It was one of those moments that seem to punctuate the history of the technology industry. Ray Ozzie, the creator of the enormously successful Lotus Notes software package, was having a shower when suddenly it came to him: office workers who want to collaborate electronically should not have to upload all their files into a shared area on a central server, they should just make their files accessible in real time to others over any network – from the office LAN to the global Internet.

Ozzie did not have to jump out of the shower to record his ideas – or so the story goes – he simply reached for his waterproof pad. Later, it was all set down in a business plan and shortly after, Groove Networks, the first major software company based on peer-to-peer (P2P) technology, was born.

Four years later, in June 2001, Ozzie bumped into Steve Ballmer, the CEO of Microsoft, at a US west coast industry event. For Ozzie, it may have been an equally significant moment. During Groove’s four-year history, the company had at various times been one of the industry’s most talked about hot technology companies. But now the project was running into trouble. Groove’s products were barely out of development, money was tight, and few investors were willing to invest in speculative ventures.

Ballmer, however, was persuaded by Ozzie that the technology had great potential. The first version, Groove 1.0, had been launched in April, and while there are few real customers, it was getting corporate interest. More importantly, both men thought that P2P technology could help boost the function and competitive positioning of Microsoft’s .Net web services architecture. Projects were already underway to integrate Groove with Microsoft Office and the XP operating system. Microsoft took the bait and invested $51 million (€58.2m) for a 19.9% stake, valuing the company at $250 million (€285.8m).

The investment, finalised in October 2001, has helped to re-invigorate not just Groove, but a whole software market sub-sector – collaborative working. It is a sector that, dominated by IBM’s Lotus Domino, had lost its fizz since Lotus Notes first created the market for groupware in the early 1990s.

Groove uses P2P technology to enable group chat sessions, team working and document collaboration, with the promise of making inter/intra company communication cheaper, more effective and quicker. But it is not alone: several other young companies are active in this area (see Other peer-to-peer start-ups), many with lower profiles but higher revenues. And this wave of innovation is now spreading, both to the groupware giants and to related technologies such as portal software.

Groove, helped by the Microsoft investment, has been widely identified as one of the likely leaders in the emerging market. But that view may be simplistic – the company is still very early stage and many analysts are beginning to wonder if it will ever live up to its early promise.

After four years, for example, revenues are a long way from the levels that could justify the $250 million (€285.8m) valuation, and cash burn continues. It has only launched one major release of its software, and its presence outside US software circles is low. So far, it doesn’t have a single customer broadly deploying the software, and only 20 pilot customers in various stages of experimentation.

Against all this, the company argues that the potential is huge, the technology complex and, moreover, that it required other technologies to mature before it was likely to be accepted. GlaxoSmithKline is now piloting the software with a view to rolling it out to 10,000 users.

One problem for Groove is that there appears to be considerable scepticism about P2P software. The Gartner Group, for example, believes that only 30% of corporations will have even experimented with P2P applications by 2003, despite the fact that products have been on the market since 2000.

This is partly architectural. P2P products usually require that a program is installed on the client, running against the current fashion for ultra-thin browser-only clients. In Groove’s case, this is an 18MB file, which some analysts think is too large.

Using Groove, the only time the server is involved is when the computer running Groove connects to the Internet. It then registers with a ‘presence server’. Other users are then notified and each can interact. In this way groups can work as if they are in meetings, communicating over the network, sharing files, documents and data.

This model is very different to Groove’s precedents in groupware. Lotus Notes and Microsoft Exchange use a server-centric architecture. But some of the ideas are catching on: Ed Brill, Lotus’ worldwide manager for messaging software, recently said that Lotus is breaking up its monolithic code base into Java applets that can be run on local clients, or embedded in other applications.

Integration with enterprise applications is a key goal for Groove – although it has so far only signed up a few partners. Ideally, customers of, for example, SAP, would be able to stay within that environment while they work collaboratively – accessing, for example, instant messaging or watching colleagues’ documents as they change. At present, they must work in Groove and import the relevant documents and data from SAP.

Groove does have some partners, though, including high-end document management company Documentum; knowledge management and groupware firm Intraspect; collaborative design software specialist PTC; and search engine firm Autonomy. PTC is the only vendor to embed underlying Groove technology into its collaborative design suite and thus achieve the full integration required for true group working with Groove. The rest only link with the P2P functions via connectors.

Over the next year, analysts are likely to be watching Groove closely. At present, the company only makes money by distributing enhanced versions of its client software. However, this will change with the launch of three server applications as part of Groove 2.0 in April 2001: Relay Server, which adds store-and-forward message queuing; Management Server, which makes the deployment and management of the Groove client easier; and Enterprise Integration Server, which enhances interoperability with web-based applications from SAP, PeopleSoft, JD Edwards and others.

This may at last bring Groove the sales and recognition it needs. But if Groove’s revenues fail to point very sharply upwards, investors may conclude that its apparently powerful technology needs a home elsewhere.

The most obvious place, of course, is Microsoft – although the way it complements other key technologies, such as portals and knowledge management, could make it appealing to others.