O2 is apparently considering "adopting the i-mode technology standard". Calling I-mode a set of technical standards kind of misses the point: it's a business model based on generous revenue shares and an expression of an ecosystem that allows DoCoMo to define standards, as well as those standards themselves (most of which are based on open Internet technologies). But that's by the by...
An alliance with NTT DoCoMo would undoubtedly bring some learning to O2 on how to produce popular services... but what is this worth? I-mode has been documented by its originators and analysed to death by outsiders over the last few years, and many of its strengths are completely independent of any technology: they're commercial or procedural. O2 have made a significant investment in WAP, and (like all European operators) have a large installed base of WAP users who they need to support. They'll also have pre-existing infrastructure internally (e.g. billing systems) which they can't just junk.
So O2 can't bring I-mode over without supporting additional technical standards; and they're unlikely to copy the revenue model wholesale (by offering a 91/9 revenue share in favour of content owners)... so err what do they get, other than a chance to bask in the reflected glory of DoCoMo? It's not a wealth of content - only a miniscule percentage of Japanese I-mode services are in English and appropriate for UK distribution; and it's not better handsets either: they're here already.
Is I-mode really magic pixie dust with a secret ingredient? Or is it a set of practices which, like most really great ideas, are obvious in hindsight?