A Jones <[email protected]> wrote on Fri, 11 May 2012 at 20:26:37: > >"Tony Proctor" <.net> wrote in message >news:[email protected] > >> When it was launched, there was a lot of talk about the very inaccurate >> transcriptions. However, has anyone confirmed whether that's just an >> obfuscation of the short transcription summary it presents in the search >> results? > >There are no "transcriptions" as such - merely OCR [Optical Character >Recognition] text running in the background and struggling with the >inevitable problems posed by the nature of the original materials, which can >pose particular difficulties. There is provision for users themselves to >make instant corrections to the OCR text whilst they work. The more people >who avail themselves of this opportunity, the better it will be for >everyone. > >> It seems to find most surnames I've looked for, but the context around it >> in the few lines that the search-results hold are usually junk. I wondered >> if that was simply to avoid people getting too much of an answer for free. > >No, I don't suppose that to be the case. > >I certainly wouldn't say that I'd been disappointed with my subscription - I >simply accept that the available tools have their limitations, and that they >need to be used thoughtfully, carefully and methodically in order to get the >most out of the facility. In most cases it's still a lot more practical >than having to travel to the appropriate repositories, or having to make do >without any kind of index at all. > Their current offer -- they seem to occur moderately frequently -- is a year's subscription for half price, ie £39.98. It's open until 23.59 on Sunday 4 January. I've seen two promotion codes: 50XMAS and XMASDEAL. http://www.britishnewspaperarchive.co.uk/ -- Iain Archer