The economics of networks implies that a product or a technology becomes more valuable to each new user the more widespread its use has become.

Both consumers and producers benefit from the ease of a single standard.

Frequently, a "tipping point" is reached when - as with Microsoft Windows - a single standard becomes so dominant that it is in more or less everyone's interest to use it.

Such standardisation can be benign, as with the universality of the VCR format in videos which provides a common platform for viewers and media companies alike.

But when the network technology is owned by a single company, as with Windows, the benefits of standardisation have to be traded off against the potential for that company to exploit its monopoly position.

Network effects are not new: literal networks like railway tracks or gas pipelines, in which competition is virtually impossible, have long been subject to heavy regulation or nationalisation.

But for regulators, the Microsoft situation is complicated further by the fact that in dynamic, fast-moving markets like computer technology, monopolists may use their market power in unusual ways.

A traditional monopolist in a market with a mature technology is often easy to spot: it will gain a large share of the market, try to stop other companies entering and then raise prices to maximise profits.

Critics of Microsoft, by contrast, object to too low prices - such as the cheap or free "bundled" or "tied" related products such as browsers and media players - which they say aim to protect the long-run monopoly profits in the operating system that underlies them.

While some economists believe regulators should err on the side of giving companies free rein, on the grounds that even tech monopolists face incentives to continue innovating rapidly, some think the extraordinary power of tech standards means that regulators should err on the side of restraining companies.

Or, one side is frightened of killing the goose that lays the golden eggs, the other of letting the fox run riot in the hen coop.

Some economists argue that the threat of predatory pricing could kill markets not yet born and technologies not yet invented. Future Googles and Adobes might not bother developing innovative new products if they fear being undercut by Microsoft aiming to protect its near-monopoly in operating systems.

"Normally there is a first-mover advantage in these tech markets but Microsoft may have a second-mover advantage which has the effect of stifling innovation," says Jay Pil Choi, an economist at Michigan State University.

Defenders of Microsoft say that no monopoly can be safe for long in the fast- moving tech field, and that the waves of "creative destruction" posited by Joseph Schumpeter, the Austrian economist of the mid-20th century, are an ever-present challenge.

Nicholas Economides, an academic at New York University's Stern School of Business, thinks that courts and regulators have interpreted the economic arguments to take an overly aggressive antitrust stance.

"The argument that network effects can result in anti-competitive behaviour has been understood," he says. "But the argument that network effects can lead to Schumpeterian competition where a dominant firm is suddenly overthrown by another has not."

And in practice, a tech company undertaking predatory pricing to deter future competition can look remarkably like one using aggressive but legitimate "penetration pricing" against its current competitors to win the battle to become the industry standard.

Simple rules like preventing companies pricing below the marginal cost of their production will confuse the two.

So far, courts have reached only tentative and ad hoc conclusions about the economics.

With masterly understatement, the appeals court in the US Microsoft case concluded: "There is no consensus among commentators on the question of whether, and to what extent, current monopolisation doctrine should be amended to account for competition in technologically dynamic markets characterised by network effects."

The court set aside the lower court's decision that tying Windows to other Microsoft products violated antitrust law - also part of the European Commission's case - but warned against extrapolating a general rule from this decision.

The European appeals courts will have the chance to build on the scant case law that exists. But they are highly unlikely to be able to set rules that definitively settle the extraordinarily complex questions of regulating dynamic network industries.