<?xml version="1.0"?>
<rss version="2.0.">
  <channel>
    <title>Advogato blog for ignatz</title>
    <link>http://www.advogato.org/person/ignatz/</link>
    <description>Advogato blog for ignatz</description>
    <language>en-us</language>
    <generator>mod_virgule</generator>
    <pubDate>Wed, 9 Jul 2008 09:42:47 GMT</pubDate>
    <item>
      <pubDate>Fri, 8 Apr 2005 14:04:31 GMT</pubDate>
      <title>8 Apr 2005</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=37</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=37</guid>
      <description>Note to self: To get the &lt;a href="http://xml.apache.org/security/Java" &gt;XML Security API&lt;/a&gt; to work with the Axis Happiness Page, place the Xerces and Xanlan jars in the CATALINA_HOME\common\endorsed directory. </description>
    </item>
    <item>
      <pubDate>Sun, 3 Apr 2005 17:49:50 GMT</pubDate>
      <title>3 Apr 2005</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=36</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=36</guid>
      <description>&lt;b&gt;AJAX XMLHttpRequest Helper Class&lt;/b&gt;

&lt;p&gt; Just put up a finished &lt;a href="http://cvs.ad1455.com/cgi/viewcvs.cgi/ajaxlib/AjaxRequest.js?rev=1.1.1.1&amp;content-type=text/vnd.viewcvs-markup" &gt;XMLHttpRequest JavaScript class&lt;/a&gt; to simplify working with the things. It's based upon stuff glommed from a lot of sources. The main thing is that I wanted to create an OO wrapper around the code and clean up how it can be called. 

&lt;p&gt; USAGE:

&lt;p&gt; &lt;blockquote&gt;&lt;pre&gt;var http = new AjaxRequest()
http.setTarget("/ping.php")
http.setForm(new URLBuilder("word", "ping"))
var synchronousGet = http.get()
var synchronousPost = http.post()
// Asynchronous
http.get(closureHere) 
http.post(closureHere)
&lt;/pre&gt;&lt;/blockquote&gt;

&lt;p&gt; See also: &lt;a href="http://cvs.ad1455.com/cgi/viewcvs.cgi/ajaxlib/#dirlist" &gt;ViewCVS&lt;/a&gt;, &lt;a href="http://www.ad1455.com/lib/tars/js.ajaxlib.tar.gz" &gt;tar&lt;/a&gt; and &lt;a href="http://www.ad1455.com/lib/js/ajaxlib/test/testAjaxRequest.html" &gt;test page&lt;/a&gt;.
</description>
    </item>
    <item>
      <pubDate>Mon, 14 Jan 2002 23:50:31 GMT</pubDate>
      <title>14 Jan 2002</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=35</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=35</guid>
      <description>Walked off the job today early. Just had to get out! Caused 
a major firestorm when people found out that I was using 
CVS. "Oh dear! That is not authorised. You must use 
PVCS." "Ummm how do I do that." "Oh, your manager has not 
authorized you to have a copy of PVCS installed on your 
computer so you cannot use it." So, how to I maintain 
version control?" "You must use PVCS. We will all get in 
trouble if they find out that we had CVS installed on the 
web server." When one developer heard that someone had CVS 
dozens of other ones started asking around. "CVS? CVS? We 
can use CVS? Where can I get it???" Had to nip that in the 
bud. Probably end up writing a script in Perl to do 
something like it. Poor man's retarded CVS. 

&lt;p&gt; I don't understand how Fortune 500 companies survive. Back 
to dice.com.

&lt;p&gt; It's been a while. Finally got the family moved out to 
HicksVille. </description>
    </item>
    <item>
      <pubDate>Tue, 13 Nov 2001 17:01:18 GMT</pubDate>
      <title>13 Nov 2001</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=34</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=34</guid>
      <description>Interesting: as a way to make money &lt;a href="http://www.sleepycat.com/" &gt;Sleepycat&lt;/a&gt; charges for 
prebuilt binaries as a "support contract". Does that 
actually work as a revenue stream? </description>
    </item>
    <item>
      <pubDate>Wed, 7 Nov 2001 03:31:57 GMT</pubDate>
      <title>7 Nov 2001</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=33</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=33</guid>
      <description>trying out &lt;a href="http://argouml.tigris.org/" &gt;argoUML&lt;/a&gt; 
with java web start. This is so kewl!</description>
    </item>
    <item>
      <pubDate>Fri, 2 Nov 2001 14:04:50 GMT</pubDate>
      <title>2 Nov 2001</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=32</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=32</guid>
      <description>Decided to focus on UML for a while. Need to get that down. 
Preparing docs for work gives me a good excuse for focusing 
on it.
</description>
    </item>
    <item>
      <pubDate>Mon, 29 Oct 2001 03:00:20 GMT</pubDate>
      <title>29 Oct 2001</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=31</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=31</guid>
      <description>Banging my head against the wall about the type of abstract 
data type to use for controling Xd builds and then it's 
like, just use the DOM, retard. (yes, I often call myself 
retard when I'm thinking) Why the heck should I build all 
of these complex structures when it's right there in the 
DOM. It's like I'm kicking and screaming my way into 1999. 

