In my opinion, the problem with allowing "user judgement" in included
license variability can lead to disagreements of what a license really
is, or even worse, misunderstanding of what the license of a file is.

Say hypothetically, you read a license and for you it is zlib, and for
me it is not, and I prefer to refer to it as a zlib-variant, because
for me the differences are strong enough to worry.

I would prefer that there was a single place at the beginning of the
SPDX file where such two variants of the license are located, and then
I can look at it and decide if it is equal or not. Rather than
trusting your judgement.

Perhaps I am just beating a dead horse, and nobody really cares about
such differences (think MIT/X11 and BSD-variants not this zlib
example).

--dmg

On Thu, Sep 30, 2010 at 1:19 PM, Peter Williams
<peter.williams@...> wrote:

3. Even thought there is no perfect textual comparison of the license
(aside from whitespace) the licenses have been considered to be
equivalent.

This is the only sane thing to do.  Unfortunately, there are situations in
which reasonable people could disagree about whether two license texts are
really the same license or not.

These are very good reasons why standardizing text of licenses by
inclusion seems to me like a bad idea.

Here i disagree.  I think standardizing some license texts is a Good Thing.
 No one will be force to reference those standard licenses.  If you find a
license that you believe is materially different from the all the texts in
the public repo that license can be included in the spdx file as a
non-standard license.  Having a set of licenses with standardized names
allows much more efficient communication and greater interoperability.

--
--dmg

---
Daniel M. German
http://turingmachine.org