Retraction Request to Gill concerning arXiv:1507.00106

Foundations of physics and/or philosophy of physics, and in particular, posts on unresolved or controversial issues

Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Tue Apr 06, 2021 1:53 pm

Richard Gill has written an article which he has posted on arxiv and also preprints.org and who knows were else.

https://www.preprints.org/manuscript/202001.0045/v1
https://arxiv.org/abs/1507.00106

The first version of the article was submitted to arxiv on Jul 1st, 2015 and the latest version there is from January 6, 2020.
I have evidence suggesting that Richard knows these assertions are false or should have known before posting them. I have requested a retraction in the other thread but I want to post it separately here as notice.

The current abstract says the following, which contains several demonstrably false assertions about me:

Richard Gill wrote:In this note, I analyze the code and the data generated by M. Fodje's (2013) simulation programs "epr-simple" and "epr-clocked". They were written in Python published on Github only, initially without any documentation at all of how they worked. Inspection of the code showed that they made use of the detection-loophole and the coincidence-loophole respectively. I evaluate them with appropriate modified Bell-CHSH type inequalities: the Larsson detection-loophole adjusted CHSH, and the Larsson-Gill coincidence-loophole adjusted CHSH (NB: its correctness is conjecture, we do not have proof). The experimental efficiencies turn out to be approximately eta = 81% (close to optimal) and gamma = 55% (far from optimal). The observed values of CHSH are, as they should be, within the appropriately adjusted bounds. Fodjes' detection-loophole model turns out to be very, very close to Pearle's famous 1970 model, so the efficiency is close to optimal. The model has the same defect as Pearle's: the joint detection rates exhibit signaling. The coincidence-loophole model is actually a clever modification of the detection-loophole model. Because of this, however, it cannot lead to optimal efficiency. Later versions of the programs included an explanation of how they worked, including formulas, though still no reference whatever to the literature on the two loopholes which Fodje exploits, not even to the concept of an experimental (i.e., in principle, avoidable) loophole. The documentation available now does make a lot of the "reverse engineering" in this paper superfluous. I plan to incorporate its results in a new paper with a wider focus.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby Joy Christian » Tue Apr 06, 2021 2:12 pm

.
He has also posted that preprint here: https://www.preprints.org/manuscript/202001.0045/v1

I have successfully had him correct some of the lies in one of his papers published in the journal Entropy, of which he is one of the editors: https://www.mdpi.com/1099-4300/22/1/61

It is worth noting that another one of his papers, more relevant to your work, is also published in Entropy and contains many falsehoods: https://www.mdpi.com/1099-4300/22/1/1
.
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Tue Apr 06, 2021 2:19 pm

I'm aware that he tried to publish this article in Entropy also but failed. But it is quite difficult to understand how Richard can claim that I provided no documentation how the code worked, when the github record shows exactly when the documentation was posted, and Richard himself quoted the documentation in discussions about the model?

See for example: viewtopic.php?f=6&t=11&start=20#p113
Note the date of the post. And at the bottom he says "This description is very clear, quite unambiguous" :shock: Then he goes on years later to claim, documentation was non-existent "until now"? :shock:

He says in the abstract that
Later versions of the programs included an explanation of how they worked, including formulas

Later versions? What is the basis for this lie. The above text appeared in version 6 which was published in 2020. Note that Richard himself quoted the text with an explanation of how they worked including formulas back in 2014!!!
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Tue Apr 06, 2021 2:41 pm

In the introduction Richard says the following:

Richard Gill wrote:The programs are written in the Python programming language and are freely available at https://github.com/minkwe/epr-simple and https://github.com/minkwe/epr-clocked. Descriptions are given at https://github.com/minkwe/epr-simple/bl ... /README.md and https://github.com/minkwe/epr-clocked/b ... /README.md. In the original postings, the description
of how the programs worked was non-existent
. Even the comments in the code did not give anything away. However I read the code and reconstructed the formulas so that I could find out what Fodje was doing.

:shock: (see viewtopic.php?f=6&t=11&start=20#p113, and viewtopic.php?f=6&t=61&p=2899#p2899)
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby FrediFizzx » Tue Apr 06, 2021 4:38 pm

FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby gill1109 » Wed Apr 07, 2021 5:49 am

I suggest that Michel writes me an email and tells me what he would like me to change in the present version of my arXiv paper on his simulation model. I did submit the paper to "Entropy" but it was rejected. The referees and editor did not find it interesting. In my memory, when I was first studying his code - which I downloaded from Github - I did not find any description of what the code was doing. But it was nicely written and so I could figure out his model from his code. Now the Github page does contain a short description of the model, so that is fine, and it made my own work even more superfluous than it was to begin with.

