T H E  N E T W O R K  O B S E R V E R

  VOLUME 3, NUMBER 6                                     JUNE 1996


     "You have to organize, organize, organize, and build and
      build, and train and train, so that there is a permanent,
      vibrant structure of which people can be part." 

          -- Ralph Reed, Christian Coalition


  This month: Ralph Reed's new book
              Running a filter list on the Internet
              Software for professional networks


  Welcome to TNO 3(6).

  This issue begins with a review of Ralph Reed's new book, "Active
  Faith".  Sorting through all of the debate about the Christian
  Coalition, my conclusion as always is that they are doing good
  basic social organizing and that everyone should study and learn
  from them.  I have mentioned the Christian Coalition frequently
  in TNO because they provide so many lessons in the practical
  mechanics of politics.  These mechanics include technology, but
  the technology only makes a difference if it's an integral part
  of a coherent strategy and organizational form.

  I've also summarized the lessons I have learned in three years
  of running the Red Rock Eater News Service, the mailing list to
  which I send whatever I find interesting.  The general category
  that includes RRE is called, I gather, a "filter list".  Filter
  lists are a low-cost, low-effort way to get useful information
  out to large numbers of people -- large, anyway, by Internet
  standards.  RRE's membership, for example, is approaching 4,000
  (perhaps 75% of them in the United States).  On the other hand,
  these excellent people only amount to one Internet user in
  10,000, one English-speaking human being in 200,000, or a few
  percent of the circulation of an American political magazine.

  This month's wish list sketches a family of programs and data
  conventions for building and maintaining professional networks.
  I have no idea if such a scheme would actually work.  The
  interest of the exercise lies in the patterns of congruence and
  tension between the workings of the technology and the workings
  of the social system -- in this case, the social system of the
  research world.

  A footnote.  Let us once again praise "Dilbert".  When I came
  across a book of Dilbert comics that had been translated into
  Norwegian (subtitle: "Nerdenes Konge"), I decided that I had to
  understand the deeper significance of, as the French would say,
  the Dilbert Phenomenon.  Fortunately for me, this imperative
  coincided with the publicity for Scott Adams' hardcover Dilbert
  book.  I conducted my research by scrutinizing the relevant
  primary sources while sitting on the floor in the humor section
  of the bookstore waiting for a friend.  I realized something
  that Adams himself had more or less said in interviews, that
  his early strips were unfocused and often unfunny, and that he
  only hit its stride once he decided to focus on workplaces and
  base his comics in large part on reports he gets on the Internet
  from his readers.  This is clearly as an important discovery.
  Everybody knows that employees in large companies want to do
  a good job, and indeed that their human dignity is tied up
  with the integrity of their work, but that they often *can't*
  do good work because of the four layers of managers above them,
  all of whom pursue their individual career strategies and speak
  their endlessly changing languages without much knowledge of
  the reality of the work.  But we've always known all of this
  in an abstract, piecemeal fashion.  By channeling these folks'
  war stories in syndicated cartoon form, Adams has turned himself
  into a regular voice of the people.  Yeah, okay, so it's totally
  goofy to discuss Dilbert in academic terms; I won't carry on
  analyzing Scott Adams as an organic intellectual, lest I become
  the inspiration for a cartoon myself.  But I do think we should
  figure out how to generalize the model that he represents.  Note
  that Dilbert's success is not just about the Internet -- it's
  about a synergy between two media, the Internet and the newspaper
  -- and that it is not just about the voices of the employees or
  the voice of Scott Adams -- it's about the closed loop between
  the two voices, with each playing its own distinct role in
  shaping the other.  Tell you what: let's declare a three-month
  moratorium on discussions that treat this thing, "the Internet",
  as the monolithic prime mover of history.  Instead, let's get
  practice naming the synergies and tensions between the Internet
  and other media, and let's remind ourselves that media have
  value for a civilized society when, and only when, they provide
  the means for new voices to emerge and to say things that matter.


  Assessing the Christian Coalition.

  Given my boundless respect for the political skills of Christian
  Coalition executive director Ralph Reed, I snapped up his new
  book, "Active Faith: How Christians Are Changing the Soul of
  American Politics" (Free Press, 1996), the moment it came out.
  It is a fascinating and sophisticated text that repays close
  reading.  The first thing that struck me is that it provides a
  powerful argument for the religious left.  That's very much the
  way the book is constructed: it portrays religiously motivated
  political action as the norm rather than the exception over
  the last hundred years of American life, with special attention
  to the temperance movement, the civil rights movement, and a
  series of social justice movements based in the working class.
  Reed places some emphasis on the scope and sophistication of the
  temperance movement, articulating lessons for his own movement
  from the tactical reasons for its successes and ultimate demise.
  But he is not eager to identify the Christian Coalition with
  the temperance movement, whose agenda he regards as discredited.
  Nor, much less, does he wish to identify with the historical
  religious right, which he treats, with persuasive eloquence, as
  a swamp of racism.  He offers little historical detail on racist
  uses of religion, though, and among other Christian conservative
  political movements he gives passing mention only to a 1950's
  anti-communist group.

