My thanks to those of you who commented, tweeted or wrote to me about my post on the customer perspective yesterday.
Some of the questions raised were such that I felt a follow-up post was of value, so here goes. I’ve tried to structure it as a small number of points that clarify and simplify what was stated yesterday, while still remaining readable as a stand-alone post.
1. Subscriber-driven: The real shift taking place is one of control, passing from me-the-publisher to me-the-subscriber. In future the smart money is on environments that empower me-the-subscriber, and on environments that help me-the-publisher offer a better service to me-the-subscriber. For example, a blog platform that allows me-the-publisher to offer me-the-subscriber the ability to subscribe to the blog feed by tag. So someone should be able to say “I only want JP’s posts on cricket and on cooking”. Or, as is far more likely, “I want all of JP’s posts except for the ones he does on cricket and on cooking.” The ability to get granular control of the feed will become more important. A note of caution: we shouldn’t waste our time trying to standardise tags, it’s the sort of thing we’ve all wasted our time doing before.
2. General different from specific: As a result of the shift, we will land up with two types of habitats that me-the-subscriber will frequent. One is a general habitat, a meta-habitat for aggregating everything. The second is a topic- or tag-specific habitat. Aggregation will take place in both habitats, but the type and nature of the aggregation will be different. Twitter and FriendFeed are general habitats. Last.fm and Flickr and blip and dopplr and even seesmic are specific habitats. The habitat of choice will determine the etiquette required. You do not fill a general habitat with an overload of specific information. You cannot write on cricket in the Times as if you are writing in The Cricketer, the depth and frequency have to suit the habitat. It’s called being courteous to your subscribers.
3. Social objects need graphic equalisers: Facebook learnt this lesson soon after the news feed was released, but it’s worth repeating. General habitats need sliders, need mixing desks, need graphic equalisers, in order to allow me-the-subscriber to manage the balance of what’s coming in. This is despite the fact that I choose whom and what to subscribe to. Me-the-subscriber needs the sliders, the graphic equalisers, because me-the-publisher is sometimes inconsiderate and makes too much noise of a particular type. Like telling me where he is every 10 minutes. I’m interested in where he is, but not that much. Like telling me when he turns the page while reading a book. Like telling me about each and every song he is listening to. What I want to know is where I can find out the detail if I want it. In the meantime, I only want him to give me a sample of what he’s doing, show some judgment.
4. Me-the-publisher needs tools in the specific habitats that allow this sampling to take place. So for example I need to be able to say, in a general habitat like Twitter: “Hey guys, I’m over at blip.fm DJing as http://blip.fm/jobsworth if anyone cares” And then people who are interested can do something about it. And as me-the-publisher I can choose to send over a tweet for every tenth song I play. And you, as me-the-subscriber, can choose to change that frequency up or down. And the publishing/subscribing platforms will have to deliver as needed.
5. Signals become more important: The general habitats become aggregators of aggregators, where as publishers we signal our availability to subscribers. Come join me I’m cooking at. Come join me I’m listening at. Come join me I’m watching at. The choice of participation is always the subscriber’s. [Image courtesy ecoustics forum].
6. Visualisation becomes even more important, as do the tools used for visualisation. [PS top image above attributed to manyeyes and IBM; bottom image attributed to www.wordle.net] As subscribers, we will have better and better tools to convert firehoses into capillaries. That is really the only way we can avoid the potential overload risk of aggregation.
7. Mobile devices rule. There will be a number of general habitats, based on subscriber preferences. Some (most?) will be designed natively for the mobile device. As with general habitats, specific habitats will not be monopolies. People will choose one over the other, but retain the freedom to move from one to the other. Any attempt at lock-in will either fail (by being subverted) or atrophy to death (because it won’t be adopted).
8. Unsubscribe: We will see a major rise in usage of “unsubscribe” facilities. Clay Shirky famously said that wikis worked because the cost of repair is kept as low as the cost of damage. The same is true of how we subscribe to, and unsubscribe from, people and feeds. The cost of unsubscribing can and will drop. It must drop. As will the cost of unfriending, unfollowing, unwhatever-ing.
None of these ideas is new. All I’ve tried to do is to provide some context and some narrative to my idle ramblings about what it means to move to a subscriber driven world. Comments as usual welcome.
I’d love the ability to turn up the granularity on some people in my social sphere, and down on others. I’m keen to see what music my IRL friends are listening to but I only want to know the big news (Jobs, marriage, house moves etc.) for some of my extended family.
However, I worry that having this would cause me to miss some of the wonderful little connections that a chance tweet or status update can give.
