<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7638.1">
<TITLE>RE: [Distutils] [Catalog-sig]  Metadata-Version in PKG-INFO</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->

<P><FONT SIZE=2>&gt;&gt; For licensing, I would say we could update the PEP to say that the<BR>
&gt;&gt; 'licence' argument could be used to clarify the classifier (e.g., to<BR>
&gt;&gt; spell the version).<BR>
&gt;<BR>
&gt;That works for me.<BR>
<BR>
>From my point of view, in order to achive that humans and machines could read by themselves the license of a package, why not create a proper classifier?<BR>
<BR>
BTW, I asked today a friend who is involved in license issue and she explained me that: if the version of a license is not declared in a software product, that means that the license applied is the last one.<BR>
<BR>
Regards,<BR>
<BR>
Carlos Tejo</FONT>
</P>

</BODY>
</HTML>