Re: Ancestry's Drastic Changes Dash Hopes of Finding Connections #dna


Dahn Cukier
 

A few words about computers.

All the former "matches" are still on the computers as are all
non matches. Unless a person has demanded their DNA be removed,
Ancestry should not be removing any results.

We, the users, see only the data deemed relevant to us. I do not see
any of my uncle's wife's relatives, but did find an 5-8th cousin in
common, or so say statistics.

Since Ancestry removed the ability to jump from page to page, I
have not seen as many matches. I tried to scroll down when the new
display first came out about a year ago, but after an hour, I was
no where near the 150,000 matches I saw before the display change.

I would appreciate Ancestry supplying a utility to request a spreadsheet of
matches as MyHeritage does/did in batch form. "Batch" means it is produced during
slow hours and a file is prepared for the user by request that can be downloaded.

With so much endogamy, the results on any database are less than perfect.
I have access to 7 direct relatives raw data DNA at Ancestry. When a new
1st-3rd cousin shows up, I always look at he person from my mother's data,
my father's brother's data and my sister's data. If only I am a
relative, it is most likely a false positive. By looking at 2nd cousins data,
I can find if the person is related to my mother's father or mother, or
my father's father or mother.

As I write this, I begin to suspect that Ancestry may not be so much about
genealogy as connecting living people. As families started to move around
more and more beginning in the 1960s many have lost touch with
2nd generation relatives. While I knew my aunts and uncles, I have never met
many of their 1st cousins.

Dani Cukier
Cukier/Zucker/Zukrowicz, Brif/Brieff, Sklawir/etc. Lisoecki/Lisobitki/etc.


When you start to read readin,
how do you know the fellow that
wrote the readin,
wrote the readin right?

Festus Hagen
Long Branch Saloon
Dodge City, Kansas
(Gunsmoke)


On Friday, September 4, 2020, 05:27:02 AM GMT+3, Teewinot <teewinot13@...> wrote:


I beg to differ. In all my shared matches until about 7 days ago, I had
matches down to 6.0 cM. So did my cousins I've been working with. If
we didn't, we wouldn't have found some of the links between us. I also
wouldn't have found other critical matches in the shared DNA.

I know what shared matches are. I've been using them for the past two
years, since I tested.

I'm a retired medical professional and know about DNA, genetics and
inheritance. In more distant relationships, DNA is more a guide, not an
absolute. Due to the way DNA is inherited, you can have two siblings
with vastly different DNA inheritance from even great great
grandparents. Also, the estimation of relationship can often be way
off. Someone with 8.0 cM could be as close as a 4th cousin or as
distant as an 8th cousin. It all depends on how the DNA was
inherited/passed down. (Ancestry had two of my 1st cousins once removed
listed as 3rd to 4th cousins.)

I paid for the data I was given (down to 6.0 cM). I did not in the
least appreciate it being taken from me without even asking me. I
worked feverishly for the last week to save as many matches as I could
below 8.0 cM. Obviously, everyone else was, too, because the servers
were sluggish, crashing constantly, and even going down completely for
two hours at a time. They hadn't been prior to that. On Sept. 1st,
they were back to normal. I managed to save just under 7,500 matches.
I dread to think of all the valuable data I lost in the matches I
couldn't save.

I personally believe that Ancestry has done all this because they can't
handle the storage of the massive amount of data that is being generated
as more and more people get tested. I also personally believe that what
Ancestry did was disgraceful and just plain bad business. I have never,
in all my years, seen a business take away something from a customer
that they had paid for. If Ancestry wanted to make changes, they should
have started with the new customers as of Sept. 1st, and left alone all
the data of customers who had paid for the service before that date.
They've just made it far harder now to trace links between families.

I had a long talk with someone in the corporate HQ today. He agreed
with me, and is going to look into returning all matches to the shared
DNA. As for the data below 8.0 cM, it's all been dumped. So now it's
just wait and see.

Jeri Friedman

On 9/3/2020 5:54 PM, Adam Turner wrote:

Only the 8.0 cM cutoff is a recent change. The bit about the shared
matches was never announced because it is how AncestryDNA has /always/
worked, at least since I tested.

Your /main match list/ (the screen you are brought to when you click
"DNA matches", which shows all of your matches) showed, and continues to
show, all matches: everyone from "close family", "2nd-3rd cousins,"
"4th-6th cousins" (3500+ cM down to 20.0 cM) to "Distant Cousins" aka
"5th-8th cousins" (20.0 cM down to 8.0 cM). The change that AncestryDNA
made in August is that the cutoff used to go down to 6 cM, and they took
out all matches between 6.0 and 8.0 cM.

The /shared matches tab/ is what you're brought to when you click on the
profile of one of your matches, and are trying to triangulate the
results and see /other /people who match /both/ you and that match. This
section has /always/ had a cutoff of 20.0 cM; you cannot drill past 20.0
cM from this tab within a match's profile. But you could, and still can,
see matches from 8-20 cM in your main match list.

There's an interesting debate to be had on whether revising the cutoff
from 6 cM to 8 cM was beneficial for users (apparently, until 2016, it
used to go down as low as 5 cM), as well as the degree to which some of
these changes are really motivated by the desire to improve accuracy, as
opposed to AncestryDNA's engineers demanding this from within because
their job of running a gigantic match database while minimizing bugs is
hard. (Seen the "our backend servers are overtaxed at the moment"
message lately? I know I have.) But they didn't take 8-20 cM matches out
of Shared Matches; they were never there in the first place.

Adam Turner
--
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Join main@groups.jewishgen.org to automatically receive all group messages.