  Having dissociated himself from the history of conservative
  political uses of religion, then, Reed invests considerable
  effort in portraying the Christian Coalition as the inheritor
  of the American religious left.  He starts by going out of his
  way to emphasize the continuity of religious motivations on the
  left through the 1970's.  This requires him to argue, not at all
  convincingly, that Franklin Roosevelt was inspired by Christian
  faith to promulgate the New Deal.  It also requires him to assert
  that the religious left basically disappeared in the 1970's,
  and that the liberal movement became essentially secular at
  that time as opposed to a badly managed coalition of tendencies.
  He makes no mention of the religious left's role in the 1980s'
  peace movement, or of its role in supporting resistance to
  the savage oligarchies in Central America that some Reagan-era
  conservative Christian organizations heavily supported.

  His idea in editing history in this way is that Pat Robertson's
  1988 presidential campaign and the Christian Coalition have
  picked up the baton of religious work in politics that the left
  had dropped.  He explicitly argues that the old religious left and
  the new religious right (a phrase that Reed curiously excoriates
  as an expression of bigotry without explaining why) are equally
  concerned with social justice; he uses the leftist phrase "social
  justice" repeatedly, although he does not provide any detailed
  accounting of what exactly the left and right have meant by
  this phrase -- not to mention the Catholic Church, whose members
  the Christian Coalition has begun recruiting through a newly
  formed organization called the Catholic Alliance.  He does an
  interesting and very complicated dance with the legacy of the
  civil rights movement, often drawing parallels between it and
  modern-day conservative evangelicals and then quickly backing off
  from any suggestion that this white social movement's experiences
  and grievances can be compared with the racist oppression that
  motivated the likes of Martin Luther King.

  (Although Reed is particularly cautious about it, this effort
  to dress conservative evangelical issue campaigns, specifically
  the anti-abortion movement, in the clothing of the civil rights
  movement is not at all unique to him.  Listen to a song called
  "Breakdown" by the Christian musician Michael W. Smith, from his
  new album "I'll Lead You Home", in which a speech by Martin Luther
  King about the repression of civil rights marchers is interleaved
  with electric guitars and a social-decay message.  Smith isn't in
  the same league as Steven Curtis Chapman, who I recommended in TNO
  3(3), but he exemplifies an important principle that the classic
  labor movement and the American Communists at their height knew
  well, but that the modern left is forgetting: a successful social
  movement must provide symbolic forms into which people can pour
  a wide range of feelings and experiences, not just alienation.)
  Reed's book is a monument to modern publishing technology.  It
  discusses events that occurred as recently as March and April,
  covering the South Carolina primary and the wave of fires at
  black churches.  Some glitches testify to the speed with which
  it was sent to press -- superscripts are missing for the first
  several endnotes, the name of the Promise Keepers organization
  is repeatedly mangled, and the text includes at least one garbled
  sentence -- but it is definitely a quality production, very well
  thought out as an intervention in the political summer before the
  1996 elections.  Just about every criticism ever made against the
  Christian Coalition is addressed, and it is here that the book's
  most disturbing feature lies.  Although he quotes every last
  hard word ("hate", "stealth", etc), Reed virtually never recounts
  the concrete evidence upon which these charges have been based.
  He does allude to the occasional passage from a fund-raising
  letter than gets endlessly recycled in the fund-raising letters
  from the other side, but we do not learn which passages he has
  in mind or whether he still stands by them.

  (This is a standard public relations tactic: reduce the issue to
  a vague formula and adduce three facts that seem to spin things
  in a different direction while distracting attention from valid
  criticisms by picking invalid criticisms to stand for the whole
  bunch.  As such, it is further evidence that American politics,
  as a professionalized practice, has been completely absorbed by
  PR.  Reed is himself very much a professional, completely under
  control; the contrast with Pat Robertson could not be greater.)

