Subject: Re: "Well, I want to switch over to replace EMACS LISP with Guile."
From: Erik Naggum <erik@naggum.no>
Date: 28 Sep 2002 03:52:44 +0000
Newsgroups: comp.lang.lisp
Message-ID: <3242173964750774@naggum.no>

* Bulent Murtezaoglu
| Reinvention sometimes takes less time than understanding the state of the
| art.  This perception is usually misleading, but not always.

  Right now, I am working with project to evaluate the applicability of Dewey
  for automated classification, compared to the Open Directory stunt.  My main
  objection to such stunts it that they get can easily get very enthusiastic
  people at the start of the project, but over time need paid staff to maintain
  them and it develops into the dreaded "committee work" when it becomes
  successful.  My take on this is actually deeply social: It is not successful
  until it becomes a "democratic" committee project.  The idiot individualist
  with megalomania will think that others are his inferiors, that people who
  work in groups never get anything done while he can get a lot done alone.

  A system needs to be alive and workable even when other people than the first
  enthusiasts start using it.  Reinvention and revolution are enthusiast stuff.
  Invention and evolution are engineering.

| Am I wrong in assuming that with XML you are roughly as safe as you would be
| with a _documented_ binary format?

  It is a myth that XML is documented.  You have no idea what the elements
  /actually/ mean (and have warped into over time) until you see the source code
  for the application.  XML becomes /more/ application-dependent over time than
  binary formats because it provides a false security and an appearance that
  belies its true nature.  XML /is/ a binary format, it is just that it is the
  kind of binary formats that line printers and raw text editors can use, too,
  and it is no less dependent on the exactness and precision that other binary
  formats require.  At least when you have a binary format, you know that you
  need to have a version field and proper version management.  People who think
  SGML or XML will save them tend to forget version management and rely on
  stupid human tendencies to believe that that which they can "read" must also
  be understandable to the machine.  The combination of ignorance of computing
  principles and programming with the kind of fuzzy thinking we find in people
  who have never paid attention to details is seriously deadly to information.

| It is unclear what 'expert' means any more.  The blind has been leading the
| blind for so long that sight is becoming irrelevant in signifacnt parts of
| the marketplace.

  It saddens me to find that I think this is very good summary of the situation.

| It is _worse_ than that.  The semantics of knowing anything well is shifting.
| I am noticing that (and Tim B. sometimes reminds me) that most of our grumpy
| observations (rants?) here are observations about _literacy_.  You are
| implying a conscious decision above -- I am not so sure about that.

  You may be right.  I am far more conscious in general than other people.
  But with these saddening words, I shall go and be unconscious for a while.

-- 
Erik Naggum, Oslo, Norway

Act from reason, and failure makes you rethink and study harder.
Act from faith, and failure makes you blame someone and push harder.