Moving this thread to the discuss list ... plase follow it there :), Cheers, Graham Bob Phillips wrote: > > Hi Steve > Could you please tell me more about page range checking and page range > checking program. I think I understand as far as "Hunslet only had 500 > births that quarter" therefore a 600 must be wrong. > Bob Phillips (John Slann Institute of Transcribers) > > ----- Original Message ----- > From: "Steve Gaunt" <[email protected]> > To: <[email protected]> > Sent: Tuesday, October 09, 2001 3:50 PM > Subject: Handwritten scans > > > I am following with interest the discussion about the quality of scans, > also John's suggestion for training transcribers. > > I know we would all love for scans to be better but at least the scan > quality for the most part isn't so bad, though I can't speak for every > quarter of course. > > Our syndicate has split into two teams (along gender lines strangely), > one enjoys the challenge of the handwritten stuff and consider them a > learning experience. They are about 80% of the syndicate. > > To help I transcribed a faded page of scans from each recognisably > different style (three). I made sure they were as correct as humanly > possible and sent them out with my transcriptions to compare. Following this > I took a sample of each number for each style and made a recognition chart > which I also sent out. It didn't take long and provided the team with useful > tools for difficult scans. This team are very happy to work with handwritten > and I have just been allocated another quarter to cope with the demand. > > The other team are working on the 1861 typed scans. They know about the > fantasy district names, the m-n problem and the original errors that plague > this source. Again a number recognition chart and page range checking > program helps keep page number errors to a minimum. > > I have asked both teams to 'return' any scans which are partially > illegible. Returned scans for our first handwritten quarter (Sept 1845 > Births) are only 29 of over 3000 scans. These I pick through myself. > > I started transcribing in 1999 with Dave Mayall's syndicate on handwritten > scans and they were often quite dreadful. The ones we have now are an > improvement. But having started on the bad stuff and serving my > 'apprenticeship', I much prefer the handwritten scans, warts and all, to the > typed ones or even the luxury of the typeset. > > Steve Gaunt > > > > > > > > > > --- > > Outgoing mail is certified Virus Free. > > Checked by AVG anti-virus system (http://www.grisoft.com). > > Version: 6.0.280 / Virus Database: 147 - Release Date: 11/9/2001 > > > > > > ==== FreeBMD-Admins Mailing List ==== > > Need to get a fast answer to your transcribing problems? Go to the > > Transcribers Knowledge Base at http://FreeBMD.RootsWeb.com/vol_faq.html > > > > ============================== > > Search over 1 Billion names at Ancestry.com! > > http://www.ancestry.com/rd/rwlist1.asp > > > > ==== FreeBMD-Admins Mailing List ==== > Want to help FreeBMD? > Go to http://freebmd.rootsweb.com/Signup.html to find out how. > > ============================== > Search over 1 Billion names at Ancestry.com! > http://www.ancestry.com/rd/rwlist1.asp

In message <[email protected]>, Graham Hart <[email protected]> writes >Moving this thread to the discuss list ... plase follow it there :), > >Cheers, > >Graham > >Bob Phillips wrote: >> >> Hi Steve >> Could you please tell me more about page range checking and page range >> checking program. I think I understand as far as "Hunslet only had 500 >> births that quarter" therefore a 600 must be wrong. I'll leave Steve to explain it but I will say that it is excellent and has allowed me to correct many silly errors. -- Philip Powell