  Lacking evidence, then, the charges against his organization and
  movement start to seem mysterious, and he proposes to resolve the
  mystery by repeatedly suggesting that the real motivation behind
  the charges is antireligious bigotry.  It is certainly true that
  some stereotypes of Christian conservatives have been motivated
  by antireligious bigotry, and that certain charges against the
  Christian Coalition have been unfair, for example some people's
  lazy habit of equating them with the Klan.  Reed's professed
  antiracism may be good politics, but I have seen no evidence
  that it is a lie.  At the same time, it's grating to hear Reed
  complain that religious conservatives have been "demonized" when
  Pat Robertson routinely uses his television program to insinuate
  that his opponents are literally agents of Satan.  And I found
  it really remarkable that Reed could go on so indignantly about
  charges of stealth political tactics against his organization
  without discussing the evidence that has been brought forward to
  support them.  This evidence, as is well known, includes his own
  famous description of himself as a guerilla operating at night,
  the explicit description of the stealth strategy at Christian
  Coalition meetings several years ago, the testimony of people
  in communities where the strategy has allegedly been practiced,
  and academic analysis documenting severe distortion of opposing
  politicians' records in Christian Coalition voters' guides --
  which, by the terms of the Coalition's tax status, are supposed
  to be nonpartisan.  He does mention in a different context that
  war metaphors have proven inexpedient, but he does not recant the
  substance of what he has used those metaphors to say.

  Having said this, it should also be said that some portion of
  the "stealth" problem is the fault of liberals who remain almost
  completely unaware of the whole parallel universe in which Steven
  Curtis Chapman and Michael W. Smith are big stars and in which
  Ralph Reed is a pragmatic moderate.  If you want to know what
  the religious right is up to, my friends, you can go right ahead
  and attend a conservative evangelical church.  And this, I think,
  is by far the most important point to be made about Reed's book:
  even after all of the rhetorical tricks and tactical dodges have
  been laid out and weighed up, the bottom line, as Sara Diamond
  constantly and correctly emphasizes, is that the social movement
  that Reed represents is winning fair and square.  The reason
  for this is very simple, and it can be found in the quotation
  from Reed that is found at the top of every issue of TNO: the
  Christian Coalition invests a large proportion of its resources
  organizing its institutions, training its people, and building
  its movement.  They use technology, they build infrastructure,
  they learn from their mistakes, they repress self-indulgence,
  they keep refashioning their rhetoric and agenda to present their
  opponents a smaller target, they extend their coalition, they
  minimize the ad hominem, they reach out to make personal contact
  with the people they want on their side, and they pick fights
  they can win.  And they do all of this on their own resources.

  Given my interest in such matters, I was disappointed at first
  that Reed's book did not present more detail on the mechanics
  of the Christian Coalition's organization.  But of course he
  had no reason to write about this: his people know all about it
  already.  He writes largely from his own perspective as the head
  of the organization, and it is certainly instructive to hear his
  political analyses of various situations and to read his accounts
  of particular conversations, for example the meeting at which
  (he claims) he got the House to put the Communications Decency
  Act up for a vote despite having to rush off across town in
  a waiting car to appear on the Coalition's television program
  in five minutes.  Although nobody in human history has ever
  had the political acumen that Reed presents himself as having,
  I don't fault him for making things seem easier than they could
  possibly have been.  My point is that opponents of the Christian
  Coalition who might read this book looking for hints and clues
  in reconstructing a darker backroom story are largely misguided.
  If you're looking for shadows, you can certainly find them:

    Reed protects his organization's tax status by glossing over
    the moments where the Coalition explicitly sets out to support
    a specific candidate in an election.

    Several passages make more sense once you're aware of the
    internal factional politics of the anti-abortion movement.

    The reader gets little realistic sense of the coordination
    and negotiation that goes on among different organizations
    and factions in the conservative coalition.

    Reed offers no answers to the atheist group who observed that
    the Christian Coalition's claimed membership is several times
    larger than the officially stated circulation of its membership

    In response to charges that the Christian Coalition implies
    that people who disagree with its political positions are not
    good Christians, he replies with the non sequitur that the
    Coalition regards its positions as justified on non-religious
    grounds as well.

    The title "Active Faith" suggests that something is passive or
    otherwise deficient about religious faith that is not expressed
    through political action.

    He stretches reason to argue that Paul's instruction in Romans
    to render "taxes to whom is due taxes, honor to whom is due
    honor, respect to whom is due respect" constitutes a "biblical
    injunction" to "register to vote, become informed on the
    issues, and go to the polls".

    Anybody who has actually read Pat Robertson's "New World Order"
    will laugh at the bland account of it in Reed's book.

    Finally, he finds himself in the unenviable position, given the
    wide backing that Pat Buchanan received among the anti-abortion
    rank-and-file despite Reed's pragmatic preference to align
    with Bob Dole, of having to defend Buchanan against charges of
    anti-Semitism and pass over in silence Buchanan's abundantly
    documented racism.

  These are, of course, all matters that people should know about.
  But in my opinion the essential truth about Ralph Reed and the
  Christian Coalition is out in the open for anyone to see.  Reed
  is grasping a historical moment to do something that's just about
  as straightforward as it could be: organizing, building, and
  training a social movement and political coalition based on a
  coherent, consciously elaborated ideology that speaks to genuine
  concerns and grievances that a large portion of the electorate
  share and that other contemporary movements are not adequately
  addressing.  So long as Reed's opponents remain in denial about
  these facts, preferring to fight the shadows of their enemies
  from olden days, they will continue to lose.


