<?xml version="1.0" encoding="iso-8859-1"?>

<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
<title type="text">To Be GNU/Hurd</title>
<subtitle type="html"><![CDATA[
lee aylward's INTERNET BLOG
]]></subtitle>
<id>http://www.tobehurd.org/blog/index.atom</id>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog" />
<link rel="self" type="text/xml" href="http://www.tobehurd.org/blog/index.atom" />

<author>
<name>Lee Aylward</name>
<uri>http://www.tobehurd.org/blog/index.atom</uri>
<email>lee@laylward.com</email>
</author>
<rights>Copyright 2005 Lee Aylward</rights>
<generator uri="http://pyblosxom.sourceforge.net/" version="1.3.2 2/13/2006">
PyBlosxom http://pyblosxom.sourceforge.net/ 1.3.2 2/13/2006
</generator>

<updated>2008-09-26T15:45:00Z</updated>
<!-- icon?  logo?  -->

<entry>
<title type="html">new version of RPC::XML, yay</title>
<category term="/development/rtorrent" />
<id>http://www.tobehurd.org/blog/2008/09/26/xmlrpc</id>
<updated>2008-09-26T15:45:00Z</updated>
<published>2008-09-26T15:45:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/rtorrent/xmlrpc" />
<content type="html">&lt;p&gt;
It looks like a new version of the RPC::XML Perl module that supports the &apos;i8&apos; 
type is out. This is what was holding me back from further developing smellatorrent,
because rtorrent uses &apos;i8&apos; if you are running it on certain machines.
&lt;/p&gt;
&lt;p&gt;
It is rather frustrating that it took so long to get fixed. In my bug report I
included a working patch, admitedly missing tests and docs, but it took a request
from a &quot;high profile&quot; client for anything to happen. Oh well, I guess downloading
movies just isn&apos;t important to some people!
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Using Apple&apos;s X11 with a .Xresources file</title>
<category term="/mac" />
<id>http://www.tobehurd.org/blog/2008/07/14/applex11launchd</id>
<updated>2008-07-15T00:32:00Z</updated>
<published>2008-07-15T00:32:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/mac/applex11launchd" />
<content type="html">&lt;p&gt;
Because I am a horrible masochist I have been trying to use my 
second favorite window manager, &lt;a href=&quot;http://www.xmonad.org&quot;&gt;xmonad&lt;/a&gt;, to handle the terminal
windows on my new Mac. This was surprisingly easy after compiling
GHC with MacPorts. I also had to compile xmonad, xmonad-contrib, and
ghc-x11 by hand, but that was relatively painless.
&lt;/p&gt;
&lt;p&gt;
The real pain came when I dug into Apple&apos;s deceptively simple
X11.app. I foolishly assumed that clicking X11.app would start
X, instead I found that it launched a vanilla xterm. After some digging I found
out that Apple&apos;s X11 uses launchd to listen on a special socket for
any requests from a X client. 
Apparently X11.app simply launches xterm, which in
turn makes an X request, which in turn triggers launchd to start X. Simple enough, but
because of the fact that X11 is started &lt;em&gt;after&lt;/em&gt;
xterm, none of the settings from .Xresources are in place when xterm
launches. So not only am I required to launch xterm if I want to launch X,
I am also forced to use xterm with the default settings.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">use Hazel instead of folder actions</title>
<category term="/mac" />
<id>http://www.tobehurd.org/blog/2008/07/14/hazel</id>
<updated>2008-07-15T00:11:00Z</updated>
<published>2008-07-15T00:11:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/mac/hazel" />
<content type="html">&lt;p&gt;
After fighting with Apple&apos;s folder actions for hours last week, my friend 
&lt;a href=&quot;http://blog.clintecker.com&quot;&gt;Clint&lt;/a&gt; suggested I try out &lt;a href=&quot;http://www.noodlesoft.com/hazel.php&quot;&gt;Hazel&lt;/a&gt;
instead. Hazel is basically what folder actions should be. It 
