virus: Epistemology of Email (version 2.0)

Eric Boyd (
Sun, 13 Jun 1999 12:34:14 -0400

Hi Virians,

As promised, here is my Epistemology of Email, including my proposed solution, which is the area now where serious work must be done.

You can read this essay online at

although this email contains imbedded comments to Virians, while the web page does not.

{begin include: word count 5151. Sorry about the length.}

An Epistemology of Email


Long term experience with email has revealed certain /dynamics/ which always operate in any list-based email forum. These dynamics -- described below in detail -- range from

(1) misinterpretation (lack of bandwidth or context) to

(2) motivational problems (why people write emails) to

(3) constant interruptions (leading to lack of resolution) to

(4) isolation of discussion (caused by an inherent one-on-one
discussion style).

Many of these effects can be highlighted with a memetic viewpoint, and I do so – but regardless of the lens through which you view them, the influence of these dynamics is so strong that one is reminded immediately of McLuhan's "the medium is the message". One gets the sense that the dynamics control the discussion and lead it away from what is wanted (knowledge creation) and towards

(1) memetic fitness (e.g. flamewars) and

(2) cycles of unproductive discussions revisited time and again.

It is the latter problem that plagues Virus -- to the point where nothing is ever decided. The problem is thus one of knowledge creation: we need an epistemology of email; a systemic understanding of how knowledge is (or could be) created via email discussion. Only then can we break the cycle forever.

My theory is that such a system will rely on a deep understanding of the dynamics of email, and a conscious approach to email writing which will attempt to change the dynamics -- hopefully leading eventually to

(1) a more "human" discussion forum,

(2) knowledge creation in any field of mutual Virian interest and

(3) an explosion of the potential of email into actuality.

Any criticism and comments on this essay are welcome, indeed invited.


Anybody who has spent much time with email knows that a certain dynamic is always present – that communication via email has a distinctly different "feel" from that of face to face conversation, or letter writing. As stated above, I have categorized the dynamic into four major sections: misinterpretation, motivation, interruption, isolation. A memetic view of the situation is also presented, as well as a business meeting perspective, on the view that the more angles we have of viewing the problem, the better we will understand it.

Misinterpretation: "That is not what I said!" Email messages have a distinct tendency to be read in the wrong way. I have identified many causes for this phenomenon. The most important is that email is text based. As text, people approach what is said in emails with certain assumptions – those that text in it’s more common incarnations, such as books and signs, have given them. Rational. Authoritative. Purposeful. It follows that if email posts are not written as such, then they will commonly be misinterpreted. A typical result is the conversion of discussion into debate – and then, according to a dynamic I am about to explain, into fight. Each person involved interprets the content of the others post as being more rational, authoritative, negative, and purposeful than the writer intended, and the cycle, via positive feedback, escalates into an explosion: the flamewar.

There is a second cause of misinterpretation, which also plays into the cycle above, is the lack of emotion or tone in the text. A text-based medium simply does not carry sarcasm, irony, emotion or any of the other modes of implicit communication that humans are capable of. The scope of this problem is so large that partial solutions already exist to solve it: emoticons, such as the smily face :-), and bracketed comments (irony alert!) both serve to clarify the writers intent. When one remembers to include them, this solution is workable, although often the inclusion ruins the effect in the mind of the reader and writer. I am not terribly concerned with this particular effect, since it doesn’t relate to knowledge creation, but it is worth including in a survey of email dynamics.

Finally, there is the issue of context. How much explanation, how many words do you need to use to communicate your point? How much of your intellectual background is shared by your readers? Again, because of the divorce between people over the internet – they can only communicate by words – and because of the impersonal nature of text, it is often hard to judge people and their background. The international nature of much of the internet is also a hindrance: one cannot count on shared experience of culture. For long term discussions and email lists, there is an issue of history: "this has been said here before". Even recent history (like two or three posts ago) is often long enough for people to forget what was said – and although quotation of the last message helps, extensive quotation itself causes problems. The development of FAQ’s is like an outsourcing of this problem; with common discussions curtailed by reference to a compiled list of answers. But anyway, this paragraph is clearly too rambly and broad to serve as the final draft... there are ideas which need development here.