  Starting a filter list.

  For the past three years, I have been running a mailing list to
  which I send whatever I find interesting.  Called The Red Rock
  Eater News Service (RRE), it currently has about 4000 subscribers
  in 60 countries.  It requires little work, and sometimes it
  does some good.  Such lists, I gather, are called filter lists.
  Sometimes people ask me how to start such a list.  I am afraid
  I have little technical advice to offer, since Mike Corrigan
  wrote all the code.  (It's all homebrew, so I doubt if it would
  port anywhere else.)  But I can report some hard-earned lessons
  from three years of running my own list.  I've formulated them
  as general rules, but obviously you should take what you need
  and leave the rest, following your own evolving sense of things.

   * Send a steady trickle of good stuff.  Two issues are foremost
  in potential subscribers' minds, based on bad experiences they've
  had with other mailing lists clogging up their mailboxes.  First
  they want a high signal/noise ratio, so you must promise to send
  only good stuff to the list.  Second they want a limited number of
  messages, so you must promise not to send too many -- ten messages
  per week seems to be the limit.  These two points are absolutely
  crucial, and they are the main points to emphasize in describing
  the list.

   * Define the scope of the list both in terms of a particular
  subject matter (e.g., "social and political aspects of networking
  and computing") and in terms of your own subjective interests
  (e.g., "whatever I find interesting in my guts").  That way
  you can wander from your official topic occasionally.  You will
  also have an excuse when someone wants something forwarded to
  your list that doesn't feel right to you.  The subscribers are
  subscribing to *you*, not to an abstraction, and they will stay
  on the list if they happen to share your interests.

   * Take time to build an audience.  Don't expect that you can
  blast out an e-mail announcement to the net and get an instant
  humongous subscriber list.  It will take time, and you shouldn't
  start unless you're willing to sustain the effort.  Social
  networks are not very interconnected, even on the Internet, and
  you can easily have high visibility in one community and zero
  visibility in another.

   * Never add anyone to a mailing list without their permission.
  If you think someone would like your list, write them a personal
  note inviting them to join.  But don't push it; don't become
  possessed by a desire to get this, that, or the other person on
  your list.  It will grow naturally if you let it take time.

   * Take some initiative to find good stuff to send.  Once my
  list had about 2000 subscribers and two years of happy operation
  behind it, I could rely on the subscribers for my primary source
  of good stuff to send to the list.  That probably won't work
  for a new list, though.  One approach is to subscribe to mailing
  lists X, Y, and Z, and send your subscribers the good stuff
  from those lists.  Assuming that your subscribers share your
  sense about what's good, maybe you can save them the trouble of
  subscribing to those lists themselves.  Decide whether you want
  to promise to filter those particular lists for all eternity or
  whether you want to leave it less defined.  You can also, if you
  have time for it, seek out worthwhile stuff by actively prowling
  the net, or by noticing announcements of conferences etc in print
  media and writing away for the electronic versions.  This will
  seem like a lot of work at first, but it gets easier as you build
  an audience who can help you find good things to distribute.

   * Make sure you can turn off mailer errors.  This requires
  that your messages include the appropriate field to direct
  error messages to an address which is not your personal mailbox.
  Perhaps that address is an alias that directs the messages into a
  file in your directory.  You will get an amazing number of mailer
  errors, absolutely no matter what you do.  Believe me: you will
  drive yourself insane if you try to reduce your incoming flow of
  mailer errors to zero.  Many of these error messages, moreover,
  will be badly formed, mailed to the wrong place, or otherwise
  useless.  If you can shunt them into a file then you can look
  at them at your leisure.  Even better, arrange it so that the
  messages can also be directed into the bit bucket (on Unix this
  is called /dev/null) for a few months.  It's a good responsible
  practice to weed bad e-mail addresses off your list occasionally,
  but not every day.  Every few months is plenty.

   * Do not ever give out your mailing list to others.  This
  is probably obvious, but I'll say it anyway.  If you are using
  Listserv, turn off the feature that lets anyone retrieve the
  mailing list by sending a command to the server.  On the other
  hand, you cannot make a blanket promise to protect the list.
  Your system administrator probably needs to see it occasionally,
  and someone might hack into your system and steal it.  Just
  promise to do your best.

   * Do not send mail to individual subscribers unless you really
  have to.  People generally want to feel a degree of anonymity
  when they subscribe to a mailing list, so you may invite bad
  interactions if you send a subscriber a personal note saying,
  "oh, hey, good to have you on the list" or whatever.

   * Try to get software that requires new subscribers to send a
  second message confirming their intent to subscribe to the list.
  This solves two remarkably common problems: forged subscriptions
  intended to flood a victim's mailbox with unwanted mail and badly
  formed subscription messages that create nonfunctioning entries
  on the mailing list.

