Anything You Could Ever Want

PJ Harvey is at the Hay Festival this year. I cannot think of a more perfect day but I cannot go. Before we were married, my wife and I went to Hay, not for the festival, just to look at books and let’s face it there is not much else to look at in Hay-on-Wye. It was about this time of year and we combined it with a visit to my parents over in Malvern. My wife was petrified meeting them for the first time, worrying that they would not like her. I hope she considers that this was a groundless fear. A few years later we took my dad over there so he could look for a magazine cover he remembered from when he was a kid, a factual thing like Look and Learn but with a picture of two children gazing out over the various delights of technology that could be found within. And we found it! I wish I could remember what it was called because I am sure I could find a scanned image of it somewhere.

I have been reminded of this because today has the same weather as on that first trip down. Weather makes so much of memory, the smells and atmospheres, the humidity and light – they all seem to be so much more evocative than plain sights and sounds. At the time they do not get noticed as our brains are overwhelmed by the wider bandwidth of information from our eyes and ears but this seems to be filtered out until all inputs result in the same amount of information ending up in long-term storage. Just thinking of this makes me realise how much processing of our input information gets done before it ends up backed up. The process has never been defined but is the ultimate information system, an elegant flow from photons and air-compressions to a matrix of memory distributed and backed up in our brains. It always appears better to be able to finish a complete design document and then code up a finished system. Of course, this hardly ever happens despite the best efforts of the project managers and the specifiers and coders perform a sort of dance, a heuristic approach with two participants. The evolutionary process which has turned out the complexity of life has produced an elegant result without any original design and is the ultimate heuristic loop. I do realise that the systems designed by human beings are mostly far simpler than even the most basic processes of life but in that case shouldn’t we use those processes to test some sort of evolutionary design?

I am currently working on a complex spreadsheet which has many data items all interrelated, with an almost chaotic output based on tiny changes to the inputs. On top of this, the organization has many copies of the spreadsheet, one for each of many entities and though there is no physical link to the spreadsheets, each one is logically a node of a subjective network of entities. The individual sheets already remind me of a neural network, albeit a very fuzzy one and then on top of this, the whole network could be seen as a meta-network. We are all too busy concentrating on the details of the local processing to take a look at the whole thing to see if anything useful could be done. I am conscious of the fact that I am trying to explain this without going into any detail of what the process is actually for. Even a single sentence describing what organization this relates to, would give you a much clearer picture of the real-world situation. Now I am a c# man, the neural network I have always wanted to code is more of a possibility. What with the graphics of the fractal program, something to recognize handwriting may be forthcoming. We shall see.

I finished The Last Three Minutes and for some mad reason decided to start on Catcher in the Rye. I was going to read A Moment of War , the last in Laurie Lee’s trilogy of autobiography, having finished listening to Cider with Rosie and As I Walked Out One Midsummer Morning, but reading the book and not hearing the power and passion of Lee reading it himself seemed weirdly empty. I will have to wait a bit before starting it again.