Was it there from the start? I don't know.

Michel has known about my arXiv paper for many, many years, so I am surprised he suddenly has problems with it now. But if he does have problems he can just share them with me and I will happily do what I can to fix them.

Michel never explained the connection between what he was doing with the detection loophole and the coincidence loophole. Perhaps he should add some references to that literature on his Github page. I think I made a useful contribution by relating Michel's work to the existing literature. The other contribution of the paper is to show that the versions due to Jan-Åke Larsson of the CHSH inequality which take account of detection rates are not violated. One should use the Bell inequality which is appropriate to the experimental design.
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 6:29 am

gill1109 wrote:I suggest that Michel writes me an email and tells me what he would like me to change in the present version of my arXiv paper on his simulation model. I did submit the paper to "Entropy" but it was rejected. The referees and editor did not find it interesting. In my memory, when I was first studying his code - which I downloaded from Github - I did not find any description of what the code was doing. But it was nicely written and so I could figure out his model from his code. Now the Github page does contain a short description of the model, so that is fine, and it made my own work even more superfluous than it was to begin with.

Was it there from the start? I don't know.

Michel has known about my arXiv paper for many, many years, so I am surprised he suddenly has problems with it now. But if he does have problems he can just share them with me and I will happily do what I can to fix them.

Michel never explained the connection between what he was doing with the detection loophole and the coincidence loophole. Perhaps he should add some references to that literature on his Github page. I think I made a useful contribution by relating Michel's work to the existing literature. The other contribution of the paper is to show that the versions due to Jan-Åke Larsson of the CHSH inequality which take account of detection rates are not violated. One should use the Bell inequality which is appropriate to the experimental design.

I won't write an email. This thread is enough notice. Your paper contains lies about me, and you know the lies. I've also pointed out some of the lies above. If you do not retract it, I will take further action on my own and you may not like the outcome.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 6:54 am

gill1109 wrote: In my memory, when I was first studying his code - which I downloaded from Github - I did not find any description of what the code was doing.

You referenced the documentation I our discussions about the code from the beginning on this forum. I've provided the relevant threads. If your memory is failing, you should stop making factual claims like these. Nevertheless, you must still retract the paper.

Was it there from the start? I don't know.

The threads show that you know they were there from the start.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby gill1109 » Wed Apr 07, 2021 7:13 am

The paper has been corrected. I can't retract an unpublished paper. I'm glad you got around to reading it at last.

Indeed, my memory is not what it was, but I could have sworn that when I first downloaded your Python code there was no explanation on the GitHub pages of how the program worked. As far as I know you still do not connect your model to the already existing detection and coincidence loophole models.
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 7:55 am

gill1109 wrote:The paper has been corrected. I can't retract an unpublished paper.

First of all, the paper is published. Publication is different from peer-review. Just because your preprint is not peer-reviewed does not mean you haven't published it. In fact, you have published it on many pre-print servers. Secondly you can retract a pre-print: https://arxiv.org/help/withdraw.
Indeed, my memory is not what it was, but I could have sworn that when I first downloaded your Python code there was no explanation on the GitHub pages of how the program worked.

This cannot be true.

On Feb 10, 2014 you quoted from the same documentation you claim did not exist. See viewtopic.php?f=6&t=11&start=20#p113 and viewtopic.php?f=6&t=61&p=2899#p2899. I have multiple other examples proving conclusively that you knew the documentation existed and were aware of its contents.

You can also check the GitHub history of the documentation here to see when it first appeared and every change made to it: https://github.com/michel4j/epr-simple/ ... /README.md. You will note that the full documentation was added on Oct 28, 2013. The same day as the "epr-simple" code was published on GitHub. The same can be seen for "epr-clocked" here: https://github.com/michel4j/epr-clocked ... /README.md and anyone can see that the full documentation was included in the original posting of the code on Feb 26, 2014.


Failing memory is not an excuse. Whatever the reason for your misstatements, you have now been made aware of them, therefore it is not just a matter of posting a new version that does not contain the misstatements, The previous versions containing the misstatements, which are already published must be retracted.
I don't care if you submit a new version that does not contain the misstatements, I care that the versions which contain the misstatements are retracted. If they cannot be completely removed, a new cover page must be added explaining that those versions are no longer reliable and providing the reasons. This is possible on all the preprint servers where you have published them. This is the minimum that is required to correct this wrong.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 8:10 am

