Most of the Internet has been getting better during the past few years. In much of the world, it is now good enough for all but the most demanding applications. This improvement hasn't depended on ISPs implementing fancy Quality of Service mechanisms. Paradoxically, some ISPs might see this news as a threat to future financial health.
There are a number of research groups currently studying Internet performance, although it still is not easy to get good data, as KC Claffy detailed in one of her talks.
Claffy is the main investigator of Cooperative Association for Internet Data Analysis (CAIDA).
Members of the physics community are also studying the Internet. The International Committee on Future Accelerators has had working groups thinking about Internet performance since at least 1997.
One such group was formed in 2002 and published a paper on the state of Internet performance in January (DocFinder: 6529). I'm not sure why the physicists are studying Internet performance, unless it's to figure out if they can use the Internet to deliver the very large data sets that their experiments produce. In any case, their work is very good.
Their latest report mostly deals with packet loss in data transmissions, round-trip times and data throughput between the Stanford Linear Accelerator and global testing points.
The countries where the points are located represent 78 per cent of its population and 99 per cent of the world's Internet users. The test results show that by the end of 2003, the packet loss rate to countries with 77 per cent of the world's population was low enough that VoIP would work with good or acceptable quality. This is up from 48.8 per cent in 2001. One example is reliability within the US - packet loss rate fell from more than 10 per cent in January 1995 to less than 0.5 per cent in January 2004.
Round-trip times have fallen and data throughput has increased. These improvements have been in the standard Internet service.
As Vonage and other overlay-VoIP services have shown, VoIP works for much of the world most of the time. You don't have to pay the carriers extra for better service to make VoIP work well enough to be very useful. This fact might be a real threat to the financial well-being of carriers that plan to make more money by charging extra for better quality service - and that includes most of the traditional telcos.
These carriers will be forced to try to make money selling a commodity service, unless more of them purposefully try to mess up their networks to disrupt VoIP, as Vonage has claimed some already do.
These carriers could be in for a rough ride.
Disclaimer: Harvard claims not to be in the commodity service business but has not expressed an opinion on carriers that may be forced to be so - thus the above is my own opinion.