The author should disclose the age of his test equipment, it makes a big difference. The methods the author used will only works vs equipment at least 2 years old.
Since the initial panic about WEP in summer 2001, 802.11 manufacturers made changes to their firmware. Fast WEP cracking comes from capturing enough "weak IV" frames, something that occurs randomly. Manufactures have apparently altered the algorythm to not use any weak IVs -- this technology is sometimes known as WEP-plus. For example, Orinocco interfaces with firmware older than fall of 2001 can be cracked. Change the Orinocco firmware to an early 2002 version, Airsnort and Kismet never see weak IVs, thus can not crack the WEP key. Note that either side of a WLAN connection (access points or client interfaces) can send weak IVs if they are using old interfaces.
My impression is that manufacturers have implemented WEP Plus fixes in equipment manufactured since late 2001/early 2002. Typically, old equipment can be upgraded to newer firmware. At a large trade show early this year I fired up kismet and Airsnort. Hundreds of WLAN interfaces were visible, but I collected maybe 2 weak IV frames. I was manning a both at this show, and run the tools all day for 2 days.
Articles like this were important 2 years ago, but really they distort the problem today. Keep up with the firmware updates and you are reasonably secure. I have some many people insisting WEP can can be broken in 15 minutes, it drives me nuts.
AND ... Even if you are attacking a 2-year old WLAN, collecting millions of frames can take many days depending on the traffic load! You can't do a ping flood until you have that WEP key.