I understand your viewpoint - we are all "pragmatic perfectionists" at
heart. Any index is only as good as its compiler, and every single one
benefits from a second pair of eyes.
Whether you subscribe is your call. But to reach the standards that we
would all like to have, would mean the work has to done locally like
with the 1881 Census - and that will not happen in real time in any
realistic time-scale to be useful to the Internet revolution sweeping
the world - demand-driven by the USA.
You can now do online in about 6 months or less, what it took me 20
years to do.
What you cannot get online is experience. What you can get is (fairly)
instant gratification - and new research possibilities.
Brigstocke is an unusual surname - like my Picton interests in Wales.
But try it for Evans or Davies - and you cannot find a better way to
pick your particular person out of endless background "noise" by any
Usual story - depends what problem you are trying to solve.
From: Patrick Brock [mailto:firstname.lastname@example.org]
Sent: 28 October 2005 15:55
To: Brian Swann; DYFED-L(a)rootsweb.com
Subject: Re: [Dyfed] 1851 Census - Ancestry
Thank you Brian for that explanation and Emma for the description of how
to correct errors
My experience has been
I scanned through the 1851 Census during the Ancestry free trial a few
I found 12 spelling errors in the 19 names of my Brigstocke family at
Walwyns Castle (correctly provided to me previously by Pat Powell and
others a few years ago).
Because of that poor showing, I decided that it was not worth
subscribing, and so I am not in a position to correct their mistakes.
I am glad to hear that my experience is not representative, but I still
am not convinced that it is worth subscribing.
----- Original Message -----
From: "Brian Swann" <bps(a)norvic.force9.net>
Sent: Friday, October 28, 2005 3:25 AM
Subject: RE: [Dyfed] 1851 Census - Ancestry
I would be grateful if anyone out there does know how Ancestry.com
produces these indexes. I am sure it must be described (somewhere).
The 1881 Census Index was produced using local Family History Societies
in England and Wales - so it tends to be of high quality - as local
people know the local names.
I would be pretty certain that the indexes to the subsequent ones are
produced commercially in the Far East somewhere [India, Philippines,
etc.]. Double data entry will be used - but these establishments can
type anything - literally - any language. Deciphering foreign
handwriting of 1851 may be more difficult than the average task they get
- but they still probably get 95+% correct.
It is when the initial letter is wrong that real problems can arise -
distinguishing capital 'B' from 'K', for example. It is the keying
standards used and the allowed error rate, or [any] proof reading which
counts. Wrong initial letter makes it very difficult on Soundex
However better something than nothing. And we all know the IGI is not
From: robertsc(a)southwestern.edu [mailto:email@example.com]
Sent: 28 October 2005 07:02
Subject: [Dyfed] 1851 Census - Ancestry
I looked up several families on Ancestry.com
1851 census a couple of
days ago and did run into the "only English counties" during one of the
searches. I also ran into some amazing transcription problems,
where a family of Lewis, became Servis and then Service, all within
the same family household. There was also a lady listed as being from
St. Clears Ireland, but when I looked at the census image, it said
nothing at all about Ireland. So there are definitely problems. With
that said, I did find the information I was looking for and I am SO
GLAD that it is available to us now! Does anyone know if there is a
procedure for letting Ancestry know about possible transcription
problems we find or if they even are interested in having that
==== DYFED Mailing List ====
Cardiganshire FHS http://www.cgnfhs.org.uk