On Mon, 17 Sep 2001 09:05:26 +0800, you wrote: >--------------- >To all Transcribers of 1862 Birth Scans, > >I recently complained to Administration about the Quality of the >scans >we are using, and advised that I was not willing to upload >guesses as to >what I think might be the correct details, suggesting that >incorrect >data should not be uploaded, and I got the following replies. >What is your feeling about the quality of these scans and should >they be >rescanned before we progress any further I hope that Harry doesn't feel that the answers from Alan and Graham were in some way a cop-out. We are aware that the first batch of Scans from Ancestry are not of the quality that we would wish (indeed, In loading them to the server I rejected a whole year as *completely* unusable). I would make the following points; 1) Moving the discussion here *is* the best thing. We know about the problem, and clogging up the admins list with it isn't achieving much. Far better to bring it here and see if we can work out a way round the problem. 2) We don't have (and can't get at short notice) other better material. What we have isn't the best, but it is all we have. 3) The choice is between making what progress we can with these scans or stopping for several weeks to get better scans 4) There is a big difference between incorrect data and incomplete data. If something is unreadable and we mark it so, that data is incomplete but correct. We very much do not want people to guess anything. We are seeking to obtain better material, but it will take time. -- Dave Mayall

Hi, I too have been taking a look at some of these scans - in particular the Qtr currently being transcribed by my syndicate (1861 Q4 Births) and they really are awful. Many of the pages that I looked at had an "unreadable" character on virtually every line, most lines had a number of them. The page numbers and volume numbers seem to suffer the most but many names are equally bad. And yet, despite telling everyone in my syndicate to do so where they feel it is warranted) I have so far only had one page referred back to me for re-scanning and looking at some of the uploaded files there are a surprisingly small number of underscores etc. My conclusion is that most people either have much better monitors than mine (unlikely), much better eyesight (possible but unlikely), or they are taking best guesses and it seems to me that this is the most likely. I suppose that my point is that no matter what anyone says people are going to guess or perhaps perceive a badly reproduced letter as something else without realising, and as such the accuracy for which we strive could well be taking a nosedive. The poorer the quality of the scan the more guesses people are going to make. As accuracy is our goal then maybe Dave M's idea of stopping for a few weeks while better scans are obtained is not such a bad idea. In addition it is possible that such better scans would allow people to work much faster (I have heard many complaints that people's transcription rate has rapidly dropped with the poor quality) and could even make up the lost time. My vote (if we have one) is to stop. Regards Ian ----- Original Message ----- From: "Dave Mayall" <[email protected]> To: <[email protected]> Sent: Tuesday, September 18, 2001 12:50 AM Subject: Re: Scan Quality Concerns > On Mon, 17 Sep 2001 09:05:26 +0800, you wrote: > > > >--------------- > >To all Transcribers of 1862 Birth Scans, > > > >I recently complained to Administration about the Quality of the > >scans > >we are using, and advised that I was not willing to upload > >guesses as to > >what I think might be the correct details, suggesting that > >incorrect > >data should not be uploaded, and I got the following replies. > >What is your feeling about the quality of these scans and should > >they be > >rescanned before we progress any further > > I hope that Harry doesn't feel that the answers from Alan and Graham > were in some way a cop-out. > > We are aware that the first batch of Scans from Ancestry are not of > the quality that we would wish (indeed, In loading them to the server > I rejected a whole year as *completely* unusable). > > I would make the following points; > > 1) Moving the discussion here *is* the best thing. We know about the > problem, and clogging up the admins list with it isn't achieving much. > Far better to bring it here and see if we can work out a way round the > problem. > > 2) We don't have (and can't get at short notice) other better > material. What we have isn't the best, but it is all we have. > > 3) The choice is between making what progress we can with these scans > or stopping for several weeks to get better scans > > 4) There is a big difference between incorrect data and incomplete > data. If something is unreadable and we mark it so, that data is > incomplete but correct. We very much do not want people to guess > anything. > > We are seeking to obtain better material, but it will take time. > > -- > Dave Mayall > > > ============================== > Join the RootsWeb WorldConnect Project: > Linking the world, one GEDCOM at a time. > http://worldconnect.rootsweb.com > >