   * Be careful about copyrights.  If you send copyrighted material
  to your list without permission, and you're located in a country
  that enforces the laws, then you can get in trouble.  Maybe
  not *that* much trouble, yet.  Still, you should assume that
  everything is copyrighted unless there's a good reason to believe
  that it is not.  Articles from newswires are always copyrighted.
  Treat messages that people have sent to other mailing lists as
  copyrighted, and get permission before you forward them.  (I just
  send the author a quick note enclosing their message and asking
  "May I forward this to my mailing list?  For details on the list,
  see ..." followed by the URL for the Web page that describes
  the list.  Then I say "forwarded with permission" at the top of
  the message when I sent it out.)  The lack of a copyright notice
  absolutely does not imply that the material is not covered by
  copyright.  A message that feels like a professionally written
  magazine article, for example, probably is; it may have been
  typed or scanned in, violating copyright, by someone besides
  its author.  Some genres, such as conference announcements, are
  obviously intended for wide distribution.  But that's a kind of
  implied license, not a lack of copyright.

   * Be careful about political action alerts.  If you send out a
  badly designed or misinformed political action alert then you can
  cause an awful lot of havoc.  Consider refusing to send out any
  action alert that does not conform to a set of guidelines that I
  laid out in the very first issue of TNO.  One particularly subtle
  guideline is that every action alert should have a stop date
  (e.g., "take this action until 15 October 1996 and then stop").

   * Decide whether you endorse the material you send out.  Unless
  you clearly specify otherwise, people will assume that you intend
  to endorse every word of every message you send to your list
  -- even messages that are obviously silly.  Consider including
  a disclaimer in the boilerplate message that new subscribers

   * Tell people how to unsubscribe.  Many people will not save
  your explanation of how to unsubscribe from the list.  So you
  need to provide them with several obvious ways to find out.
  No matter what you do, a trickle of people will send you little
  messages saying "please unsubscribe me".  If you've given them a
  real chance to find out how to unsubscribe, then you can ignore
  these messages in good conscience.  Alternatively, you can set up
  a keyboard macro or some other quick means of replying to such a
  message with an explanation of how to unsubscribe.

   * Don't let yourself be baited.  You will probably get messages
  from people who wish to recruit you into mind-games of various
  sorts, particularly if you send out messages with political
  content.  Refuse.  If you are an angel then you can send out
  polite responses to everybody.  Otherwise you might have to
  delete these messages.  I found that I suddenly got many fewer
  baiting messages after I wrote a long article about baiting in
  TNO 2(11), though I cannot be sure that it was the article that
  did it, and not a coincidental increase in Internet civility.

   * Have an agenda.  In sending a stream of messages to a filter
  list, you effectively speak in a distinctive voice.  Even if you
  don't add many annotations of your own, your choice of materials
  to send out makes a statement.  While I do recommend being guided
  by your gut feelings when choosing what to send to your list, I
  also recommend having an intellectual life that gives your guts
  something to work with.  For example, after the appeals court
  in Philadelphia struck down the Communications Decency Act, I
  sent my mailing list the press release from a prime supporter of
  the CDA, the Family Research Council.  Why?  In part because it
  seemed too obvious to send out the court's decision, which was
  already flooding the net by other means.  But more importantly,
  because I am trying to get the Internet enthusiast community to
  listen to the language and arguments of their opponents -- who,
  court decisions notwithstanding, are vastly better organized
  in plain political terms.  I didn't say this when I sent out
  the press release (though it's a theme I've harped upon in TNO),
  but it's what I had in mind.  I cannot guarantee that everyone
  will understand my reasons for sending out particular items, but
  I do think that everyone gets a sense of my voice and judgement
  and agenda if they care to remain on my list over the longer
  haul.  By continually trying to articulate to myself what I'm
  trying to accomplish, I help the list maintain focus and express
  a coherent personality.  And by rationing my own commentary as
  much as I can, I let people draw their own conclusions.


  Wish list.

  The world of research, like all social worlds, has elaborate but
  largely tacit rules.  Many of these rules concern the workings
  of professional relationships and networks.  The social networks
  of research worlds are tremendously important.  Sociologists
  commonly refer to them as "invisible colleges" (see Diana
  Crane's book by that title), since they are rarely written
  down and only become visible to outsiders through bibliometric
  research and the like.  They are not at all invisible to their
  participants, though, who treat them as real and deadly serious.
  In "Networking on the Network", I tried to help out graduate
  students by writing down, as best I could, some the rules that
  govern them.  But it also seems to me that networked software
  could support the development and maintenance of professional

  Here's how it might work.  Let us imagine that every researcher
  in the world has their own unique "web object".  This is a common
  enough idea, though it's not yet in the newspaper.  A web object
  resembles a home page, except that it has much more standardized
  structure.  And it resembles an entry in a distributed object
  database, except that it has various visible manifestations,
  one of which might look like a home page.  A researcher's
  web object would include contact information, interesting web
  links, and other such home-page stuff.  It would also be hooked
  into a digital library, providing ready access to citations
  (and, in many cases, full text) of that author's publications.
  Some people might put a lot of work into maintaining their web
  objects, whereas others might ignore them once they have been
  created, except for major transitions like job changes.  Some
  organizations might simply pay someone to create a web object
  for each member of their research staff.  It's crucial that
  everybody have an interest in creating a basic web object, even
  if they don't plan to think about it again.