lets you easily select a folder, define some conditions, and finally create 
an action. In a matter of minutes Hazel let me create two simple rules 
that move nzbs and torrents to my file server. Hazel even retries the action
if it fails the first time (this is helpful if the network share is not mounted.)
&lt;/p&gt;
&lt;p&gt;
Apple really should just buy them and get rid of the current folder action 
implimentation.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Folder actions in Leopard are a pain in the ass</title>
<category term="/mac" />
<id>http://www.tobehurd.org/blog/2008/07/07/folderactions</id>
<updated>2008-07-07T05:40:00Z</updated>
<published>2008-07-07T05:40:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/mac/folderactions" />
<content type="html">&lt;p&gt;
	I recently got a Macbook, and ran into the need to save all .torrent files
	to a specific directory. Safari saves all files to a single directory
	by default, which made this impossible. To get around this I attempted
	to make a Folder Action for the Downloads folder. It all sounded quite
	simple. Unfortunately, Apple hasn&apos;t quite finished this part of the 
	OS yet.
&lt;/p&gt;
&lt;p&gt;
	My first mistake was assuming that the Automator script would be passed
	the files that had changed, when in fact the script is passed
	only the Folder Action directory. After figuring this out
	I tried saving everything as an Application and then using that for the
	Folder Action. Nope! Apparently you have to save your Automator script
	as a &quot;Folder Action plug-in&quot;, which actually creates 2 files:
	&lt;ul&gt;
		&lt;li&gt;~/Library/Scripts/Folder Action Scripts/actionname.scpt&lt;/li&gt;
		&lt;li&gt;~/Library/Workflows/Applications/Folder Actions/actionname.app&lt;/li&gt;
	&lt;/ul&gt;
	The generated Applescript (scpt) does not actually work and must be modified
	as described &lt;a href=&quot;http://www.peachpit.com/articles/article.aspx?p=1215430&quot;&gt;here&lt;/a&gt;.
&lt;/p&gt;
&lt;p&gt;
	Even after figuring all of this out I am still pretty unhappy with how
	it works. Every time a file is downloaded the script is run, which
	steals focus repeatedly until it is finished. I think I will have to
	go ahead and write a python or perl script to do such things. Hopefully
	there is something akin inotify or FAM that I can use.
&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Follow-up: I found a program called Hazel which handles all this much better.&lt;/em&gt;&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">WWW-Discogs initial release</title>
<category term="/development" />
<id>http://www.tobehurd.org/blog/2008/07/06/www-discogs</id>
<updated>2008-07-07T00:26:00Z</updated>
<published>2008-07-07T00:26:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/www-discogs" />
<content type="html">&lt;p&gt;
	In an effort to improve my &lt;a href=&quot;http://covers.laylward.com&quot;&gt;cover art tool&lt;/a&gt;,
	I have started development on a new perl module that talks to &lt;a href=&quot;http://www.discogs.com&quot;&gt;
	discogs.com&lt;/a&gt;. This lets me search the site, and get art related to artists, labels,
	and specific releases. The site is very comprehensive (e.g. multiple versions of albums
	based on release format), so I am going to have to add some methods to make searching
	easier.
&lt;/p&gt;
&lt;p&gt;
	Here is what I have so far: &lt;a href=&quot;http://search.cpan.org/~leedo/WWW-Discogs-0.02/lib/WWW/Discogs.pm&quot;&gt;WWW-Discogs-0.02&lt;/a&gt;
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">automatic per-user multiple-site setup on apache2</title>
<category term="/development" />
<id>http://www.tobehurd.org/blog/2008/06/10/usersites</id>
<updated>2008-06-10T21:39:00Z</updated>
<published>2008-06-10T21:39:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/usersites" />
<content type="html">&lt;pre&gt;

