May 23

Social Media Systems (1)

(Update – swapped Sam’s live UStream embed for Michael’s recorded YouTube embed)
This is an expansion on my Brrism talk on Systems Theory and how it can be applied to social media (systems).

There are a couple of benefits of using something like systems theory when designing a social media system, plus a couple of drawbacks. Perhaps most importantly, it helps with the big picture before worrying about API calls, jscript vs php, etc. The main drawback is that is won’t tell you how to make your social media system actually grow and thrive.

If you’re interested in the history of systems theory, General Systems Theory, Bertalanffy, et al, then my previous post touched on that, plus there are good wikipedia pages to read (linked to in this sentence). This article will concentrate on the overview, the next article will introduce to a particular systems approach and I’ll conclude this mini-series with an example application.

What is a System?

The word ‘system’ has become somewhat diluted and it’s meaning confused. However, systems are relatively easy to spot and describe.

Patrick Coin, Feb 1996

Firstly a system has components. A ruler is not a measurement system, it’s just a straight bit of metal or plastic with some marks on it. A feather is not a flight system, it’s very well adapted to direct airflow, perhaps with display markings, certainly helps with insulation, but it’s not a system.

All those components are contained within a boundary. Sometimes that boundary is fairly obvious, like the Yellow-Bellied Sap-Sucker. Sometimes the boundaries are more arbitrary, this website might be considered a system but there’s no physical boundary. Ultimately, the boundary is where you want it to be, but should be relevant to the system you’re looking at. But remember, the wider the system the more complex, too tight and you might not have a system at all!

There is a theoretical construct called the closed system, in reality all systems are open. This means there are inputs and outputs that cross the boundary. What this stuff is that crosses the boundary will depend on the system. You might have information, raw materials, written articles, even abstract concepts like trust. The point is that stuff crosses the boundary and you can describe it.

There is an important caveat to all this stuff crossing the boundary, there has to be a transformation between the input and output. Otherwise you’ve got a pipe! A pipe is not a system.

A communications system is a system because although part of the input (your message) is hopefully the same as the output (your message), there is a pile of associated data about the sender and receiver that is transformed in moving your message to your recipient. There are also internal transformations of the input message through encoders, compressions algorithms, decoders, etc.

Systems are also in a hierarchy of systems that form their environment. My little Sap-Sucker lives in the rainforest that is its ecosystem, that is part of a wider global system, and so on. It also has a digestive system, neural system, and so on down the scale. Systems within systems.

And lastly, but most importantly, systems display something called emergent properties. In social systems this is the law of unintended consequences. Behaviours that arise from the interaction between all the components and sub-systems, variations in inputs, changes in the environment. The point being that you can’t define them all when you analyse the system.

So far, so mechanistic. How do we apply this to social media? More >>

If you’re interested in what was actually said in my talk, the video is up on Facebook (sorry, not on an embeddable site, will have to talk to Michael about that). Sam Downie (@samdownie) was streaming on UStream and the slides are on Slideshare.

May 11

The performance of junior doctors in applying clinical pharmacology knowledge and prescribing skills to standardised clinical cases

I have recently submitted a thesis for a masters of philosphy, entitled: The Design and Validation of assessment tools for use with Junior Doctors in Applying Clinical Pharmacology.  Part of this work has been written up and will be published in the next edition (June 2010) of the peer reviewed academic journal British Journal of Clinical Pharmacology.

Entitled: The performance of junior doctors in applying clinical pharmacology knowledge and prescribing skills to standardised clinical cases. This paper builds on the fact that safe prescribing is a core competency in undergraduate medical education. That a large proportion of undergraduate medical students and recently graduated doctors in the UK are not confident in their ability to effectively and safely prescribe and that errors are common in all healthcare settings and prescribing errors are the most common type.

This study produced twelve valid and statistically reliable assessments of Clinical Pharmacology and Therapeutics (CPT) knowledge and prescribing skills in areas that pose a high risk to patient safety. The findings show that a large proportion of Foundation Year 1 (FY1) doctors fail to demonstrate the level of CPT knowledge and prescribing ability judged by a subject matter expert (SME) panel to be required at this stage of their careers. My co-authors (Prof Nicky Britten & Dr David Bristow) suggest strategies and areas where teaching can be focused to improve the safety and effectiveness of FY1 doctors’ prescribing.

If this is an area of interest then below is the abstract related to the article and a link to the journal.



Recent studies suggest a worryingly high proportion of final year medical students and new doctors feel unprepared for effective and safe prescribing.  Little research has been undertaken on UK junior doctors to see if these perceptions translate into unsafe prescribing practice.  We aimed to measure the performance of foundation year 1 (FY1) doctors in applying clinical pharmacology and therapeutics (CPT) knowledge and prescribing skills using standardised clinical cases.


