I’d marked the article for comment but just haven’t had time (and still don’t) to really do a response justice so here are a few preliminary thoughts.

Graphs have been around an awfully long time, like hundreds of years but their use in computer science really was a child of the 80s & 90s.  Their current framing is in the social context, a map of all the relationships one has within a social network for example.

However as Chris points out this is limiting.

The thing is ALL my “social” relationships across all contexts, sites, online and offline are my social graph.  So my Twitter followers/following may indeed represent the portion of my social graph that is interests however that context spreads across any number of other properties.

And this is where I think Chris’ thinking butts heads with my own – but only slightly..

Chris talks about “the rising importance of other types of graphs” and gives examples of graphs for Taste, Financial Trust, Endorsement, Local(e).

To me these are all the same graph.  It’s just they represent differing (social) relationship types.  If each were to be represented on separate graphs then the power of graphs in general would be lost – at least without serious jumps forward in semantic technology that is.

What do I mean?  Well, take me for example;

My relationships to my family take on many types (father, husband, son, brother) but they also occur in other contexts such as “financially who do I trust” or “who am I local to.”

To break these into separate graphs would mask the true picture of me.  It would be an administrative nightmare for me to maintain relationships this complex across all and every property in a way meaningful to others.

Far better in my head is an overarching graph that contains all my relationships (this does not mean centralisation of everything as distributed graphs are fine) marked up appropriately with context.

That’s all I have time to write now but this does need more thought.

BUT (and I really shouldn’t have been that surprised given the Facebook tie in) that the default settings for the installation are to share anything and everything from installation.

So anytime you create a new playlist it gets shared.  Unless of course you go and manually disable automatic updates.

Given all the flak Google got over Buzz and it’s presumptions on automatically opting people in, given all the grief Facebook gets for it’s over sharing it is such a shame to see Spotify falling into such a simple trap.

Oh and don’t even get me started on seeing adverts re-appear on my desktop version – I am a paid up member of the premium subscribers gang which was supposed to be non-advertising!

UPDATE: It gets worse.  After a few minutes use adverts are popups, and also taking over other areas in the UI.  On a netbook this is unacceptable as pace given over to my music is already squeezed and now it’s even worse.  Also audio adverts have re-emerged.  Not happy at all

Bosses are now using the popular social networking site as a tool to double check how likely it would be that their new worker would take a sick day for being hung-over or on drugs the night before.

And job seekers were being found out for lying about their qualifications, with employers checking their Facebook pages to see if their online details matched their resume.

No great surprises there.  After all background checks, references and such have been the bread and butter of the Human Resource industry for yonks and let’s be truthful here; business and HR in particular has never been great advocates of treating people as people.  Liri Anderson highlights some of the absurd thinking in her post here.

But the article had me thinking, especially in light of Mark Zuckerberg’s recent Crunchie Awards statements on privacy and sharing.  With open sharing of very personal information rapidly becoming “normal” (at least within a certain and growing portion of society) businesses are being offered up a far greater insight into who people really are, their true identity.

I recently spent a day being psychometrically tested, a practice I have had little respect for in the past.  But this time it was different.  After an hour of online tests prior to even leaving home, I spent the best part of 9 hours being subjected to a battery of tests, exams, questionings all culminating in a fairly probing interview with an industrial psychologist.

Throughout the whole process I was very conscious of the various (seeming) inconsistencies in my responses, my body language, volume, level of language – the whole performance. The psychologist then blew me away by not only articulating back to me all of those traits but painted a picture of me that was so close to my own view that I could not fail to be impressed.

And of course the whole exercise is designed to see through performance, misdirection and untruths.

With the rapid increase in sharing of personal information HR practioners now have the ability to undertake much of the due diligence that would be accurately be shown up by the above process themselves, in-house with no context, response or even the applicants knowledge.

I’m not going to argue the rights and wrongs of this surreptitious behaviour (although I give a nod towards Deep Packet Inspection) but I do want to pose a couple of points;

1) Are we going to see third party agencies now remotely scanning peoples online behaviour in order to offer up a “professional” opinion of that candidate based on nothing more than what is actually shared as opposed to that which is not expressed?  Where will the oversight come from and can these businesses build a credible model?

