Subject: Re: Java is really convenient. Re: Sun thinks about switching Java      to S-expression syntax: Lava
From: Erik Naggum <>
Date: 1999/02/18
Newsgroups: comp.lang.lisp
Message-ID: <>

* (Rainer Joswig)
| Sorry, but this whining about CL-HTTP really pisses me off.

  some of us are sufficiently legally trained or experienced to see danger
  signs in legal material.  I don't know that it pisses anyone off, but it
  sure says "no go".  some of us believe in the sanctity of contract, and
  sometimes a very legitimate reaction is "I can't honor that contract".
  some of us work very hard to ensure that everything we do is according to
  all applicable laws, not because we're afraid of punishment, but because
  by breaking some laws, we would also lose its protection.  if we broke a
  contract or were even challenged for it, our own contracts would lose
  legal force in the minds of very many people.

  like, does anyone _really_ respect Microsoft's license agreements?  I
  sure don't, but this is one of my best arguments against ever installing
  anything owned by Microsoft on my computers -- people who do actually and
  usually unwittingly allow Microsoft officials or anyone acting on their
  behalf to search their entire computing environment for illegal copies of
  unrelated software, except that you _cannot_ agree to this in any country
  on earth and still be held legally responsible for it, but Microsoft will
  try and crush you regardless of your legal standing and rights.  as long
  as their opponents go away, Microsoft is happy, and if the law come into
  play, so much the better.  threats and force actually work.  that's why
  most countries have severe penalties for fraud and make fraud a crime,
  instead of just a matter of agreements.

  yes, the issue is this big if you actually take agreements seriously.

| I'm not really sure if other people really appreciate ...

  now you're being destructive and paranoid.  drop it.

| I really don't like CL-HTTP to go the route of some other "closed" or
| "open" Lisp packages -> "dead end".

  I don't think you appreciate that others may have the same goal.

| Lisp users have the luck to have access to a system that incorporates
| some real cool stuff and a large library of networking code.  Some users
| have built even more cool applications on top of that.  Actually CL-HTTP
| is thought as an enabler to develop these cool applications and to make
| Lisp usable for web applications - whereas most people think you can't
| write a web server in Lisp.

  all of this is true.  all of this is extremely irrelevant to the terms of
  the license agreement.

| If people have real problems with the license and they have a real need
| for a web server in Lisp: write your own.

  well, that _is_ the problem with the license.  thank you for agreeing.

  Rainer, let me be blunt: that license is a toy license, written by a
  legal incompetent.  e.g., "by retaining a copy of the distribution, you
  acknowledge that you accept the terms and limitations of this license",
  which means that just picking up the distribution is an encumbrance upon
  the user.  in most countries, this clause is unenforceable -- I don't
  know that it is unenforceable in the Commonwealth of Massachussetts, but
  I would be greatly surprised if it weren't.  and the right to produce an
  "executable" is reserved.  what does that mean?  does it mean compiling
  it?  does it mean dumping it with your Lisp image?  but what if there is
  no executable mentioning, only various loadable files?  and what _is_ the
  actual _license_?  "personal, educational, and commercial uses are
  allowed", it says.  this is freaking _meaningless_ in the precise legal
  terms required for licensees to know _exactly_ what they are allowed to
  do -- I have no idea whether what I'd like to do is considered a
  "commercial use" or not.  _real_ legal documents define all these things.

  finally, and this is really a fundamental issue; where is MIT's grant to
  John Mallery that he can claim ownership of this in the first place?  my
  understanding is that it has been developed over a long time while he has
  been a graduate student at MIT.  in case you didn't know, MIT retains a
  number of interesting rights to _anything_ students or employees do, but
  they routinely waive them -- they idea is simply to default to MIT
  ownership.  therefore, I expect to see a waiver and grant from MIT to
  John Mallery.  I haven't found it, and since I'm not a legal doofus who
  can reasonly claim ignorance of the law, as if that were even a valid
  excuse, or having acted in good faith, the whole thing is in danger of
  being invalid and someone might come along and claim it, including
  anything I might want to retain rights to.  this may not be a real issue,
  but it sure is a hell of danger sign.

  my own endeavors to publish software other people can use, have faltered
  on the legal work to do it right, which alone would set me back tens of
  thousands of dollars, which I have a very hard time defending, so I don't.

  I'm sure CL-HTTP is great, but the license agreement is useless, and so
  it is _irrelevant_ to me whether the product is great or not.  it would
  not cross my mind to argue against the quality of the code based on the
  lack of quality of the license agreement.  although programming is a lot
  like legal matters, few programmers think it's worth the effort, and thus
  the computer industry, and especially the amateur scene, is littered with
  unenforceable licenses that would make a lot of people _really_ unhappy
  if they were broken.  I'm unhappy that they are breakable without peril
  of legal consequence, because it means those who think nothing of such
  issues can have a field day if they want it.