apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Roy T. Fielding" <field...@gbiv.com>
Subject Re: MD4/MD5 implementation is non-free
Date Tue, 17 Oct 2006 04:59:16 GMT
On Oct 16, 2006, at 9:02 PM, Jeffrey Thompson wrote:
> "Roy T. Fielding" <fielding@gbiv.com> wrote on 10/16/2006 03:44:20 PM:
>> On Oct 16, 2006, at 3:38 AM, Joe Orton wrote:
>>
>>> An argument has been made that the third-party MD4/MD5 code in APR
>>> (specifically, APR-util) is licensed such that it is not
>>> permissible to
>>> distribute modified works.
>>
>> AIUI, copyright law has separate restrictions on "to make ...
>> derivative works" from the restrictions on reproducing works, and
>> thus the text is merely reflecting each of the permissions needed
>> in turn.  In other words, if you have a license to copy and a license
>> to make derivative works, then you have a license to redistribute
>> the derivative works as well, since the derivative work is
>> covered either by the original's license-to-copy or by the new
>> copyright of the entity that created the derivative work.
>>
>> The intent of this "licensed" interpretation is further evidenced
>> by the requirement that derivative works contain a notice that they
>> are "derived from ...", since such a notice would serve no useful
>> purpose if the person was not allowed to make copies.
>>
> Roy, but the issue here isn't copies, its distribution.  Remember that
> copyright was originally designed for books where the copying step was
> separate from the distribution step and was usually performed by  
> someone
> other than the distributor.  Just because I'm allowed to make an  
> unlimited
> number of copies of a piece of software, that does not mean that I'm
> allowed to provide a single one of those to another party.

Ouch!  You are, of course, quite right -- I completely missed that
"copy and use" was not equivalent to distribution.  For others, you
can see the distinction in

    http://www.copyright.gov/title17/92chap1.html#106

It didn't occur to me that they might only be licensing personal
use copies.  Blech.

> So, that leaves us with a situation where, if the text presented is  
> the
> only means of getting the right to distribute, we may have a problem.
> There may be another license somewhere that covers this code.  Then  
> again,
> there may not.  A clarification from RSA might be in order.

The origin is

    http://www.ietf.org/rfc/rfc1321.txt

which may or may not help.

....Roy


Mime
View raw message