How to buy a printer

I saw an interesting post at 37signals complaining about how hard it is to decide which All-in-one printer to buy.

The process is actually about as simple as anything in computers, but like most things in technology you need to use filters.

For example, if you're going to be buying a new printer and you run Linux on your desktop, it's a safe bet to throw a dart on a list of HP All-in-one printers, go to the OpenPrinting database and see how well the printer is supported.

In any event, the steps below should help you to find the right printer for you regardless of the OS you run.

1. Identify what you want to use it for.

- Will you be scanning many pages at the same time regularly? (you need an automatic sheet feeder)
- Do you care about color quality? (besides looking at color dpi you'll want to search customer reviews for "quality")
- Does speed matter to you? (If you will never print out more than 4 pages at a time and the printer is just for you, likely no).
- Will you be printing out photos on a regular basis? (a photo printer might be a better bet than an All-in-one printer, though my old HP PSC 1610 All-in-one prints photos remarkably well)

2. Go to newegg.com. You don't have to buy it from Newegg, but their site tends to make finding what you need much easier than other online stores.

From there go into Computer Hardware | Printers and Faxes | Select Inject or Laser Printers | Select Advanced Search if you're not already there | Select what kind of printer you want under recommended use (All-in-one in this case). Filter by Best Rating if you aren't already there.

3. Look at the top 10 recommendations that aren't made by Brother or Lexmark, see which ones meet what you're looking for in terms of quality and speed. You will be left with around 5 in most cases.

4. See which printers have drivers for your OS. Search for forum and blog posts of people complaining about installing or using the drivers (For example, for an HP F4280 search Google for "HP F4280 all-in-one linux problem". Search the customer reviews on Newegg for the OS you run. Ignore any printer that has poor support for your OS.

5. Buy the remaining printer that fits your budget.

Great color scheme for Konsole (Tango Console)

I'm currently using the color scheme Tango Console with Courier New bold 9pt in Konsole on KDE 4. You can see a couple screen shots of how it looks below.

Groups that your user needs to be a member of in Arch Linux

In the weekend that I've been playing with Arch Linux on my laptop I've found that I needed to add my main user account to the groups below:

[scyon@blackbird ~]$ grep scyon /etc/group | cut -d: -f1
network
audio
optical
storage
power
vboxusers

Getting Tor, Privoxy and Torbutton working on Ubuntu

This was tested on my laptop running Ubuntu 8.04 (Hardy Heron).

1. From a terminal run

sudo apt-get install tor privoxy

2. Edit /etc/privoxy/config
3. Add the line (including the period at the end):

forward-socks4a / localhost:9050 .

4. Comment out the line:

logfile logfile

5. Restart Privoxy:

sudo /etc/init.d/privoxy restart

6. Install the Firefox torbutton extension.

Right clicking the torbutton and selecting Preferences | Proxy Settings | Test settings should work out of the box now.

The best thing I've read all day

Interoperability, transparency, extensibility, and storage or transaction economy: these are the important themes in designing file formats and application protocols. Interoperability and transparency demand that we focus such designs on clean data representations, rather than putting convenience of implementation or highest possible performance first.

From The Art of Unix Programming.

Free (as in beer) PDF to Word Conversion

Normally I try to avoid talking about my job here (this is my personal blog and nothing on it is attributable to my employer) but I want to make an exception in this case.

The company I work for (Solid Documents) just launched a free PDF to Word conversion service.

Usually when I'm at home I have to fire up a VM running Windows (and Solid Converter PDF) to convert a file (I typically convert to RTF as our software doesn't require MS Office for that conversion), but now I don't have to (and neither do you). Nice.

Stupid (but useful) one liner for Visually Monitoring Free Disk Space

I'm sure there is a better way of doing this (or a prebuilt tool that does) but I felt like doing some oldschool shell scripting while I update my local Ubuntu mirror for Intrepid Ibex.