Motivation: Why do individuals choose to write emails? I have two compatible and complementary views on origin of motivation. The first is that motivation stems from the perception of a problem: that action is the result of a conflict between theories. This general theory reveals itself in email writing mostly via replies: the writer to be reads something she disagrees with, and writes an email to point out this fact. She thereby hopes to solve the problem (conflict between theories). My second theory, which is actually more of a special case of the first, is that people have needs-which-seek-filling. This often explains the origin of new discussion topics, as people bring up something they need or want to talk about. So, for instance, Karl writes an email to his friend Janis because he wants to inform her about recent events in his life. Karl had a need-which-seeks-filling, and he filled it by writing an email. Janis receives the email, and writes back because she cannot believe that Karl has moved to Ottawa! She has a problem (conflict between theories), and she is seeking resolution by emailing Karl back to be sure.

Now, in the case of personal correspondence, motivation is not generally a problem. On the other hand, for email-based lists in which the ultimate goal is knowledge creation, the above motivational system is woefully lacking. What motivates a person to post a new theory? What motivates a person to make constructive criticisms, rather than destructive ones? What motivates a person to spend the time necessary to write a good post? The answer is commonly nothing – and the result of that is silence on the email list, or low quality posts. This is the number one problem to be solved.

There are a few other details of email writing related to motivation. The first is the perceived immediacy of email. An email which is five days old is – five days old! and is thus clearly not worth replying to... I cannot count the number of times that I have wanted to reply to an email, but put it off for so long that, eventually, I say: "to hell with it... they’ve forgotten it anyway by now", and into the garbage it goes. This is a big problem for posts which have difficult subject matter, or for which much thinking will be required to craft an intelligent reply: the time limitation is crippling. Another effect of the perceived immediacy of email is the gut response reply, which often serves more as an incitement to flamewars than anything else. Essentially, a response is banged out and sent before critical reflection has a chance to act, because people are treating email writing like speaking. Since it’s not, and in fact is not treated like this when read (see above), discussions encounter problems when people shoot from the hip.

Another real-time related problem is actually motivating yourself to read emails. Usually, this is not a problem, but at times – with high list volume, or really long messages, it can be difficult. It just doesn’t seem that email, such a quick medium, should take that long! I know from personal experience that any email beyond about two thousand words is simply not going to get read. Taking with others has revealed that my limit is rather generous – even five hundred words is enough to justify "skipping" the message for some. There may be ways around these limits, and I will talk about them later. (note that I do not, yet)

Interruption: This is the most important and most crippling of all the dynamics for list based communication via email. You have received the email, you have read and (mis)understood it, you are motivated to write a reply. Now what? You press reply, and the computer begins a new message, usually automatically including the text of the old email. Your reply, then, consists of inserting comments at appropriate points in that text, or making comments which refer to it. You are interrupting the email every time you insert space for your reply; you are making comments on exactly the details you have a problem with. It is a very powerful technique, this interrupting. But what you almost always forget, when you "interrupt" an email in this way, is that in doing so, you have not interrupted the writer -- they finished the thought and the message before you ever saw it. Have you possibly missed the point of the email, in your concentration on specific pieces of it?

By itself, a few interruptions are not particularly harmful – indeed, like I said, it is a powerful technique for clarifying exactly what the problem is. However, emails in a thread tend to be built on each other, which is to say, the fourth email is a reply to the third, which is a reply to the second, which is a reply to the original
(thread introducing) email. Because of this series of emails, locally
harmless interruptions can become global problems, because while each interruption is only a little shift in attention away from the topic, the person who replies to that interruption shifts the discussion further away again, and in a few generations, the header topic is no longer appropriate: the conversation is now about some aspect of the former discussion which has only tangential relevance (if any at all). Worse than that, even if agreement is reached in this tangential discussion (which is itself unlikely, due to the same effect), the participants are left without any easy way to link it back to the original topic; and this leads to lack of resolution. Problems are posed, and people jump merrily on board to talk, but no actual resolution of the topic at hand occurs, because the conversation diverges and then peters out. Knowledge creation fails. Eventually, somebody re-proposes the topic, and the cycle begins again.