  The web object would serve much the same purpose as a home
  page and online vita.  But it would also provide the starting
  point for much more elaborate functions, and these functions
  would provide a motivation for individual researchers would
  be motivated to construct a web object for themselves.  These
  functions start with a program called a professional relationship
  manager (PRM).  The PRM would resemble, in broad outline, the
  contact management software that sales people use -- although it
  would probably not be wise to point out the resemblance.  My PRM
  would include an entry for each researcher with whom I have any
  kind of relationship.  These relationships might run the full
  spectrum, from someone whose book I glanced at once to someone
  with whom I have collaborated extensively.  Each entry would
  include a pointer to that person's web object, further personal
  information such as home phone number and spouse's name, full
  text of my correspondence and notes on other contacts with the
  person, bookmarks and annotations on their writings, and so on.
  It needn't be elaborate, and most of the entries might even be

  As such, the PRM simply consolidates several existing functions,
  and one might worry that maintaining it will become a compulsive
  substitute for doing real work.  The real point of the system,
  though, is the basis it provides for various novel (or newly
  practical) functions.  It would be most excellent, for example,
  for each field to pool its members' online citations to provide
  a ready source for authors who need a citation quickly without
  going to the library.  It would also be terrific if each graduate
  student, as a requirement of advancement to candidacy, had to
  write a critical survey of research in a field -- not just as a
  text but as a hypertext with connections to all of the relevant
  web objects.

  Or imagine a PRM specifically to support the editor of a journal.
  It could keep track of all past and present submitted articles,
  help in selecting potential referees, and generally track the
  workflow on the whole editorial process.  This information would
  also be selectively available, courtesy of familiar security
  mechanisms, to various classes of people: guest editors, members
  of the editorial board, the publisher, authors who have papers
  under consideration, librarians, subscribers to the journal, and
  so on.

  Other PRM's might be built to support the staff of a professional
  society, a publisher, a department chair, a research grant
  administrator, a program manager for a funding agency, a library
  bibliographer, a sociologist studying a particular field, a
  public relations person for a university who brokers contacts
  between journalists and researchers, or the occupants of any
  other identifiable institutional role that relates to the
  research world.  Each specialized PRM could be organized with
  hierarchically nested domains, so that each individual has
  their own personal information and each group (e.g., members of
  a research team, publishing house, etc) could have information
  that is "inherited" by each member of the group.  This sort of
  structure is what object-oriented databases are all about.

  Another potential function of a PRM is automatic notification.
  The University of California library catalog, for example, is
  willing to automatically notify me whenever a new publication
  appears that satisfies a given predicate (e.g., it was written by
  my department chair).  This feature, though well-intentioned, is
  completely useless.  I have to type a separate command for each
  author and each database I want to search in, and then I have to
  update it by hand every six months.  With a PRM, though, I could
  say something more interesting like "please inform me weekly of
  new publications in any format by anybody whose work I have cited
  in my own publications, and daily of new publications by everyone
  in my PRM who resides at my own university".

  As individuals' web objects become integrated with other sources
  of information, other notification functions become possible.
  Publishers, for example, would have an incentive to produce their
  announcements of new and forthcoming books in a database that
  links to the authors' web objects, so that people who care about
  those authors can be notified about the books before they appear.
  Likewise, the universal event calendar that I wished for in TNO
  3(4)could refer to the web object for whoever is speaking in
  a given event, and people with relationships to that person in
  that geographic area could be notified automatically of upcoming
  speaking engagements.

  Other functions could start from citation information, to the
  extent that this information is available online.  The world is
  full of potentially useful techniques for bibliometric analysis
  of research communities, for example, and programs that implement
  those techniques could be running full-time.  Likewise, to the
  extent that full text is available online, existing systems
  for automatic conceptual mapping could be used to notify me of
  publications that might relevant to my research.  Experience
  with such systems suggests that it is crucial to provide me with
  extensive control over how I get this information, and when I get
  it, and how much of it I get.