A subject matter expert (SME) panel constructed a blueprint and from this twelve assessments focussing on areas posing high risk to patient safety and deemed as essential for FY1 doctors to know. Assessments comprised six Extended Matching Questions (EMQ) and six Written Unobserved Structured Clinical Examinations (WUSCE) covering seven CPT domains.  Two of each assessment types were administered over three time points to 128 FY1 doctors.


The twelve assessments were valid and statistically reliable. Across seven CPT areas tested 51-75% of FY1 doctors failed EMQs and 27-70% failed WUSCEs. The WUSCEs showed three performance trends; 30% of FY1 doctors consistently performing poorly, 50% performing around the passing score, and 20% performing consistently well.  Categorical rating of the WUSCEs revealed 5% (8/161) of scripts contained errors deemed as potentially lethal.


This study shows that a large proportion of FY1 doctors fail to demonstrate the level of CPT knowledge and prescribing ability required at this stage of their careers. We identify areas of performance weakness that pose high risk to patient safety and suggest ways to improve FY1 doctors’ prescribing.

Mar 17

Blast from the Past

Unfortunately I can’t be at this months Brrism (sorry Micheal @kobb). However, one of the topics will be around ancient scrolls of wisdom (or anything over 5 years in Internet time).

Which got me thinking about what I was up to 10 years ago.

At the time I was a junior PhD researcher at the University of Plymouth. Bizarrely my old homepage is still live, though mercifully I’d updated it in 1999 with fewer flashing GIF’s and roll-over image maps, though the obligatory animated email gif is still there (the email isn’t live however).

A large part of my research (1997-2000) was looking at systems theory as it applied to ‘human activity systems’, so I’m going to cheat (slightly) and reference some work that wasn’t directly about Social Media (or even web-based technologies). Though I was based in the School of Computing, my background is in engineering and the research group was mostly engineers, economists and psychologists. We were interested in how systems theory could be applied to particular social groups (mostly engineering companies in this case) and in particular the processes that those social groups used to achieve certain aims (generally converting some specification into a manufactured product). However, I believe there is good reason to think that much of that research can be applied to social media application in other business endeavours.

Squared Circle Mosaic: Fibonacci Spiral with Hue Twist - Uploaded on January 27, 2005 by krazydad / jbum

Prof Peter Checkland (Lancaster University) is a Chemist that worked in industry on complex engineering problems and eventually moved to the most complex systems of all, those involving humans (unfortunately his seminal work is a book, 1981, not an online article).

Systems are generally recognised by some fundamental principles;

  • Boundaries – there’s a bunch of stuff that’s “inside” the system, and a bunch of stuff that’s “outside” the system (there will usually be an argument over where to draw the boundary but that almost defines the fact that you’ve got a system)
  • Inputs and outputs – stuff crosses the boundary, this can be physical or non-physical [you can have a ‘closed’ system but they’re generally rather boring and hypothetical]
  • There’s some transformation, i.e. difference, between the inputs and outputs
  • There are components within the system; a single component is not a system
  • Systems are nested; within large systems are smaller systems

and most importantly

  • Emergent Properties – you can’t describe the performance of the system just by analysing the component parts

Checkland is important because he was one of the first people to try and describe the messy company and organisational situations he was working within from a systems perspective (building on much of Bertalanffy‘s work between 1934 & 1969) . He identified boundaries and within those boundaries the sub-groups that actually made the company work. He identified information and flows of power within the organisation, and across those boundaries. He was able to sketch out the ‘actual’ human activity system, rather than the business or computer information system. It was the systems characteristic of emergent properties that led to them not performing as planned, and gave rise to the law of unintended consequences (previously identified in social sciences by Robert Merton (1936) but not explained).

But what does this mean for the social media strategist?

Well it means that, despite our shiny shiny toys, there is quite a bit of good research and clear thinking about how people work in groups and in particular how we can design such systems. No matter how carefully we design them, there will be emergent properties; but that doesn’t mean we shouldn’t design them in the first place.

There will always be boundaries, some stuff will be “inside”, some stuff will be “outside”. You don’t want everything to be “inside“; even planet earth isn’t a closed system.

Stuff needs to cross your system boundary, and it needs to be transformed en-route to becoming an output. I put my details into Facebook (as do 350m others) and a whole bunch of RSS feeds, and I get a ‘useful’ homepage about my ‘friends’. I can put photos, comments, stories, whatever in, and people can further transform them with additional comments, links to other people, etc. True, you can’t automagically get that data out of Facebook, but you can log in and read what’s there. And reading what’s there is one form of taking information out of the Facebook system, as are social networks, etc.

Of course there are lots of component software chunks and sub-systems within Facebook. Each casual game on Facebook is its own system, nested within Facebook. Each fan community is a nested system. Each discussion board is, potentially, a sub-system, nested within a fan page, nested within Facebook, nested within the Internet, etc. Depending on where you draw the boundary, everyone that’s on Facebook is also part of the system…

And of course there’s a ton of emergent behaviour that wasn’t predicted (or predicable).

