Genealogy Transcription

This isn't a new idea as I've seen it done before on a limited scale, but perhaps it is new for the Church. It's my understanding that the Church intends to put the bulk of its family history archives online. It's also my understanding that they intend to have most or all of it transcribed from images to text so you can access the data in more useful ways. I'm sure that at this point, putting the images up is far easier than getting it transcribed and putting the text up. How about throwing the images up and allowing people to access the website to do the transcribing from the images and submit it? This does NOT have to result in inaccuracy. You simply have multiple people transcribe the same images and compare the results. If there's a difference you either have it transcribed again and compare again, or toss it out for manual review. This should allow the Church access to far more volunteer effort than what it probably has on hand at the moment.

When the Church released the 1880 census I _LOVED_ it. However, it was rather depressing when I found how long they had to work on it. Genealogy is something that's lighting a fire under a great many people (more outside than inside the church from what I can tell) and many of us would be absolutely thrilled to contribute to a project like this where we knew the results of our labor would be available for free to the public. The current cost of memberships to various genealogy sites is often beyond the reach of many people wanting to do genealogy research.

I don't know if this is what you're looking for, but the Church has an extraction program whereby members can take time to transcribe names and vital data from census records and other records in digital images. Recently, this has become a process that anyone can do online. Take a look at the FamilySearch Indexing website. I just volunteered yesterday, I got an account this afternoon. I chose out of a list of 10 or so ongoing projects, downloaded a batch and transcribed 50 names from a census record in an hour or less. I imagine this will become part of a free public database at some future time.

Actually this is already being done. Take a look at http://www.familysearchindexing.org/ and they are doing it that way already.

In fact, projects that used to take years are taking only months, and projects they thought would take months are done in only weeks.

Some stakes that were involved in the beta went from doing only maybe doing a few thousand names to maybe 20,000 names a year in the old extraction program did in excess of a million names in the indexing program while it was in beta.

A stake extraction director in Mesa Arizona reported that one indexer did 16,000+ names in one month.

A report at the Utah Genealogical Society's 'Institute for Genealogy' or whatever it was, said they had 12,000 indexers by that date, mid-January of this year, and expected shortly for that number to double, and by the end of the year it might be in excess of 100,000 people, member and nonmember alike, doing indexing.

Data will be tied to images from other things I've heard about the program, so if you look up a name once they put the images online, you will also view the image that name is associated with.

As another volunteer indexer, I can say that this is what you are looking for. The old extraction program (I was a ward extraction director) did this but the overhead made it near impossible to keep fellow extractors (who also were full-time students) motivated to work on the data. (I think we were only able to complete one batch in an entire school year) This allows for smaller batches to be downloaded per volunteer and they get the batches back very quickly.

There is a due date of a week after you download the batch to keep you working hard. but the usual batch is only 20 to 50 names. Also you can set goals for yourself, like how many names you are going to extract in a month. It is a very cool program and I'd suggest to everyone to volunteer a little time to indexing.

Currently we're working on the 1900 census.

Yep. This is exactly what I'm looking for. Thanks!
I guess the part that bothers me now though is why I didn't know about it before. Out of curiosity, how did each of you find out about it?

I don't remember exactly how I heard about it. There are various sites I occasionally look at for what's coming on the horizon. I've listed some of them below.

• FamilySearch
  • They have a News section, though it's rarely updated.
• BYU Center for Family History and Genealogy
• FamilySearch Labs
  • FamilySearch Labs Blog
  • Taking Genealogy to the Common Person (Dan Lawyer - FCH PM blog)
• BYU Family History Technology Workshop

• BYU Computerized Family History & Genealogy Conference
• BYU Family History & Genealogy Conference
  • These two conferences are held in March and August respectively and often have Church representation regarding the new things coming up. There are often archives available from past years.

Hope some of these interest you!

I heard about while it was still in beta. It was told to the ward extraction directors so that they could see the cool things on the horizon. Of course we tried to get into the beta but to no avail. So I found out about the indexing being available to all through familysearch.

Here is a new release I received by being on the Family History Consultant E-Mail list. This is a project which has been completed by FamilySearchIndexing and posted online.

News Release was sent Feb 7, 2007

Utah Death Certificates Now Viewable Online
250,000+ certificates from 1905 to 1954 linked with index and images

SALT LAKE CITY, UTAH - FamilySearch, in conjunction with the Genealogical Society of Utah and the Utah State Archives and Records Service announced today that the state’s free online index to death certificates is now linked to original images of the historic documents. The integration of the index with free digital pictures of the death certificates issued from 1905 to 1954 by the state will open doors to additional information for family historians and genealogists with Utah ties. To search the index and view the certificates, users need to go to [url=mhtml:{F6725479-018B-4A2A-9D17-E947559281F7}mid://00000016/!x-usc:http://historyresearch.utah.gov/indexes/index.html]http://historyresearch.utah.gov/indexes/index.html[/url].

The online index to 250,000+ Utah deaths was created by the State Office of Vital Records and Statistics and has searchable information limited to the name of the deceased person, their date of death, sex, and where they died. The Utah State Archives turned to FamilySearch to help them get the digital images online. FamilySearch digitized the images and provided the technology to link the images of the certificates to the state’s online index. The linking process was completed in just a few weeks, incredibly fast for a project of this nature and magnitude. The names of Utah’s deceased are now very much alive, searchable, and viewable online and for free.

