norby), is preceded by a " norby>", and similarly, every line that i sent him, once it appeared on his screen, was preceded by a " tabloyd>" (my login name). This simple innovation makes it possible to carry on long-term conversations while simultaneously doing other things; being able to suspend an operation to do something else is quite helpful. Additionally, it is quite possible to carry on multiple nwrites, or to have conversations that simply keep going for hours but with fairly long pauses. When I log into soda, I tend to receive at least three or four nwrites within the first few minutes, and the majority of the time that I'm logged in I will have at least one conversation going, if not more. Nwrite, then, is simply another aspect that allows interaction and communication within the soda community.

Other aspects of communication which proliferate on the net include the "talk" command. Talk allows users from different systems to communicate real-time with each other; when someone talk requests you, assuming you are accepting messages, an alert appears on your screen telling you who talk-requested you and what to type to "talk" to them. If you choose to do so, your screen will divide in half; what you type appears on the top of your screen, and what the other person types appears on the bottom. The talk function can be very useful when you want to communicate with friends on another system; however, it has also become the refuge of people who, often bored, randomly choose people to talk to just to pass time. This is rather easy to do, by just fingering a certain system to see who is logged in, selecting a person based on the information that appears when you do this. An example of a finger@ command would be the following:

5:07pm soda% finger @csua
Login    Name                 Tty  Idle  Login Time   Office     Office Phone
aaron	 Aaron C. Smith        pv     8	 Dec 13 16:40 CEA 2N	 643-7295
adoruk	 Jon Masami Kuroda     pP	 Dec 13 16:45 2035 Chann
appel	 Shannon D. Appel      rp    19  Dec 13 14:46 CEA	 643-5650
ausman	 James Ausman	       p3     1  Dec 13 16:21 3rd Floor	 642-0720
bob	 Bob Chang	       rM     5	 Dec 13 14:54 D614-Unit1
chiapet	 David San Chia	       qr	 Dec 13 17:07 343 Soda	 642-7453
choice	 Andrew Choi	       pV	 Dec 13 15:21 BARRA 	 649-4223
clee	 Cornel Lee	      *pY	 Dec 13 16:53
coganman Andrei Cogan 	       pj    22	 Dec 13 15:11		 642-7453
cristine Christine Lee 	      *q2 	 Dec 13 16:55 CKC East	 664-3248
dim	 D. Gerasimatos	      *pr	 Dec 13 16:33
dpassage David G. Paschich     rN	 Dec 13 14:55		 642-7453
dsiu	 Danny Siu	      *p7    36	 Dec 13 16:23
Notice that the command gives both the login and the person's real name, which sometimes makes it possible to tell things like gender and ethnicity. With this information, you can send a talk request to a person, and, assuming the person responds, start a conversation. This can be useful; however, notice also the fact that out of twelve people logged in (actually, far more were logged in, but space constraints led me to select the first twelve), only one is obviously female (Not only is Shannon a gender-neutral name, but I also know Shannon and he's definitely not female.). This means that if the person looking to talk with someone wants to talk to a woman, he or she will have a much smaller pool to draw from than that of men. Because of this, then, women online tend to get talk-requested much more frequently than men do, to the point where they are sometimes driven to turn their messages off or put a message in their .plan (the file that appears when an individual is fingered) explaining why they are sometimes antisocial, like
yumin's .plan. Some women who don't know much about the Internet may choose a name that attracts quite a bit of attention, for one reason or another; they are often the ones who end up refusing to receive talk requests, who ask people not to talk-request them (assgirl), or who rapidly get their login names changed (there was a sex@uclink2, but she got so much harassment that she fairly quickly had her login changed).

Although other aspects of life on the net do affect women, the different methods of communication discussed above seem to contribute most strongly to the quality of life for women on soda. For that reason, then, these are the aspects of the net which will be drawn on most frequently for specific examples. Before looking at my specific project, however, it is important to have an idea of some of the theoretical background for such studies. Obviously, the study of interaction between people of different genders has been very popular historically and remains so today; tomes upon tomes have been written on the subject. Examining gender on the net, though still in its nascent stages, looks to become an equally common topic for academics. The focus of Lewis and Clark College's Eleventh Annual Gender Studies Symposium -- The Electronic Salon: Feminism meets Infotech -- reflects this new concentration in academia, as does the proliferation of other articles, sites, and servers relating to women and gender issues, many of which specifically focus on the net.


Because I am most familiar with soda, as I mentioned before, I focused upon the soda community. The majority of my direct research, then, occurred via a
survey which I wrote and kept in my home directory. I posted a message in the unofficial message of the day (soda's motd has a world-writable section which is used as a way of communicating current events affecting members of the soda community.). From this, and from both walled and nwritten requests that people fill out the survey, I received 48 survey responses, ten of which were from women. Though this may seem like a very high male/female proportion, it's actually the exact opposite; the only reason that I got so many female respondents is that I actively sought out women who I did not know to get them to answer the survey; I did not work so actively to get male responses. The plain fact of the matter is, though, that men far outnumber women on the net, and soda is no exception. People quite obviously put different amounts of time and thought into their surveys; this can be seen in the fact that some of the surveys had long, involved answers, while in others people skipped all the "difficult" questions.

