Most of the people I know missed the launch announcement and I’ve been bringing it up every chance I get. My friends over at Cards.ly just couldn’t find anything that worked well for them to keep all their ideas and tasks in order. I showed them Trello and they thought it looked great, and after using it for a while, now swear by it.

I myself hadn’t used it until this week when I needed to coordinate Jay’s visit down here. I wanted a place to keep track of what he wanted to do and what I needed to do to prepare for his visit. He also has a project that he’d like some help on so we wanted a place to coordinate on that as well. Trello has been great for all of these things, providing a responsive way for us to keep each other accountable and keep track of what needs to be done.

Joel Spolsky recently did a post on his blog about how Trello is different (for those that don’t know, his company wrote Trello). In it, he explains that the reason Excel became so popular is that despite it’s intended use as a spreadsheet tool for doing a bunch of math, it’s really just a data structure. That is, it’s a simple way to make tables. Hell, I’ve used Excel to do website mockups before, so I get that. Joel claims that Trello is also just a simple data structure, a simple way to make lists. I think he’s spot on with that, but it’s the tools that you get with those lists that makes Trello in particular really valuable. Click on a card and you get access to a broader description, checklists, notes, due dates, activity lists and even votes.

Since signing up three days ago, I’ve built 4 boards (collections of lists) and I intend to build a lot more. I’ve also gotten another developer friend of mine to start using it to manage his community project and he too really seems to enjoy it.

tl;dr If you want to manage tasks and anything else that could be a list, go sign up for Trello right now!

• Identify a potential problem you’d like to solve
• Create a landing page that states the problem and that you have the solution
• Give little or no insight in to your solution, just that it works
• Collect email addresses on your landing page
• Run ads tailored to your idea and landing page
• Tweak as needed to get conversions (sign-ups)

It’s a simple plan, but it’s pretty valuable. It also jives well with the whole Lean Startup movement. The theory goes that if you can’t even get people interested enough in your idea to give you an email address, there’s no way you can expect them to give you money. If your conversions, that is, sign-ups, are low, then you have two options; tweak the landing page and problem statement to better match what you think customer needs might be, or scrap the idea and go focus your efforts elsewhere.

I have a few ideas I’d like to build but I’m definitely going to start here. First things first, I’d like to find a good landing page creation/hosting site. So far, I’ve tried LaunchRock but I wasn’t too impressed with their options and integration. The article suggests using Unbounce but while I don’t have a problem paying for a solution, I’d rather start cheaper and scale up. (This is also why I’m currently in love with Heroku, btw).

I haven’t dug too deep, but I’d love to hear about other solutions. I may just end up building my own though…

In order to get your application on their service, you need to install their command line tool, which requires Ruby. No problem, just install it all with apt-get and you’re good to go, right? Short answer, yes. Long answer, No. You’ll be using an unsupported version (the version of ruby in the Ubuntu repos is aged), so while it might work now, it may not in the future. If you do like I did, you’ll also end up with 1.8 and 1.9 on your machine, and while that didn’t cause any problems, it seemed silly.

Poking around, I found this article which lays out how to use RVM to both install a newer version of Ruby as well as easily update it in the future. So, here’s the whole shebang:

$ mkdir ~/src
$ cd ~/src
$ sudo apt-get install build-essential libssl-dev git-core git # some dependencies before we start
$ git clone https://github.com/joyent/node.git
$ cd node
$ git checkout v0.4.7 # heroku only support 0.4.7 now, change this as that changes
$ ./configure && make && make install # I opted to use checkinstall, but this keeps it simple # now we move on to RVM and Ruby
$ bash < <(curl -s https://rvm.beginrescueend.com/install/rvm)
$ echo '[[ -s "$HOME/.rvm/scripts/rvm" ]] && source "$HOME/.rvm/scripts/rvm"' >> ~/.bashrc
$ . ~/.bashrc
$ rvm requirements # towards bottom, you will see "Additional Dependencies" and "For Ruby" -- copy everything from apt-get on
$ sudo apt-get install build-essential openssl libreadline6 libreadline6-dev curl git-core zlib1g zlib1g-dev libssl-dev libyaml-dev libsqlite3-0 libsqlite3-dev sqlite3 libxml2-dev libxslt-dev autoconf libc6-dev ncurses-dev automake libtool bison subversion # likely a bit different for you, but you get the idea
$ rvm install 1.9.2
$ rvm --default use 1.9.2
$ ruby -v # make sure ruby is in fact working, should give you some form of 1.9.2
$ gem install heroku

And that’s it. You now have a local version of Ruby and Rubygems for you to use. If you want to keep your version of Node local as well, you can use the –prefix when compiling, as explained in the Node wiki.

