Louis Kessler’s Behold Blog     The Behold User Forum

Thank you, Dana, for reposting your comment on my blog post. Dana originally made this comment on the Facebook group: Genealogy and Artificial Intelligence (AI) when I referred to this blog post. This was my response to Dana: Hi Dana. I took a SLIG course on reading Russian documents a few years ago, and ...

Hi Dave, Yes, you have outlined several of the challenges that AI will have interpreting handwriting. These are all likely one-day solvable, for example with regards to your point on penmanship, an algorithm may have to be trained on the writings of individual people first before it can be generalized. ...

Bob: Personally, I would prefer a 25 year old standard that is used by 98% of the intended audience, than an "updated" standard that obviously doesn't have enough advantages to be able to attract more than a handful of developers to implement after more than 2 years. So something's not working. If ...

Bob (Coret): Main reason is that 5.5.1 is supported by 98% of genealogy vendors, whereas the number who have implemented some form of 7.0 can be counted on one hand. There are many reasons why I don't think 7.0 will ever take hold. See my newest blog post Is Updating the GEDCOM Standard Necessary?

jandrea: Well it is a recursive procedure. The problem is that if you have loops, then the recursive procedure is infinite. So you have to figure out how to stop. The normal stopping method in a recursive procedure is to mark a node (person) as "visited" once they've been checked so that you don't process them ...

Thanks EJ, for your confirmation about the issue. I came up with a new idea that I'm going to try out, which may result in my next blog post.

Thank you k1w1. The link is now fixed.

Elaine: I think in my case, the text was so small that the Hebrew OCR had trouble recognizing the characters. That's why scanning at 600 dpi helped. I suspect if the text would have been larger, the translations would have been much better. It also was difficult for Google, Bing and Yandex to differentiate ...

Steve: Thanks, but I don't have an affiliate link for Shotbox. If you want to purchase a Shotbox, you might still be able to take advantage of their RootsTech special: https://rootstech.shotbox.me/

Yinwang: I compared both by RSID and by position. In one comparison between two files, for example, I noted to myself that 299 RSIDs had changed at the same position, and that 3 RSIDs had different positions. But mentioning that in the article would have been off-topic and would only have complicated what I ...