Free publishers weekly forecasts software for linux

Podcast::Publisher is a Perl module for creating and managing podcasts.

SYNOPSIS

use Podcast::Publisher;

my $podcast = Podcast::Publisher->new;
$podcast->set_logger( sub { my $msg = shift; print $msg; } );
$podcast->set_error_logger( sub { my $msg = shift; print STDERR $msg; } );
my $xml = "./podcast.xml";
$podcast->set_file( $xml );
$podcast->set_remote_root( "http://localhost.localdomain/podcast/publishing/" );
$podcast->set_db_connection( { driver => "mysql",
username => foo,
password => bar,
host => localhost,
database => podcast } );

# If we change podcast information, synchronize this information in the MP3 file itself
$podcast->set_synchronize( 1 );

$podcast->set_metadata( { title => "Chris Podcast",
description => "All About Chris",
"docs" => "http://localhost",
"editor" => "podcastmanager@localhost",
"webmaster" => "podcastmanager@localhost",
} );

# This adds an item to the database, and synchronizes the
# MP3 Tag information in the file with the database
$podcast->add_new_episode( { title => Some title,
author => Chris of course
category => Jazz
description => First in a series of many
mp3 => /home/foobar/mp3s/episode1.mp3 } );

$podcast->set_upload_settings( { host => localhost.localdomain,
username => someuser,
password => somepass,
path => podcast/publishing/,
remote_root => http://localhost.localdomain/podcast/publishing/ } );
$podcast->upload();

Better Weather contains a couple of scripts using normal KDE and Linux tools to provide better display of weather info than KDEs weather applet. It downloads the information from the US national weather service and maps from weather.com. Two launchers on the main menubar bring up the window separately. Full details and instructions are in the readme.

This is a solution that worked for me, not being a coder. You have to make it work for you. Its an idea and a set of tools. Its also US-centric, but someone may take the idea and make it work just as well anywhere. I hope someone finds it as useful as I do.

WHAT IT DOES

The tools I wrote are two *very* simple bash scripts. The script, localwx.sh, invokes the python weather utility, and downloads the current weather conditions and forecast in text format from the weather service. It stores the information temporarily in a text file (I chose the desktop as the directory in which to save it.) It then "reads" the text file into a kdialog box. When the "ok" button is clicked, the temporary file is deleted.

The other script, wxmap.sh, downloads a current local weather map image using wget, saves it temporarily to the desktop, and calls Kview to display it. When Kview closes, the temporary image file is deleted.

HOW I DID IT

The man pages for "weather," "kdialog," and "wget" were useful for determining which options to use. The script for weather forecasts has to be edited to receive the data for the users current location. My nearest airport is PDK, so my ID was "KPDK." The rest is straightforward enough. To make the weather map script work, you can go to weather.com or your local tv station website and copy the image location for the weather map they use. It can be pasted into the script.

Once the scripts were edited to suit, I made them executable (chmod +x filename), copied them to /usr/local/bin, and created a launcher for each in my main menubar. Describing the process is a lot more difficult than actually doing the work!

PieSpy is an IRC bot that monitors a set of IRC channels. It uses a simple set of heuristic methods to infer relationships between pairs of users. These inferrences allow PieSpy to build a mathematical model of a social network for any channel. These social networks can be drawn and used to create animations of evolving social networks.
PieSpy has also been used to visualize Shakespearean social networks.This page got slashdotted on 11 March 2004, with the site getting 250,000 hits per hour. Thanks to Notnet for making sure it all stayed alive!
PieSpy was presented at the Information Visualization conference (IV04) in July 2004. Read the full paper online. It has also appeared in Computer Weekly, ct magazine, and I was interviewed live on BBC Radio Kent.
Enhancements:
- Now tracks nickname changes. Generates images five times faster. Removes formatting and colors from messages. The source code has been refactored to make it easy to add new InferenceHeuristics, which are used to work out who is talking to whom - if you create any good ones, let me know! The config file lets you apply different weightings to each InferenceHeuristic.