&amp;lt;Perl&amp;gt;
my $basepath = &apos;/home/&apos;;
my $vhosts = &apos;/sites/&apos;;
my $logs = &apos;/logs/apache2/&apos;;

opendir my $users, $basepath;
my @users = readdir $users;

for my $user (@users) {
	next if $user =~ /^\./;
	next if ! -d $basepath . $user . $vhosts;

	opendir my $sites, $basepath . $user . $vhosts;
	my @sites = readdir $sites;
	close $sites;

	for my $site ( @sites ) {
		next if $site =~ /^\./;

		push @{ $VirtualHost{ &apos;*&apos; } }, {
			ServerName	=&gt; $site,
			ServerAlias	=&gt; &quot;www.$site&quot;,
			DocumentRoot    =&gt; $basepath . $user . $vhosts . $site,
			ErrorLog    =&gt; $basepath . $user . $logs . &quot;$site-error.log&quot;,
			CustomLog   =&gt; $basepath . $user . $logs . &quot;$site-access.log combined&quot;,
		};
	}
}
&amp;lt;/Perl&amp;gt;

&lt;/pre&gt;

&lt;p&gt;I keep this in &lt;span style=&quot;font-family:monospace&quot;&gt;/etc/apache/sites-enabled/usersites&lt;/span&gt;. Users create
sites by creating a directory of the format: &lt;span style=&quot;font-family:monospace&quot;&gt;/home/$username/sites/$domain/htdocs/&lt;/span&gt;.
The only problem is that apache requires a restart each time one adds a new site.&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella 0.11 fixes broken newzbin searching</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2008/06/09/hellasmella011</id>
<updated>2008-06-10T01:02:00Z</updated>
<published>2008-06-10T01:02:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella011" />
<content type="html">&lt;p&gt;Newzbin recently switched to v3 for the default interface, thus breaking the screen scraping that was being done. Here is a new version that forces the v2 interface.&lt;/p&gt;
&lt;p&gt;&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.11.tar.gz&quot;&gt;HellaSmella-0.11.tar.gz&lt;/a&gt;&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella 0.10</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2008/06/07/hellasmella010</id>
<updated>2008-06-08T00:37:00Z</updated>
<published>2008-06-08T00:37:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella010" />
<content type="html">
&lt;p&gt;This release will hopefully fix some small bugs people where running in to. I have also
added a speed plot to the status bar. Enjoy!&lt;/p&gt;
&lt;p&gt;&lt;img src=&quot;/~leedo/hellasmella/hellasmella010.png&quot; /&gt;&lt;/p&gt;
&lt;h4&gt;Changelog&lt;/h4&gt;
&lt;ul&gt;
	&lt;li&gt;restart status refresh timer on queue resort&lt;/li&gt;
	&lt;li&gt;html updates and fixes&lt;/li&gt;
	&lt;li&gt;queue is not updated unless there are changes&lt;/li&gt;
	&lt;li&gt;don&apos;t use a cookie file for newzbin model&lt;/li&gt;
	&lt;li&gt;more html and url escaping in templates&lt;/li&gt;
	&lt;li&gt;add &apos;url&apos; filter to Jemplate, fix &apos;uri&apos; filter&lt;/li&gt;
	&lt;li&gt;fix progress bar in FF3 and probably other browsers&lt;/li&gt;
	&lt;li&gt;speed plot&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.10.tar.gz&quot;&gt;HellaSmella-0.10.tar.gz&lt;/a&gt;&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella 0.09</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2008/02/27/hellasmella009</id>
<updated>2008-02-27T06:43:00Z</updated>
<published>2008-02-27T06:43:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella009" />
<content type="html">&lt;p&gt;
I went ahead and put my lighttpd config in the ex/ directory. Hopefully that will
help some people get this set up. I also updated the README with some more details.
Here is a list of the latest changes:
&lt;/p&gt;
&lt;ul&gt;
	&lt;li&gt;nzb upload field&lt;/li&gt;
	&lt;li&gt;ie6&amp;7 rendering fix&lt;/li&gt;
	&lt;li&gt;fix initial style for completed items&lt;/li&gt;
	&lt;li&gt;consolidate code&lt;/li&gt;
	&lt;li&gt;get status update with resort&lt;/li&gt;
	&lt;li&gt;restyling of search results&lt;/li&gt;
	&lt;li&gt;add &quot;fetch info&quot; option to search results&lt;/li&gt;
	&lt;li&gt;more descriptive time remaining field (Mark Glines)&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;
I need to do a little more styling of the extended information that pops down,
but I think it is time for a new release.
&lt;/p&gt;
&lt;p&gt;&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.09.tar.gz&quot;&gt;HellaSmella-0.09.tar.gz&lt;/a&gt;&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella 0.08</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2008/01/07/hellasmella008</id>
<updated>2008-01-07T07:05:00Z</updated>
<published>2008-01-07T07:05:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella008" />
<content type="html">&lt;ul&gt;
&lt;li&gt;added javascript functions to pause/continue/clear so
the search results are not lost with a page refresh&lt;/li&gt;
&lt;li&gt;changed &quot;clear (all)&quot; down to just &quot;clear&quot; for simplicity&lt;/li&gt;
&lt;li&gt;all queue controls return a status update if made as an
ajax request, this lets us update the UI faster&lt;/li&gt;
&lt;li&gt;removed [rar] password control, not used or testable since
nothing posted to usenet is password protected&lt;/li&gt;
&lt;li&gt;removed &quot;uptime&quot;, pointless and not accurate&lt;/li&gt;
&lt;li&gt;added category filter on newzbin search&lt;/li&gt;
&lt;li&gt;added HTTP::Cookies to build depends&lt;/li&gt;
&lt;li&gt;refined styling of queue and search&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;
&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.08.tar.gz&quot;&gt;HellaSmella-0.08.tar.gz&lt;/a&gt;
&lt;/p
</content>
</entry>

