Papers vs. Email & Books (was: language)

Robin Hanson (hanson@econ.berkeley.edu)
Thu, 14 Jan 1999 15:30:39 -0800

dave gobel wrote:
>As I read the various threads, I am struck by one thing - the written word,
>english, human language is very hard pressed to convey sufficiently accurate
>and precise meanings. ... Its like seeing the OZ of language hiding
>behind the curtain and he's having a devil of a time keeping up with you ...
>Many of the conversations go thru an evolution of : an idea, assertion, or
>announcement, followed by simple interesting reactions/coments, followed by
>secondary reaction/beliefs, followed by rebuttal, then semi-personal attacks
>then fracturing of the subject, then attenuation of interest. ...

The related thing that most strikes me is the unfortunate lack of paper-length contributions on the topics which frequently appear on this list.

There is a natural spectrum of "length" of contributions on a topic, ranging from off-the-top one sentence comments to detailed book-length or longer analyses. In between are email posts and paper-length discussions.

Each of these lengths has its place, but contributions lengths should increase as topics become more important and as conversations about them mature with time.

Shorter contributions are more flexible and responsive, allow discussants to more quickly adapt to changing feedback about what others find interesting and understandable. Longer contributions, however, can support deeper and more elaborate examination of topics, and can make conversations more accessible to wider audiences.

On many novel topics on this list I think discussants too quickly begin with email, and would do better initially with the flexibility of a phone conversation. On many older long-discussed topics, however, I think conceptual progress is hindered by the failure of most of you to move on to paper-length contributions.

Now a few people have, over the years, written books on the topics we discuss. But these mostly come off as idiosyncratic personal visions, instead of part of a conversation within a larger community. They mostly give the impression that the author thought up all this stuff by him/herself, and rarely cite others who they are responding to or building on.

Why? It seems to me it is because of jumping too quickly to book-length contributions, where paper length contributions would be more appropriate.

When you get a new idea, call up a friend and talk. When that gets stale, write it up as a short post, and try it out on an email list or newsgroup. But after the topic gets hashed out a lot in email, and discussions there don't seem so productive anymore, if you think the topic still worth pursuing, please, MOVE ON TO WRITING PAPERS!

Now maybe the first paper-length contribution on a topic will seem an idiosyncratic personal vision, but after that good papers should cite each other and respond to each other, just as good email posts do. And only after the conversation of papers seems to be getting stale and limited by the depth of analysis possible in a paper, only then should you really think about writing a book.

I think you can infer some things about a topic from the length of contributions made on it. If you only ever see short contributions on a topic, you might reasonably infer that people don't really take it that seriously. Either they don't care enough about it to take the next step, or the topic doesn't withstand careful scrutiny; those people who tried to write longer analyses realized it's all bunk and gave up.

So what do you guys want other folks to infer about the topics that come up over and over on this list? The "it's science fiction, not science" critique can be in part translated as "sure they'll throw out a few half-baked ideas in a novel, but anyone who thinks about it carefully will realize it doesn't make sense."

Well, do these ideas make sense or not?