NOVA-SCOTIA-L Archives
Archiver > NOVA-SCOTIA > 2003-07 > 1057186529
From: Lindsay Patten <>
Subject: Re: [NS-L] 1901 Census Indices Complete!>100,000 lines transcribedin NS!
Date: Wed, 02 Jul 2003 19:55:29 -0300
References: <3F031420.3070208@netscape.net> <3F03383D.923494BA@freenet.carleton.ca>
Hi Bill,
My thinking is that an online index is fundamentally
different from a paper transcription. The primary
purpose of an index should be to help you locate people
in the database of images at the National Archives. If
you view it that way I think it makes sense to have
the index available as it is transcribed, as even with
transcription errors the index is tremendously helpful
in locating people. Once a user finds a person in the
index the image of the original is literally "a click
away". It is true that some people will not verify the
transcribed information against the original but I
can't see depriving everyone of access in order to
protect some people from themselves. Besides, you need
to consult the image to examine the untranscribed fields
such as place of origin, occupation, etc.!
A second rationale for providing immediate access is
that the people using the index can help with proof
reading. Submitting correction suggestions is pretty
easy and people looking up their relatives may have
more information than a generic proofreader, which
as you point out, is handy in terms of being able to
make out seemingly illegible entries.
A third rationale is that I don't think we could get
as many people involved and/or get anywhere near as
much done if the index weren't immediately available.
There are over 300 registered volunteers and I would
venture that it is because the site is up and running
and useful that they decided to help out. Most of them
would never have even heard of us if we didn't have
the index available in useful form. Remember when I
first posted on the NS list? People were rather
suspicious of my motives and I think the fact that the
site was there and freely available was what gave
people confidence. I doubt as many people would have
joined in on the promise that their work would only
become available later, after proofreading.
Do you find these arguments convincing? In my mind,
given the direct links to the images, the above factors
outweigh the arguments for delaying release until after
proofreading. Again, the key thing is to recognize the
difference between an online index and a printed
transcription.
I do need to work on adding more efficient mechanisms
for cleaning up the field notes as they are dealt with.
I'm working out how best to do that. In the meantime,
user's should take the presence of a lot of notes on a
page as an extra indication of the wisdom of verifying
the info by consulting the images.
Cheers,
Lindsay
Bill Tufts wrote:
>Hi Lindsay,
>
>I think it is great that so much is getting done so quickly,
>but when I look at some of the transcripts where I know the
>families, there are sometimes a whole lot of mistakes.
>Wouldn't it be better to get these pages proofread before
>they get put on the site? Some pages are going to be a mess
>before all the corrections are made.
>
>Case in point: Someone made the right correction by saying a
>certain family surname was misread. McLeau should have read
>McLean. The first entry for that family seemingly now reads
>"McLean, Robert Mr." But that isn't right either and I made
>a 2nd correction. It was "Mrs." Robert. Someone is going to
>come along, see the flag/tick to know there's a correction
>below, but they could miss that there's a 2nd correction.
>Why not get it right the first time around.
>
>I'm not complaining about the volunteers. I *know* how long
>it takes to transcribe even one page & sometimes, I
>recommend that people take a bottle of murine with them <g>
>because I know the census taker moved his lips when he
>wrote. These transcribers have made honest mistakes and I
>can show examples where you really can't read whether the
>blighter was 38 or 83. One census taker wrote his 4 and I
>promise you that you will read 7, instead. Having a
>birthdate for the family member on my computer helps to read
>better & I'm sure some people will be nodding their heads in
>recognition as they read this. "Yup," the guy is right. Been
>there, got the T-shirt.
>
>There's a 2nd problem, too. A lady in NB wrote a rather
>short complaint on the NB list about errors and it has to
>do, in part, with not just reading what's on the page, but
>what was it the census taker intended? As a transcriber, you
>have to be honest and copy "as is," but some of the
>oldtimers writing the census returns were "old school," even
>then, and when they wrote names like Sharp and Clark, they
>used a little curlycue at the end of the name and it gets
>transcribed as Sharpe and Clarke. The transcriber becomes,
>all of a sudden, like a baseball umpire. Whether he calls it
>a strike or a ball, he's 50% wrong before he opens his
>mouth. The transcriber has to make decisions, sometimes, and
>the only way you can honestly handle it is to read the
>original. Then, if a mistake is made, you can blame none but
>yourself.
>
>I think, in the long haul, if we move forward more
>carefully, the end result will be better and good, accurate
>transcripts will come sooner.
>
>Bill
>
>Lindsay Patten wrote:
>
>
>>Oops! Gee golly if I didn't forget the URL:
>>http://AutomatedGenealogy.com/census/
>>-Lindsay
>>
>>Hi Folks,
>>
>>Having passed a couple significant milestones, I think a status update
>>on the Automated Genealogy 1901 Census Index Project is in order!
>>
>>The lines transcribed in Nova Scotia has passed the 100,000 line mark
>>and Annapolis and Kings Counties are complete! At the national level
>>there are more than 620,000 lines transcribed including:
>>234,000 in New Brunswick (68% of the entire province!),
>>209,000 lines in Ontario, and
>>50,000 lines in Quebec!
>>
>>Viewing the index is completely free for personal use and the site
>>contains surname indices at the national, district, and polling district
>>levels. Each page includes a link to the image of the original census
>>form for verification and finding information that isn't included in the
>>index.
>>
>>All the transcription is being done by volunteers and more volunteers
>>are always needed! Transcribing is easy, all you need is a web browser,
>>and internet connection and some spare time. The site has an automated
>>system for reporting errors (be sure the error is in the transcription
>>and not the original please!) and more than 4,100 user reported
>>corrections have already been incorporated into the index! The surname
>>indices include a list of other names with the same soundex code also
>>has a new feature to allow you to find people who were recorded as
>>enumerated under the wrong name.
>>
>>Happy researching!
>>
>>Cheers,
>> Lindsay
>>
>>==== NOVA-SCOTIA Mailing List ====
>>To unsubscribe
>>LIST mode: Send a message to NOVA-SCOTIA-L-
>> that contains (in the body of the
>>message) the command: unsubscribe
>>and no additional text.
>>DIGEST mode: Send a message to NOVA-SCOTIA-D-
>> that contains (in the body of the
>>message) the command: unsubscribe and no additional text.
>>
>>
>
>
>
This thread: