p*******m
发帖数: 24 | 1 here is an example.
BBC claimed:
Will Smith’s father sent in a vote in 2017 city elections, but died the day
before election day
That vote was counted, in accordance with city rules, but there is no record
of any vote by him afterwards
https://twitter.com/bbcworld/status/1326504252777390080
sounds good right? but the problem is
Smith’s father died of cancer, on Nov 7, 2016.
https://www.cheatsheet.com/entertainment/will-smith-on-fathers-death-
released-me-to-be-me.html/
Still believe those fake news? |
p*******m
发帖数: 24 | 2 internet never forgets, fake news BBC!
day
record
【在 p*******m 的大作中提到】
: here is an example.
: BBC claimed:
: Will Smith’s father sent in a vote in 2017 city elections, but died the day
: before election day
: That vote was counted, in accordance with city rules, but there is no record
: of any vote by him afterwards
: https://twitter.com/bbcworld/status/1326504252777390080
: sounds good right? but the problem is
: Smith’s father died of cancer, on Nov 7, 2016.
: https://www.cheatsheet.com/entertainment/will-smith-on-fathers-death-
|
I****s
发帖数: 1167 | 3 看看tucker已经为把一个死人投票搞错的道歉了事，还被老川promote。
https://www.newsweek.com/trump-promotes-false-fox-news-story-about-dead-
voters-that-tucker-carlson-has-already-apologized-1548230
President Donald Trump on Tuesday promoted a debunked Fox News opinion story
about dead election voters days after host Tucker Carlson apologized and
retracted claims in the report. |
p*******m
发帖数: 24 | 4 BBC is fake news, and you read those crap think they are real.
story
【在 I****s 的大作中提到】
: 看看tucker已经为把一个死人投票搞错的道歉了事，还被老川promote。
: https://www.newsweek.com/trump-promotes-false-fox-news-story-about-dead-
: voters-that-tucker-carlson-has-already-apologized-1548230
: President Donald Trump on Tuesday promoted a debunked Fox News opinion story
: about dead election voters days after host Tucker Carlson apologized and
: retracted claims in the report.
|
R******6
发帖数: 609 | 5 “Fake news”是指机构，不是指一条新闻。比如CNN再没底线，通俄门被共和党人踢爆
这件事，它也如实报的。但是不影响CNN是fake news
新闻机构故意说几次谎，就是fake news了，不管它以后说什么。That’s why CNN is
being sold right now. |
I****s
发帖数: 1167 | 6 Tucker搞错了死人投票并为此道歉这个事也是fake news吗？ 老川还在twitter上
promote这个谣言，是fake news吗？
【在 p*******m 的大作中提到】
: BBC is fake news, and you read those crap think they are real.
:
: story
|
I****s
发帖数: 1167 | 7 Same applies to 老川和川粉们。
爆这件事，它也如实报的。但是不影响CNN是fake news
is being sold right now.
【在 R******6 的大作中提到】
: “Fake news”是指机构，不是指一条新闻。比如CNN再没底线，通俄门被共和党人踢爆
: 这件事，它也如实报的。但是不影响CNN是fake news
: 新闻机构故意说几次谎，就是fake news了，不管它以后说什么。That’s why CNN is
: being sold right now.
|
R******6
发帖数: 609 | 8 BBC只是程度不同，都缺乏credibility |
a****1
发帖数: 634 | 9 这就有点鸡蛋里挑骨头了。这件事有什么好 Fake 的？有什么特别的政治利益？或者特
别给什么人打掩护？尤其是个英国媒体，没有什么特殊原因有什么好 Fake 的？可能只
是报道错误，不必因为川普仇视媒体，你就要被川普牵着走 |
p*******m
发帖数: 24 | 10 https://www.usatoday.com/story/entertainment/tv/2020/11/15/fox-news-host-
tucker-carlson-apologizes-false-claim-voter-fraud/6303120002/
you are misleading here again. Tucker said
A BUNCH of dead people voting except one, who he got wrong. It does not
dispute the fact that there are dead people voting.
【在 I****s 的大作中提到】
: Tucker搞错了死人投票并为此道歉这个事也是fake news吗？ 老川还在twitter上
: promote这个谣言，是fake news吗？
|
|
|
I****s
发帖数: 1167 | 11 看看bbc是怎么debunk 死人vote的. 但是就怕认真。
https://www.bbc.com/news/election-us-2020-54874120
10,000 'dead absentee voters' in Michigan?