So what can we do about/with it?

The first thing to note is that, from my experience, most people aren’t very good at meta-cognitive thinking about systems theory. That is, they are used to living within systems (social, organisational, leisure) but they don’t actually spend a lot of time thinking about those systems, and even less thinking about how they are thinking about them.

This means that people will usually try to apply existing social behaviours and norms to on line systems, and if that doesn’t work they get frustrated/angry/disillusioned/etc. You either build your on-line social media system exactly like ‘real’ world (but then why would anyone be interested in your system?) or educate people  into the operating of the new system. That’s why all games have tutorial / training built in.

In order to develop a training programme you need to understand, and be able to communicate, the designed purpose and functioning of your system. One way to do this is draw a picture of it, not a UML diagram or a wireframe, but a human activity system diagram. You don’t need to use any ‘standardised’ modelling nomenclature, so long as you and your team understand it and it covers the basics above.

Michael, Rick Chapman, and I spent some time recently thinking about modelling social media systems. We tried to cover the basics, without employing a formal systems modelling methodology. Its not perfect but I think it’s a good start.

You should know where the boundaries are, what the expected information flows are going to be, the transformations and components that will do the transforming, and what the wider emergent property will be. That will all change once the system begins to operate but at least you’ll have a blueprint and can either take action to bring the system back into the original concept or decide to take things in a different direction.

We did quite a lot of this within my old research group, and its surprising how good a consensus you can arrive at for generic systems diagrams.

This early draft is far from ‘perfect’ but I think there is something of value if you’re building, or thinking about social media, to have a model similar to this in your toolbox to refer to.

Conclusion & Caveat

In conclusion, some thinking time about the network you’re trying to build is valuable. The tools you employ should come afterwards; twitter is not a social media strategy. There are lots of good, well established frameworks to think about social networks, systems of activity, etc. You don’t need to follow slavishly the minutia of their particular quirks and peculiarities but you should understand why they are there and why you are ignoring them.

The caveat: feedback loops are a feature of systems. The huge difference that digital technologies have brought is the near frictionless feedback loop. There is almost no transactional cost to publishing a comment and for that comment to then be republished to +350m people (it happens both automatically via services like posterous and twitter-bots, and through the retweet/comment feature in all social media services), and re-re-published endlessly. That is something that we haven’t modelled effectively yet. The good social media marketeers amongst us know how to achieve this, even if they don’t fully understand the why.

Jan 12

Fresh coffee & Opportunities

Uploaded on October 23, 2006 by Hamed Saber

Well it’s been quiet externally for jbsh the last couple of months but there’s been plenty going on. This post is a cross-post from Open Coffee Bristol where we welcomed in the New Year this morning.

Well 2010 kicked off in the UK with snow, ice, sub-zero temperatures and general chaos as public services ground to a halt.

But not Open Coffee and the entrepreneurs of Bristol.

Fortified by the best coffee that the Boston Tea Party on Park Street has to offer we gathered on their first floor to catch up after the break and discuss the future. By the end Steve Cayzer (HP Labs, LinkedIn), Rupert Russell (Carmen Data, LinkedIn), Helen Davies (For Effect, website), Sam Machin (Orange, personal website), Nigel Legg (Katugas Social Media, website) and Andy (who surname I’ve unforgivable forgotten, sorry).

Conversation covered the various tax implications of company car ownership, developing new brand images for the new year (and the difficulty finding a good printers these days), online marketing for small tourism companies and the challenge of getting good geo-location data, and that was just at my end of the tables!

The general opinion was that while the weather and economic climate might be a bit inclement (or just down right awful) there was business to be done and opportunities to be exploited. Business cards were swapped and a couple of new collaborations initiated.

So the New Year is off to a great start and looks to get better.

Look forward to seeing you at the next Open Coffee Bristol on Tues, 26 Jan from 8.30am in The Boston Teaparty on Park St.

View Larger Map

Jan 08

Brrism – Social Media in Bristol

Always a good evening, the next Brrism looks to be a real barnstormer (6.30pm on Wednesday 20 Jan in the Pervasive Media Studio, Bristol).

There promises to be some great presentations on social media tools (Colin Rainsforth, twitter) and the fine line between time management & time wasting (Lee Cottier, twitter) but I’m guessing the hot topic will be the Digital Economy Bill.

Paul Smith (Labour Prospective Parliamentary Candidate for Bristol West) is leading the discussion. The bill is about to enter the Committee Stage in the House of Lords, so there’s no guarantee that it’ll be law before the election, but this is a critically important bill for the UK and Bristol’s digital economy so come along and can make a positive contribution.