One can also liken this divergence process to a childhood game called "telephone". An individual, usually the teacher, originates a message. She whispers it to the first child, who whispers it to the second, and on through the whole crowd. By the time the message has reached the end of the line, it bears almost no resemblance to the original message: it has diverged because each communication is built on the last. Communication via email is different in that "the same" message is not being repeated, but the essence of the analogy holds because one wishes to maintain the topic of discussion, but successive "whispers" inevitably lead it astray.

Isolation: Email, even in discussion groups, is an individual endeavor. Each participant sits at their desk, alone, reading and writing emails. Even more than this, email is not tailored to a group effort; the structure of email and program which implement it, assume you are writing to one person, or at most sending a message to multiple people. There is no easy way to reply to multiple people. Therefore, each message written is only a reply to one person; regardless of the number of people participating in the discussion. Given this, it’s not surprising that a conversation beginning with three or more people eventually diverges into two or more conversations with two participants each – that is just an artifact of the interrupt dynamic. Conversations thus become isolated from each other, even though list based communication is supposed to be a group effort.

To clarify the isolation dynamic, I have created an elaborate example of the typical email dynamic, as well as showing an achievable dynamic, if the writers are willing to expend a little effort. (But I anticipate...)

See Figure 1 and 2.

#include html link to jpeg image: >>

Emails are denoted by circles, the numbers inside are authors. Each "conversation" contains exactly the same content (any "reply" in figure one has it’s counter part in figure two), but the content is formatted and sorted differently among the messages by a different dynamic. It is my contention that, in real life, these conversations would have evolved quite differently, with figure two representing a much fuller and more complete analysis of the subject.

Figure 1: The current email dynamic, the result of interruption and isolation. The "one-on-one" structure is clearly visible. The lack of resolution is apparent. There are twenty-one separate emails.

Figure 2: The modified email dynamic, the result of a conscious "reunification" and goal directed writing process. The unity of the conversation is clearly visible. There are fifteen separate emails.

I intend to do an empirical study on an actual Virus thread – I’m thinking the recent discussion on Virus Maxims would be a prefect example of the interruption and isolation combo. Certainly discussion was unproductive... Such a study will be attached to the essay as an appendix – it will not replace this analysis, as it’s likely to be too messy to carry the points clearly. It will however, be a real demonstration of the email dynamics in operation, and if a solution is ever implemented, it would be interesting to revisit the topic to see what can be done with conscious writers in a goal directed thread. {note: I have completed about 60% of this project -- but I lack a program to graphically display the results. Right now, I have a large list of emails in Excel with details about which message was a reply to which message. I wish to create something like my current figure
(only much bigger, of course), but am not willing to do all the menial
circle and line drawing that that figure took. Anybody know of a program which will do what I want?)

The four email dynamics mentioned – misinterpretation, motivation, interruption, isolation – serve as a complete analysis of email from the individual writers perspective. However, there are several other perspectives which also yield insights into email dynamics, most notably memetics and business meetings.


Memetics is the science of idea propagation. At it’s core, memetics consists of an evolutionary theory of information transfer – with the meme playing the role in memetics that the gene plays in biology. A meme is thus a unit of information which, in some way, influences events in such a way as to cause it’s own replication. Common examples of memes include fads, jingles, and email forwards. Each of these items ‘causes’ the humans who receive it to act in such a way that the meme is spread – in the case of email forwards, a particularly good meme will be forwarded again by each user who receives it.