The story starts with a list of around 10,000 names posted on Twitter by a
Trump-supporting activist.
It purports to be of people who have died, but who have also voted in the
presidential election in Michigan.
Claims such as this have been repeated many times on different social-media
platforms, including by Republican legislators.
The list of 10,000 contains the name, zip code, and the date a ballot was
received. It then lists a full date of birth and a full date of death. Some
of the people supposedly died more than 50 years ago.
Michigan has a database that lets you enter someone's name, zip code, month
of birth and year of birth and allows you to see if they voted by absentee
ballot this year. So you can easily check whether people on the list voted.
There are also several US websites that include databases of death records.
But there's a fundamental problem with this list of 10,000.
With an exercise like this you are going to find false matches - somebody
born in January 1940 voted in Michigan in the election, and there was
somebody born somewhere else in the US in January 1940 who has the same name
and is now dead. This will happen a lot in a country as big as the US (328
million people), and particularly with common names.
To test the list, we picked 30 names at random. To this we added the oldest
person on the list.
Of this list of 31 names, we managed to speak directly to 11 people (or to a
family member, neighbour or care home worker) to confirm they were still
alive.
For 17 others, there was no public record of their death, and we found clear
evidence that they were alive after the alleged date of death on the list
of 10,000. A clear pattern emerged - the wrong records had been joined
together to create a false match.
Finally, we found that three people on the list were indeed dead. We examine
these cases later.
What we discovered
The first thing we did was to check the official Michigan electoral database
to see whether our 31 individuals had sent in ballots - they all had.
We then looked at the death records and quickly became suspicious on seeing
that the vast majority did not die in Michigan, but elsewhere in the US.
We wondered whether we could find people of the same name currently living
in Michigan.
Checking Michigan state public records, cross-referencing voter postal codes
, we were able to find precise dates of birth for those who had voted - and
as we had anticipated, they failed to match the dates of birth on the death
records.
So we could be confident that we were dealing with two sets of people -
those who had voted and those with the same name and age who had died
elsewhere.
But what we really wanted to do was to speak to the voters themselves.
'I'm alive!'
We called Roberto Garcia, a retired teacher in Michigan. He told us: "I'm
definitely alive and I definitely voted for Biden - I would have to have
been dead to vote for Trump."
We also found a 100-year-old woman who, according to the "dead voter" list,
had died in 1982. She was alive and is currently living in a nursing home in
Michigan.
But the results of our search weren't always so straightforward.
When we looked for another centenarian, who according to the list had died
in 1977, we found that she had still been alive when her postal ballot was
returned in September. However, a neighbour told us the woman had died just
a few weeks ago. We also found a matching obituary from October to confirm
this.
If a voter dies before election day after submitting their ballot, the
Michigan authorities say the ballot will be rejected .
We have not been able to establish whether her ballot was counted.
For those we couldn't reach by phone, we wanted to use other means to
confirm they were alive.
These included public records of, for example, business activities, from
state and local authorities.
For one woman who was supposed to have died in 2006 we found an annual
company statement signed under her name from January 2020.
Two other men on our list of 31 died some time ago, yet votes had been cast
in their names - with the correct postcodes and years of birth - according
to the voting database.
We found that for both men, there were sons with the same name currently
registered at the same address as their deceased fathers.
In both cases, a ballot was sent in for the dead fathers.
Local election officials told us that one of the votes had been counted but
there was no record of the son having voted.
In the other, it was the son who actually voted, but it had been recorded as
the father's due to a clerical error.
It's simply a matter of statistics'
Our selection of 31 cases is only a small sample of the 10,000 names on the
list, but it has clearly revealed the flaws in the database shared by Trump
supporters.
From our investigation it's clear that in almost all of our 31 test cases,
the data for genuine voters in Michigan has been combined with records of
dead people with the same name and birth month and year from across the
United States to yield false matches.
"If the lists are linked based on name and birth date alone, in a state the
size of Michigan, you're guaranteed to get false positives," says Prof
Justin Levitt, an expert on the law of democracy.
It's known as the birthday problem - the high probability that two students
in the same class share the same birthday.