Now you’re ready to continue along with the Heroku docs and your Node app deployed on their server.

Google will give you a plethora of guides for compiling the drivers (this one is probably the best), but I just couldn’t get things to load. It was also never clear what version of the drivers I should be trying to run. Most tutorials made reference to 1.0.0.9, some to 1.0.0.10, but the current version (at the time of this writing) is 1.0.1.9. In the end, though, that’s the version I did manage to get working. Here’s how I did it:

• Grab the driver from the Atheros site Dead link, copy is here
• Locate the file on your computer, then make a new directory (name it anything, AR8131 will work) and move that file there
• Decompress the file
• Enter the resulting src directory

Once you’re in there, it’s time to start that terminal magic! One note; if you’ve already added another network card to your machine, the Atheros card may be eth1, not eth0. This was the case for me, and I’ll treat all the instructions here as though that’s the case for you too. If it’s not, just change eth1 to eth0.
make && sudo make install
sudo modprobe atl1e
sudo ifconfig eth1 up
sudo dhclient eth1


At this point, the machine will try to assign an IP address for the card. This may or may not work; for me, it did not. To get mine working, I simply restarted the networking init script. Before that though, let’s add the following to /etc/network/interfaces
# The secondary network interface
auto eth1
iface eth1 inet dhcp

And lastly, restart the network on the machine (you CAN reboot, but there’s no need).
sudo /etc/init.d/networking restart

At this point, you should have your network card up and running with an IP address assigned by your DHCP server or router. If not, keep hunting I guess, but you should at least be very close!

I was discussing this with my friend Jason and he recommended playing with the font sizes. Sure enough, that did the trick! It seems that the reason everything is so big is that the default font sizes are 10pt. I shrank them down and MAN does it look nice now!

You can edit these values one ways; via gconf-editor or via gconftool-2 on the command line. I won’t post the gconf-editor direction since, if you know your way around it, you can extrapolate the parts you want to edit via the gconftool-2 commands. The following settings worked extremely well for me, but you can adjust the font faces and sizes as you see fit.

gconftool-2 --type string --set /apps/metacity/general/titlebar_font "Sans Bold 8"
gconftool-2 --type string --set /desktop/gnome/interface/font_name "Sans 9"
gconftool-2 --type string --set /desktop/gnome/interface/monospace_font_name "Monospace 9"

This will adjust the title bar, all normal window text and all monospace text, respectively. Again, these numbers looked the best to me, but you can make them even smaller (or bigger) to fit your needs. I did these adjustments on my 1680×1050 display as well, and it looks amazing. Then again, I love small text!

Another way to free up screen space is to auto-hide the panel(s) you have. Now, I know this is cumbersome with the default settings because it’s slow to respond and expand, but you can tweak that stuff too. Here’s what I have set:

gconftool-2 --type boolean --set /apps/panel/toplevels/top_panel_screen0/auto_hide 1
gconftool-2 --type boolean --set /apps/panel/toplevels/top_panel_screen0/enable_animations 1
gconftool-2 --type string --set /apps/panel/toplevels/top_panel_screen0/animation_speed fast
gconftool-2 --type string --set /apps/panel/toplevels/top_panel_screen0/hide_delay 500
gconftool-2 --type string --set /apps/panel/toplevels/top_panel_screen0/unhide_delay 100

Now, it opens much faster, animates much faster and goes away much faster. the hide_delay is fast enough that if I’m sloppy with the pointer, it still doesn’t disappear when I don’t want it to. Still annoying? play around with the hide and unhide delays until it works for you. Still too slow to pop in? Set enable_animations to 0 and it’ll come on the screen instantly.

As an added piece of advice, if your panel is over-crowded with junk, you can add another panel (or 2 or 3…) to separate out tasks. For example, I have a small, hidden, non-expanding panel I keep in the bottom left of the screen on which I have my timer-applet and my hamster-applet. Additionally, if you have something on your panel that you always want to see on your screen, add it to another panel that you don’t hide.