What does memetics have to do with email dynamics? Everything, of course! One of the most interesting memetic effects on email is the "flame war". I have given a few reasons above, but a memetic viewpoint reveals another – since people are more motivated to write a response to negative, attacking emails than to positive, supportive emails (and what they write is itself often a negative email), it follows by memetic theory that negative emails will out propagate positive ones – thus leading to "flame wars" in the extreme case. But even in non extreme cases, the negative-tone begets negative-tone dynamic is clearly visible – and quite hard to avoid, even when one is aware of it happening. (if you’ve ever tried to reply to a truly negative email without yourself being negative, you know what I mean)

A list based email forum is dominated by the effects of idea-evolution over time (in successive posts). As such, one might expect that memetics would have a fair amount to say about the dynamics of email. But the nuts-and-bolts of memetics has already been spelled out above: memetics predicts that the ideas transferred via email will evolve to enhance their own transfer, unless artificial selection is employed. It is my contention that as long as people are conscious of the role they play in idea-evolution, the memetic disutopia of nothing but button-pushing self-replicating ideas will rarely dominate discussion.

One of the major differences between memetic theory and the biological theory of evolution is that with ideas, Lamarkian evolution is the rule, rather than Mendellian. What this means is that many ideas are "hopeful monsters", large mutations of present ideas which lack a breeding population in the global meme-sphere. They may fail not because they are bad ideas, but because they do not spread! The lesson here is that if we want unusual memes to live, we must provide an environment in which they can replicate. The purpose of this project is then seen to be a modification of the environment of email to support the existence of memes we like; i.e. true, useful or valuable memes, instead of viruses, trojan horses and parasitic memes of all flavours.

Business meetings:

Tim Rhodes writes "as anyone who has every conducted a meeting knows, the hardest part is keeping the discussion alive while trying to keep it from degrading into a conversation. Keeping the participants ‘on topic’ is the role of the leader in a meeting, but in a mailing list there is no leader, so who keeps these discussions ‘on topic’?" As my dynamics above reveal, nobody keeps discussions on topic, and because of that, discussions do not stay on topic! A possible solution, then, is clearly obvious from the business meeting metaphor of email lists: a leader is needed. But how should this leader be designated, and how is s/he to keep discussion on topic? I address these issues in The Solution, below. (or, rather, I will, although I think it's obvious what I'll say)

Extra stuff (left over from version 1.5) which I intend to integrate:

Finally, I want to talk a little about silence in emails – about how not responding is often taken as agreement. This results, I think, from the view that, if they don’t write back, they don’t have a problem with your post, i.e. they agree with it! An interesting strategy, which I have seen used, and which I call the "factual snow job", is to simply bury the "opponent" in information; so much so that they will be simply unable to reply to it all. It is a cruel and unusual strategy, and "works" only so long as silence is taken as "victory". However, in the proper light, silence is seen not as victory, but as failure – no problem resolution has occurred, no synthesis is present; instead, communication has been halted! Silence is, of course, the ceasing of the process of knowledge creation – and can be viewed as success only if acceptance or rejection of dogma is taken as the standard of knowledge acquisition.

One proposed solution to this problem on Virus was the (so-called) Reed principle, later developed into The Discipline of Translation . The idea was that, just before the topic veers out of hand and people move on, each participant would calmly review the discussion, and present both their viewpoint and what they see as that of their opponents. In general, I think this principle itself is sound – that spacing oneself from the discussion and writing a summary often results in the necessary refocusing, as well as the equally necessary input of creative thought (variation) – in essence, both parties sit back and write a synthesis – then further discussion can occur based on the new positions. However, because the process is "forced", as it were, and alien to the normal dynamic of email communication, it requires a certain consciousness of purpose – which is often hard to invoke precisely at the times it is needed most. Something more natural is needed.

THE SOLUTION: (which is still version 1.1 material)

Well, as I stated above, I believe that any solution worthy of the name begins with a more conscious approach to email communication.