For example I am FB friends with a lot of guys I used to work with when I was a student. I’ve not seen or spoken to many of these guys for over 10 years but it’s nice to see when they have kids, or go on holiday or the like. However there have been updates that have sparked more in-depth engagement and I may not have got these had I dialled down the detail on these guys.
I wonder if there is a way to set a comfortable level of detail but still keep the serendipitous sparks of connection.
I like the equalizer idea, particularly when applied to given topics. One particular peeve of mine is that it’s so hard to keep track of blog posts that I comment on – so that I can follow up on subsequent comments. I touched upon this in the ‘interest feed’*, but still haven’t figured out an elegant solution. I guess that what I want my aggregator to do is a sort of micro-subscription to every post that I comment on, and once again we hit the context change issue that’s inherent in replies.
Looking back at my original post I suggested that the solution would be server side and identity centric, but if we turn things around to be subscriber focussed then once again we’re faced with a need for some client side magic (in JavaScript [using GreaseMonkey or similar?]).
* http://thestateofme.wordpress.com/2008/01/29/the-interest-feed/
James, the solution may be to have *two* general habitats. Like Twitter *and* Friendfeed. You go to one on a directed basis, and you go to another just to surf.
Another possibility is to have the ability to “stumble” in your own aggregator, have the ability to surf your own space much like you would with StumbleUpon and the Web.
Chris, I think our concepts of client side and server side are being challenged anyway by at least two sets of happenings, the explosion in mobile devices and the growth in RIA development environments, especially Flash/Flex.
JP – I think there is another challenge that isnt remotely technical and that is THE WAY THINGS ARE TODAY.
The commentators above and below on both posts (complete genius by the way), are by people like ‘us’.
You know what happens when you run this stuff by the people who control most of the commercial companies.
Technologically there is a major challenge – but I believe the over-arching issue is emotional.
I mean that in the very literal sense of each and every word.
There are an almost endless variety of ways to technically deal with with filtering both content and social connections. What is needed is for really intuitive apps and aggregating interfaces to be built and tested.
The best of these will be the ones that users form an emotional bond with, the app becomes the window to their favorite things. They get to see, and interact with everything they want to, easily and pleasurably. A highly customisable stack of simple controls for the equalisation. This also creates a unique signature of preference data.
I believe that a combination of this data, and that from search, transaction and location will form it’s own currency on mobile. I’m looking forward to the day I can advocate a music artist and get credit to purchase more music from the same publishing channel for doing so. This is something my musical peers and myself have done informally for a very long time. Mobile will vastly expand these peer networks.
This will go way beyond subscription once customers begin to understand the power they hold, and are rewarded with true value for making their preferences publicly available.
JMac, Dominic, as you say, the emotional aspect will get important, with all its bag and baggage. For some it will be inertia and resistance to change. For some it will be the user experience as we learn about what is needed.
But the payoffs are great, as true personalisation emerges, with the empowerment and the choice-making capacity to go with it.
I don’t particularly like ads, but they can fulfil a useful purpose in aiding discovery. What I deeply dislike is the cowpath-paving from analogue to digital ads, where people are settling for things like 1% clickthrough conversions to sales. What nonsense. We should have real quality personalisation of ads, with 75% click throughs.
The graphic equalizer isn’t a bad metaphor, but the cybernetic term for this is “attenuation”. This notion is very important within Stafford Beer’s Viable Systems Model, for example.
I see this as a question of complex (postmodern) identity.
“JP-talking-seriously-about-social-networking” is not the same person as “JP-talking-humorously-about-cricket”. So I may want to subscribe to one person and not the other.
The telecoms analyst Martin Geddes wants to speak to his wife if and only if she isn’t putting the baby to sleep. I think this falls under the same general class of requirements.
See my blog at http://rvsoapbox.blogspot.com/2008/12/subscribe-and-follow-towards-complex.html
Well said JP, it’s much clearer to me now that you’ve laid out the basic message of the last post into the 8 “simple” points, though I think there is heavy overlap between them:
Subscriber-driven publishing at more granular levels (topic-led and person-led) would effectively mean less need to separate the general content from specific, there are twitterfriends who are cricketfriends and others who are cookingfriends, I’d prefer to read the posts on their subjects of expertise. But even with the best graphic equalisers and result visualisation tools, there is a corresponding need for tailored data-mining algorithms – at its simplest level perhaps just correlation matching that attributes scores to various combinations of friend criteria and content criteria.
Actually, that rambling paragraph probably just shows you were right to neatly split the general principles into those 8 areas. The point about advertising is important too, a “majority” clickthru rate should be the target measure that shows you’ve really added value to the average ad viewer.