Interview with Charles Roth, 2004, about the beginnings of Caucus

Lisa Kimball:

Charles,

Some folks here have been involved with Caucus since it was a gleam in your eye, while others are relatively new to the community. 

Many other tools have come and gone since Caucus was released originally.  The Internet and the Web have changed the environment substantially. 

I think it would be fun and interesting to capture some Caucus history here so everyone has more of an appreciation of how Caucus has evolved.  I'm hoping you're willing to be 'interviewed' Caucus-style about the beginnings and middles of Caucus. 

Charles, can you remember how and when you first got the idea to write Caucus?  What did you think was missing from the family of tools available then?  What experiences had you had with other software that inspired you to create Caucus?  What need in the world were you hoping to fill? 


Charles Roth:

Hmmm . . .  Caucus goes back quite a ways.   In 1980-82 I was a grad student and instructor in computer science at Wayne State University, and I ran across Confer, an early on-line conferencing system.  I had already written several e-mail systems at GE, and was fascinated by the power of communicating with people over a computer. 

(Mind you, this was the dark ages of electronic communication.  Everything was text; 300 baud modems were the norm.  That's about 2 seconds per line of text!  I had people come to visit me just to see my state-of-the-art 1200 baud modem.  There was no internet, and just the beginnings of e-mail that could go from one computer to another. 

You could pay $7-$15 an hour via long distance phone or Telenet network, using a CRT like the ADM3a, with a huge screen of 80 characters by 25 lines... just to talk to a computer that was located somewhere else.  Etc. )

At the time, I thought my professional calling was to be the prophet of portability.  Back then, it seemed like every computer manufacturer had their own proprietary operating system: VM/CMS, Vax/VMS, Primos, MTS, GECOS, Multics, RMX-11. . .  and the upstart that escaped from Bell Labs, Unix.  Windows wasn't even a glimmer in Bill Gates' eye. 

So I suddenly saw the power of many-to-many conversations on a computer, and had a vision of what it could be like if only it ran on everybody's computer.  Confer was limited to MTS; I think there were only ever 13 MTS systems in the world!  So I put my passions together and decided to write something that could easily be made to run on all these different computers. . . 

. . . and that's how Caucus was born. 

More about the passion of communication. . .  there are many science fiction stories about "group minds", the idea that many minds somehow connected together would be more intelligent than any one of them.  (For a classic, beautiful, and occasionally disturbing example, see Spider Robinson's DeathKiller. )

That's what I began to see happening, in (of course) a very simple, crude, (and real) way.  At first it was just a massive information filter:

  1. People can read very fast
  2. They write slowly
  3. Everyone has a passion and knowledge about something. 
  4. Thus many people can quickly scan a lot of questions and discussion -- at a low cost in time.  When someone has knowledge to contribute, it may take them a little more time to type it -- but the net gain for everyone is very great. 

Beyond that, I started to see actual collaboration -- where something new was actually being created by people coming together online.  As a tech person, I saw this most clearly in technology, of course.  But I saw glimmers of creation happening in other areas, too. 

(I think my favorite story was from later on -- somewhere around 1988, I think.  I was working with three small groups in New York, London, and Tokyo, to make a version of Caucus that would handle Japanese.  We used Caucus -- of course -- to handle the project.  Not only did we have a blast, but we effectively had a 24 hour work day.  Each time a problem came up, it rotated around the globe, and often "our" next morning saw a solution. )

An aside on the "family of tools" part of the question. . .  remember that this was 'way before the Internet.  E-mail was largely something that happened on a single computer system.  The idea of moving e-mail around between computers was still somewhat novel (and mostly involved sets of unofficially cooperating computers dialing each other up late at night, rather like teenagers exchanging gossip). 

So there really wasn't a family of tools to speak of, at all.  A couple of forward-thinkers like Jacob Palme, Murray Turoff, and Bob Parnes, who were just a step or two short of science fiction, had been pointing the way to some sort of centralized discussion tool that wasn't just emails or postings flying around the primitive net. 

So it was a very do-it-yourself world then, with a very different set of expectations about how software was developed, and how many people would use it. 


Lisa Kimball:

Charles, you've shared some interesting stuff about what we might call the first phase of Caucus history . . .  What are the things that have influenced you most in the past 5 years? 


Charles Roth:

That's a tough one -- there have been so many influences that I wish I could take advantage of, it's frustrating to even think about!  The biggest one is the rise of good tools for object-oriented programming -- I wish I could rewrite everything in Python or Java.  Of course, every year I wait, the better and better the tools get!  Sort of like waiting to buy a new PC (or Mac <g>). 

I can pin some of my work in the last couple years on two general trends, however:

1.  Linux and Open Source
I finally bit the bullet and went completely linux-only for my development work (i. e.  my laptop).  Part of that is because linux finally got good enough to be a reasonable desktop environment -- especially with tools like OpenOffice and CrossOver Office

But the real influence there was the ongoing success of the unix/linux philosophy: write tools to be really good at what they do, and let them "play well" with other tools.  This is completely contrary to the Microsoft philosophy of building everything into one single monolithic structure.  (For a really good extended read on the unix philosophy, see In the Beginning was the Command Line, by Neal Stephenson. )

Anyway, I've always liked this kind of approach, and I've written 'freeware' from way before the days when this was ever any kind of philosophical issue.  But the point is that this approach was clearly succeeding, like the Gandhi quote: "First they ignore you.  Then they laugh at you.  Then they fight you.  Then you win. "

2.  Other Good Tools
Once the dot-bomb madness (in which I cheerfully/frightfully participated) was over, and no-one was trying to dominate markets or own eyeballs and all that crap, a lot of really good tools that did a specific thing well started to crawl up out of the slime.  (OK, I've got a metaphor mixmaster going here, bear with me.  )

Let's name a few.  Apache.  OpenOffice (ok, I used it twice).  LDAP.  MySQL.  More chat tools than the Hydra had heads (but I finally found one that I liked and could easily extend).  Several WYSWYG editors (though they still have rough edges).  Wiki.  Blogs.  RSS. 

Every one of these is an opportunity to leverage what I can do, within Caucus, with a tool that yields enormous added power.  At the same time, each requires more work than it would take to do the one piece I need, by myself.  But the longer term rewards are so much greater. . .  it is enormously delightful and quite frustrating, all at the same time. 

But, like Twain's (? ? ) quote about impending execution, it can help to sharpen one's focus!