The survey questions began with simple demographic information, asking for name, gender, age, occupation/major, Internet accounts, etc. This was aimed at determining the variance in user background. I also asked for information about how much time people spent online, and how they used their time while online. I then moved to more complicated questions, less easily-answered, asking about people's perceptions of the different roles men and women fill on the net; how they behave and how they're treated. Finally I asked for any overall comments about gender on the net. Though people had problems with the phrasing of each of the questions, the specific questions seemed to vary and even the most vague produced a few very valuable comments.


Though there are many different ways to report the results for a survey such as this, the vast differences in people's responses and the fact that the answers are not simply multiple choice but often almost essay-style prevent any sort of quantitative description. Instead, I have been forced to rely upon a more qualitative methodology. Nevertheless, it is possible to discuss some general trends among respondents, again aiming toward giving some background information about the population I am considering.

Respondents ranged in age from 18 to 28, with the majority being between 19 and 22. As previously mentioned, most were male, and the majority are either currently studying some aspect of computer science (including EECS, L&S CS, etc.) or working in some sort of computer-related field. Interestingly, the women who responded were much less likely to be CS majors; there was one female EECS (electrical engineering and computer science) major, and one each in the chemistry, physics, and chemical engineering majors, while the remainder were English (three), psychology, mass communications, and other liberal arts majors. Men, however, were much more computer-science related; 23 were either EECS or L&S CS, and very few were liberal arts majors. The CSUA is an ethnically-mixed group, though the majority of members are either Caucasian or Asian. Most respondents seem to have several e-mail accounts, which makes sense because undergraduates tend to get uclink or OCF (Open Computer Facility) accounts first and later possibly get a soda account if they begin to feel constricted with the rules and regulations that constrict the other systems. Respondents had used computers for between three months and fourteen years, with the average being around three years.

Responses differed greatly, from people who insisted that there is little, if any, difference between the way men and women act and are perceived on the net to those who see soda society as a microcosm of the culture in general and thus subject to the same stereotypes, prejudices, etc. Finally, some respondents very vehemently point to discrepancies in treatment and behavior, and consider the on-line environment very harmful to women in general. Generally, women saw a larger gap in their treatment and the expectations of them than men did, for what some might consider obvious reasons. But the particularly interesting part of this survey comes not in a general recital of overall trends, but from hearing people's own responses, and looking from the specific to the general. That, then, is the focus of the next section.

A Day In The Life Of Soda (or, the faces behind the logins)

In this section, the
survey responses really speak for themselves. I was most surprised by the major discrepancies among what people thought (though perhaps I shouldn't have been); there was little consistency even in the areas where I would have expected people to agree for the most part. For example, the question "What is the difference between man and women's uses of the net?" generally received one of two very different answers; a fairly sizable minority answered something along the lines of:
"Women are MUCH more productive than men, because they actually do something. Men just sit there and act like a total moron"
(from a male user), but the majority of respondents gave some variation on the following: Elaborating upon that perception, one female user wrote,
"Hmm . . . men are perceived as knowing what they're doing, it seems to me. They have some legitimate reason for being on the net; they're working on some vital project. Women, though they may be just as serious and knowledgeable, seem to be perceived as being in male territory.... I'd say that men are expected to produce more, to create more new programs, etc. Whereas if a woman comes up with some amazing new way of doing things, it's a huge wonderous event, and she'll get lots of media attention because she's female and she's thought up this new thing, not just cause she thought up this new thing."
People did seem in general to agree, though, with the idea that, as one respondent put it,
"On soda, the active males tend to be very much into the 'I have clue, I am a power-user, I can COMPILE THINGS' mode of thinking; women are less like this."
Men, they believed, tend to be very proud of what they can do and to work to learn more; as a corollary to that, many women seem to simply want to accomplish a task. On a very personal note, last week I was with three friends, two men and a woman, and the conversation turned, rather naturally, to computers. The men immediately began comparing who had more accounts, who had root on more systems, who had access to more disk space, and who had compiled more programs; my female friend and I simply started laughing, as we both recognized it as a stereotypically male conversation. Thus, whether or not the observation is universally true (and that I strongly doubt), it does have a certain amount of resonance for many of the people who use the net, or at least those on soda.

