Yes, it’s nearly time for the next recap. Tempus f. and all that jazz, but I hope to complete the recap before I go on vacation with the family in early August. So I thought I’d share a few things buzzing around inside my head, see what you think.
The first theme is about client-side and server-side software and how they’re evolving. More and more, as web services and SOA and virtualisation become part of our lives, we get the opportunity to look at what happens at client and server level with a slightly different perspective. Some old problems go away, and new ones emerge.
What I’m mulling over is this: As client installs become thinner and smaller from an end-user application perspective, we may get two significant benefits. One, we can make real progress on (client) platform and device agnosticism, with sharply reduced rollout costs and longer device lives, and even lower maintenance costs; and two, we have this by-product of real diversity at the (client) device level, a diversity that acts as a natural brake on virus propagation.
The second theme is about caching versus long-tail. A lot of the arguments about net neutrality tend to focus on “Someone must pay for all the upgrades we must do, in order to let all of you download all these videos that are going to clog up the tubes and make sure Senator Stevens never receives his internet“, while the real arguments may be about something else altogether: See Doc’s recent post on the subject, and Gordon’s follow-up.
What I’m mulling over is this: There’s a lot of talk about some form of local “neighbourhood” caching to solve the “problem” of video downloads (while happily skipping over the forced asymmetry with respect to uploads); I’ve even heard tell of trucks being deployed as mobile mega-caches. But cache what? I thought there was a very long tail of things people watched, as Chris Anderson quite clearly demonstrated. The caching discussions I’ve seen all tend to believe that the concept of “hits” will remain, which obviously makes caching useful. But I can’t reconcile the long-tail argument with the cache argument. [That’s one more reason for me to stay Confused].
The third theme is that of customer information versus DRM. Dick Hardt et al have done wonders in educating all of us about “It’s the customer’s data, stupid”. And Doc and Steve Gillmor et al have done similar wonders in getting us to understand attention and intention. So we’ve got to grips with the idea that the customer owns his/her intentions, purchasing behaviour, preferences, the lot.
What I’m mulling over is this: What happens to “content” and DRM hawks if the customer says no, you can’t have my data, it’s an invasion of my privacy? Aren’t those behaviours and profiles and clickstreams worth much more (to the content “owner”) than the apparent loss of revenue as a result of no DRM? What would the content “owners” do if someone suddenly turned the tap off. A sort of You Can’t Mine My Data Because the Data’s Mine.
The fourth theme is about synthetic worlds and their value to enterprises, particularly if Second Life met Tivo. You’ve already seen me get started on some aspects of this.
Blogging is provisional, it’s a sharing of nascent thoughts and ideas and kernels and snowballs, trying to see what happens if enough eyeballs see the thoughts and ideas. So, before I do the next recap, I wanted to get your opinions on these themes, see where I’m going wrong.