<entry>
<title type="html">hellasmella 0.07</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/12/08/hellasmella007</id>
<updated>2007-12-08T18:59:00Z</updated>
<published>2007-12-08T18:59:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella007" />
<content type="html">&lt;p&gt;
I took some time and created a Newzbin model for HellaSmella. This makes things
a little less ugly, but they also let you configure your Newzbin login info
in the hellasmella.yml config file.
&lt;/p&gt;
&lt;p&gt;
&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.07.tar.gz&quot;&gt;HellaSmella-0.07.tar.gz&lt;/a&gt;
&lt;/p&gt;
&lt;p&gt;
I was very close to switching to WWW::Newzbin for this release. Unfortunately,
after some tinkering around I realized that the file searching provided by
the Newzbinx FileFind API does not let you search &quot;reports&quot;. In the world of 
Newzbin, reports are the collections of files that make up a single TV show, 
Movie, etc. Instead of searching reports, the FileFind API only lets you 
search for individual files (part001.rar, part002.rar). This is fairly useless 
for my purposes, so it looks like I&apos;ll have to stick with screen scraping.
&lt;/p&gt;
&lt;p&gt;
Here&apos;s hoping they don&apos;t shut down v2 of the site!
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">This is progress?</title>
<category term="/miscellaneous" />
<id>http://www.tobehurd.org/blog/2007/11/10/elitist</id>
<updated>2007-11-11T03:32:00Z</updated>
<published>2007-11-11T03:32:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/miscellaneous/elitist" />
<content type="html">&lt;blockquote&gt;
Because dwm is customized through editing its source code, it&apos;s pointless to make binary packages of it. This keeps its userbase small and elitist. No novices asking stupid questions
&lt;/blockquote&gt;
</content>
</entry>