2) In Zuckerberg’s ideal world we all share more and share more openly.  Given this scenario when will the tipping point come where candidates are equally exposed and deemed inappropriate; what then?  Does this point surely not create a Zero-Sum game for this practice of pseudo-psychology, one where employers realise that the process will not actually highlight potentially “bad” employees but that people are just people.

The premise of the service is that by asking all new followers to jump through a few basic hoops (captcha’s and such) TrueTwit can validate that the profile belongs to a proper person rather than a spambot.  Seems a smart enough idea providing some provenance but it got me thinking…

1. Do I really care if accounts following me are real people or bots enough to ask new followers to place a barrier to them following me?
2. By not validating oneself as a person how does TrueTwit preclude that account from following other than by simply applying a “block”?
3. Even blocking a profile does not prevent an account on Twitter from @ replying anyway as it is not follow/following dependant.
4. What about those bots I actually want to follow me, those which I use for automated functions?

What would be more useful to me would be the ability to validate those that I wish to follow, or at least selectively.  Of course the problem there would be akin to the first point above, “do I care if you follow me enough to validate myself to you?”

Any thoughts on Twitter or any other SocNet validation usefulness?

As yet I am not yet convinced I have a sound argument for their thinking other than to subvert process is human and that many social media tools provide an path or opportunity to do just that. 

Hierarchies are great, they provide control, a sense of order to things.  They have a very sound place in life.  Indeed as do silos of information, segmentation provides a wealth of benefits from privacy to ease of management.

It struck me late last night that I can draw a direct parallel between social media and a lot of the work I have been involved with over the last 13 years in data navigation. Back in the day pretty well all data was stored in some form of relational database modelled hierarchically.  This was, as mentioned, great for control but lousy for integrating multiple and disparate data sources (which led to a whole industry built on assimilation – data warehousing).  Worse still, hierarchically modelled data was a nightmare to navigate. 

Let me provide a simile.

In an organisation built upon traditional management structures with departments and the like, rigid reporting lines often make for poor communication channels and  awkward cross department interactions.  Those very structures designed to provide human resource control actually prevent humans from doing what humans do best – connecting.  How on Earth does one quickly & easily connect to the right person in another area of the company for help when constrained into following hierarchical chains of reporting?  This has been long recognised and working groups, committees and project focussed groups containing staff from across  a number of departments or skill bases are commonplace nowadays.

Dr Karen Stephenson, a corporate anthropologist and lauded as a pioneer and "leader in the growing field of social-network business consultants” (Business 2.0 2006), and her company NetForm have been publishing work on social network (think social graph web peoples) analysis for years which quite clearly shows that no matter how one tries to enforce structure on people informal networks of people will emerge – normally based around a specific context.  Yet the structure, the hierarchy prevails

So back to social media.  Are the tools of which social media experts talk not just enabling this networking behaviour, by providing easier more human, more informal navigation (for a particular task or context) across the hierarchies or silos? This to me makes far more sense than any talk of the structures being destroyed.

So everyone has seen the news Marketing (in an interview with Biz Stone, co-founder of Twitter) broke about Twitter possibly having decided upon a route to revenues.

This is what Stone reportedly said:

“We are noticing more companies using Twitter and individuals following them. We can identify ways to make this experience even more valuable and charge for commercial accounts.”

There has been a lot of talk recently on where Twitter’s revenues could/should come from, everything from the lacklustre advertising models done to death to premium accounts offering SMS notifications (please please bring SMS back for the UK).

Conversations I was having with Robert O’Brien, a friend of mine, back at the end of 2007 early 2008 revolved around our thinking that Twitter had really missed an opportunity in monetising their namespace (not a revolutionary idea as we are in the digital identity space).  After all it’s a well understood model and one with which most digital life-stylers are quite comfortable to a level.  Even if Twitter had offered up leases for names say for US$5 a year with the first 2-3 years free just think how many of the 5 million plus active users today with more than a few months of interaction would be more than happy to pay up to maintain their account.