The specific problem addressed in this analysis – how to create theoretical knowledge via email communication – can admit of a solution only if all participants work together. A most important step – and one which hardly ever occurs via email – is the "Statement of the Problem", i.e. what is our motivation in attempting this theoretical development? Lacking this, I feel that any effort will fizzle out quickly; the initial excitement of exploring the idea-space dies out long before the necessary development of ideas can take place. In order for this to have any long term effect, however, this Problem must be remembered even as the local problems – small disagreements inside of the framework of discussion, those that often prompt the next email in the series – are dealt with, so that the work continues even after successful resolution of the issues immediately at hand. (it is this focus which the Reed Principle helps to stimulate). This type of "big picture" thinking is especially needed in email because of the "interrupt" dynamic explained above. Finally, a conscious approach to individual knowledge creation must also be present – if someone advances a thesis, it is not enough to criticize what seems to be wrong; one must also propose an antithesis, or at least suggest wherein the solution to the perceived problem might lie, so that the process of knowledge creation can continue. (without variation, selection is useless). What one must always remember is that an email is an individual creation – that, despite the "community" feel of an email list, all knowledge creation eventually occurs at the hands of one person, alone at their terminal – the email writer. Therefore, buckling down and actually doing some thinking
(instead of responding from the hip) is absolutely essential to
knowledge creation via email (and probably more generally as well). [Virians: this "solution" is rather wordy and vague. It’s main points are: (1) Focus on the Purpose (big picture) of the Email and (2) Put some Effort and Substance into writing emails – knowledge does not create itself. Tim Rhodes is currently re-writing it (are you still, Tim, or should I do it?), and below, in THE CHECKLIST, I have created a more action oriented version]


To implement the solution, I think an "email writing checklist" is ideal. The plan is that the core members of Virus (and we know who we are) will begin by consciously implementing a better email dynamic – one which includes the key insights derived in the above analysis. If enough of us do it, for long enough, it should become second nature to us – and the control we exert on Virus will keep the positive dynamic we have engineered in place and working. It will be necessary, from time to time, to "train" those new members of Virus who obviously intend on staying around.

So anyway, the checklist / questionnaire:

Before one begins writing, ask yourself this:

(1) (a) Why am I writing this email? What is my motivation? What
problem am I attempting to solve? What is the Big Picture?

(b) Based on the above, is this email worth writing? Do I need to clarify the topic / problem with the other participants?

Now, write the email, and ask yourself these questions:

(2) (a) Are there other posts on this same topic to which I should
also reply? (if so, it should be done in this email, to avoid balkanization of the conversation)

(b) Is there any discussion in this email which would be better with a different thread name? (if so, place it in a new/different email or thread, to avoid misdirection)

And, finally, when all the writing is done, consider these questions:

(3) (a) Have I clearly expressed my solution / communicated my
message? Is my tone positive and my email creative rather merely destructive? Have I placed sufficient background (quotation) and context (explanation) in the email for it to make sense to Virus readers in general?

(b) Is this email worth sending, or do I need to revise it in order to change my answers to the above questions? (possibly even scraping it, and beginning the process again)

The Call to Consciousness:

(4) Have I honestly answered each of the above questions? Do my
answers make me feel happy about sending this email to the list? If not, why not, and what can I do to fix that?

Finally, the last step:

(5) Send the highly polished – happy – email, and return to step one
for the next message. (joy oh joy!)


This section is just for random thoughts that I think will be useful later.

(1) Message headers. Is a code system a good idea? Can we use
headers to direct conversation to desirable goals? To keep our attention on the topic? To communicate our goals? How about an "extended header" in the body itself, detailing the messages purpose and a one line statement of thesis?

(2) "automatic mechanism of originality". Can we intentionally make
message writing difficult, so as to *force* originality? If so, how?

(4) How to keep the techniques (solution) fresh and conscious in
peoples minds? "the question (in part) becomes, how do you get people concerned enough about the culture of their virtual community to make that kind of an investment in it?" (Prof Tim)

(6) As to training newbies, I think that would only become necessary
if they remain around any length of time. Provided the core of Virus has the techniques, any newbie cannot have enough of an effect to alter the positive culture of Virus. At least, that's my hope. Otherwise, we would have to work up a real system for new posters -- which we could probably get included in the "sign up" emails that majordomo sends to new subscribers. That might also be a good idea... We need this. The core group of regulars is *not* sufficient – as witnessed by recent (may99) events on Virus.

(14) Email allows us to combine the exact, systematic and reasoned
nature of *writing* (think philosophy books) with the rapid stimulus and variety of the "beerhall", to use your term. This combination could be *explosive* -- and if we can harness that explosion, it may usher in a "new age" of knowledge creation, accelerating our growth of knowledge even beyond the dreams of the Extropians.