<entry>
<title type="html">initial release of smellatorrent</title>
<category term="/development/rtorrent" />
<id>http://www.tobehurd.org/blog/2007/10/28/smellatorrent001</id>
<updated>2007-10-29T00:57:00Z</updated>
<published>2007-10-29T00:57:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/rtorrent/smellatorrent001" />
<content type="html">&lt;p&gt;
Here is the first release of my catalyst rtorrent front end. Right now it is more or less a recreation of the curses frontend in HTML. I don&apos;t plan to do too much more with it apart from bug fixes and add the ability to start new torrents (through upload or URL).
&lt;/p&gt;
&lt;p&gt;
&lt;a href=&quot;/~leedo/smellatorrent/smellatorrent001.png&quot;&gt;&lt;img src=&quot;/~leedo/smellatorrent/smellatorrent001_thumb.png&quot; /&gt;&lt;/a&gt;
&lt;/p&gt;
&lt;p&gt;
I don&apos;t see any point to a &quot;detail&quot; view for individual torrents. What other clients display in those views is largely information that user does not need, such as number of connected clients or graphs of completed parts.
&lt;/p&gt;
&lt;p&gt;
&lt;a href=&quot;/~leedo/smellatorrent/SmellaTorrent-0.01.tar.gz&quot;&gt;SmellaTorrent-0.01.tar.gz&lt;/a&gt;
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">some rtorrent debs with xmlrpc support</title>
<category term="/development/rtorrent" />
<id>http://www.tobehurd.org/blog/2007/09/29/rtorrentdebs</id>
<updated>2007-09-29T16:14:00Z</updated>
<published>2007-09-29T16:14:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/rtorrent/rtorrentdebs" />
<content type="html">&lt;p&gt;
&lt;a href=&quot;/~leedo/rtorrent-svn_20070928-1_amd64.deb&quot;&gt;rtorrent-svn_20070928-1_amd64.deb&lt;/a&gt;&lt;br /&gt;
&lt;a href=&quot;/~leedo/libtorrent10-svn_20070928-1_amd64.deb&quot;&gt;libtorrent10-svn_20070928-1_amd64.deb&lt;/a&gt;&lt;br /&gt;
&lt;a href=&quot;/~leedo/libtorrent-svn-dev_20070928-1_amd64.deb&quot;&gt;libtorrent-svn-dev_20070928-1_amd64.deb&lt;/a&gt;
&lt;/p&gt;
&lt;p&gt;
I can put source packages up if anyone wants them. It does require the &lt;a href=&quot;https://launchpad.net/ubuntu/+source/xmlrpc-c/1.06.17-0ubuntu4&quot;&gt;libxmlrpc-c3&lt;/a&gt; packages from Ubuntu Gutsy, though.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">i got a new shirt from dinosaur comics!</title>
<category term="/miscellaneous" />
<id>http://www.tobehurd.org/blog/2007/09/26/sweetshirt</id>
<updated>2007-09-27T00:20:00Z</updated>
<published>2007-09-27T00:20:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/miscellaneous/sweetshirt" />
<content type="html">&lt;p style=&quot;font-size:0.8em;font-style:italic&quot;&gt;
&lt;img src=&quot;/~leedo/photos/failure.jpg&quot; style=&quot;margin:0&quot;/&gt;&lt;br /&gt;
Failure is just Success rounded down&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">oops, hellasmella 0.06 fixes more stuff</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/09/25/hellasmella006</id>
<updated>2007-09-26T02:45:00Z</updated>
<published>2007-09-26T02:45:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella006" />
<content type="html">&lt;p&gt;
I accidentally removed some redirections that happen when you pause,
continue, or remove a queue item. This new version fixes that, and 
in a way that works with non-root path installs.
&lt;/p&gt;
&lt;p&gt;
&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.06.tar.gz&quot;&gt;HellaSmella-0.06.tar.gz&lt;/a&gt;
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">beginning work on a rtorrent frontend</title>
<category term="/development/rtorrent" />
<id>http://www.tobehurd.org/blog/2007/09/22/rtorrent</id>
<updated>2007-09-22T19:35:00Z</updated>
<published>2007-09-22T19:35:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/rtorrent/rtorrent" />
<content type="html">&lt;p&gt;
I recently discovered that &lt;a href=&quot;http://libtorrent.rakshasa.no&quot;&gt;rtorrent&lt;/a&gt; has a very powerful XMLRPC interface. Now that I am so used to having a nice web interface for usenet I figured I would take a stab at something similar for torrents.
&lt;/p&gt;
&lt;p&gt;
Unfortunately, the rtorrent package in Debian is not built with XMLRPC support because libxmlrpc-c3 is &lt;a href=&quot;http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=317979&quot;&gt;painfully out of date&lt;/a&gt;. I managed to find newer versions of the &lt;a href=&quot;https://launchpad.net/ubuntu/+source/xmlrpc-c/1.06.17-0ubuntu4&quot;&gt;library package&lt;/a&gt; in Unutu &quot;gutsy&quot;, which works fine in Debian unstable. I then rebuilt the Debian package with this line added to the debian/rules file.
&lt;/p&gt;
&lt;pre&gt;DEB_CONFIGURE_EXTRA_FLAGS := --with-xmlrpc-c&lt;/pre&gt;
&lt;p&gt;
Once I got the new rtorrent built and installed everything went very smooth. Right now I have a page that lists the current torrents and some status info for each. It should be trivial to add some simple controls thanks to &lt;a href=&quot;http://search.cpan.org/~fmerges/Catalyst-Model-XMLRPC-0.04/&quot;&gt;Catalyst::Model::XMLRPC&lt;/a&gt;.
&lt;/p&gt;
&lt;p&gt;
And in case you are actually interested, here is a screenshot of my progress:
&lt;/p&gt;
&lt;a href=&quot;/~leedo/smellatorrent/smellatorrent000.png&quot;&gt;&lt;img src=&quot;/~leedo/smellatorrent/smellatorrent000_thumb.png&quot;&gt;&lt;/a&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella 0.05 fixes stuff!</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/09/22/hellasmella005</id>
<updated>2007-09-22T19:19:00Z</updated>
<published>2007-09-22T19:19:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella005" />
<content type="html">&lt;p&gt;
I put together a new release of HellaSmella this morning that includes a number of changes. Some very important!
&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;fixed fatal bug in &quot;time remaining&quot; calculation&lt;/li&gt;
&lt;li&gt;html fix on status panel&lt;/li&gt;
&lt;li&gt;behaves better in non-root directory setups&lt;/li&gt;
&lt;li&gt;added C::P::Auth::Credentials::HTTP to build requirements&lt;/li&gt;
&lt;li&gt;added &quot;Recently Completed&quot; list to queue&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.05.tar.gz&quot;&gt;HellaSmella-0.05.tar.tz&lt;/a&gt;&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella 0.04</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/08/19/hellasmella004</id>
<updated>2007-08-20T02:56:00Z</updated>
<published>2007-08-20T02:56:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella004" />
<content type="html">&lt;p&gt;
Here is a new version of HellaSmella. Changelog is below:
&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&quot;time remaining&quot; added&lt;/li&gt;
&lt;li&gt;html rendering fixes&lt;/li&gt;
&lt;li&gt;non-javascript browsers search properly&lt;/li&gt;
&lt;li&gt;focus search box on load&lt;/li&gt;
&lt;li&gt;show a &quot;currently processing&quot; section while hellanzb is processing something&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.04.tar.gz&quot;&gt;HellaSmella-0.04.tar.gz&lt;/a&gt;&lt;/p&gt;

