Developing Usable Experiences

Symbols, words, meaning…you figure it out

September 23rd, 2008 10:47 pm —  559 views

Traffic wasn’t bad going into work school today and the parking garage had 400+ spots when I arrived. A nice change of pace from last semester when there would be 1 or 2 open spots on the display as I’d turn the corner.

I don’t remember who it was, but years ago I caught a television segment where some guy with either big curly hair or a pony-tail was lecturing on how the Internet and the information accessible therein is meaningless without a brain to interpret, give it meaning. Otherwise the data is virtually meaningless.

Sitting through lectures today on how to scrape web pages for chunks of data and later how to analyze shopper patterns to maximize conversion, I couldn’t help but think about how all the data we access is nothing but a bunch of magnetized bits distributed in a nearly impossible to follow manner…and we don’t care really about it. How it’s stored that is. We ultimately see patterns, put sentences together, assemble meaning based on arrangement of words etc. Pictures add a few words I suppose…

But let us not forget the subtle but unmistakable body language, smell, sound, even silence. When you’re around someone and nary a word is spoken (or IM’ed), is there still communication?

Inspired by a classmate, I took on the task of writing a Perl script to scrub an exported xml file of this blog so I could upload it to ManyEyes and Wordle to see what it had to say. Here is the script for the curious out there who want to appreciate the Perl programming language. I’ve no doubt this is a lame example, could be improved upon, and shows my general level of ignorance with Perl…I’d be inclined to agree.

open(IN,"wordpress.2008-09-23.xml") || die "FAIL. Could not open wordpress.2008-09-23.xml for reading. \n";
while($line = <IN>) {
    $some++;
    chomp;
    if  ($line =~ /\<title\>.+\<\/title\>/) {       
        $line =~ s/\<title\>//;
        $line =~ s/\<\/title\>//;
        $title = $line;
        next;
    }
    if ($line =~ /=\"\w+\">http:\/\/blog\.interspike\.com\/\?p=/) {
        $getanother = 1;
        while ($getanother) {
            <IN>;
            $line = <IN>;
            if ($line =~ /<\!\[CDATA\[/) {
                while ($line !~ /\]\]><\/content\:encoded>/) {
                    chomp($line);
                    $line =~ s/<.+[\"\/]>|<\/.+\>|<.+\[|<.+\>/ /g;
                    $line =~ s/[^a-zA-Z0-9 '\.,\?\!]*//g;
                    $title =~ s/[\t\n\r\f]+//g;
                    $posts{$title} .= $line;
                    $line = <IN>;
                }
                $getanother--;
            }
        }
    }
    #last if ($some > 2000);
}
for $title (keys %posts) {
    print "$title\n$posts{$title}\n\n";
}
close(IN);

Image: This is the result of a “wordle” based on all the words appearing in all my posts since fall of 2006. See anything interesting?

Related Posts

2 Comments 

Comments

2 Responses to “Symbols, words, meaning…you figure it out”

  1. Erin on September 24th, 2008 9:12 am

    Nice! I’m glad to see that “Get” has a prominent appearance in your word cloud as well. Way to take it to the next level with Perl. It’s sort of a running joke in our family to write a script to do anything – “hey, I spent 5 hours writing the script when you did it by hand in 2. But if I want to do the same thing again, all I have to do is push this button.”

  2. Dawn on September 24th, 2008 9:39 am

    I think this shows that you LIKE WORK! :) And that classes and weekends are about the same to you… And it’s pretty! Would make a good piece of wall art.

Leave a Reply




Subscribe without commenting