One liner:

while true; do clear; date; df -ah; sleep 10; done

The output looks like:

Wed Nov 12 20:10:43 PST 2008
Filesystem            Size  Used Avail Use% Mounted on
/dev/hda1              72G   49G   20G  71% /
tmpfs                 506M     0  506M   0% /lib/init/rw
proc                     0     0     0   -  /proc
sysfs                    0     0     0   -  /sys
procbususb               0     0     0   -  /proc/bus/usb
udev                   10M   44K   10M   1% /dev
tmpfs                 506M     0  506M   0% /dev/shm
devpts                   0     0     0   -  /dev/pts
/dev/hdb1             459G   59G  377G  14% /var/spool/apt-mirror

Ping a range of IP addresses using Python (or whatever else you need to do)

#!/usr/bin/env python
#
# IPRange. Do stuff with IP addresses.
#
# this is code is designed to parse a range of ip addresses and
# do something with them. Example uses are ping, nmap, ssh, etc.
#
# have fun, but don't do illegal things and if it breaks everything
# around you it isn't my fault. Use this only for good, not evil.
#
# contact author(Sam Levine):
# host = darkscyon
# domain = geemail.com (you know, google mail)
# no spam or complaints please. patches are welcome. feedback
# (good or bad) is very welcome.
#
# Be excellent to one another.

def IPRange(octets, func=""):
    # octets = "192.168.50-55.1-100"
    # func = (some function you've written)
    # objects passed to func are strings
    
    if func == "":
        def func():
            pass

    octets = (octets.split('.'))
    ranges = []
    loop = 0
    for octet in octets:
        # this is ugly
        # basically the idea is that if the user enters a range of ip
        # addreses delimited by '-' it parses out the first part of the
        # range and the second part.
        # it works, just not pretty or readable
        if octet.find('-') != -1:
            spot = octet.find('-') + 1
            octets[loop] = int(octet[:octet.find('-')])
            ranges.append(int(octet[spot:]) + 1)
        else:
            octets[loop] = int(octet)
            ranges.append(int(octet) + 1)
        loop += 1
    CurrentAddress = ""
    loop = 0
    output = []
    for one in range(octets[0], ranges[0]):
        for two in range(octets[1], ranges[1]):
            for three in range(octets[2], ranges[2]):
                for four in range(octets[3], ranges[3]):
                    for item in (one, two, three, four):
                        CurrentAddress += str \
                            ((one, two, three, four)[loop]) + "."
                        loop += 1
                    CurrentAddress = CurrentAddress[:-1]
                    output.append(func(CurrentAddress))
                    CurrentAddress = ""
                    loop = 0
    # return a list of the output
    return output

if __name__ == '__main__':
    import os
    # self test code
    def func(x):
        print x

    test = IPRange("192.168.1-2.10-15", func)
    print "The output should be none", test

    # example usage
    def pinger(x):
        input, output = os.popen4('ping -c2 ' + x)
        return output
    pinglines = IPRange("192.168.1.1-254", pinger)
    for streams in pinglines:
        for line in streams.readlines():
            print line,

A python script to find and play all the videos on your computer

#!/usr/bin/env python
#
# a silly script I wrote to find and play
# all the movies on my hd.
#
class FileLister:
    def __init__(self, suffixes, directory=None):
        self.suffixes = suffixes
        if directory:
            self.cwd = directory
        else:
            self.cwd = os.getcwd()
    
    def list_files(self):
        list = []
        items = os.listdir(self.cwd)
        for item in items:
            for suffix in self.suffixes:
                if item.endswith(suffix):
                    list.append(item)
        return list

    def change_dir(self, directory):
        self.cwd = directory

    def change_filter():
        pass
    
    def walk_dir(self):
        movies = []
        for root, dirs, files in os.walk(self.cwd):
            self.change_dir(root)
            list = self.list_files()
            if list:
                for item in list:
                    movies.append(os.path.join(root, item))

        return movies

class MoviePlayer:
    def __init__(self, movie_list, player='vlc'):
        self.player =[player]
        self.mlist = movie_list

    def play(self):
        command = self.player
        for movie in self.mlist:
            command.append(movie)
        
        subprocess.call(command)

if __name__ == "__main__":
    import os, subprocess, time
    # replace the magic string below with the location you store your videos
    dir = os.path.expanduser('/media/storage')
    # you may wish to edit the list below to add filetypes you use
    lister = FileLister(['.avi', '.mpg', '.flv', '.mkv'], dir)
    movies = lister.walk_dir()
    movies.sort()
    # mplayer can be substituted with gmplayer or left out to use
    # vlc by default
    playme = MoviePlayer(movies, 'mplayer')
    playme.play()

My current python "IDE"

The script below is one that I'm currently using to quickly setup my python coding environment.

