This is a nifty example of how reality works on the interweb. I saw the poster on facebook and, like many others, for a moment thought, 'well, that's the sun all over, innit?' However, I've been gotcha'd too often by the Guardian's April 1st stories. They generally aim for a core of plausible, while tweaking the average Guardianista's sandal-wearing liberal nipples in just the right way. More than once, April 1st has seen me having a foaming rant, before eventually realisation hits, followed closely by acute embarrassment and sitting in a corner sulking at my own stupidity. (Chris Martin openly supporting David Cameron seemed particularly likely, I recall.)

It occurs to me, though, there's a parallel to how `markets' are talked about. To paraphrase Obi Wan, `market forces are are an energy field created by all living things. They surround us and penetrate us. They bind the galaxy together.' Which is to say, they're a mystical nonsense prayed to daily by people who believe in the confidence fairy. It's cargo cult gibberish. It's like turning up to some vital, knife-edge diplomatic negotiation teetering on the edge of war, cracking open a beer and saying, `chill - language will save us.'

Similar claims are made of the interwebs, and they're wrong for the same reason. There's nothing intrinsic to its structure that will produce truthiness or optimal outcomes. It can be used as a platform for doing that - but only if you actually develop tools to achieve specific aims. A trivial but suggestive example: guitar tab. The net's full of the stuff, and there are many sites with tabs for specific songs. But the vast majority of the time, they turn out to be precisely the same. That's not because they're correct, it's because the first person to dump it on the web got copied and recopied. No effective mechanism exists for improving the accuracy of particular songs. An echo-chamber is the unsurprising result, despite the fact that the net should be the perfect vehicle for some form of guitartab wikipedia. The machinery for that hasn't been built, though.

There are projects attempting to get meta on the net's inability to manage `peer review' effectively, like hypothes.is. That's got to be on the right track if there's no magic force that can do it for us. And the same applies to markets: effective ones are quite specific structures. Often cobbled together haphazardly, they can nevertheless be tweaked and developed for specific purposes, even though - just as with good code development - you'd best stay away from attempting all-encompassing gargantuan rebuild projects (compare the approach of NHS hack day to CSC's work on NHS records).

There's so much more to write about truthiness in the wake of current Republican goings-on, all the Assange gubbins and the ongoing mismatch between the physical and political reality of climate change... another time.