My Aeron Chair

A good ergonomic chair is a wise investment if you’re going to spend a lot of time at your computer. One of the better known ergonomic models is the Herman Miller Aeron Task Chair.

Picture of Aeron Chair

What Other People Say About the Aeron

Jeff Atwood (from Coding Horror) says:

In fact, after browsing chairs for the last few years of my career, I’ve come to one conclusion: you can’t expect to get a decent chair for less than $500. If you are spending less than that on seating – unless you are getting the deal of the century on dot-bomb bankruptcy auctions – you’re probably making a mistake.

I still believe this to be true, and I urge any programmers reading this to seriously consider the value of what you’re sitting in while you’re on the job. In our profession, seating matters:

Chairs are a primary part of the programming experience. Eight hours a day, every day, for the rest of your working life – you’re sitting in one. Like it or not, whatever you’re sitting in has a measurable impact on your work experience.

Cheap chairs suck. Maybe I’ve become spoiled, but I have yet to sit in a single good, cheap chair. In my experience, the difference between the really great chairs and the cheap stuff is enormous. A quality chair is so comfortable and accommodating it effortlessly melts into the background, so you can focus on your work. A cheesy, cheap chair constantly reminds you how many hours of work you have left.

Chairs last. As I write this, I’m still sitting my original Aeron chair, which I purchased in 1998. I can’t think of any other piece of equipment I use in my job that has lasted me ten full years and beyond. While the initial sticker shock of a quality chair may turn you off, try to mentally amortize that cost across the next ten years or more.

Joel Spolsky (from Joel On Software) says:

Let me, for a moment, talk about the famous Aeron chair, made by Herman Miller. They cost about $900. This is about $800 more than a cheap office chair from OfficeDepot or Staples.

They are much more comfortable than cheap chairs. If you get the right size and adjust it properly, most people can sit in them all day long without feeling uncomfortable. The back and seat are made out of a kind of mesh that lets air flow so you don’t get sweaty. The ergonomics, especially of the newer models with lumbar support, are excellent.

They last longer than cheap chairs. We’ve been in business for six years and every Aeron is literally in mint condition: I challenge anyone to see the difference between the chairs we bought in 2000 and the chairs we bought three months ago. They easily last for ten years. The cheap chairs literally start falling apart after a matter of months. You’ll need at least four $100 chairs to last as long as an Aeron.

There was a post with a large set of comments on Hacker News about chairs and cheap alternatives to Aeron Chairs, with a lot of people saying that there is no cheap alternative to a good chair.

What I Say About the Aeron

I just got a refurbished Aeron chair for Christmas. I am really pleased with it. It comes in three base sizes, being larger than normal I went for the Size C. Once it arrived it was a good fit, but it was easily configurable to make it into an excellent fit. I prefer my chair not to recline, so I can’t comment on whether it is not a good chair for reclining, as Jeff Atwood says, but I will say it felt comfortable and stable reclining.

The real benefit so far has been on my posture; I did not know how much back pain I was having before with my really cheap office chair. Only when I got up from a 4 hour session on my Aeron chair did I notice how much better my back was feeling. It moulds your back into a good position, and I have noticed my back clicking and popping into its proper shape, after using the chair for a couple of days.

I ordered my chair from, who I am very happy with, as they provided an MK2 Size C chair with lumbar support for less than £400, which is an absolute bargain compared to the retail price of a new chair. This also included delivery! It seems like new – there are no scuff marks and everything works perfectly.

I highly recommend getting an Aeron chair or an equivalent ergonomic chair recommended in the above resources, as I have really noticed a significant difference.

Tailing a log file and Running an Application at the Same Time


A quick tip this, but a useful one. You can tail a log file in the background while running a script in the foreground. So for example, I frequently execute the following commands:


tail -f /var/log/httpd.log &
/etc/init.d/apache restart

2. (The log file will spool onto the terminal as Apache is restarted.)

3. Once you are finished viewing the log file, foreground the log file process and kill it:


Then terminate the foregrounded log tail with a control-c.

With this technique you can run as many commands as you want, and see the real-time effects on your log file, without having to open a new terminal. You will also see your program output interspersed with your log file output, which can be helpful when tracing down particular problems.

Monitoring a Slow Internet Connection in OSX


I am currently on holiday in Tenerife, and although I really like it here, one thing I do not like is the internet connection we have in our resort. Sometimes networked applications will just hang with no warning and there will be minutes where it’s not clear what is going on. Here are some ways you can find a little bit more about what is happening when an application is slow or seems to hang when you have a poor internet connection. Execute the following commands each in a separate terminal window.

Log Files

tail -f /var/log/*

This will give you an indication of what is happening in OSX. For example, I was installing the XCode Command Line Utils from within XCode. The installation progress information is severely lacking, it just shows a bar which moves from left to right. However I was able to find out what was happening by tailing the log files in /var/log, which provided me with an updated breakdown of the installer progress. You can exit tail by using Control-C which will return you to the shell.