#!/usr/bin/python
import os
os.popen4('gnome-terminal -e python --working-directory ~/scripts/python -t PYTHON')
os.popen4('gnome-terminal --working-directory ~/scripts/python')
os.popen4('xterm -e "cd scripts/python && /bin/bash"')
os.popen4('firefox http://docs.python.org/ http://www.pygtk.org/pygtk2tutorial/index.html')

You may be asking yourself why I'm opening 3 terminals. I personally like having lots of text in front of me at once so I can focus on reading/thinking and not using the mouse (or the keyboard).

Xterm is there because it is really really fast. Gnome-terminal is there due to it's superior copy/paste support (I grew up on Windows, sue me).

determining your visible IP address using perl and whatipmyip.com

A while back Ashish Shukla put together
a great hack for using perl an whatismyip.com to determine your IP address. My modified version of this is below:

#!/usr/bin/perl -w
#whatismyip.pl - This script uses http://www.whatismyip.com/ service.

use LWP::Simple;
use strict;

my $url = "http://www.whatismyip.com/automation/n09230945.asp";

my $content = get($url);
print "$content\n";

fun with nmap

I had some fun with nmap today.

Should you care to see some of the horrible pipe-fitting I did please see the commands below:

nmap -A -vv 192.168.1.1-254 > nmapoutputwithosdetection.txt  
grep "open port" nmapoutputwithosdetection.txt | awk '/[0-9]+\.[0-9]+\.[0-9]\./ {print $6}' | \
sort -u > iplisttoscan.txt
nmap -T sneaky -iL iplisttoscan.txt

What does this do? nothing productive whatsoever really. I wanted a script to go through and determine the hosts up on my local network, then simulate someone being "sneaky" when using nmap so I could watch the scan using tcpdump (note to self: start using a better sniffer like wireshark).

my .fluxbox/keys file

Although I love gnome and kde, they are are both way too heavy for my EEE PC 4G 700. Speed is extremely important on a device with limited cpu power, more so on a mobile device (hence why putting Vista on a UMPC is insane). My choice for a window manager on my EEE PC is Fluxbox. It is easily configurable (with a little reading) and is very responsive, even on slower hardware.

Below is the contents of my .fluxbox/keys file, hopefully someone will get something out of seeing it.

OnDesktop Mouse1 :HideMenus
OnDesktop Mouse2 :WorkspaceMenu
OnDesktop Mouse3 :RootMenu
OnDesktop Mouse4 :NextWorkspace
OnDesktop Mouse5 :PrevWorkspace
# 100 = left 102 = right 61 = / 98 = up 104 = down
Control Mod1 100 :PrevWorkspace
Control Mod1 102 :NextWorkspace
Control Mod1 61 :ToggleDecor
Control Mod1 98 :MaximizeWindow
Control Mod1 104 :Exec gnome-do
Mod1 F1 :Workspace 1
Mod1 F2 :Workspace 2
Mod1 F3 :Workspace 3
Mod1 F4 :Workspace 4
Mod1 F5 :Workspace 5
Mod1 F6 :Workspace 6
Mod1 F7 :Workspace 7
Mod1 F8 :Workspace 8
Mod1 F9 :Workspace 9
Mod1 F10 :Workspace 10
Mod1 F11 :Workspace 11
Mod1 F12 :Workspace 12
Control Mod1 f :Exec firefox
Control Mod1 t :Exec xterm
# turn on bluetooth and connect to my n75
Control Mod1 i :Exec gksu netz
# turn off bluetooth
Control Mod1 n :Exec gksu nonetz
Control Mod1 p :Exec pidgin
Control Mod1 c :Exec gcalctool
Control Mod1 k :Exec conky
Control Mod1 e :Exec FBReader
Control Mod1 0 :Exec gksu /etc/init.d/networking restart
Control Mod1 9 :Exec gksu ifconfig ath0 down
Menu :RootMenu

Projects I'm currently working on (or will be soon)...

1. Creating an Ubuntu mirror.

2. Building a Squid Web Cache Server.

3. Installing OpenWrt on my Linksys router.

4. Reading O'Reilly books on OpenSSL, LDAP Administration and Kerberos.

Setting up FirePGP on Ubuntu 8.04 (In 5 seconds)

Although I found several guides for installing GnuPG and FireGPG (on Ubuntu no less) I didn't find anything that put all the instructions together in one place.

Below are the steps needed (at least at my end) to set up FireGPG on Ubuntu 8.04. It's really more of an overview (and it takes more than 5 seconds), but it should be helpful in getting you where you need to go.

