[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Another Habitual Lurker confesses--with (hopefully) useful transcribing tips



I, too, am a Habitual Lurker on the list, and when I "Un-Lurk" I usually send responses directly to the questioner.  In February I sent a reply to Claire McGuire's query about methods for transcribing documents that reached her too late to be included in her summary for the list (I was out of the office for a few days and saw the post late).  Since I haven't seen anything similar to what I describe below, and hoping that this information will be as enlightening to others as it was to us, I decided to post a modified version of the War-and-Peace-length message I sent to Claire (it's still long, so consider yourselves warned):
 
We transcribed a collection of some 318 family letters (1860s to early 1900s) for LC's American Memory website two years ago,  and I can tell you a little about how we did the project and A LOT about how NOT to do it!

The first thing to do is to get two transcribers working on each letter.  It sounds labor intensive and potentially time wasting, but here's the payoff:  after each person types the same letter you compare them electronically using a standard feature found in Microsoft Word [pull down the "Tools" menu, select "Track Changes" then "Compare Documents"].  The places where the computer says the two transcriptions differ are the places that you need to check for mistakes, as these will only show up where BOTH transcribers make the SAME mistake.  A final read to check for such places is helpful, but the electronic comparison boosts accuracy in an amazing way.
 
Unfortunately for us, we didn't learn about the two-transcriber-on-the-same-document benefit until after we finished our project, and we ended up reading, and re-reading, and re-reading each letter 4 or 5 times for accuracy, which is much more time consuming than typing something twice and comparing it electronically.  The typing should be done by two DIFFERENT people, as a person familiar with a text will tend to skip their own mistakes or think they see what is supposed to be there, etc. 
 
We felt pretty silly when I found out about this method while listening to a lecture by David Seaman of the University of Virginia Electronic Text Center, and there was much mourning for the time we had wasted, not to mention wailing and gnashing of teeth!  Live and learn--the hard way--but let others benefit from our mistakes... 
 
Anyone contemplating an electronic text or document mark-up project should definitely visit the excellent UVA Electronic Text Center Website:  http://etext.lib.virginia.edu/standard.html .

Also, when creating electronic text documents do not, under any circumstances, decide to use three spaces to separate sentences that do not have periods.  This is a common editorial practice that works fine for *printed* transcriptions, but many computer editors read a space as a space, three spaces as one space, ten spaces as one space, etc. You will end up with paragraphs that look like one, long, continuous sentence.  Decide in your editorial policy how to handle lack of punctuation--we now recommend judicious additions of periods where necessary--and state your policy clearly for readers.  Again, this is a voice speaking from the "Learned It the Hard Way School."
 
I hope some folk find this information useful.  So now, having Un-Lurked with a Vengeance this week, I will retreat to Comfortable Lurking Mode. 
 
Karen "Still Lurking but feeling better now that I've 'fessed up" Koka
 
Curator of Manuscripts
Nebraska State Historical Society
P.O. Box 82554
Lincoln, NE   68501-2554