</content>
</entry>

<entry>
<title type="html">new netflix facebook application &amp;ndash; because the others all look like shit</title>
<category term="/development" />
<id>http://www.tobehurd.org/blog/2007/08/19/netflixcompact</id>
<updated>2007-08-19T19:46:00Z</updated>
<published>2007-08-19T19:46:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/netflixcompact" />
<content type="html">&lt;p&gt;
I recently renewed my &lt;a href=&quot;http://www.netflix.com&quot;&gt;netflix&lt;/a&gt; account
and was interested in what netflix facebook applications existed. It turns
out that there are two up there now, but they are both hideously ugly. Each
application displays a rather large cover image along with a full description for each
movie in your queue. As you might imagine this takes up a lot of real estate on the page, and is more information than most people are interested in.
&lt;/p&gt;
&lt;p&gt;
To remedy this problem I decided to make a smaller netflix application that
only shows the covers of your queue, highlighting the movies that are
currently being viewed. I got a basic version working, but I have to say
that the facebook api is one of the most over-engineered pieces of shit
I have ever worked with. It really shouldn&apos;t be that complicated to get
a tiny piece of dynamic HTML to display on a profile. Instead they decided
to reinvent javascript, SQL, and HTML.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella version zero point zero three</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/07/07/hellasmella003</id>
<updated>2007-07-08T01:11:00Z</updated>
<published>2007-07-08T01:11:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella003" />
<content type="html">&lt;p&gt;
This is mostly a bug fix release. The majority of the changes are just html rendering 
fixes and a few javascript changes. One thing to note is that previous releases 
wouldn&apos;t work without a patch to Catalyst::Model::XMLRPC. This is no longer the case 
with the latest version of C::M::XMLRPC. Joy.
&lt;/p&gt;
&lt;p&gt;
&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.03.tar.gz&quot;&gt;HellaSmella-0.03.tar.gz&lt;/a&gt;
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">here is hellasmella 0.02~</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/06/24/hellasmella002</id>
<updated>2007-06-25T00:32:00Z</updated>
<published>2007-06-25T00:32:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella002" />
<content type="html">&lt;p&gt;
	This version adds some sw8 stuff:
&lt;/p&gt;
&lt;ul&gt;
	&lt;li&gt;draggable queue&lt;/li&gt;
	&lt;li&gt;set rar passwords&lt;/li&gt;
	&lt;li&gt;server info in top right corner&lt;/li&gt;
    &lt;li&gt;switched to JSON to update entire page with one request&lt;/li&gt;
	&lt;li&gt;document title updated with download percent&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;
	&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.02.tar.gz&quot;&gt;HellaSmella-0.02&lt;/a&gt;
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Here is the initial release of HellaSmella.</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/06/19/hellasmellarls</id>
<updated>2007-06-19T23:52:00Z</updated>
<published>2007-06-19T23:52:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmellarls" />
<content type="html">
&lt;h4&gt;Features&lt;/h4&gt;
&lt;ul&gt;
	&lt;li&gt;Inline newzbin search and queueing&lt;/li&gt;
	&lt;li&gt;Queue controls such as pause, clear, and cancel.&lt;/li&gt;
&lt;/ul&gt;

