Don't laugh at someone because he disagrees with you--that is puerile in the extreme. It's a far better idea to take a look at how he might be persuaded to agree.
This guy obviously did his homework, taking a look at the FSF and OSI websites. He recognises that OS/FS is not a new idea and has `been around for a long time.' He admits that Open Source works `if a
sufficient number of programmers are willing to donate their time improving software,' although he judges this as doubtful. So the first order of business is to demonstrate that it is not doubtful, but likely. Exhibits A and B: Linux and the GNU tools which atop it. The former provides an OS in many ways better than some commercial offerings, and the latter provides software which is much better than the commercial alternatives.
He holds the `there's no-one to sue' line, and states that a multitude of evolving products makes it difficult for `consumers...[to] identify the products they need.' The counter to this is to argue that software is not really so much a product as it is a method of dealing with data. No-one argues that every car manufacturer should re-invent a measurement system--the sane ones use the imperial system, and the foolish ones the metric:-) Seriously, software is more a matter of public standards on data formats and agreed-upon ways of dealing with data formatted therewith than it is something akin to a gear, cog or wrench.
He seems to believe that opening the source leads to a chaotic number of differing versions (forks, in our parlance) of software. We can easily counter this argument with the fact that forks don't happen very often. Indeed, it's actually the free market at work: anyone is free to fork at any time, but it is not generally in one's best interest to do so. And the multitude of versions actually encourages adherence to standards, and adherence that is discouraged in the proprietary market.
He misunderstands our view of ownership. Some of us are no doubt lunatic-fringe socialists believing owenrship is theft &c. Others of us, though, view ownership as a good thing, but believe that it has its limits. How can one own an idea? One can only be given power over it by the state--unlike property, and idea has no boundaries and cannot itself be patrolled. It makes perfect sense for a hardware vendor to open-source its drivers; it makes its money from said drivers, and any improvements to them (such as might be made by the legions of OS/FS programmers) increase the value of its hardware, for free.
The best argument is in practice. In practice OS/FS works--programmers do contribute their work, and we do advance the state-of-the-art. We are in support of openess and standards. Obfuscation hurts innovation--it is as though Fords used a system based on 17.456 inches to make the parts for their autos, thus tying the consumer into them as the sole provider of parts. Mechanics would not be able to work on several different brands at once. We've seen what the prevalence of two competing standards does in the car repair market: normal mechanics and European-auto mechanics. The same Balkanisation occurs with software, but due to the incredibly low barriers to entry (anyone can make a standard cheaply, and he can change it with each version of his software) it has become far worse.
As a last thought, he argues against forcing Microsoft to support standards. This can easily be counter-argued by the simple existence of the Internet. Due to a common set of protocols (SMTP, HTTP, FTP, BGP &c.) on top of a common set of protocols (TCP, IP, Ethernet &c.), the Internet has achieved much greater heights of value to the end-consumer than any of its closed predecessors. AOL, Compuserve, Genie and Prodigy at the height of their power never equalled the magnificence of what OS/FS has created. We have demonstrated our superiority.