  Certain automatic functions could use individuals' private PRM
  information in a way that does not violate their privacy.  It
  would be nice to told, for example, about individuals who are
  very close to me in the professional network, but who are not
  yet in my database.  Researchers could also use the private
  information in an anonymized fashion, much as epidemiologists
  use medical records to perform statistical analyses without
  having to identify the individuals whom the records are about.

  Once these automatic functions are in place, individuals would
  have easy ways to maintain up-to-date home pages.  I gather that
  literally hundreds of firms are building software to generate
  web pages from underlying databases, and this would be a good
  application.  We're really just talking about a new-age report
  generator, after all.  People could build templates that generate
  a decent-looking home page from the database entries that mention
  them (for example: display my name and contact information in
  these fonts with this indentation, then use our professional
  society's citation format to display all of my publications
  for the last two years that are listed in these databases,
  then a bullet for each of the classes I am teaching this term
  with pointers to syllabi, followed by this hand-written HTML
  code for my favorite links and other random stuff).  Then other
  people could grab those existing templates, perhaps modifying
  them slightly, as a basis for their own automatically generated
  home page.  The important point is that home pages would now be
  largely virtual constructions that are built on the fly, either
  daily or whenever someone fetched them, and not just a batch of
  uninterpreted HTML code.

  Automatic systems are fun to talk about, and they might even
  be useful, but they should not distract attention from the
  lived work of building social relationships.  I would propose
  that the real test of a PRM comes when I attend a conference.
  The point of attending conferences is to talk to people, talking
  to people takes preparation, and a PRM is useful if it supports
  that preparation.  A designer would want to ask: what do people
  talk about at conferences, and what information would it be
  useful for them to review beforehand?  It would be nice to know
  who is planning to attend a given conference, for starters: this
  information is generally available through attendee lists, so it
  would be reasonable to make it available in an object database
  ahead of time.  Before talking to someone, I want to know what
  they have published lately -- especially books.  I also want to
  review our correspondence quickly, to help ensure that I connect
  a person's disembodied e-mail persona with their face.  If I am
  organizing a book or conference in which that person is involved,
  or might be involved, then I want to have the materials relevant
  to that project organized in a handy form.  If I will have a
  computer handy at the conference then I can just prepare a file
  that makes all of this material mouse-click handy.  Or else I can
  print out relevant stuff and have it bound to take on the plane.

  The point of a PRM, then, would not be to automate research but
  to support the often laborious and haphazard process of creating
  and maintaining relationships in a research community.  What
  would happen in practice?  Not all of the results might be happy.
  University administrators these days are talking among themselves
  about how to rationalize the administration of universities and
  research laboratories, and standardized online systems could play
  a role in personnel evaluations.  How often have you been cited?
  How many other researchers have you listed in their databases?
  How many people have read your online publications?  How many
  articles have you refereed for journals?  The PRM database might
  also be used to select suitable authors of evaluation letters at
  promotion time.  As soon as the database is used in this fashion,
  of course, everybody will start skewing their online personae in
  whatever direction the administrators want to see.  It is already
  common in many fields -- computer science is an extreme case --
  for researchers to publish their results in many small increments
  to pump up their publication count.  And mutual citation mafias
  are common as well.  A world of networked object databases could
  accelerate these effects and create new ones.

  The effects on individuals' careers could be even greater.  It is
  already routine for departments and research units to track the
  careers of potential future "hires", for example the students who
  are finishing their degrees in other programs.  It is commonly
  forgotten that much of the point of affirmative action, despite
  all of the hype about quotas and standards, was to formalize
  and generalize this process, so that hiring was based less on
  informal networks, and so that institutions had incentives to
  extend their informal networks to include potential sources of
  hires from traditionally underrepresented groups.  The use of
  PRM's could make job candidates "visible" to the job market in
  a strategic way, starting well before the formal hiring season
  begins.  A PhD advisor could use the web-object network to help
  each PhD candidate build a professional network in preparation
  for the job hunt, and departments and research units expecting
  to hire in the future could engage in advance intelligence about
  various candidates.  The whole hiring process could thus shift
  more toward active recruitment.  Job candidates, likewise, could
  prepare much better for their interviews than they can now, since
  they would have ready access to publications and other relevant
  information about the members of the departments where they are
  applying and interviewing.  And all of the same arguments apply
  to graduate school recruitment and admissions as well.

  It is worth noting the engine that makes all of these gears
  turn: multiple relational databases that are interlinked and
  thus searchable as if they formed a single database through the
  use of a common object that represents the same person for many
  different purposes.  That's how we're accustomed to thinking
  about information on computers.  It's a good thing, if a bit
  rigid sometimes, when it supports generally good applications
  like the ones I am describing here.  Applications tend to be
  good when the data is under the control of its owners, or under
  the collective control of the community that it's about.  But
  the same interlinking of databases can also be at the root of
  serious privacy problems when data about individuals is collected
  and used involuntarily, or without notification and rights of
  correction and control of secondary uses.