&lt;h4&gt;TODO&lt;/h4&gt;
&lt;ul&gt;
	&lt;li&gt;Move items up/down in the queue&lt;/li&gt;
	&lt;li&gt;Single prototype.js PeriodicalUpdater for page, rather than one for each widget&lt;/li&gt;
	&lt;li&gt;Move Net::Newzbin to standalone module&lt;/li&gt;
	&lt;li&gt;Move Net::Newzbin config to hellasmella.yml&lt;/li&gt;
&lt;/ul&gt;

&lt;a href=&quot;/~leedo/hellasmella/HellaSmella-0.01.tar.gz&quot;&gt;HellaSmella 0.01&lt;/a&gt;
</content>
</entry>

<entry>
<title type="html">hellasmella: a hellanzb web interface</title>
<category term="/development/hellasmella" />
<id>http://www.tobehurd.org/blog/2007/06/18/hellasmella</id>
<updated>2007-06-19T03:29:00Z</updated>
<published>2007-06-19T03:29:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/hellasmella/hellasmella" />
<content type="html">&lt;p&gt;
So I spent some time this weekend working on a web interface for 
hellanzb. I was amazed at how fast it went using Catalyst with the
Catalyst::Model::XMLRPC module. I got a working version finished 
within an hour, and spent the remaining time working on HTML/CSS
stuff. Here is a screenshot with code to follow soon.
&lt;/p&gt;
&lt;a href=&quot;/~leedo/hellasmella/hellasmella.png&quot;&gt;&lt;img src=&quot;/~leedo/hellasmella/hellasmella_thumb.png&quot; /&gt;&lt;/a&gt;
</content>
</entry>

