Newer Media

Jorn’s been goaded enough to throw some opinion around. Not just his 5 word opinion, but a really irritated multiple-paragraphs of opinion.

And so we come full circle. Jorn Barger, coiner of the term ‘blog’, who in retrospect was really writing a linkblog the whole time (though we lacked the context to give it such a relative label), joins the linkblog/linklog fray. Before taking a swipe at boingboing, he rightly points out that the best linklog tool is del.icio.us. And having watched Jorn’s site over the years, growing from manual HTML maintenance, to blogger, via third party RSS scrape-and-bake services, his current del.icio.us-based solution is perfect.

Abstract: Closed social network and web utility sites represent the old way of using the web. Today’s web is more agile and flexible, building on open syndication, RSS and APIs to allow the creation of new platforms and functionality which not just substitute but exceed the current offerings, while addressing the unspoken issues of data ownership and transparency.

Linkblogging is just the tip of the iceberg

The web has changed. It’s not where you read blogs any more - that happens in your RSS reader of choice, whether web-based or not. You use the web to run applications or access services (the aforementioned RSS reader, forum systems, your social networking tools, or go shopping). Every blog I visit I do through RSS, and rarely do I see the actual, original site (something I am increasingly happy for; with google’s insidious text advertising, and federated media’s bloatserv technology, I resent 200kb of adverts for a paragraph of copy - did no-one tell these guys that a 400×35 banner is worth 1000 words, not 20?). As Google reader, embedded RSS readers (Windows Mobile, iPhones, Nabaztag rabbits, sidebar widgets/gadgets/newyorkgets, you like what I did there?) and every modern browser and mail client pushes more and more people into adopting RSS, whether or not they realise they are doing so, this will become the norm. People don’t even bother crafting layouts any more; in one day, I saw approximately 4 corporate blogs using exactly the same off-the-shelf theme; it’s the content, it’s google presenting your text in a white-on-black, it’s Firefox’s chrome, it’s Outlook - this is your blog’s theme most of the time. Maybe you did spend a lot of time, or money, making a wild theme, and power to you - but ultimately, I’m there for your content, your output.

Which reminds me - I had some direction for all this. Del.icio.us is a superb linkblogging tool. It has organisation features (tags … and bundles, if anyone uses them), social and peer research and validation tools (30 other people bookmarked this site, but no-one else described it as ‘arousing?’) and a slew of publication tools and access clients. It does a single thing, and does it exceptionally well. It manages your bookmarks from around the web, and lets you share them - it linkblogs for you. But what’s this?

The Daily Mail, and I pick (on) them just because they’re on the top of my mental stack, make their own linkblog tools. You click a little blue button (blue - the spectral opposite to RSS orange) and the site remembers the article for you. The memory, however, is trapped inside the Daily Mail site. I have to go there, as me, and log in, to be granted access to these bookmarks. They can’t be shared. They can’t be backed-up, exported, consumed by other services, or subscribed to. They are a dead end; a metaphor for human memory, intangible, personal, private. Tears in the rain which expire when the account, the fleeting cookie, or computer dies.

But is this really so different from current darlings of the social media word, Facebook and MySpace? They generate (or rather, solicit) reams of experiential data; who knows who else, what they are doing, and what real or imaginary affiliations they wish to present. All of this data is then consumed internally, munged about, and fed back out to users through pages on either site. Exhibit A may help clarify things.

How Facebook Eats Your Data

This is an approximation of how Facebook plays in the social data world. We’ll use a few standard protagonists here, namely

• Alice
• Bob
• Clive

who visit the sites

1. Social Platform Du Jour
2. Twitter
3. Other Service

Clive is friends with Alice and Bob (they’re all into cryptography). They’re all signed up to services 2 and 3, and don’t mind sharing data between them.

Facebook operates a walled garden of user data. Alice goes to Facebook and says something, as does Bob. They leave a message on Twitter, and also do something with Other Service. When Clive wants to see what they’ve been up to, he has to visit the Facebook site, log in, and check the feed - it’s centrally owned by Facebook, who don’t provide an aggregate outbound RSS feed. This is the Daily Mail’s blue icon approach to gathering user attention and activity - it’s ringfenced and shared, but only on their terms, through their site.

The orange strip describes what Clive can see of his friends in this scheme. He has to be on Social Platform Du Jour’s site, he can see what Alice and Bob have said and done on that site, which may include activity from Twitter and Other Service, as long as they have a service presence that’s been built by those provider and hosted there. He can’t grab it from anywhere else - he absolutely has to be on that site.

Decouple behaviour data from access

The true power of linkblogging (as described by Barger) lies in service agnosticism, of not relying on a single central path to access your data. Here’s where my interpretation of linkblog expands on Barger’s, to the point where any generator of a consumable feed is a linkblog (ical entries, flickr posts, and tweets all coming under my extended wooly definition). Here come our friends the protagonists, this time under Exhibit B, the new model swarmy.

This time, there’s no central location where data is held. Twitter and Other Service are accessed independently, without the middleman of Social site Du Jour. I can tweet from my phone and operate Other Service through carrier pigeon, if they support it. They both export platform-neutral RSS, so I don’t care. If I wanted, I could subscribe to Alice and Bob’s RSS activity feeds from both sites, but when my friend pool extends to the rest of the alphabet, it starts getting unwieldy. When my alphabetical friends sign up to 20 services each, it’s definitely overload.

