<?xml version="1.0"?>
<rss version="2.0">
  <channel>
    <title>Advogato blog for ppatters</title>
    <link>http://www.advogato.org/person/ppatters/</link>
    <description>Advogato blog for ppatters</description>
    <language>en-us</language>
    <generator>mod_virgule</generator>
    <pubDate>Sun, 19 May 2013 18:09:10 GMT</pubDate>
    <item>
      <pubDate>Wed, 30 Sep 2009 19:50:53 GMT</pubDate>
      <title>30 Sep 2009</title>
      <link>http://www.advogato.org/person/ppatters/diary.html?start=1</link>
      <guid>http://www.advogato.org/person/ppatters/diary.html?start=1</guid>
      <description>For a while now, my &lt;a href="http://www.carillon.ca/" &gt;PKI in Aerospace&#xD;
Consulting&lt;/a&gt; company has been working on finding ways to&#xD;
make PKI more usable. From the "Relying Party" side of the&#xD;
equation, I think we're getting fairly close. Between some&#xD;
of the advances that others have made (such as Microsoft&#xD;
CAPI now doing Path Discovery and Validation mostly&#xD;
correctly), and our own work on writing an open source &lt;a href="http://www.carillon.ca/tools/pathfinder.php" &gt;Path&#xD;
Discovery and Validation Daemon&lt;/a&gt; that can be used by&#xD;
programs like &lt;a href="http://www.carillon.ca/tools/pf_apache2.php" &gt;Apache&lt;/a&gt; and&#xD;
&lt;a href="http://www.carillon.ca/tools/pf_freeradius.php" &gt;Free&#xD;
Radius&lt;/a&gt;, I think there is very little reason why&#xD;
someone could not actually build a site and fully use&#xD;
certificates for authentication (especially with the &lt;a href="http://www.carillon.ca/tools/apache_sslinfo.php" &gt;certificate&#xD;
information patches&lt;/a&gt; that we've just published for&#xD;
Apache).&lt;br&gt;&#xD;
&#xD;
&lt;p&gt; Now, the problems we're seeing are on the pure client&#xD;
side, such as in browsers, mail&#xD;
clients, VPN clients, or wireless clients in the open source&#xD;
world. The &lt;b&gt;nice&lt;/b&gt; thing about the proprietary world of&#xD;
Microsoft and Apple is that they, for the most part, all use&#xD;
their platform certificate store (CAPI on Microsoft, and the&#xD;
KeyChain on Apple). In the open source world, certificates,&#xD;
keys and trust anchors can be just about anyplace. And, most&#xD;
annoyingly, even applications built by the same projects&#xD;
don't even use the same certificate stores (I'm looking at&#xD;
you, Firefox and Thunderbird, and you too KMail and&#xD;
Konqueror). So, consider this a call for someone (maybe the&#xD;
LSB folks) to come up with a full standard that everyone can&#xD;
adopt for both trust anchors and user keys/certificates, and&#xD;
then please, please, everyone use that&lt;sup&gt;1&lt;/sup&gt;.&#xD;
&#xD;
&lt;p&gt; [1] - Yes, I know WHY Firefox and Thunderbird have their own&#xD;
store: so that they don't have to implement per-platform&#xD;
solutions, thereby easing their FIPS validation. At the&#xD;
very least, they COULD implement a common certificate store&#xD;
used by both (and any other LibNSS-using application). At&#xD;
least then I wouldn't have to install all of my certificates&#xD;
twice. If the Mozilla folks wanted to really endear&#xD;
themselves to the community, they would also, once a single&#xD;
store is in place, at least give an install-time&#xD;
option for those that need it to use the system certificate&#xD;
stores, instead of the LibNSS specific store.&#xD;
&#xD;
</description>
    </item>
    <item>
      <pubDate>Fri, 14 Mar 2003 22:01:18 GMT</pubDate>
      <title>14 Mar 2003</title>
      <link>http://www.advogato.org/person/ppatters/diary.html?start=0</link>
      <guid>http://www.advogato.org/person/ppatters/diary.html?start=0</guid>
      <description>For some people, this won't make any sense...

&lt;p&gt; Blogging is about to become VERY popular at Net Integration Technologies - that crazy place where I work (along with &lt;a href="http://www.advogato.org/person/pphaneuf/" &gt;pphaneuf&lt;/a&gt; and &lt;a href="http://www.advogato.org/person/apenwarr/" &gt;apenwarr&lt;/a&gt;). A bunch of us have been active in varying capacities around the Net for quite a while, with some being more active than others.

&lt;p&gt; Now, in the hopes of garnering recognition for NITI (as we generally shorten it to), and the cool stuff we're working on, and of course the shameless plug of ourselves... we're hopping on the Blog bandwagon. We're already pretty open about giving stuff we do away (check out &lt;a HREF="http://open.nit.ca" &gt;http://open.nit.ca&lt;/a&gt; sometime), but I guess we just need to talk about it more.

&lt;p&gt; So, expect more noise (not sure whether it will be signal yet) out of us at the very least.

</description>
    </item>
  </channel>
</rss>