gill1109 wrote:arXiv writes: Your replacement is scheduled to be announced at Thu, 8 Apr 2021 00:00:00 GMT. The abstract will appear in the subsequent mailing as displayed below,
In this note, I analyze the data generated by M. Fodje's (2013, 2014) simulation programs "epr-simple" and "epr-clocked". They were written in Python and published on Github only. Inspection of the code and program descriptions showed that they made use of the detection-loophole and the coincidence-loophole respectively. I evaluate them with appropriate modified Bell-CHSH type inequalities: the Larsson detection-loophole adjusted CHSH, and the Larsson-Gill coincidence-loophole adjusted CHSH (NB: its correctness is conjecture, we do not have proof). The experimental efficiencies turn out to be approximately eta = 81% (close to optimal) and gamma = 55% (far from optimal). The observed values of CHSH are, as they should be, within the appropriately adjusted bounds. Fodjes' detection-loophole model turns out to be very, very close to Pearle's famous 1970 model, so the efficiency is close to optimal. The model has the same defect as Pearle's: the joint detection rates exhibit signalling. The coincidence-loophole model is actually a clever modification of the detection-loophole model. Because of this, however, it cannot lead to optimal efficiency.

See also
https://researchers.one/articles/20.01.00001
Michel, I posted *preprints* of that paper. It did not pass peer review. It is of almost no interest to anyone, though it should interest you.

I did all that years ago. I'm glad that you have at last got around to reading it. I have fixed the sentences you object to.

See above. You introduced many misstatements in last year's versions. It is not enough to publish a new version. You must retract all the errant versions and if they can't be completely removed, the pre-print archives must add a new cover page indicating that those versions contain misstatements of facts.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby gill1109 » Wed Apr 07, 2021 8:27 am

minkwe wrote:
gill1109 wrote:arXiv writes: Your replacement is scheduled to be announced at Thu, 8 Apr 2021 00:00:00 GMT. The abstract will appear in the subsequent mailing as displayed below,
In this note, I analyze the data generated by M. Fodje's (2013, 2014) simulation programs "epr-simple" and "epr-clocked". They were written in Python and published on Github only. Inspection of the code and program descriptions showed that they made use of the detection-loophole and the coincidence-loophole respectively. I evaluate them with appropriate modified Bell-CHSH type inequalities: the Larsson detection-loophole adjusted CHSH, and the Larsson-Gill coincidence-loophole adjusted CHSH (NB: its correctness is conjecture, we do not have proof). The experimental efficiencies turn out to be approximately eta = 81% (close to optimal) and gamma = 55% (far from optimal). The observed values of CHSH are, as they should be, within the appropriately adjusted bounds. Fodjes' detection-loophole model turns out to be very, very close to Pearle's famous 1970 model, so the efficiency is close to optimal. The model has the same defect as Pearle's: the joint detection rates exhibit signalling. The coincidence-loophole model is actually a clever modification of the detection-loophole model. Because of this, however, it cannot lead to optimal efficiency.

See also
https://researchers.one/articles/20.01.00001
Michel, I posted *preprints* of that paper. It did not pass peer review. It is of almost no interest to anyone, though it should interest you.

I did all that years ago. I'm glad that you have at last got around to reading it. I have fixed the sentences you object to.

See above. You introduced many misstatements in last year's versions. It is not enough to publish a new version. You must retract all the errant versions and if they can't be completely removed, the pre-print archives must add a new cover page indicating that those versions contain misstatements of facts.

I'm not going to request old versions to be removed. Go ahead and do that yourself if you think it's necessary.
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 9:19 am

https://researchers.one/articles/20.01.00001 wrote:Version 2 (2021-04-07)

Abstract
In this note, I analyse the code the data generated by M. Fodje's (2013, 2014) simulation programs epr-simple and epr-clocked. They are written in Python and were published on Github with little documentation. Inspection of the code and program descriptions showed that they made use of the detection-loophole and the coincidence-loophole respectively. I evaluate them with appropriate modified Bell-CHSH type inequalities: the Larsson detection-loophole adjusted CHSH, and the Larsson-Gill coincidence-loophole adjusted CHSH. The experimental efficiencies turn out to be approximately eta = 81% (close to optimal) and gamma = 55% (far from optimal). The observed values of CHSH are, as they must be, within the appropriate adjusted bounds Fodjes' detection-loophole model turns out to be very, very close to Pearle's famous 1970 model, so the efficiency is very close to optimal. The model also has the same defect as Pearle's: the joint detection rates exhibit signalling. The coincidence-loophole model is actually an elegant modification of the detection-loophole model. Because of this, however, it cannot lead to the optimal efficiency.