So it really comes as no surprise that Twitter are now considering paid for accounts, in this case for commercial use.  But there are a couple of obvious speedbumps:

• The proverbial gate of free accounts has been left open so long that perhaps the monetising namespace horse has already bolted.  Will commercial entities even now be able to realise and evaluate value in maintaining a presence on Twitter?  When Marketing contacted Bob Pearson, VP of communities and conversations at Dell, with that exact question and got a telling response: “If it becomes complicated and costly, our instinct would be to move elsewhere.”  Sadly I suspect that Dell and a raft of other commercial entities will be quicker to jump ship than they were to board in the first place.
• Personal accounts will be exempt (at the moment) from this charge.  Obvious really but for businesses to be enticed into using Twitter as a marketplace it is vital that Twitter continues to grow it’s user numbers.   But when does a personal account become a commercial account?

Is Barack Obama a personal account or (now) a government account?  What about Stephen Fry – sure it’s a “personal” account but the business of Stephen Fry is, well selling Stephen Fry.  How about one man bands, the Whatleydude’s and Jonathan MacDonald’s of the world just when (and who decides) do these personal accounts become something more?

I’m not suggesting for a second that there needs to be a process for deciding, just that this is an area dotted with mines.

And of course there has as yet been no mention of cost.

An alternative approach

In October 2008 I had about an hours chat with Jon Bishop at a meeting of the Tuttle Club in London where we talked about where Twitter could derive income.  An idea forming in my head at the time (and it probably was based on those prior conversations with Robert) was around the usefulness of Twitter as a B2B & B2C platform, a messaging platform if you will for applications using the simple form of source based routing we all know and love on Twitter.  The ideas weren’t fully formed at the time (an old school mate Nick Halstead recently discussed the API route to revenue over on his blog) until just before Christmas when during one particularly busy afternoon I stopped receiving updates periodically.

For any serious Twitterer the bane of one’s tweeting existence is the current API limit.  I’m not a developer and will not claim to fully understand REST of any of the other goodness that has gone into the creation of the Twitter API, but I do get API’s what they do and why.   The volume of messages going across the Twitter API daily must be huge (anyone got stats?) and growing weekly.

Of course scaling this is easy is the system was designed to do so from the ground up, but I suspect that like most organically created systems the  Twitter API groans and strains like a 14 year old boy.  Remember 2008 – the year of the fail whale?

Twitter’s current limit of 20,000 calls per hour for whitelisted apps and 100 calls (?) for individual accounts places some control on this growth but seriously hampers active users.

For commercial operations this restriction needs to be either lifted or better still paid for by volume.

Rather than having say Dell pay for an account username, have them pay for messages.  A bit like how we all used to pay for SMS before all-you-can-eat packages were available.  In fact why not have an all-you-can-eat service level for the seriously verbose users?

As long as the cost is proportional to traffic and the business provided can easily access, buy, maintain and analyse their costs/benefits and that the money is driven back into helping to create the infrastructure needed then everyone wins.

To my thinking this user-pays approach if applied makes the system fairer, placing cost on those who i) place the most strain on the system and ii) derive the most direct/indirect monetary benefit from their Twitter presence.  Not at all dissimilar to web hosting, it’s easy enough to get a free account but if you generate a lot of traffic you’ll have to pay at some point.

Of course I may have assumed too much.  Maybe the smarts at Twitter have realised some of this and are already heading down the API  route – until they announce more I guess we must but speculate.  I welcome thoughts and feedback.

Related articles by Zemanta

Have a read it’s interesting and what is does highlight to me again is just how important some form of trust engine is to the internet as a whole. This I have not fully thought through by a long way but to my thinking we need tools and information exposure online that better enables us to replicate the conscious & subconscious checks we make on others before establishing and maintaining an offline relationship.

Before I have mentioned that trust is a personal viewpoint on another and their ability/likelihood to undertake a certain action. That trust is really just an aggregate view on a number of factors that matter at that point within context to the trustor.

The problem I see right now on social networks is that the marking up of the relationship is i) too simplistic, ii) not transparent enough & iii) not supported by any form of reputation metric.

For example; in the article had the potential victims of the scam had access to more information than the standard “Joe Bloggs want to add you to their network / be your friend…” then in most likely there would be fewer victims and therefore less incentive to scam.

What information do I mean? Well simple stuff like when the sender’s account was created, reputation metrics like number of posts or interactions, maybe one day even the ability to explode the sender’s profile out to view their identity across a number of social networks.

When did you last give a guy you met in the pub once $500 to bail them out of a Nigerian jail?

I think it’s more a case of “Why you shouldn’t automatically trust ‘friends’ on social networks”