So if you compare millions of voters in Michigan with a database of deaths
across the United States you're bound to find cross-over, particularly if
the voter database doesn't include the day of the month on which a person is
born.
"It's simply a matter of statistics that if you cross-reference millions of
records with millions of other records, you'll get a sizable number of false
positive matches. We've seen this before," says Prof Justin Levitt.
With her vote safely cast, and counted, Maria Arredondo tells us she's
looking forward to the new administration.
"He was a great vice-president under Obama. I'm so pleased. A weight has
lifted off my shoulders." |
p*******m
发帖数: 24 | 12 again, you can cling to your fake news BBC, even I have already shown you
how fake they are and i don't really care what you believe.
media
【在 I****s 的大作中提到】
: 看看bbc是怎么debunk 死人vote的. 但是就怕认真。
: https://www.bbc.com/news/election-us-2020-54874120
: 10,000 'dead absentee voters' in Michigan?
: The story starts with a list of around 10,000 names posted on Twitter by a
: Trump-supporting activist.
: It purports to be of people who have died, but who have also voted in the
: presidential election in Michigan.
: Claims such as this have been repeated many times on different social-media
: platforms, including by Republican legislators.
: The list of 10,000 contains the name, zip code, and the date a ballot was
|
I****s
发帖数: 1167 | 13 bbc认真的fact check川粉提供的10000个死人vote的名单，在你眼里就是fake news。
明白了...
【在 p*******m 的大作中提到】
: again, you can cling to your fake news BBC, even I have already shown you
: how fake they are and i don't really care what you believe.
:
: media
|
p*******m
发帖数: 24 | 14 fake news seriously fact checked trump russian collusion for 4 years, so
what?
【在 I****s 的大作中提到】
: bbc认真的fact check川粉提供的10000个死人vote的名单，在你眼里就是fake news。
: 明白了...
|
R******6
发帖数: 609 | 15 人自己都说就sample check了31个，到你这就是“ 认真的fact check川粉提供的10000
个死人vote的名单”，你更不靠谱好不好
【在 I****s 的大作中提到】
: bbc认真的fact check川粉提供的10000个死人vote的名单，在你眼里就是fake news。
: 明白了...
|
I****s
发帖数: 1167 | 16 从那个1万人随机抽出了31个人，通过对这些人的调查，证明大部分人都是活着的，并
且找出了川粉数据科学家提供的死人投票名单的问题。川粉数据科学家写sql不太会做
join，缺了几个关键的join clause，所以这个名单就是一个谣言。
10000个死人vote的名单”，你更不靠谱好不好
【在 R******6 的大作中提到】
: 人自己都说就sample check了31个，到你这就是“ 认真的fact check川粉提供的10000
: 个死人vote的名单”，你更不靠谱好不好
|
R******6
发帖数: 609 | 17 哎呦，数据科学家、sql join clause都出来了，说的跟你参与了一样，看的“真真”
的吧……接着编
【在 I****s 的大作中提到】
: 从那个1万人随机抽出了31个人，通过对这些人的调查，证明大部分人都是活着的，并
: 且找出了川粉数据科学家提供的死人投票名单的问题。川粉数据科学家写sql不太会做
: join，缺了几个关键的join clause，所以这个名单就是一个谣言。
:
: 10000个死人vote的名单”，你更不靠谱好不好
|
I****s
发帖数: 1167 | 18 我贴的那个文章里写的清清楚楚，除非你看不懂英文或者不懂数据库。
It's simply a matter of statistics'
Our selection of 31 cases is only a small sample of the 10,000 names on the
list, but it has clearly revealed the flaws in the database shared by Trump
supporters.
From our investigation it's clear that in almost all of our 31 test cases,
the data for genuine voters in Michigan has been combined with records of
dead people with the same name and birth month and year from across the
United States to yield false matches.
【在 R******6 的大作中提到】
: 哎呦，数据科学家、sql join clause都出来了，说的跟你参与了一样，看的“真真”
: 的吧……接着编
|
R******6
发帖数: 609 | 19 你怎么说的啊？Sql、join clause、数据科学家怎么出来的？
the
Trump
【在 I****s 的大作中提到】
: 我贴的那个文章里写的清清楚楚，除非你看不懂英文或者不懂数据库。
: It's simply a matter of statistics'
: Our selection of 31 cases is only a small sample of the 10,000 names on the
: list, but it has clearly revealed the flaws in the database shared by Trump
: supporters.
