Tom Sutcliffe: Hooked on the net's national treasures

I've written about the Public Catalogue Foundation before in these pages, a wholly admirable enterprise to identify and log every oil painting in public ownership in the country. It's a kind of Pevsner of fine art and its efforts have resulted in an accumulating library of volumes which, county by county, itemise every single painting and offer a matchbox-sized illustration of it. Browsing through these almost invariably turns up something interesting that you would never have expected to find, not to mention striking juxtapositions of the incompetent and the masterly. When I have written about the endeavour, though, it's often been with the faint sense that these handsome library volumes are the last gasp of a dying form of scholarship. That is, the information is of durable value but the form it arrives in is already beginning to look distinctly antique.

In fact, the enterprise was always destined to end up online – and yesterday saw the launch of the website that makes that possible. Your Paintings (www.bbc.co.uk/your paintings), created in partnership with the BBC, will eventually offer online access to all 200,000 art works on the Public Catalogue list, some 80 per cent of which aren't currently on public show. Far more significantly though it will make it possible to search that database by the content of the pictures – something that would be effectively impossible in book form, and without the participation of the public as volunteer indexers.

I tried out a beta version of the website's tagging software this week and very addictive it is too. Once you've signed up as a tagger, a painting is delivered at random and you're asked what "things or ideas" you can see in the painting. You can be stolidly literal and write things like "table" and "hat" and "flower". Or you can get a little more metaphorical and tag the painting with abstractions such as "expectation", say, or "motherhood". Provided enough other taggers agree with you the software will eventually add those terms to the artwork, and once enough taggers have looked at the painting it will be added to the Your Paintings website. Further stages allow you to identify people and places and, for some users, genres and painting styles. Ultimately – it's hoped within 18 months – it will be possible to type in a request for paintings that represent, say, motherhood in Southampton containing cameo brooches, and find out almost instantaneously whether any such thing exists in the public collection. Alternatively, you could look at every painting in the country that depicts Pendle Hill.

Quite what use you would put this facility to I'm not sure. That's probably for art historians to say. But what's intriguing is that the act of tagging itself makes you look at paintings in a slightly different way. You're more than usually aware that any representational painting is an accumulation of "things" (to use the site's studiously non-scholarly language) that together amount to an idea. "Things" become more apparent to you – such as the umbrella clutched in a Victorian lady's hand. And ideas that were transparent become suddenly salient. You can't get too carried away, of course. Looking at a portrait of a mid-20th-century Durham divine, I tagged it with the term "self-satisfied". Then I felt guilty, since I had no idea who this man was. And then I remembered that unless this term was endorsed by others it wouldn't go through anyway. Because of the randomness with which paintings are served up to taggers and because a certain degree of consensus is required, it's effectively impossible to game the system with false or malicious descriptions.

Once the paintings have been tagged and added to the Your Paintings website, a further refinement is added – almost literally vulgar in the satisfaction it delivers. You'll be able to give a painting an approval rating – meaning that in time a ranking of the nation's favourite paintings should be available. And alongside that crude assay of public taste there will be a more vaporous sense of what concepts and emotions animate – a kind of collective Rorshach test based on pictures of real things. I would recommend you try it out – before the unusual opportunity to contribute to a great work of national scholarship has gone – and I'd recommend that you think hard about concepts. Software can already recognise a face in a painting. Only you can identify "grief" or "love".

Fazed by the family's unfamiliar faces

On the radio the other day, Alan Hollinghurst revealed that his editor on 'The Stranger's Child' had drawn up a genealogical table for the two families who feature in the novel. I wasn't at all surprised, frankly, since the novel quite often makes you feel like an outsider at a family wedding, vainly trying to map out the exact connections between all the people present on the basis of eavesdropping alone. Since Christian names, surnames and nicknames often crop up in close succession, this isn't always very easy, and just when you think you've got everything squared away the action shifts by a generation and a crop of new arrivals turn up, talking sixteen to the dozen about people you've never heard of before. By chance, I read it immediately after Aravind Adiga's new novel, 'Last Man in Tower', which offers a similar profusion of characters, though in his case there's an added twist, in that it's quite handy to know which flat each character occupies in the Mumbai apartment building in which the book is set. Helpfully, the publishers provide you with a floor plan to which you can turn back at moments of doubt so that you know which party – and which party wall – is being described. With the Hollinghurst, you get no such crutch, partly, I suspect because some of the confusion is deliberate. Being an outsider is partly what the book is about. But I still think they should consider adding a map to the paperback edition.

Artists shouldn't be in the running

I confess that I had mixed feelings on reading that distinguished British artists were going to be designing posters for the 2012 Olympic Games. On the one hand it seemed churlish to complain about the fact that artists such as Martin Creed, Rachel Whiteread and Howard Hodgkin will get a public showcase outside the relatively rarefied space of a gallery. And having seen – from the Olympic logo – what horrors professional designers can perpetrate, it also seemed an excellent idea to get serious artists in on the act. But I couldn't suppress a faint feeling that those involved would be collaborating with the enemy. When I was at school you could be sporty or you could be arty but very few people were both. As our year of compulsory sports looms, I'll be looking to artists to take a lead role in the resistance, rather than join in with the cheer-leading squad. This is partly a serious point too. Surely the only interesting art about the Olympics will be art that acknowledges some ambiguity about this festival of physical elitism and patriotic fervour. But ambiguity isn't what successful posters are about. I hope all those involved will murmur one name to themselves as they work on their designs: that of Leni Reifenstahl. And that they remember which side they're on deep down.

t.sutcliffe@independent.co.uk