Constant Ping


When I have problems with my internet connection, I always keep a ping running in the background in a terminal. The interesting information here is the ICMP RTT time shown as the milliseconds next to the ‘time’ label, and how many packets were dropped shown by the number of ‘request timeout’ messages. Google does not mind you pinging it, just like hundreds of thousands of other people do, and so you can keep this up constantly, monitoring problems with your internet connection. When you get ‘no route to host’ printed, this usually means that your gateway or wireless connection is down, which means you usually have to reestablish a connection manually.


sudo tcpdump en0 -vvv

Do you really want to see what is happening on your computers network connection? Turn the floodgates on then, and use tcpdump. This will output information on each packet that your computer sends out and is received in a slightly Matrix-style torrent of information. If you are downloading something via an application or have a number of active web connections such as AJAX Facebook pages loaded, you would expect to see a lot of traffic. If you don’t have a lot of traffic, and you’re expecting a lot, then something may be wrong. You can use tcpdump to get a general feel of what data is being passed around, and to what IP address, which you can then look up later for more clarity. You can also use grep and some basic TCP/IP networking knowledge to find out what exactly is happening on the network level.

Network Connection Status of Each Application

sudo lsof -i tcp

Want to find out information about applications are using your internet connection, and the connection state of each TCP connection? Use lsof. You will be given the name of the application that is using each TCP connection, the IP address to which it is connected to, and the TCP connection state (established is good, time wait can be a problem sign). Run this regularly to check on the connection state of your programs. This won’t monitor UDP connections, but should cover your web browsers.

Hopefully this information will give you a bit more insight into what is actually happening on your OSX machine when your internet connection is being unreliable and you want more information about what is going on. Once you have this information, you can use it to inform actions such as toggling the wireless off and on again to reestablish a connection, reloading webpages that have hung, restarting application downloads, or possibly finding a new hotel or resort with a better internet connection 🙂

Reverting back to a previous version in CVS – the magic “undo” feature

If you’ve committed some code into to CVS, and made a mistake on that commit, you will want to know how to revert to a previously saved version. Here is the command line command for CLI versions of CVS:

$ cvs update -D '1 week ago'

Run this command in the main directory of your checked out working copy. This will revert your working copy to the version of the code that was checked in ‘1 week ago’ from the present date. You also use commands like “1 day ago” and “5 days ago”.

Then simply commit the changes with a log message:

$ cvs commit -m "Oops! Made a mistake, had to revert back to the 21/1/2011 version"

Netbeans for simple Java GUI Applications

I’ve been writing some simple Java GUI applications using the Netbeans IDE. It allows you to quickly make event-driven GUI applications, and generates a lot of skeleton code that you’ll need, but don’t necessarily want to type out. It reminds me of the IDE designer of Visual Basic 6, which allowed you to mock up simple GUIs with code in almost no time at all, although the VB language itself often proved difficult. With Netbeans you are using Java, and so you can make some powerful software with little effort.

Applications I Reccomend

Software I use on my macbook & PC:

DVDRipper Pro for Mac – DVD ripping, can also rip to ISO
Handbrake for Mac – Transcoding from DVD rip to iPhone-playable file
iMovie for Mac – Video editing
BabasChess for Windows – Best chess client for internet play
Hypercam 2 – Best screencapture utility
Skype for both – For reliable messaging as well as voice and video chat
Virtual Clone Drive for Windows – For mounting ISO images
iTunes for Mac – Best music player, and keeps media synced with iPhone
VLC Player for both – For watching movies
DVD Player for Mac – For watching DVDs

iPhone Apps:

Skype – Best messenger
iBooks – Best ebook reader
London Buses – Best London transport router, can route via tube, bus, cycle path and foot
Tube Status – Displays the status of all lines, with any disruptions summarised
NextBuses – Great app that gives you lots of info on the buses and bus stops in your area.
Apple Remote – Apple remote, allows you to control the music on any wi-fi linked iTunes library’s Chess – Great chess game for vs. computer play
TasteKid – Type in a film, author, tv series.. and it will give you similar recommendations
Google Earth – Brilliant for navigational help, although I use iPhone’s inbuilt Maps first, for most things.
SomaFM – Chilled out relaxing electronica

Tip for watching the completion of a large file copy

Forget the wonderful windows progress bar, and imagine I’m in the world of command-line Linux, and I want to copy a 484MB file, called VMware-server-2.0.2-203138.i386.tar.gz, from my home directory to a remote server. But I want to figure out how long it’s going to take.

1. First I can run a “du -m” command to get the total MB size of the original file:

du -m /home/david/VMware-server-2.0.2-203138.i386.tar.gz


david@believe:~$ du -m VMware-server-2.0.2-203138.i386.tar.gz
484 VMware-server-2.0.2-203138.i386.tar.gz

Now I know it is approximately 484MB.

2. Then I run the copy. I’m copying the file from /home/david/ to /opt/remote/myserver, which is a remotely mounted directory on a server somewhere in Canada.

david@believe:~$ cp ./VMware-server-2.0.2-203138.i386.tar.gz /opt/remote/myserver/

At this point cp will just hang until it’s finished. There is normally no progress indicator or anything. But I want to figure out how much of the file has been copied, so I can figure out how much is left to copy, and get a rough idea of the progress.