: From our investigation it's clear that in almost all of our 31 test cases,
: the data for genuine voters in Michigan has been combined with records of
: dead people with the same name and birth month and year from across the
: United States to yield false matches.
|
I****s
发帖数: 1167 | 20 看来你真的不懂数据库。算了，我再怎么说也是对牛弹琴。
【在 R******6 的大作中提到】
: 你怎么说的啊？Sql、join clause、数据科学家怎么出来的？
:
: the
: Trump
|
|
|
R******6
发帖数: 609 | 21 懂数据库就有权利添油加醋的编吗？
【在 I****s 的大作中提到】
: 看来你真的不懂数据库。算了，我再怎么说也是对牛弹琴。
|
I****s
发帖数: 1167 | 22 你如果懂一些relational数据库的知识，就会知道我是不是在编。你可以问问你们川粉
数据科学家。比如那个dime同学。
【在 R******6 的大作中提到】
: 懂数据库就有权利添油加醋的编吗？
|
R******6
发帖数: 609 | 23 这么说，你看见人写sql了？
【在 I****s 的大作中提到】
: 你如果懂一些relational数据库的知识，就会知道我是不是在编。你可以问问你们川粉
: 数据科学家。比如那个dime同学。
|
h**********c
发帖数: 4120 | 24 How to quantify 证明大部分人.
Do you mean all of them are still alive?
Or, still there are some of them are dead.
【在 I****s 的大作中提到】
: 从那个1万人随机抽出了31个人，通过对这些人的调查，证明大部分人都是活着的，并
: 且找出了川粉数据科学家提供的死人投票名单的问题。川粉数据科学家写sql不太会做
: join，缺了几个关键的join clause，所以这个名单就是一个谣言。
:
: 10000个死人vote的名单”，你更不靠谱好不好
|
I****s
发帖数: 1167 | 25 不用我亲眼看见。这1万个名单明显就是用sql把两个表做join出来的，只不过join的不
够好，所以把很多活人当成了死人。这种水平的sql根本没法通过ds或者analyst的面试。
【在 R******6 的大作中提到】
: 这么说，你看见人写sql了？
|
R******6
发帖数: 609 | 26 既然不是你亲眼所见，那不就是你脑补出来的吗，一个字“编”吗
试。
【在 I****s 的大作中提到】
: 不用我亲眼看见。这1万个名单明显就是用sql把两个表做join出来的，只不过join的不
: 够好，所以把很多活人当成了死人。这种水平的sql根本没法通过ds或者analyst的面试。
|
I****s
发帖数: 1167 | 27 31个人里有28个都还活着。两个确实死了的，是跟他们同名的儿子投了票。
【在 h**********c 的大作中提到】
: How to quantify 证明大部分人.
: Do you mean all of them are still alive?
: Or, still there are some of them are dead.
|
I****s
发帖数: 1167 | 28 你去问问dime就知道我是不是在编了。技术上有些东西一看就知道怎么回事。不懂技术
的人会产生这样的疑问，我不怪你。
【在 R******6 的大作中提到】
: 既然不是你亲眼所见，那不就是你脑补出来的吗，一个字“编”吗
:
: 试。
|
h**********c
发帖数: 4120 | 29 刑事案件司法部，跨州联邦调查局都可以介入。看SQL不是什么难事。
FBI welcomes tip in.
周末或感恩节的话，我可以研究一下。做了半辈子财务IT，对这个比较感兴趣
试。
【在 I****s 的大作中提到】
: 不用我亲眼看见。这1万个名单明显就是用sql把两个表做join出来的，只不过join的不
: 够好，所以把很多活人当成了死人。这种水平的sql根本没法通过ds或者analyst的面试。
|
I****s
发帖数: 1167 | 30 你可以问问dime同学，前一阵他下载了类似的数据，在做分析。看看他有没有什么结果。
至于这个川粉提供的所谓1万个死人投票的名单，已经被证明是false了。我贴的文章里
写的很清楚他错在哪里了。你仔细读一下就应该知道。
tip in.
【在 h**********c 的大作中提到】
: 刑事案件司法部，跨州联邦调查局都可以介入。看SQL不是什么难事。
: FBI welcomes tip in.
