Search is perhaps one of the most magical-seeming things about using the web – the power at your fingertips (no wand required), the speed of response (faster than a speeding bullet, surely?), the amazing diversity of information that you can summon up from the vasty deeps. It’s enough to make you think you can find anything that you want – if only you get the original spell (search-term) right.
But it’s worth being in mind a few things that perhaps aren’t so obvious – on the surface at least:
- lots of information resources – the so-called ‘deep web’ – aren’t indexed at all by the main search engines – good as they are. Not all of these are behind a paywall. Web crawlers need to be told where to look;
- what we find and the prominence that it is given, is the object of a great deal of hard thinking and design work by both search engine providers, and by those who seek to influence the results – the Search Engine Optimisation industry is worth a substantial amount in its own right – humans design and tweak the principles that machines then build upon, to create the algorithms that present results back.
The point is that search is a human construct. Just as Mr. Dewey originally designed the categorisation system, used by physical libraries the world over, that was highly influential for the way generations viewed the way bits of knowledge fitted together, it’s humans who moderate what our magical search engines give back to us.
Meantime, ALISS will be developing the means to support our community of contributors and users to a) moderate and b) filter stuff that they take from the Engine.
We’re reckoning to draw a distinction between the two.
At the moment we’re seeing ‘moderation’ as a collective process which will deal with obvious scams, only. So a consensus that a resource has been flagged for the Engine, for nefarious purposes, will mean that it will be set aside. We’ll need to work out what counts as a ‘nefarious purpose’, and we’ll want to support a good discussion about this – but at the moment we’re thinking that this definition will be pretty tightly drawn, rather than loosely.
We’re seeing filtering as a different matter. We reckon this is something that intermediaries of all types will apply to the flow of information from the Engine to where they make it available as part of their own information services.
There are a couple of things that underpin this.
- One is our stance that we are working with information intermediaries rather than setting up yet another source on the web. We want to work as part of the long-term conditions family. Just as in the production of news, Reuters helps by linking people who issue press releases with those who mould them into news stories: it isn’t there to run its own newspaper.
- The other is our understanding that filtering happens now – what we are developing isn’t really that new.
Most LTC organisations include a list of useful links somewhere on their site. Someone decides which links to include. That’s filtering in practice.
At another level, different organisations may well have a stance on certain sorts of information resources. For instance:
- You won’t see many homeopathy resources promoted on mainstream NHS information sources.
- In Mental Health, what counts as an effective therapeutic intervention is sometime the subject of forthright debate – different MH organisations can take a quite different stance, and these are reflected in what they signpost people towards.
- Then again , a group of GPs developing social prescribing might well take a pretty severe look at the ‘quality’ of the resources they were prepared to refer people towards – as independent practitioners, they could be excused for thinking that they have more skin in the game than the rest of us.
ALISS intends to be studiously neutral in all this. We think that any individual organisation’s decisions on what they filter out shouldn’t pre-empt the choices made by others.
So if the NHS doesn’t want to show anything about homeopathy in what it takes from the Engine, that choice shouldn’t prevent others from showing those resources in their information provision.
They’ll reside in the Engine.
Whether or not individuals see them will be part of the decisions that individual members of the LTCs information family make, about the information services they variously want to provide to their constituencies.
In this way ALISS will be supporting a distributed editorial model, as befits the mixed economy of organisations and points of view that are part of the LTCs domain. We see that this distributed editorial model is an important support for the army of intermediaries that do currently – and will in the foreseeable future – help people access the information they are looking for in some way, whether they are web-enabled or not.