1.1 --- a/make-rss.py Wed Mar 19 13:49:53 2008 -0700
1.2 +++ b/make-rss.py Wed Apr 23 14:05:32 2008 -0700
1.3 @@ -27,7 +27,7 @@
1.4 <title>%s</title>
1.5 <description><![CDATA[%s]]></description>
1.6 <link>http://apgwoz.com</link>
1.7 -<guid isPermaLink="false">%s</guid>
1.8 +<guid isPermaLink="true">%s</guid>
1.9 </item>"""
1.10
1.11 for li in lis:
2.1 --- a/templates/index.html Wed Mar 19 13:49:53 2008 -0700
2.2 +++ b/templates/index.html Wed Apr 23 14:05:32 2008 -0700
2.3 @@ -4,6 +4,9 @@
2.4 <h2>News & Updates</h2>
2.5 <ul class="news-list">
2.6 <li>
2.7 + <span>23 Apr 2008</span>—<a href="http://apgwoz.com/sketch/whats-in-a-vote/">What's in a Vote?</a>
2.8 + </li>
2.9 + <li>
2.10 <span>10 Mar 2008</span>—<a href="http://newshoured.com/" rel="external">Newshoured</a>, a site that provides "Local News Without Fuss" has launched. It's not that exciting now, but maybe soon.
2.11 </li>
2.12 <li>
3.1 --- a/templates/mainlayout.html Wed Mar 19 13:49:53 2008 -0700
3.2 +++ b/templates/mainlayout.html Wed Apr 23 14:05:32 2008 -0700
3.3 @@ -36,7 +36,7 @@
3.4 <div id="footer">
3.5 {% block "footer" %}
3.6 <p>
3.7 - <a href="/copyleft/" title="Copyleft 2007"><img src="/images/cleft.gif" alt="©" /></a> 2007 Andrew Gwozdziewcz – <a href="/about/">Who is Andrew?</a>. This page is hopefully <a href="http://validator.w3.org/check?uri=referer" rel="external nofollow">valid</a>.
3.8 + <a href="/copyleft/" title="Copyleft 2008"><img src="/images/cleft.gif" alt="©" /></a> 2008 Andrew Gwozdziewcz – <a href="/about/">Who is Andrew?</a>. This page is hopefully <a href="http://validator.w3.org/check?uri=referer" rel="external nofollow">valid</a>.
3.9 </p>
3.10 {% endblock %}
3.11 </div>
4.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
4.2 +++ b/templates/sketch/whats-in-a-vote.html Wed Apr 23 14:05:32 2008 -0700
4.3 @@ -0,0 +1,93 @@
4.4 +{% extends "mainlayout" %}
4.5 +
4.6 +{% marker "title" set "What's in a Vote" %}
4.7 +
4.8 +{% block "content" %}
4.9 +
4.10 +<h2>What's in a Vote</h2>
4.11 +<h4>April 23 2008</h4>
4.12 +
4.13 +<p>With the Pennsylvania Primary behind us (congratulations Mrs. Clinton), it
4.14 +seems like a great opportunity to share my views on the election process.