: 周末或感恩节的话，我可以研究一下。做了半辈子财务IT，对这个比较感兴趣
:
: 试。
|
|
|
R******6
发帖数: 609 | 31 我不用问任何人，也知道你说的有鼻子有眼，都是你脑补的结果，你也认同啊
【在 I****s 的大作中提到】
: 你去问问dime就知道我是不是在编了。技术上有些东西一看就知道怎么回事。不懂技术
: 的人会产生这样的疑问，我不怪你。
|
I****s
发帖数: 1167 | 32 随你。希望你对川粉们传播或者编造的谎言也能有这样的态度就好。
【在 R******6 的大作中提到】
: 我不用问任何人，也知道你说的有鼻子有眼，都是你脑补的结果，你也认同啊
|
h**********c
发帖数: 4120 | 33 那你结论是里面完全没有死人，还是最少有1人已经死亡?我记得FBI说你不需要证明，
just tip in.
果。
【在 I****s 的大作中提到】
: 你可以问问dime同学，前一阵他下载了类似的数据，在做分析。看看他有没有什么结果。
: 至于这个川粉提供的所谓1万个死人投票的名单，已经被证明是false了。我贴的文章里
: 写的很清楚他错在哪里了。你仔细读一下就应该知道。
:
: tip in.
|
R******6
发帖数: 609 | 34 又一个脑补的，五毛们也是拼了
【在 h**********c 的大作中提到】
: 那你结论是里面完全没有死人，还是最少有1人已经死亡?我记得FBI说你不需要证明，
: just tip in.
:
: 果。
|
h**********c
发帖数: 4120 | 35 你不要打茬，和你正经说话。还是说你没有结论。
【在 R******6 的大作中提到】
: 又一个脑补的，五毛们也是拼了
|
R******6
发帖数: 609 | 36 你们继续，脑补聊天欢乐多！
【在 h**********c 的大作中提到】
: 你不要打茬，和你正经说话。还是说你没有结论。
|
s*******9
发帖数: 1559 | 37
sb, 英国媒体跳得特别欢,
之前几次黑川普都是所谓英国情报来源
【在 a****1 的大作中提到】
: 这就有点鸡蛋里挑骨头了。这件事有什么好 Fake 的？有什么特别的政治利益？或者特
: 别给什么人打掩护？尤其是个英国媒体，没有什么特殊原因有什么好 Fake 的？可能只
: 是报道错误，不必因为川普仇视媒体，你就要被川普牵着走
|
I****s
发帖数: 1167 | 38 我的结论是这个川粉做数据处理的方法有漏洞。sql join的condition不够严谨。所以
他提供这1万个死人投票的名单是false。
至于是不是都是活着，我不知道。文章里说到随机抽查的31人里，28个都还活着，
false positive比率已经达到90%了。
，just tip in.
【在 h**********c 的大作中提到】
: 那你结论是里面完全没有死人，还是最少有1人已经死亡?我记得FBI说你不需要证明，
: just tip in.
:
: 果。
|
h**********c
发帖数: 4120 | 39 I think probably FBI will have more sober judgement than us, particularly
involving multile states jurisdiction.
Thanks,
【在 I****s 的大作中提到】
: 我的结论是这个川粉做数据处理的方法有漏洞。sql join的condition不够严谨。所以
: 他提供这1万个死人投票的名单是false。
: 至于是不是都是活着，我不知道。文章里说到随机抽查的31人里，28个都还活着，
: false positive比率已经达到90%了。
:
: ，just tip in.
|
E******k
发帖数: 1 | 40 英国的世界地图中，毛子是最大的敌人，中国是第三大经济伙伴。至于香港是个小波折
，看看怎么着地而已。
朗普孤立主义对英国极其危险，拜登上台才真正符合英国利益。 |
|
|
R******6
发帖数: 609 | 41 BBC的通俄门报道，你当时信吗？现在这么个媒体随手写一篇，你就据此脑补一大套，
这姿势……啧啧……
【在 I****s 的大作中提到】
: 我的结论是这个川粉做数据处理的方法有漏洞。sql join的condition不够严谨。所以
: 他提供这1万个死人投票的名单是false。
: 至于是不是都是活着，我不知道。文章里说到随机抽查的31人里，28个都还活着，
: false positive比率已经达到90%了。
:
: ，just tip in.
|