If you don't know what FireGPG is, you really don't need to be here. Just know that unencrypted e-mail is like unprotected sex :)

Please don't do anything crazy; if following these instructions blows up your computer it isn't my fault. :)


Setting up FirePGP on Ubuntu 8.04

1. Open a terminal and run:

sudo apt-get install gnupg gpa

2. Run gpa from the terminal. Follow the on screen prompts to create a key. If you cannot do this you should be administering another operating system.

3. Install FireGPG:

Click here.

Click Download | Allow | Install. Restart Firefox.

4. Restart Firefox when prompted. Go to gmail you should be able to encrypt and decrypt your e-mail.

Enjoy!

Using Perl to Improve Poor Spelling

#!/usr/bin/perl
# 
# written by sam levine (hn darkscyon dn gmail tld com)
# this software comes with no warranty. it may blow up your computer. you have been warned. 
# this code is licensed under the GPL.
# this is some random code I was playing with to better grasp regular expressions
# I will improve it at some point (comments, actually working, etc.), for now feel free to 
# have fun with it if you are so inclined.
#
#
sub clean {
my $input = $_[0];
print "problem text = $input\n";
$input =~ s/(.*)/\L\1/;
$input =~ s/(.*)/\u\1/;
$input =~ s/\bliek\b/like/ig;
$input =~ s/\bpwn/defeat/ig;
$input =~ s/\bteh\b/the/ig;
$input =~ s/z(o|0)r\b/ing/ig;
$input =~ s/(!|\?)(!|\?)+/\1/ig;
$input =~ s/kthank?x+/thank you/ig;
$input =~ s/\bwit\b/with/ig;
$input =~ s/\bbiye+/bye/ig;
$input =~ s/\bomg\b/exclamation/ig;
$input =~ s/\breall+y\b/really/ig;
$input =~ s/\bu\b/you/ig;
$input =~ s/\br\b/are/ig;
#print $input;
return $input;
}



while (<>) {
$output = &clean($_);
print $output;

}

Windows (Vista or XP) is actually fine, it's the lack of good 3rd party apps that kills it for the home user

Thanks to lifehacker for putting together such a good list of 3rd party software to use on Windows:

http://lifehacker.com/384545/superior-alternatives-to-crappy-windows-software

The reason why I love Ubuntu (and Linux in general) is that you can install most of the equivalent apps with a single command line operation. This alone would keep me using Ubuntu, even if I ended up buying a Macbook Pro.

Microsoft: Stop making crappy apps and bundle the good stuff with your OS (or make it easy to install them). Small groups of hackers have made great software that they give away for free. Help out your customers and help them get it without having to be nerds.

Perl To Blogger

#!/usr/bin/perl
# 
# perltoblogger
# written by Sam Levine (hn darkscyon dn gmail tld com).
# this may blow up your computer. use it at your own risk.
# this code is licensed under the GPL.
#
# This script is designed to convert perl scripts into code that can
# be posted into
# blogger with a minimum of fuss
# it does not create full html (if it did I would call it
# perltohtml), rather it creates code within
# <pre></pre> tags that may be posted as code snippets 
# on blogger (or a forum, or a mediawiki wiki, etc.)
#
# Useage: perltoblogger [InputFile] > OutputFile.html
# Example: perltoblogger myperlfile > myhtmlfile.html
#
# thanks to http://www.w3schools.com/tags/ref_entities.asp for the 
# list of ASCII entities
# thanks to http://bguide.blogpost.com/2008/03/html-entities.html 
# for showing me how to be less of a noob.
# thanks to everyone involed in making perl.
#
print "<pre>\n";
while (<>) {
s:&:&amp;:g;
s:<:&lt;:g;
s:>:&gt;:g;
s:':&apos;:g;
s:":&quot;:g;

#s:<:&lt;:;
print;
}
print "</pre>\n";

a study aid for the Linux+ exam: perlreadsome (Man Pages)

While I was studying for my Linux+ exam I was unable to
find a utility to cycle through all the administrative
man pages on my computer. I wrote the script below for
this purpose.

At some point I may put it up on sourceforge (by that
time it might even work on a non-ubuntu linux computer)
, but for now I just wanted it off of my eee pc before
I blow it away and install Ubuntu 8.04 on it.

