<meta http-equiv="CONTENT-TYPE" content="text/html; charset=utf-8">
	<title></title>
	<meta name="GENERATOR" content="OpenOffice.org 3.0  (Unix)">
	<style type="text/css">
	&lt;!--
		@page { margin: 0.79in }
		P { margin-bottom: 0.08in }
	--&gt;
	</style>

<p style="margin-bottom: 0in;">Hello,<br>I am currently interested in
working on implementing a better way to work SCM ebuilds into gentoo.
Looking at the description, I just wanted to clarify a few
things.<br><br>First of, it seems that this daemon will run on a
privileged machine that is capable of uploading files into the gentoo
infrastructure, but it also has to deal with no official ebuilds.
This does seem rather paradoxical to me. Or is it located on the
users machine, and then only downloads these snapshots?<br><br>Secondly,
I understand this as a way to have a mirror periodically check the
upstream repository for changed, and then respond by notifying the
maintainer. Ebuilds will be left using these snapshots, which seems
like a rather small improvement over the current system.</p>
<p style="margin-bottom: 0in;"><br>
</p>
<p style="margin-bottom: 0in;">Or is this a system where the daemon
sits in the in the infrastructure and creates the snapshot
automatically once an ebuild is provided, and maybe even then
automatically creates new versioned snapshots and ebuilds based on
the new versions of the source code, and places these on the master
mirror to be distributed. Maybe with some sort of maintainer having
to sign of on the new version though?</p>
<p style="margin-bottom: 0in;"><br>
</p>
<p style="margin-bottom: 0in;">And thirdly, there is currently no
mentor for this project. Would anyone be interested in working with
me on this?</p>
<p style="margin-bottom: 0in;"><br></p><p style="margin-bottom: 0in;"><br>
</p>

<p style="margin-bottom: 0in;">Also, I was interested in the �Tags�
idea. It does seem rather simple to implement these into the ebuilds
as just a TAGS variable, and parsing should not be a too big
challenge, but I was wondering how this would tie into existing
search options. Would we add an -T option to search through tags?
Would the AND keyword trigger a tag search, or should any word in the
search string be matched against the tags as well as the package
name?</p>
<p style="margin-bottom: 0in;"><br>
</p>
<p style="margin-bottom: 0in;"><br>
</p>
<p style="margin-bottom: 0in;">Thanks in advance to any clarification,</p>
<p style="margin-bottom: 0in;">Nils Schlupp</p>