Last week I posted a remark on Twitter:

Can't help thinking that the idea that Google Wave will replace email rather misses the point....

The first response to this echoed my view on this suggesting that the real nature of Wave is rather harder to explain or understand, and implying that people fall back on a frame of reference with which they are comfortable. It certainly looks as though Google have anticipated this and offered some easily digested marketing messages. However, I also saw responses which suggested that some people still seem to be missing the point. One response insisted that Wave would only be successful if it was ‘integrated’ with email. I must confess that I still don’t understand this - I can’t really imagine what impact an integration between Wave and email would really have.

It seems to me that Wave is an ambitious attempt to exploit the idea that one future for the Web lies in social networked activity clustered around shared artefacts. Such artefacts, often what we still call ‘documents’, have been given the useful label social objects. At the centre of a Wave is a social object, with a series of applied and recorded operational transforms. Wave would therefore seem to be primarily about collaboration, as opposed to email or IM which are primarily concerned with messaging. Another way of looking at this would be to suggest that Wave is 'object-centric', as opposed to email which is message oriented with a facility to attach auxiliary objects.

The idea that Wave would replace email seems to be suggesting that we won’t need apples anymore because now we have oranges. This is not to say that Wave might not better fit some use-cases currently served by email - such as the problematic mode of collaborative editing of documents by sharing copies sent as email attachments. But even as we adopt better software for collaboration, there's not much sign that we're giving up using email. I don't know about you, but my email inbox isn't getting any smaller just because I use Google Docs, IM, Twitter.... Email has been tested quite thoroughly now over a few years, and appears to work quite well for asynchronous messaging!
Wave uses XMPP as its underlying protocol which is both interesting and important, but it is also slightly misleading as it implies an important connection with 'instant messaging, which I think is illusory and unhelpful.

Wave is possible because the barrier of network latency is gradually being reduced. Real-time collaboration across the global network is now viable for many. Of course Wave is not the only game in town - other interesting approaches (mostly also using a variation on the pubsub paradigm) to the real-time Web, such as pubsubhubbub are being actively developed and experimented with. But Google Wave is important - because it's Google who are doing it. It will gain a lot of publicity, and will likely play its part in driving a culture change allowing real-time collaboration across the global network to 'go mainstream'. It should be remembered that Google's Gmail, the poster-child for Web-based email, is still significantly smaller in terms of users than Yahoo and Hotmail.

Because Wave offers APIs to developers and users out of the box, I think it is going to be difficult to say what shape this new offering from Google will take once a significant number of people are using it. The ability to federate Wave services could be significant in this respect.