Working Notes 2.0

Perhaps (surely?) there are very good tools and applications available for the support of distributed data collection, but something about the widgets being deployed (very useful synopsis here, many thanks Simon Dickson) in DIUS’ recent consultation on ‘Science & Society’ struck me.

As far as I can understand it, the widgets allow interested parties to select from a menu of (in this context) the topics being consulted upon. Very sensible - it’s unlikely that everyone will be interested in every consultation question, so why not enable interested lobbies to focus their attention.

The next handy thing that’s enabled is the easy embedding of a panel, presenting the selected topics as some form of poll, within the interested party’s publication medium of choice. Again as I understand it, this is rather like embedding a Youtube video, or a tag cloud, into the sidebar of your blog, for instance. The interested party’s constituency can then respond to the poll in (that distributed) context.

And then (hey presto!) that data is piped back to the original consultation database. Not quite sure how this element works, but the idea is brill.

Even in its own terms, this is simply wizard. It’s disruptive technology of the best sort, being a game-changer for how consultation can be done. Congratulations to Steph Gray and colleagues. His own summary of their approach is well-worth inward disgestion alone.

But I was thinking of a slightly different context. I’m just working up ideas for a wee review project, in a domain characterised by fierce (and to a certain extent divergent - they’re pilot projects, which should explain a certain amount) enthusiasm for a series of clinical data-sets. Nationally, there’s a fair amount of consistency, but it’s difficult to gain consensus over the items that aren’t part of this. And politically this is not the moment to attempt to be Stalinist.

So, would this distributed approach support this ‘agreement to differ’? And would it do this better than currently mainstream approaches to clinical data sets? Could some differentiation be usefully applied by making more of the social (web 2.0) potential, rather than thinking purely within the frame of data collection?

Certainly the ease of embedding the tailorable poll (or data-collection form) in a variety of contexts would surely be a handy wee niche thingy?

Must see if a dog-walk would either help develop this, or delete the train of thought…

a.k.a today’s wild-eyed idea. Well, not today’s, the end of last week’s, really.

Talking recently with a colleague and friend, she was unhappy about the lack of secure email between health and social care domains. I responded with the thought that a useful experiment had been sucessfully undertaken linking selected Scottish GPs with the DWP, over assessments for Incapacity(?) Benefit, using structured messaging within SCI Gateway, and secure email once the message had gone beyond NHSNet (I think). [Subsequently I found out that this successful experiment had had to be mothballed through delays in agreeing the necessary shared financial support, between partner agencies. Business as usual in joint-working then...]

But thinking about this as I was walking the dog, later, I wondered whether shared document access might have any mileage as a possible alternative to (email) messaging. I’ve used Google docs a bit so that conceptual structure (secure document - well, reasonably secure as in ‘hidden in a haystack’ - with controlled access list, for the care team) might fit quite well, I thought.

And lo & behold the next day I stumbled across a post suggesting almost precisely that, in the context of front-ending google docs with a web form. Many thanks to Liz Henry and her network for the research…and one of those ’so it’s not just me who’s a lunatic’ moments.

Of course, actually trying this out with personal data might be another matter. It would more or less certainly make our security folk have a fit of the heeby-jeebies…or do I do them an injustice?

Then I remembered that one of our data-sharing partnerships not very far away (as I look out of the window) has been actively thinking about something very similar (though probably not with the same tools) for shared Children’s Assessments. Just as sensitive, though in a different way, as Mental Welfare Reports.

Hmmmmnnn…

Everyone’s writing about the ‘Social Graph’ just now - via tracking posts about SNS, identity management, and the read/write web, my feeder is full of the stuff.

But a few things fell into place alongside one another earlier today:

• first, the concern about translating relationships from one setting (or ‘networked public’?)  into another - a variant on not wanting to have to re-input all ones contacts, but equally not wanting them copied across regardless;
• second, the flash of a diagram remembered from a recent look at the Tao of Topic Maps (where everything is connected with everything else) and where the lines representing the relationships between things were themselves objects and could be labelled..
• ..which fed into the memory of a recent post by Stowe Boyd where he said he’d much rather just be able to tag everything, rather than have to remember which silo to file something in (he was being critical of some of the rather 1.0/2.0 transitional apps at the time).

So, what if I could tag the relationship that I have with someone? Like ‘old school chum’ or ‘boss’ or ‘nephew’. Of course they’d need to be able to tag the relationship from their point of view, which might be the same (old school chum) or the other end of some binary link (subordinate/uncle) or none of these but something coming from their perception (old school chum but someone I’d prefer to avoid in RL)-(could you be able to see how the other party saw the relationship? That would be interesting).

Of course, I would want to be able to apply multiple tags.

And I would want to be able to use the grouping capability to portage various elements of my overall social network from one domain to another…

At which point I got off the bus and broke my train of thought. However, surely lots of people muct be thinking of this aspect of the social graph already? Hence the MtS - “must find out more”.

The other day I ran a simple retrieval on the NHS e-library for the term ‘child protection’. It came back with the news that there were 13114  resources. Of course, some of these will appear on more than one list, and it’s likely that a proportion of them will only be of interest to ‘dead men in white coats’ rather than the broader CP community. But it’s still a shed-load of ‘knowledge’ (OK, depending upon how you view knowledge).

This is part of some initial thinking I’m doing on trying to make the general e-library resources more available and nearby, to the Child Protection Shared Space and its various inhabitants. One simple approach would be to provide some topic lists that would drive ‘canned queries’ to retrieve pre-defined lists of material. Here’s an example, from palliative care.

But, what topics to use, and equally important, how to develop them? I remember being fearfully impressed with the card-sorting approach used by Sarah Curier in the early days of Stor Curam (now the Learning Exchange) where she did some pilot f2f work with a small number of Social Work academics, on the social care terms they used, then had one of her colleagues (Ivanna Fernandez) create a nifty Flash widget so that the other 100 or so HEI staff could all have a go on the web. She arranged for some statistical software to be connected behind the scenes, and ran the analysis. It was the democratic aspect that this introduced that seemed new and intriguing, quite apart from the sensible suitability of card-sorting, to this problem. [memo to self, must see if Sarah's notes/article on the process are available, even as grey lit.]

Then again this morning, discussions included what to do about a set of Child Protection Committee (CPC) terms for topics that the West of Scotland network would like to sort by. Once more, the process of devising and agreeing these seems just as, if not more, important than the list that emerges.

Now, there’s the risk of lurching into the whole tags, tagging, and folksonomies thing - which I hope not to do (as it’s been really, really well rehearsed elsewhere…all over the place…one item among many). But two key questions remain:

• how to avoid locking down the taxonomy that one agrees originally - so of course one needs to incorporate a ‘tag cloud widget’ or something similar into ones app.;

• how to promote the sharing of knowledge, insights, impressions, ways of looking at the world, that you can so easily get from supporting a folksonomy by tagging, and perhaps even annotation too.

Maybe a Knowledge4Practice thing?? I feel an event coming on….