&lt;p&gt; I'm going to stick to the DOM over my natural attraction to 
Jdom just becuase it's a cross language interface so I know 
that if some day when things are ported (yeah, in your 
dreams, einstein) that the APIs can match as much as 
possible.

&lt;p&gt; Built my first PEAR class based on HTML/Table for lists. 
I'm working with the guy who did that class so I'm gonna 
run it by him and see what he thinks. I just wanted to get 
my hands dirty with the library matching style.

&lt;p&gt; A strange thing that they're doing at work is coding there 
PHP in a strict Java style. Each page builds calling main
(). The how is that by doing this the transition over to 
Java will be a lot easier. I'm not sure if I'm convinced. 
Type in &lt;cite&gt;error_reporting (E_ALL);&lt;/cite&gt; at the top of 
a page and watch the fur fly. Weak typing is a real 
handicap for that kind of roll out. Still, it is very 
elegant looking code. 

&lt;p&gt; After working with JSP, I have grown rather fond of server 
side scripting with strong types. Everything that comes 
along seems to be drawn to the flexibility. Not sure why. 
E, Ruby... I'm kinda glad that some of the Perl 6 dev talk 
hints that you might be able to force types. That would be 
sweet. Getting real sick of all this &lt;cite&gt;if (!is_array
($foo)) {}&lt;/cite&gt; shit.</description>
    </item>
    <item>
      <pubDate>Thu, 18 Oct 2001 18:06:08 GMT</pubDate>
      <title>18 Oct 2001</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=30</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=30</guid>
      <description>Not a day goes by where I don't get hit with some new 
combination of letters that I've never heard of before 
referring to some "technology" that I've never heard of 
before. This time it's &lt;a href="http://www.merant.com/pvcs/" &gt;PVCS&lt;/a&gt;, a commercial 
version of CVS as far as I can tell. They're using it here 
at the new gig. Go figure. I'll stick to trusty ol' CVS 
for my own stuff.</description>
    </item>
    <item>
      <pubDate>Tue, 16 Oct 2001 19:09:01 GMT</pubDate>
      <title>16 Oct 2001</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=29</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=29</guid>
      <description>Working with &lt;a href="http://www.pear.com" &gt;Pear&lt;/a&gt;. Lots 
of good shit in it. Not sure why open source developers 
shoot themselves in the foot by developing without 
documenting. Allah knows (hi NSA!) that I've made that 
mistake. Really hard to get people into something when all 
there is is highly abstract, undocumented code. I'm sorry, 
phpdoc just doesn't cut it with PHP not being strong typed 
and all. 
</description>
    </item>
    <item>
      <pubDate>Fri, 12 Oct 2001 19:13:27 GMT</pubDate>
      <title>12 Oct 2001</title>
      <link>http://www.advogato.org/person/ignatz/diary.html?start=28</link>
      <guid>http://www.advogato.org/person/ignatz/diary.html?start=28</guid>
      <description>Notes about required features for pivot. Issues that seem
to come up alot include:
Commit: Being able to edit large volumes of content
and it being stored and editable but not commited. Similar to
a sql commit, but maintaining state over sessions.

&lt;p&gt; Unknown Auth: Being able to give people and groups access
to stuff before you even know who they are. Matching an access
id to an email or something. Email still seems to be the best
universal id for anything. no need to go beyond it.

&lt;p&gt; Wierd using the APache java Oro regex package. Perl5 regular expressions.
Works real well, just funny seeing anything Perl like in Java.
Feels almost tainted.</description>
    </item>
  </channel>
</rss>