<entry>
<title type="html">This guy is funny</title>
<category term="/miscellaneous" />
<id>http://www.tobehurd.org/blog/2007/04/25/exec</id>
<updated>2007-04-25T14:57:00Z</updated>
<published>2007-04-25T14:57:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/miscellaneous/exec" />
<content type="html">&lt;p&gt;
&lt;img src=&quot;/~leedo/random/exec_bio1.jpg&quot;&gt;
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Swearing in sci-fi shows</title>
<category term="/entertainment" />
<id>http://www.tobehurd.org/blog/2006/08/25/scifiswear</id>
<updated>2006-08-25T07:26:00Z</updated>
<published>2006-08-25T07:26:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/entertainment/scifiswear" />
<content type="html">&lt;ul&gt;
&lt;li&gt;Battlestar Gallactica - Frack&lt;/li&gt;
&lt;li&gt;Farscape - Frill,Hezmanna&lt;/li&gt;
&lt;li&gt;Red Dwarf - Smeg&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;
&lt;i&gt;... more to come!&lt;/i&gt;
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Farscape</title>
<category term="/entertainment" />
<id>http://www.tobehurd.org/blog/2006/07/22/farscape</id>
<updated>2006-07-23T02:49:00Z</updated>
<published>2006-07-23T02:49:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/entertainment/farscape" />
<content type="html">&lt;p&gt;
Remember seeing those horrible looking Sci-Fi commercials for Farscape 
a few years ago? I used to sit there and think &quot;What kind of loser would 
every watch that god-awful show?&quot; I recently downloaded the first few 
episodes off of usenet and it is actually pretty good.
&lt;/p&gt;
&lt;p&gt;
I found the initial setup on this show to be really interesting. It is 
about a scientist/pilot named John Crichton who is testing a theory
 for space travel. In the middle of his test flight Crichton is sucked 
into a worm hole, and spit out on the other side of the universe. 
I found myself much more drawn to this setup for a show much more than
 just being dropped in with a bunch of characters from 2000 years in the 
future. Having a character from the 20th century on the show allows you
 to believe more because he is just as stunned by what is happening as you
are.
&lt;/p&gt;
&lt;p&gt;
Despite the great setup, I have to say, the episodes and characters seem
to be very formulaic. It is almost as if they took characters from classic 
RPG games and stuck them in a show. There is a priest, warrior, and theif,
all who fit the stereotypes perfectly. Despite this I still have hope
that the show will pick up, and more of a continuing plot will appear.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Babylon 5</title>
<category term="/entertainment" />
<id>http://www.tobehurd.org/blog/2006/01/25/bab5</id>
<updated>2006-01-26T00:39:00Z</updated>
<published>2006-01-26T00:39:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/entertainment/bab5" />
<content type="html">&lt;p&gt;
So I am slowly discovering that Babylon 5 isn&apos;t the heaping pile of shit that
it first appeared to be. For the first few episodes all I could see what the
horribly tacky set, and rehashed sci-fi characters. Now that I am a good 13
episodes in I am starting to see a show with an excellent continuing plot, 
and some great costume design. I&apos;m not sure that I&apos;ll be able to handle five 
seasons of the show, but it is definitly worth a watch.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Irssi and gnome-terminal</title>
<category term="/development" />
<id>http://www.tobehurd.org/blog/2006/01/23/irssitimestamps</id>
<updated>2006-01-23T23:52:00Z</updated>
<published>2006-01-23T23:52:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/irssitimestamps" />
<content type="html">&lt;p&gt;
I finally figured out a way to get link-highlighting to work consistantly
with irssi inside gnome-terminal. The problem was that if the url 
ended on the far left character of the terminal, gnome-terminal would think
that they link continued down to the next line. This caused gnome-terminal 
to append the timestamp onto a url every once in a while. To fix this I simply
added a space at the beginning of the timestamp format (and after, just to be
symetrical).
&lt;/p&gt;
&lt;pre&gt;/set timestamp_format  %H:%M
/save&lt;/pre&gt;
&lt;p&gt;
Horray!&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">The Island</title>
<category term="/entertainment" />
<id>http://www.tobehurd.org/blog/2005/07/29/theisland</id>
<updated>2005-07-29T15:10:00Z</updated>
<published>2005-07-29T15:10:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/entertainment/theisland" />
<content type="html">&lt;p&gt;Holy crap what a shitty movie. I officially hate Michael Bay&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">The Trial</title>
<category term="/entertainment" />
<id>http://www.tobehurd.org/blog/2005/07/21/thetrail</id>
<updated>2005-07-21T16:00:00Z</updated>
<published>2005-07-21T16:00:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/entertainment/thetrail" />
<content type="html">&lt;p&gt;
I finished watching &lt;a href=&quot;http://www.imdb.com/title/tt0057427/&quot;&gt;The Trial&lt;/a&gt; last night. I was really fucking tired, but it was good enough to keep me awake. It reminded me a lot of Brazil. In fact, it seems like a few scenes from Brazil are taken directly from The Trial. I&apos;m really starting to dig Orson Welles&apos; films.
&lt;/p&gt;
&lt;p&gt;
I&apos;ve updated Planet NMP3s so that it properly handles Atom feeds from blogger.com now. The site is no longer indexed by google as well.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">New lyric annotation wiki thing</title>
<category term="/development" />
<id>http://www.tobehurd.org/blog/2005/07/13/lyricnotes</id>
<updated>2005-07-13T05:39:00Z</updated>
<published>2005-07-13T05:39:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/development/lyricnotes" />
<content type="html">&lt;p&gt;
Someone on a music forum I read wanted to make an MF Doom lyric discussion 
site, which gave me some motivation to get back into doing some web projects. 
I decided to see what I could come up with.
&lt;/p&gt;
&lt;p&gt;
After fiddling around with javascript to work on both IE and FF I am 
nearly happy with this kwiki module I came up with. 
&lt;a href=&quot;http://raipu.org/kwiki/&quot;&gt;The kwiki module&lt;/a&gt; lets you add notes 
to a block of text. Then when people hover over the block a popup will 
appear with your note in it. I hope the guy who originally asked for it 
finds it useful, but if not I did manage to hone my javascript skills a bit.
&lt;/p&gt;
</content>
</entry>

<entry>
<title type="html">Test, Test</title>
<category term="/miscellaneous" />
<id>http://www.tobehurd.org/blog/2005/06/23/firstpost</id>
<updated>2005-06-23T20:41:00Z</updated>
<published>2005-06-23T20:41:00Z</published>
<link rel="alternate" type="text/html" href="http://www.tobehurd.org/blog/miscellaneous/firstpost" />
<content type="html">&lt;p&gt;
	Hey I got pyblosxom setup. Prepare to be &lt;b&gt;wowed&lt;/b&gt; by my english skills (starting with my next post).
&lt;/p&gt;
</content>
</entry>
</feed>