This affected IE6, IE7 and IE8 in my testing. The way I was serving the file was through PHP via readfile. Before sending the contents of the file, I was, of course, setting some header parameters so that the browser would handle the file and the user wouldn’t just see some binary garbage on their screen. In my case, I was serving a PDF file that was being generated server-side and sent to the client. The basic header parameters are as follows:

// We'll be outputting a PDF
header('Content-type: application/pdf');
// It will be called your_file.pdf
header('Content-Disposition: attachment; filename="your_file.pdf"');
// declare Files Size here (for the sake of peoples sanity, please add this
header('Content-Length: '.filesize($filename));

This works just fine in most situations, but once you introduce a secure connection, IE fails with the above error. Of course, like most IE problems, this doesn’t happen in any other browser. Ugh.

Luckily, I noticed some additional parameters in the first example on the readfile page, namely the ‘Pragma’ parameter. Reading some comments on the header function, I stumbled upon this one. It basically, says that once you add HTTPS, you need to also tell IE how to handle caching the file. So now, my headers look like this:

// Force clear cache
header('Pragma: public'); // Fix for IE
header("Expires: 0"); //always expire
header('Cache-Control: must-revalidate, post-check=0, pre-check=0');
// We'll be outputting a PDF
header('Content-type: application/pdf');
// It will be called your_file.pdf
header('Content-Disposition: attachment; filename="your_file.pdf"');
// declare Files Size here (for the sake pf peoples sanity, please add this
header('Content-Length: '.filesize($filename));

Directly following these header calls is the readfile call and a call to exit to stop the script from executing anything else and alert the browser that the entire file has been delivered.

It seems like the Pragma parameter controls how browsers handle caching, and adding the other cache parameters to the header are just added enforcement. Now, IE works exactly as expected and I’ve got happy clients once again!

When I got it back, I immediately installed Ubuntu’s most current version, 9.10 (Karmic Koala). I was really impressed with it; even more-so than with 9.04 in fact. But I ran in to a problem when I was setting it all up; I couldn’t get the backlight adjustment to work.

I resorted to the original Ubuntu Forums article on it but I quickly found that it no longer worked. LVDS and VGA had been replaced with LVDA1 and VGA1, and changing the command still didn’t work; it complained about something which I no longer remember. But, that error (as well as replies to that thread) lead me to this Launchpad comment about disabling KMS.

Seems KMS is enabled by default for a certain chipset, specifically the Intel i915 chipset. My next questions, of course, were “what is KMS” and “how do I disable it”? Looking around, I found out that KMS, or Kernel Modesetting, could be disabled like so:

Edit: /etc/default/grub
Change: GRUB_CMDLINE_LINUX=""
To: GRUB_CMDLINE_LINUX="nomodeset"

Then run

sudo update-grub

Now, reboot and KMS will be disabled. We can now refer back to the original post here, adding the following to the startup applications (or just running it from the command line each time, you’re call I guess).

xrandr --output LVDS --set BACKLIGHT 102 --set BACKLIGHT_CONTROL legacy --output VGA --auto

This still won’t make the brightness buttons on the keyboard work, but it WILL make xbacklight work. As the article points out, you can then set hotkeys for xbacklight commands.

So what do you lose by disabling KMS? Well, according to this article, KMS offloads the graphics initialization from X to the kernel, which has some benefits. But, the fact remains that it doesn’t work on a lot of machines (the Samsung X360, The MacBook and possible MacBook Pro, MSI’s U90, U100, and U120 netbooks and the Asus EeePC, to name a handful), so for users with this hardware, you don’t lose anything ;). I’ve read about some people having success with KMS in newer kernels, but it didn’t work for me so I’m not going to bother posting it. I’m hopeful that this will be fixed when 10.04 (Lucid Lynx) is released, but I haven’t yet tried the beta release to see.

• Toss some laundry in the machine
• Start my lunch and work while it cooks
• Sign for packages when they are delivered

And that’s just the tip of the iceberg. However, to make telecommuting possible, there are some technologies I simply MST have working. VPN is a big one, so I can get at the computers back in the office. SSH and FTP access is important to manage servers and files. And of course, I need to communicate with people; for that, we use AIM and Skype. Up until recently, it’s been enough for me to use my Nokia N800 to make my calls back to the office. It’s no good for chatting with video though, so I needed to set it up on my desktop with my camera and, preferably, my headset.

Simple enough task really; both my Logitech webcam and Plantronics headset work out of the box with Ubuntu (at least in 9.04) with nearly zero configuration required. I just plug them in, tweak Skype quickly and I’m up and running. However, with the addition of PulseAudio in the more recent version of Ubuntu, this became a little trickier.

You see, in the new version of Skype, the ONLY option I have for ANY audio, input or output, is simpley PulseAudio. No control for which device to use for different tasks. Enter padevchooser, short for PulseAudio Device Chooser.