(15) What I want to accomplish with my epistemology is to
fundamentally transform the *writer* of the email, in such a way that the negative dynamics of email communication are avoided, and the potential of the medium is free to "explode".



Looking at the above solution, it seems pretty obvious that, indeed, this type of interaction and knowledge creation is alien to email communication as we know it – that only consciously striving and fighting against the grain of email can it be achieved at all. In that sense, it is not a solution to the problem – because a solution was supposed to make this type of knowledge creation easier.


Speak clearly, if you speak at all; carve every word before you let it fall. -- Oliver Wendel Holmes, Jr.

My point is -- the internet, like the theater is not a broadcasting medium, it is a narrowcasting medium. People come to you. -- Stephen Atkins <>, on the CoV mailing list

Godwin's Law of Nazi Analogies: As an online discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches one.

"You cannot solve the problem with the same kind of thinking that has created the problem." – Einstein

"Bound only by our imagination, we have endless opportunities to design and build tomorrow's global infrastructure." -- Les McCraw

"[So] long as the proponents of our best theories... have their attention turned defensively backwards, and expend their intellectual energies in futile refutation and re-refutation of theories long known to be false, the state of our deepest knowledge cannot improve." -- David Deutsch, *The Fabric of Reality*

Imagination is the beginning of creation. You imagine what you desire; you will what you imagine, and at last you create what you will. -- George Bernard Shaw

When I was a boy, I felt that the role of rhyme in poetry was to compel one to find the unobvious because of the necessity of finding a word which rhymes. This forces novel associations and almost guarantees deviations from routine chains of thought. It becomes paradoxically a sort of automatic mechanism of originality. -- Stanislaw Ulam, _Adventures of a Mathematician_, 1976

One of my responsibilities as a Memetic Vector is to condense and redistribute the data I gather from eclectic travels so others don't have to. I would never ask everyone to go and read/hear/dance to/talk with/visit all the people and places I have just to understand something I say. Time is a very limited resource (for instance, I get about 1 hour a day to play on the computer, that's it. CoV alone can take up most of that time). As a vector it is important to package ideas in a manner that is accessible to others. By giving background and defining terms I can increase the odds that my insights get communicated. This is the theory at least. -- Tim Rhodes <>, on the CoV mailing list

Let us not look back in anger or forward in fear, but around in awareness. -- James Thurber

Remember - strong and bitter words indicate a weak cause.

"You must be the change you wish to see in the world." – Gandhi

"Learning is a treasure that will follow its owner where." -- Chinese proverb

A clash of doctrine is not a disaster - it is an opportunity.

This is an example of clinging to dictionary definitions of words, as if such definitions are a static truth--an unquestionable authority from which true knowledge is derived. In this mode of thinking, knowledge is purportedly gained by putting together words in their dictionary meaning. One problem with this mode of thinking is that new knowledge is not gained by taking discrete concepts, (i.e. words) and putting them together to form a big picture, but by seeing the big picture first and then breaking it down into its components.
Another problem with this type of thinking is that there is a stagnating dogmatism when a word, or any concept (i.e. God, scriptures, etc.) is given status as ultimate authority. As far as the thinking process is concerned, there is no important difference between which "authorities" are exalted since the activity of deciding that there is an ultimate truth or authority leads to a closed mentality that is not open to different ways of seeing. -- David Rosdeitcher

Insanity is doing the same thing over and over again, but expecting different results. -- Rita Mae Brown

How different our lives are when we really know what is deeply important to us, and keeping that picture in mind, we manage ourselves each day to be and to do what really matters most. -- Stephen Covey

{end include}

Just so you Virians know -- I'm sending this in the hopes that the replies I get will re-motivate me to do some work on it. I still think the Epistemology has a great potential to improve online communication -- although I am less certain than I was at the beginning.

For comments on specific parts of the essay, be sure to include some way for me to tell where it is you would like an improvement -- quotation, or paragraph #, or whatever.