Subject: Re: new license to review
From: bruce@perens.com
Date: 7 May 1999 19:14:40 -0000

> No modifications to Server Identification Field. You agree not
> toremove or modify the Server Identification Field contained in the
> ResponseHeader as defined in Section 1.6 and 1.7.

I'm concerned about the _precedent_ here, which could be used to enforce a
more rigid adherence to some communications protocol or API in _another_
_license_.  For that reason, I'd suggest that OSI stay away from this form
of license clause.

From: Russell Nelson <nelson@crynwr.com>
> IMHO it's no worse than the Artistic License, which requires you to
> change the name of the program if you break compatibility,

OSD #4, last sentence, explicitly allows that. It's different in that the
Artistic license would have you change the name to _anything_but_ one
protected name, essentially a trademark, that protects the public perception
of the author's version. This gives you a wide latitude for modification.
The "Foobar license", however, insists that you use one specific name in a
machine communications exchange, which is essentially a total prohibition on
one form of modification, and could affect the _function_ of the software,
not just its perception of its author.

> or the BSD License, which requires you to acknowledge original authorship.

Do you mean the obnoxious BSD advertising clause? "Acknowledge original
authorship" does not affect any client-server communication protocol, though,
and is coupled with distribution, not use.

> Any suggestions on which one [existing license]?

Their intentions seem closest to the NPL.

	Thanks

	Bruce