4.15 +Let's first make it clear that, I do not affiliate myself with any political
4.16 +party and as such did not participate in yesterday's election. </p>
4.17 +
4.18 +<p>I asked my mom this morning whether or not she voted. She had, but she made
4.19 +up her mind on whose button to push just before stepping inside the booth. It's
4.20 +a tough decision, especially when you think politicians are inherently
4.21 +crooked, but nevertheless she voted for a candidate which she saw as the lesser
4.22 +of two evils. </p>
4.23 +
4.24 +<p>I asked her why she even bothered to vote, if she couldn't side with one
4.25 +candidate more than the other, and her response was that "as long as the GOP
4.26 +doesn't get in, I'm cool." </p>
4.27 +
4.28 +<p>It's disappointing that, much of the <abbr title="United States">U.S.</abbr>
4.29 +shares the same kind of view. So, what's in a vote? </p>
4.30 +
4.31 +<p>The idea of democracy is that we elect leaders who we agree with and side with
4.32 +to make informed decisions about the laws, policies, taxes and occupancy of
4.33 +foreign countries—yet we've come up with this "rule of the lesser evil",
4.34 +to combat the idea that no candidate will do what "I would do if I were in
4.35 +office." While this might be an acceptable position to take, it really doesn't
4.36 +fix the problem of getting candidates in office that actually represent what
4.37 +it is that we feel need to be represented. In other words, why are we so
4.38 +easily coerced into voting for someone who we side with in 60% of issues? The
4.39 +answer, is that we're not coerced into voting for someone who we necessarily
4.40 +<em>agree</em> with, but instead the candidate (on our side) who annoys us
4.41 +least. </p>
4.42 +
4.43 +<p>Riding home on the bus yesterday, I overheard a conversation between two woman
4.44 +about the automated messages they were receiving from the campaigns of Senator
4.45 +Clinton and Senator Obama. They both seemed very irritated and I could side
4.46 +with them, but when one mentioned that she voted for Obama on the pure basis
4.47 +of receiving less calls from his campaign, my jaw dropped. </p>
4.48 +
4.49 +<p>How many other people are voting for candidates based on appearances and not
4.50 +upon the issues at hand? In other words, how many people are voting that really
4.51 +shouldn't be voting, and how is <em>this</em> democracy? </p>
4.52 +
4.53 +<p>It seems to me that we can solve the uninformed voter problem by changing the
4.54 +way voting works. Don't cast a ballot; instead fill out a survey which ranks
4.55 +importance of the issues at hand, and build a fingerprint of your ideal
4.56 +candidate. Each candidate also fills out this survey and we create clusters
4.57 +of people with similar ideas as the candidates.</p>
4.58 +
4.59 +<p>
4.60 +Let me explain an example of a <a href="http://en.wikipedia.org/wiki/Computer_Vision" rel="external">Computer Vision</a>
4.61 +problem, in which the goal is segment the image from millions of colors into,
4.62 +say 3 for the purposes of creating <a href="http://en.wikipedia.org/wiki/Pop_art">Pop Art</a>. The idea is simple. Each pixel is composed of three values a
4.63 +<a href="http://en.wikipedia.org/wiki/Rgb" rel="external">red, a green and a blue</a>.
4.64 +When each of these values is plotted in 3 dimensions, we get a cloud of pixels,
4.65 + and the pixels which are closely related in color, are also related spatially
4.66 +in the cloud.
4.67 +</p>
4.68 +
4.69 +<p>
4.70 +If we want to turn millions of colors into 3, we plot the 3 colors we want to
4.71 +use and assign every other pixel to the closest target color. If our target
4.72 +colors are pure red, pure yellow and pure blue, then all of the shades of blue
4.73 +turn into the pure blue, all of the shades of red get changed to pure red,
4.74 +yellows become pure yellow and the colors in between, like the purples, could
4.75 +be either mapped to pure red or pure blue. It all depends on how close they
4.76 +are in the cloud to the target colors.
4.77 +</p>
4.78 +
4.79 +<p>
4.80 +The same model can be used to elect members to the electoral college, only now
4.81 +we're using N dimensional space, where N is the number of survey questions to
4.82 +be answered to cast your "vote". Each candidate's responses become the "target
4.83 +colors," so to speak, and instead of assigning new colors, we assign
4.84 +membership to the candidate's cluster. To elect a candidate, all
4.85 +we have to do is count the number of people in each of the "clusters."
4.86 +</p>
4.87 +
4.88 +<p>
4.89 +Unfortunately, there's a slight problem with this. Mathematically, it's
4.90 +possible for a "vote" to be centered directly between all N candidates.
4.91 +Fortunately, the solution to this is exactly the same as what people really
4.92 +do when they go to the polls and almost exactly the same as what my mother
4.93 +did when she voted yesterday, you just flip a coin, or roll an N sided die.
4.94 +</p>
4.95 +
4.96 +{% endblock %}