I think it would be much easier for people to discover this if there were some prominent mention of it on familysearch.org - after all, the very people visiting familysearch.org are the same people who are going to have an interest in this.

gblack wrote:I think it would be much easier for people to discover this if there were some prominent mention of it on familysearch.org - after all, the very people visiting familysearch.org are the same people who are going to have an interest in this.

I agree. The best way to get the word out would be a press release in the FamilySearch News site. Then the library catalog should be updated to have a notation that the images on these microfilms have been scanned and are available online and give a link.

thedqs wrote:As another volunteer indexer, I can say that this is what you are looking for. The old extraction program (I was a ward extraction director) did this but the overhead made it near impossible to keep fellow extractors (who also were full-time students) motivated to work on the data. (I think we were only able to complete one batch in an entire school year) This allows for smaller batches to be downloaded per volunteer and they get the batches back very quickly.

There is a due date of a week after you download the batch to keep you working hard. but the usual batch is only 20 to 50 names. Also you can set goals for yourself, like how many names you are going to extract in a month. It is a very cool program and I'd suggest to everyone to volunteer a little time to indexing.

Currently we're working on the 1900 census.

Being as my home computers all run Linux exclusively, I looked at the indexing program page to see whether it might be possible to try to get the software to run under Wine (as PAF now does). It was with a very happy moment when I saw that the software is in Java, and there were even instructions on how to use it on Linux or a Mac. I plan to volunteer for indexing as soon as the new FamilySearch beta test ends.

Somebody with the right connections, please relay a _VERY_ _BIG_ Thank You to those who made it (apparently) possible to use the indexing software under operating systems not from Redmond, WA!!!

If you are wanting to do indexing, you don't need to wait for the new familysearch to get out of beta. Since the indexing is already out of beta and anyone can sign up.

That is, of course, unless he's part of the beta and doesn't want to overwhelm himself with the Indexing. Any one can join the Indexing and as far as I'm aware there is no quota. I've only received a list of 50 names per batch with a week to do the 50, so it isn't extremely stressful for me and is pretty flexible.

rmrichesjr wrote:Being as my home computers all run Linux exclusively, I looked at the indexing program page to see whether it might be possible to try to get the software to run under Wine (as PAF now does). It was with a very happy moment when I saw that the software is in Java, and there were even instructions on how to use it on Linux or a Mac. I plan to volunteer for indexing as soon as the new FamilySearch beta test ends.

Somebody with the right connections, please relay a _VERY_ _BIG_ Thank You to those who made it (apparently) possible to use the indexing software under operating systems not from Redmond, WA!!!

I had a great meeting with the product manager over this project yesterday. I'm going to share more about this project in the near future. However being a Linux user myself, I did thank him for you and for all the Linux users out there. He did say that the don't "officially" support Linux but because it is written in java it just works. What this means is that they don't do any testing under Linux. However I have signed up as a volunteer and have already been working on batches with my Linspire box.

Tom

I also signed up for indexing a couple months ago. Very impressive system and I'm glad to see it going. I do have a suggestion though, if anyone is in a position to pass it along (this is a suggestion, not a complaint):

It would be helpful to make use of partial batches instead of just deleting them, or allow the user a little leeway in the deadline. Here's why I think that:

I spent a while working on my batch, but didn't complete it in one sitting. Then life got busy for a few days and I didn't get it finished in the allotted week. When I logged on again, it told me my time limit had expired and my batch was deleted. I was frustrated that the time and effort I had already spent was completely wasted, as no use was made of even that part of the batch I had had already completed. I have not yet convinced myself to go back and try another batch in case I don't get it completed in time. (Sometimes time is more precious than money, and just as much of a sacrifice).

If somebody does 75% of their batch, but doesn't get around to finishing it on time, an automatic email could be sent reminding them that they haven't submitted, and letting them know they have 7 more days to finalize their submission and that it will be deleted if not. They could also be given the option to: 1) submit what they have completed so far, or 2) "go ahead and give my batch to someone else now." Then they have a reminder and will likely finish up sooner than if the batch is re-assigned to someone else, or if they won't be able to finish, at least only the pages they haven't already indexed would need to be done by somebody else instead of the entire batch, since they have the option to submit what is completed so far.

Making use of partially-completed batches (along with auto-reminders) would reduce duplication of effort (and keep slow-pokes like me from feeling like their sacrifice of time was wasted if they didn't finish the entire batch in a week).

Alternatively, the user could be asked how many pages he/she wants in the batch. Then if they only have a little time, they can request only a page or two.

Thanks.

-DougK

I know you can request a batch from a certian project and I know that some projects divide the work into smaller groups, you might want to try finding groups that have smaller batches. As for the reminders, I use a calendar program to keep me on task.

I think the problem could be easily solved just by "pinging" the user in some way (e-mail?) and looking for a response from them indicating they want more time. The whole point of reassigning batches as far as I can tell is not because the person is too slow but because they don't want it to fall through the cracks. If someone can respond to a query far enough to say "hey, I'm still working on it", then leaving it with them is arguably just as efficient as reassigning it (unless of course they've been on the same batch for a year and all the other batches are done so we've got idle people everywhere wishing they could finish that batch...).

E-mail would be nice since I know all of us check that at least twice a day and maybe a lot more than that.

The ability to "renew" your batch would be great too since I've gotten real close to my batches expiring too.

I just think that personal reminders also work well. I use Outlook 07 and it places your e-mail, tasks and appointments all there for your viewing. Really, nice and keeps me up and on task. (I also never forget about the batch during the week, I just keep on putting it off in place of homework.)

Note: Outlook 07 is a for-profit program and neither the author or the church sponsers it.