This is where the next wave of social network sites steps in. I said before we had the same protagonists; actually, we don’t. In Exhibit B, Site 1, the Social Network du Jour is a jour later, and is a neutral intermediary. Rather than try to own content as our first generation social network sites have done, they aggregate and share that data out. In this scenario, Clive has told his social network provider that he know Alice and Bob. Bob has already told the site that he uses Twitter, and associated his public feed URL with his account. Bob does the same, and adds his Other Service activity feed to his account too. Should Alice tweet, our new social site finds out - it pulls in the data from external services, rather than simply owning the data, as our older social sites do.

Clive wants to know what Bob and Alice are up to. Rather than subscribe to each friend’s individual feed, he finds an agnostic social network site that Alice and Bob have already registered with. Because the site is subscribing to these other friend’s feeds, he only needs to subscribe to one new feed (or linkblog)- the aggregate feed provided by site 1 in Exhibit B. Alice and Bob can decide what they want to share, adding and removing services as they see fit. No-one has ringfenced the content, and should the social site become disagreeable, each user can migrate away and source their friend’s activity through alternative services. Everything is open, transparent, and running free in the cloud, to mix our river-of-news and internet-representation metaphors.

These new social sites are more like traditional RSS aggregators, although operating at a different layer in the aggregation stack.

People join sites such as flickr and del.icio.us, or create calendars, and go about their general business on these sites, generating domain activity. Different types of domain activity are collated together by sites (commonly known as ‘linkblogs’, such as tumblr.com) to generate live feeds for individual entities, which are usually people, but may be abstractions of some kind (such as a group of people, or search results). These groups can then be gathered together further up the stack in a FaceBook-style social network, a single feed which describes the collective activity of a number of people and their interactions. This is ultimately presented to the consumer, you, by an RSS reader (Feeddemon, Google Reader, or any of an ever expanding range of RSS consumers and output devices).

Of course, in this model, it’s still possible to bypass each level of collation; these upward flows are shown by curved, grey lines in the above image, and it’s what people have been doing for years with RSS, before aggregation services targeting different points between the bottom and top of the stack appeared. But the real power of this approach is how transparent and standardised the act of social networking and observing has become. The tools which have appeared to fill this space have only existed a short while; while they could have been created 18 months ago using Yahoo Pipes, the ecosystem has significantly geared up towards simplifying the functionality to bypass the ring-fenced social networks.

The immediate issues are of authorisation and standardisation. Authorisation is to do with who you allow to see what; if Alice doesn’t mind Clive seeing her pictures, but doesn’t want Bob to, how can she manage this when her aggregate entity feed doesn’t allow this granularity? This doesn’t seem much of an issue at the moment - pretty much anything placed on the web ends up indexed by the all-seeing G, but when these practices are adopted into mainstream business (and even social life), there will be a requirement for these levels of control. It won’t be possible to place a gatekeeper at each link in the chain - for one, the maintenance will grow exponentially with each new service or user added. At the social level, people will learn not to care so much; when you’re online, much of your life is laid bare and indexed anyway. But corporations will expect secured services behind passwords and authorisation, and each requirement that is introduced breaks the all-or-nothing sharing up the chain. It may be that aggregation services will arise that can be tuned to support higher levels of granularity, with public key authorisation at different endpoint allowing peer-to-peer permission management, and each layer sharing signed and encrypted RSS items.

Standardisation is playing out, either right now, or at the time of writing, depending on when you read this. Each ‘linkblog’ and ’social aggregator’ handles things differently; there’s some crossover between them, and Darwinian software selection is deciding which features are best (until Google can implement one offering the best fit).

The question of revenue and the business model is a different issue. With RSS, it’s more subtle than even Google’s text-only advertising tucked away in a margin. There is a huge amount of intelligence and insight offered by user behaviour and movements discernable from the inbound data, more insightful than even Google’s simplistic keyword sales market. The ethics and realisation of profitability and permission is left for another post, another day. (I suspect once the VC cash has gone, many of these sites will go the way of the webvan).

Appendix - Relevant Services

Domain-specific

• del.icio.us - bookmarking (also, social sharing and discovery)
• vimeo - video posting and subscription
• YouTube - video posting and subscription
• last.fm - playlist capture (and sharing)
• flickr - image hosting (also, social networking and discussion)
• twitter - ubiquitous microblogging (with publishing)
• Google Calendar - calendaring (with sharing)
• Plurk - twitter with a richer interface (a cruel summary)
• tumblr - microblogging
• Blogger, WordPress, TypePad - (macro)blogging?
• fire eagle - geographic location publishing
• Dopplr - geographic location publishing (with social networks)

Entity Activity

These services combine multiple domain-specific RSS feeds into a feed that represents a single person - or entity.

• soup
• Pownce

Social network aggregation

These sites make it possible to combine multiple individual profiles into a single feed.

• friendfeed - adding discussions to imported activity
• Jaiku
• LinkedIn - concentrates only on networks, ignoring activity
• Plaxo pulse

Miscellaneous

• Yahoo! Pipes - data manipulation (combine arbitrary feeds through visual programming)
• ping.fm - centrally updates a number of microblogging services, with event-driven API hooks