  These are familiar arguments, but distributed object databases
  may give them a new urgency.  Suppose that the world's major
  maintainers of personal information got together to create a
  system whereby every human being has precisely one object that
  represents them.  This object might be created at birth, if not
  before, and it would exist forever.  Any organization in the
  world could proceed to link its database to the shared object,
  provided that it supported the standard data object protocol.
  Security systems would, of course, regulate which organizations
  could conduct searches in other organizations' databases.  But
  each organization would have a great interest in licensing other
  organizations access to its databases.  Each organization would
  thus effectively have available a gigantic distributed object
  database, a sort of virtual database that effectively merges
  all of the individual databases to which it has access.  Then
  it could conduct giant relational queries that would give new
  meaning to the phrase "data mining".  It is a worrisome scenario.
  It is probably also too simple, but it is a good way to summarize
  the problem.  Perhaps the move toward a standardized object for
  everyone will start with relatively good applications, since
  those will cause less protest.  But once established, the system 
  could creep with little effort toward relatively bad applications.


  This month's recommendations.

  Andrew Feenberg, Alternative Modernity: The Technical Turn in
  Philosophy and Social Theory, Berkeley: University of California
  Press, 1995.  This book brings together several studies by the
  foremost scholar of the critical philosophy of technology.  Among
  the book's many strengths are its lucid exposition of difficult
  philosophical ideas and its combination, rare among philosophers,
  of scholarship and empirical research.  The book is organized in
  paired chapters, scholarship matched with case studies.  A study
  of Marcuse's critique of technology is applied to science fiction
  movies.  His chapter on the concept of technocracy in Adorno,
  Foucault, and Habermas will save many people some painful late
  nights of reading these obscure characters in the original, as
  well as providing the background for a study of the ethics of
  AIDS experimentation.  A study of the postmodernist philosopher
  Lyotard is applied to a case study of the Minitel system in
  France.  And a theoretical discussion the work of Kitaro Nishida,
  which suggests that theories of modernity have not been the sole
  province of European political philosophy, then provides the
  background for a study of the game of Go in Japanese culture.

  Martha Lampland, The Object of Labor: Commodification in
  Socialist Hungary, Chicago: University of Chicago Press, 1995.
  This is a historical and ethnographic study of life in a
  collectivized village in Hungary under Communist rule in the
  1980's.  Its question is, what are capitalism and communism,
  not as abstract ideal types but as forms of activity and meaning
  that people enact in their everyday lives?  The villagers, having
  lived under feudalism through perhaps the turn of the century,
  something resembling capitalism between WWI and WWII, and a messy
  approximation to communism after WWII, saw the world in a complex
  amalgam of different ways.  Individualist and collectivist values
  were both very much present, having arisen historically at the
  same time and through the same process.  Everyone saw scientific
  socialism as a corrupt farce and invested their main effort in
  their private plots of land.  Yet they generally approved of the
  efficiencies of collectivized agriculture and the predictability
  of the cooperative markets.  In this complicated middle ground,
  Lampland analyzes with great subtlety what a commodity is and
  what it means for labor to be bought and sold.  The results
  should help everyone to feel uncomfortable with the easy answers
  that come with their political beliefs, regardless of what those
  beliefs are.

  Seminary Co-op Bookstore.  This is not the biggest academic
  bookstore in North America, but I think it's the best, at least
  for the social sciences and humanities.  I can easily spend a day
  here, and I make time for it whenever I'm in Chicago.  5757 South
  University Avenue; Chicago, Illinois 60637; +1 (312) 752-4381,
  1-800-777-1456, fax +1 (312) 752-8507; books@semcoop.wwa.com;



  Several people have asked me where they can read more about
  David Noble's concept of "masculine transcendentalism", which
  I cited in TNO 3(4).  I heard him use the phrase in a talk about
  the place of religion in the history of technology.  His book,
  "The Religion of Technology", has not yet been published, but the
  last chapter of "Progress Without People" (Toronto: Between The
  Lines, 1995) is a preview.  After hearing me describe the idea
  in a talk, one person thought I meant that men are inherently
  defective, or that all men are transcendentalists.  But that's
  not what Noble or I have in mind.  The point, rather, is that
  a historically specific and contingent cultural construction
  of masculinity has reinforced, and been reinforced by, certain
  institutions and practices of technology.  I hope to write more
  about this in future issues of TNO.

  Phil Agre, editor                                pagre@ucla.edu
  Department of Information Studies
  University of California, Los Angeles         +1 (310) 825-7154
  Los Angeles, California  90095-1520                FAX 206-4460
  Copyright 1996 by the editor.  You may forward this issue of The
  Network Observer electronically to anyone for any non-commercial
  purpose.  Comments and suggestions are always appreciated.

Go back to the top of the file