The idea that men do more advanced things has far-reaching implications, because it directly impacts how people treat women when they ask for computer help. As previously mentioned, part of the function of soda's walls is to answer questions; additionally, people will very often just ask a friend or acquaintance for help when confused. Many survey respondents noted a difference in the types of question women asked ("There seems to be a trend among women to know less, or just enough to do what they want to do, as opposed to some men's almost obsessive interest in the most minute details of the systems they use. Whether or not those are really useful for what the computer is ultimately used for."), as well as the responses they were given. Indeed, quite a few of the respondents admit to answering women's questions differently than those of men. People write,

As the last few comments hinted at, along with (or possibly even independent of) people's different expectations of women's knowledge and their different treatment of women when responding to questions comes a general difference in the treatment of women online. Although this can sometimes be helpful (as when people go out of their way to help you learn how to do something), it can also be stifling (when things are done for you, you will never learn how to do them yourself); additionally, it sometimes can seem like men who help women online, or who even talk to them, sometimes assume they can have more of the woman's time or are more justified in bothering or flirting with them. And women online are flirted with constantly, on walls, through mail and nwrite, and through the use of random talk requests. The vast majority of respondents, both male and female, pointed to some aspect of this as being a problem for women on the net. One male respondent wrote that,
"women who post to are seen as 'loose' and get lots of 'wannafuck' [messages] from random net people. i've seen posts saying how they have accounts with assumed id's to spare them from people trying to mail/phone them. pretty scary. gender often becomes a scary issue. i had a female acquaintance at UCSC who told me she often got random talk requests when she logged on, and that they told her there was a "list" of asian women/logins on the net, even broken down by ethnicity. she would get quit a bit of random talks/mail from these people. i think people like this make the net a hostile place for women."
This is not unusual; women on the net do get harassed to the point that there are procedures in place at most, if not all, systems to deal with this sexual harassment. Berkeley systems are subject to the same rules and regulations as all campus departments are, as well as the campus computing regulations. As another example, the document new soda users read includes the following statement:

Soda should be a reasonably comfortable environment, so protection of user privacy is taken seriously. Any user who intentionally violates other soda users' privacy will be squished. Any user who uses soda to violate the privacy of users on other machines will be squished. Any user who makes repeated violations of net.etiquette will be squished. Your soda account is not a birthright, and it's much easier for us to squish you than to worry about Yet Another Problem User.

At least one respondent has taken disciplinary action against a male harasser; interestingly, in this case both participants were male. The respondent wrote:

"In my opinion, there is definitely some sort of harassment problem on the net. I have the misfortune of having a gender confusing first name. I am male, but because of my first name, I am often mistaken as female. After adding my middle name to my name field, I had a significant drop in the number of random talk requests. I have also been sexually harassed by another male, thinking I was a female. I had his account turned off by his local site manager, and disciplinary action was taken by their office of student conduct."
Indeed, the newsgroup alt.sysadmin.recovery had an ongoing thread about men randomly talk-requesting female users, and what could or should be done about the phenomena.

Because of the "pathetic horny geeks who try to hit on anything female, and find it easier to do if it's not face to face," many women choose to take measures like choosing a generic, asexual login name to save themselves from the harassment that some find so pervasive on the net. One woman wrote,

"You told me to change my login to my full name, instead of my initials, but I really would not like to do this. While it may be making a statement of feminity to change it, I am happy with my neuter, lurking personality. I am similarly genderless in the forums (fora?) I frequent on the net, and it has suited me just fine. I feel somewhat special to be part of the (quickly?) shrinking minority on the net, but I wouldn't want to flaunt it because I would be afraid of the responses I would receive. Which is pretty sad."

What Can/Should Be Done?

The above evidence could easily lead one to some very depressing and pessimistic conclusions. It sounds like problems between the genders are critical and inescapable; like there is no solution and no hope for any progress in the future. This is not the impression that I want to leave you with, however, and is far from the impression that I got from the surveys. Although a significant number of problems were pointed out in the surveys, quite a few included a ray of hope. Many people seem to feel like as more and more women came online, gender is becoming less and less of an issue, and the differences between men and women, at least in communication, are gradually diminishing. Discussing gender differences, they write, Hence, I am not certain if anything really needs to be done, if I can lay out a step-by-step plan for an improvement of the status of women online, or anything like that. Instead, it seems like possibly time itself will make the change for us, and indeed the fact that so many people did not even consider gender an issue would seem to indicate that this change is already taking place (however, it must be noted that the people who did not see any problems relating to gender differences on the net were almost exclusively male, and thus may lack the first-hand experience necessary to fully understand the issues -- similar to being white and trying to really comprehend racism). As more women, and particularly more knowledgeable women, make their presence felt on the net, the environment will change. (The issue of increasing the number of women on the net, however, is much larger and more complicated, and requires investigation of many interwoven issues, including schooling, social attitudes, gender aptitude and behavioral differences, if indeed such things exist, etc. Such studies have been and continue being done, and several of the sites previously mentioned contain pointers to articles discussing the issue.)


I wish I could conclude with a pretty, set conclusion: this is how it is, and this is why. But with an issue as wide and problematic as gender on the net, such a facile solution is completely impossible. All that I have been able to do in this paper is raise some of the many issues that exist in one specialized subset of the net, and report upon people's attitudes and views about the subject. Perhaps equally as important to me, however, is what I personally have learned from the study. Realizing what people's perception is of women on the net has encouraged me to go against the stereotype. Because I do not want to be just another non-clued woman, I have grown determined to learn how to do the "power" things on the net... not merely to prove to other people that women can also do "cool things" online, but also to prove to myself that I can do them. And I will.

Tara Bloyd (