#!/usr/bin/perl
#
# Perl Read Some (Man Pages) version 0.01
#
# this script was written by sam levine (hn darkscyon dn 
# gmail tld com).
# it come with NO WARRANTY at all. it is licensed 
# under the GPL (Gnu Public License)
# I am a noob programmer. this may make your computer 
# explode. YOU HAVE BEEN WARNED.
# 
# what this script is designed to do:
# Enable you to study man pages.
#
# while I was studying for my Linux+ test I read man 
# pages a lot to prepare.
# doing so was a pain in the arse. I decided to write 
# this script for fun,
# and hopefully to help other people prepare for the 
# test (this may be helpful for
# the LPIC tests as well.

# 
#@pages = qw[ man ls ping fdisk ifconfig ];
print "create the manlist? (press y then enter)
\n"; # this creates the list of pages to read
 chomp($wait = <STDIN>);
 if ( $wait eq 'y' ) {
 open MANLIST, "> manlist.sav"; # 
# this overwrites any file named manlist.sav in the 
# same folder as the script.
 select MANLIST;
 print ""; #clear manlist.sav
 close MANLIST;
 select STDOUT;
 print "cleared the manlist.sav file\n"; 
# user feedback is important


 @input = `ls /usr/share/man/man8/`;  # create an 
# array with all the admin man pages
      # need to determine the location of this 
# and make it portable.

 foreach (@input) {
  s/[\.]+.+$//; 
# remove everything after the . in the file name
  open MANLIST, ">> manlist.sav";
  select MANLIST;
  print; # add the page to the file
  close MANLIST;
  select STDOUT;
 }
 print "Just made the manlist.sav file. 
Press Enter to continue\n"; # I love feedback ;)
 chomp($wait = <STDIN>); 
}

open MANLIST, "manlist.sav";

foreach $item (<MANLIST>) {
 push(@pages, $item); # create an array with all 
the pages in the manlist.sav file
 }

# print "$pages[1]\n";
# print "$pages[2]\n";
# print "$pages[ -1 ]\n";

#$page = 'man';
foreach $page (@pages) {
 #$maning = `man $page`;
 system "man $page"; # read the man page
 #print $maning;
 push(@pages, $page); 
# copy the first item to the last item
 shift @pages; # remove the first item
 print "\nCare to quit? 
please press q then enter\n";
 chomp($wait = <STDIN>);
 if ( $wait eq 'q' ) 
   {
   print "saving the manlist\n";
   open MANLIST, ">manlist.sav";
   select MANLIST;
   print ""; # clear manlist.sav
   open MANLIST, ">>manlist.sav";
   select MANLIST;
   # recreate the manlist.sav file with the new organized list of pages
   foreach $page (@pages) {
    print $page;
   }
   close MANLIST;
   select STDOUT;
   last;
 }
 
}

Overclocking my eee pc 701

As much as I like using Ubuntu on my eee pc, I find it to be a little on the slow side when compared to the easy mode version of Xandros (their description, not mine) that the laptop ships with.

Due to this I decided to "overclock" the device back to the speed it was supposed to ship at (900mhz). Doing this will reduce your battery life and can completely destroy your eee pc, even if you do it correctly. Please don't complain if your device gets borked while doing anything mentioned in this post, this is hacky stuff.

I've had good results with the kernel module written by kiwidrew:

http://forum.eeeuser.com/viewtopic.php?id=9797&p=1

After following the instructions to download install the kernel module in the above link, I put together the script below to overclock the eee:

sh -c 'echo 85 24 1 > /proc/eee/fsb'
echo "FSB overclocked to 85MHz"
sleep 2
sh -c 'echo 90 24 1 > /proc/eee/fsb'
echo "FSB overclocked to 90MHz"
sleep 2
sh -c 'echo 95 24 1 > /proc/eee/fsb'
echo "FSB overclocked to 95MHz"
sleep 2
sh -c 'echo 100 24 1 > /proc/eee/fsb'
echo "FSB overclocked to 100MHz"

I named the script oc, made it executable (chmod +x) and copied it to /usr/local/bin.

I can overclock the eee by running sudo oc. I've been using it for a good month without any stability problems, so today I decided to make it start automatically.

To do so I made a link to it using the command below:

sudo ln -s /usr/local/bin/oc /etc/init.d/oc

I made it start automatically by running the command below:

sudo sysv-rc-conf --level 2 oc on

With any luck you'll have an overclocked eee that will be slightly faster than before. You should check the content of /proc/eee/fsb to make sure it's really running at 100 Mhz (the cpu speed is 9 times this, 900 mhz).

Thanks to Kiwidrew for writting an awesome kernel module!