It’s a pretty sweet application, and you should have no problem installing it since it’s in the repository (note: I’ve added the Medibunutu repo, and it’s possible it came from there). Once installed, you simply run it, from the command line, the run dialog, gnome-do or whatever method you choose, and you’ll get an icon in the notification area. Click on that, choose Volume Control, and you’re off and running. There are tabs for Playback, Recording, Output Devices, and Input Devices. Any current audio stream will show up in the list in the appropriate tab and you can move them around to different devices.

The easiest way to configure Skype is to place a test call. As soon as you start that call it will show up in the list. Move the “Skype: Output” stream to your headset in the Playback tab, do the same for “Skype: Input” in the Recording tab and you’re off and running. All Skype audio is now sent to your headset for easy chatting. I still haven’t figured out how to make it ring through the speakers and still use the headset for chatting, but this is good enough for now. Padevchooser made living with PulseAudio that much better!

generate.php

This script is where the user is sent when they click “download” on the parameter choice page. This is where the asynchronous magic has to happen. From here, the user is immediately sent to download.php.

download.php

This script gives the user either a “please wait” message of a “click here to download” message, depending on where the target PDF has been generated yet.

My problem was, even though was added the header redirect code (shown below) to generate.php, the browser would still wait for the PDF to be generated before sending the user to download.php. In other words, it was only synchronous PHP, which didn’t help me at all.

header("Location: download.php");

Thanks to this comment on php.net, I was able to make it work! Basically, you need to tell the browser that you are done sending it data, even though the PHP script will keep executing. Code below:

//redirects the browser to the new url, but continues processing in the background
function redirect_and_continue($sURL) {
header( "Location: ".$sURL );
ob_end_clean();
header("Connection: close");
ignore_user_abort();
ob_start();
header("Content-Length: 0");
ob_end_flush();
flush();
session_write_close();
}

So, the user is sent to generate.php, which immediately calls redirect_and_continue(‘download.php’), redirecting the user to download.php while still continuing to execute generate.php (and make the PDF file). Once the user is at download.php, the script checks for the existence of the generated PDF file (as indicated by the filename sent through the session) and uses a meta refresh tag in the HTML to keep reloading download.php. Once the PDF file exists, it provides the user with a link to the PDF file for download.

Thankfully in 9.04 (Jaunty), things seem to be working again. In fact, as I type this, my traffic is being sent from my local connection, even though I’m connected to the VPN at work. Since this still seems to be a big problem for many people, I thought I would post what worked for me.

First things first, in order to configure PPTP using network-manager, you’ll need to install the PPTP plugin for it.

$ sudo apt-get install network-manager-pptp

After installing it, you will need to at least log off to restart network-manager. For me, I had to reboot. Before I did, I kept getting an error about a lack of secrets, which basically means that it can’t connect to the password manager. Rebooting fixed that.

Next comes the configuration. You will need to know the IP address of the VPN server you wish to connect to, as well as your username and password. For me, I also needed to add routes for the internal network, the network that the VPN was on (we have the VPN on its own subnet at work) and the external network that your work has assigned to it. Without those, the connection would never route properly.

If you’re reading this and you’ve come this far, I assume you already know how to get to the configuration, so we’ll just skip to some screenshots to help you along. That last step may be optional, but it was a requirement for my configuration to work. YMMV.

Now, select the VPN connection and check out the routes.

$ route

If all went well, you should see some new routes going to the VPN network and be able to connect to the network resources. Something like this…

Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
10.242.1.0      *               255.255.255.255 UH    0      0        0 ppp0
some.domain     192.168.1.1     255.255.255.255 UGH   0      0        0 eth0
some.domain     192.168.1.1     255.255.255.255 UGH   0      0        0 eth0
192.168.1.0     *               255.255.255.0   U     1      0        0 eth0
10.242.1.0      *               255.255.255.0   U     24     0        0 ppp0
10.5.1.0        *               255.255.255.0   U     24     0        0 ppp0
default         192.168.1.1     0.0.0.0         UG    0      0        0 eth0

In my example, 10.242.1.0 is the VPN network, 10.5.1.0 is the internal network at work and some.domain is the reverse lookup of their IP address. Of course, 192.168.1.1 is the gateway for my local network. Look at all those beautiful ppp0 connections; working like a charm!

If it doesn’t work, try logging off and possibly rebooting again. If it STILL doesn’t work…. well, I’m out of ideas. There’s always Google and the Ubuntu forums. Good luck!