Just come back from a great talk by Brian Levy, HP’s Chief Technology Office for Communications & Media. Amongst many useful insights he offered were

• there is a very real possibility that current growth in video-based internet traffic will cripple internet capacity [driven e.g. by bit-torrent and similar, as used by BBC iPlayer – basically uses loads of our PCs to help supply content on to others, I think] 
• … and it’s not clear who will get paid how for resolving this – ISPs are being pushed into adding capacity at their own cost, with little or no revenue increase to pay for it
• very much web-based stuff just doesn’t work very well, relative to ‘acceptable quality levels’ – everyone is pretty tolerant, but things like MMS messaging, VOIP, video-streaming, mobile web-browsing are just not reliable enough
• … which is because too many things, from too many providers, all have to work perfectly, and usually one of them will let you down [personally, I find the amount of super-clever things that do work is just amazing!]
• we are very close indeed to an integrated ‘3-box’ solution, in which the cell-phone, TV and PC become completely integrated … which makes all kinds of even more clever solutions possible [He gave lots of examples, but just one is advertising content in programming downloaded to cell-phones of iPhone screen quality that are sensitive to both where the user is, and what time it is.

Amongst many simple strategy tips I spotted is one directly applicable to how we use dynamics … When assessing what growth may be possible, we need a clear view of the potential opportunity. Traditionally, that has been some ‘segment’ of customers, identified by social group, geography or whatever. Brian reminded me that new, virtual ‘communities’ are becoming more important to many people than the geographic, professional or other groups we have been used to for so long. So many organisations will need to be thinking about new such communities who may be their most natural target, and/or about the falling importance of segmentations they have long relied on.

PS – this is probably very obvious to many who work at the forefront of these issues, but others may not be alert to the issue.