3. So I SSH into the remote server in Canada, and run this command

david@myserver:~$ watch du -m ./VMware-server-2.0.2-203138.i386.tar.gz

the copy command by default seems to be incremental, ie: piece by piece, not all at once. Therefore with the “Watch” command, you can watch the size, in MB, of the new file as it accumulates. The watch command will refresh every 2 seconds, so you’ll be updated as the copy goes on.

You can probably invoke a progress meter with the cp command, or use rsync. Rsync is much better for large file copies, and remote file copies. But the advantage of the method above is that you can watch file copies already executed without any special arguments, which I sometimes find very useful when I remember that that file I already started copying isn’t 200MB.. it’s actually 2.5GB.

Useful OSX commands for Linux users

I wrote this list to remind me, as a newcomer to OSX, how the command line differed from the Linux commandline. I thought I’d expand on it, and share it:

To mount any iso:

hdiutil mount sample.iso

To download a file as you would using wget:

curl -o linuxmint.iso -C -

the -o specifies the output file (required)
the -C – specifies automatically resuming if possible.

To burn a bootable iso to CD, DVD or USB key:

use the “diskutil” program as described in:

Monitor disk io utilisation.. poll once per second

iostat -c 99999

will run until 99999 seconds have passed.

Monitor CPU and memory utilisation.. polling per second


Just like Linux.

Mount Windows Shares

mount -t smbfs //@/ 


mount -t smbfs //davec@SERVER/Dev samba-to-netdev

then it will appear mounted in /Volumes with the mount point name you supplied, ie: /Volumes/samba-to-netdev/.

Long Bash History Files are Great.

When I’m installing software, or doing some complicated stuff on the linux command line, which nowadays is pretty much all the time, I will sometimes want to remember exactly what I typed.

Now the normal /home/david/.bash_history file is usually fine for that. Run this command, for example, and you will see the commands you typed in before you logged out of the server last time you used it:

cat ~/.bash_history

You can also find out what you typed in this session, ie: since you logged in, by typing this:


This is great, and it’s even more useful if you add a grep pipeline, so you can search through the previous commands you typed in for a particular phrase or command, ie:

history | grep apt-get

However what I really want nowadays is an almost infinite bash_history file, so I can find out not just what I did last week, but two weeks ago, or last month or perhaps last year. Now there are obvious security risks involved with this, and to make sure you don’t accidently store mistyped passwords to other systems, or other things, you should probably make sure you never type them in on the command line. This is good practice anyway, and since I use key’d sshd logins exclusively nowadays, there is not much chance of me tripping up, typing a password into the terminal, and then forgetting about it. In theory however, using long/infinite bash_history files does mean that if anyone compromised your shell account, they’d have any passwords to systems that you mistyped.

So I’m careful with this. You can also clear your history file quite quickly if you do accidently find you’ve messed up. Log out, log back in again, and just do this:

echo  > ~/.bash_history

Then that will delete all the previous logged commands.

Apart from serving as a major memory aid to complicated install work, and a log for those increasingly complicated chained, piped, one-liners that I’m fond of but only really want to have to type once, there are other benefits to keeping a large bash_history file. The main one is that it makes it easy to convert your previous commands into a handy shell script or two, which you can set to run at a specific time of day via cron.. or even make into a system-wide command for other users to use.

OK so hopefully I’ve convinced you that it can be very useful to have a long, persistent, bash_history file. But how do you configure the shell so that it does this for you? The following is the magic customization lines that I use on my personal desktops, laptops, and any other trusted computers that I think are reasonably free from the risk of people hacking in just to retrieve my .bash_history file..:

## bash history db
# increase the history file size to 20,000 lines
export HISTSIZE=20000
# append all commands to the history file, don't overwrite it at the start of every new session
shopt -s histappend

The above will give you an (almost) infinite bash_history file. It will start deleting old commands at 20,000 lines, ie: 20,000 commands. Make sure you have enough disk space for that. My .bash_history file is currently at around 200KB, not a huge file by any means. I’d say it will grow to 400-600KB max. If you want to calculate approximatly how much it will use, then in bytes, it’s the number of characters in your average linux command x 20,000.

MicroKORG + Python = MIDI fun!

microKORG and cat

So, about a month ago I got a second-hand microKORG from Ebay. Fiddling around with the preset patches, and creating new patches is great fun, even though I only know a few chords. Recently I plugged it in to my PC via my M-Audio Uno USB->MIDI interface, and soon was using Ableton Live to program drums in time with the microKORG’s arp.

I thought I’d experiment the music libraries available in python, and see if I could send notes to the synth via MIDI. Turns out that the M-Audio Uno is supported under Ubuntu, all you have to do is install the midisport-firmware package. With the help of pyrtmidi, a set of python wrappers around the C++ audio library rtmidi I was able to recieve MIDI signals in realtime from the microKORG, and send them in realtime also. With the help of this old midi file reader/writer library that I found posted to a python mailing list, I’ve made some progress in writing a simple MIDI file player that sends notes to the ‘KORG.