https://researchers.one/articles/20.01.00001 wrote:PDF v10 (2021-04-07)
In this note, I analyse the code the data generated by M. Fodje’s (2013, 2014) simulation programs “epr-simple”and “epr-clocked”. They are written in Python
and were published on Github with very little documentation. Inspection of the code and program descriptions showed that they made use of the detection-loophole
and the coincidence-loophole respectively.
I evaluate them with appropriate modified Bell-CHSH type inequalities: the Larsson detection-loophole adjusted CHSH, and the Larsson-Gill coincidence-loophole
adjusted CHSH. The experimental efficiencies turn out to be approximately η = 81% (close to optimal) and γ = 55% (far from optimal). The observed values of CHSH
are, as they must be, within the appropriate adjusted bounds. Fodjes’ detection loophole model turns out to be very, very close to Pearle’s famous 1970 model,
so the efficiency is very close to optimal. The model also has the same defect as Pearle’s: the joint detection rates exhibit signalling.
The coincidence-loophole model is actually an elegant modification of the detection-loophole model. Because of this, however, it cannot lead to the optimal efficiency.


Keep digging.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby gill1109 » Wed Apr 07, 2021 9:21 am

You do not refer to the enormous literature on the detection loophole, or on the coincidence loophole.

So what's the problem?

And why don't you publish a paper on your own models?

Why don't we write one together?
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 9:42 am

gill1109 wrote:You do not refer to the enormous literature on the detection loophole, or on the coincidence loophole.

So what's the problem?

And why don't you publish a paper on your own models?

Why don't we write one together?

I won't allow diversion. The issue at hand is your libellous and demonstrably false assertions about me in your paper. You must retract them. You have 6 days left to do it on your own. By doing it on your own you will be acknowledging the error and making an effort to correct the wrong and in the process demonstrating good faith. I am extending you that courtesy. By refusing to take the required action, you are demonstrating bad faith. In fact your edits highlighted above already demonstrate bad faith. In six days, if you have not retracted the versions containing the misstatements, I will take action. This is not a joke https://publicationethics.org/retraction-guidelines.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby Heinera » Wed Apr 07, 2021 9:52 am

In arXiv you can't update a paper and simultaneously delete the old versions. When you update a paper it is implicitly understood that previous versions are withdrawn, even if they are still accessible.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 9:59 am

Heinera wrote:In arXiv you can't update a paper and simultaneously delete the old versions. When you update a paper it is implicitly understood that previous versions are withdrawn, even if they are still accessible.

I don't know why you repeatedly talk about things you know nothing about: https://arxiv.org/help/withdraw
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby gill1109 » Wed Apr 07, 2021 10:03 am

Heinera wrote:In arXiv you can't update a paper and simultaneously delete the old versions. When you update a paper it is implicitly understood that previous versions are withdrawn, even if they are still accessible.

Thanks! Indeed, one cannot rewrite history.

minkwe wrote:
Heinera wrote:In arXiv you can't update a paper and simultaneously delete the old versions. When you update a paper it is implicitly understood that previous versions are withdrawn, even if they are still accessible.

I don't know why you repeatedly talk about things you know nothing about: https://arxiv.org/help/withdraw

Michel, I don't want to *withdraw* the preprint. I have (rapidly) corrected it.

Even if I withdrew it, it would still exist. Too bad.

I propose we get together and write a paper together about your simulation models.

It is bad faith on your part, to accuse me of bad faith.

How about we set up a Zoom meeting to talk over these differences of opinion, face to face?
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby minkwe » Wed Apr 07, 2021 10:35 am

gill1109 wrote:Thanks! Indeed, one cannot rewrite history.

I'm not requesting that you rewrite history. I'm asking you to abide by basic scientific publishing ethics. https://publicationethics.org/retraction-guidelines

Even if I withdrew it, it would still exist. Too bad.
It would exist with a prominent note saying the author has withdrawn the version because it contains errors of material facts. The note would state when it was withdrawn. Then anyone who stumbles on it won't be misled by its erroneous content. The historical record would be preserved.

I propose we get together and write a paper together about your simulation models.

I'm not interested at this point and that is a diversion. I might reconsider if you abide by basic scientific publishing ethics. https://publicationethics.org/retraction-guidelines

How about we set up a Zoom meeting to talk over these differences of opinion, face to face?

This request is not about a difference of opinion. It is about a demonstrable misstatement of facts about me by you. I don't request retractions just because we have differing opinions.
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Retraction Request to Gill concerning arXiv:1507.00106

Postby Heinera » Wed Apr 07, 2021 11:15 am

minkwe wrote:
Heinera wrote:In arXiv you can't update a paper and simultaneously delete the old versions. When you update a paper it is implicitly understood that previous versions are withdrawn, even if they are still accessible.

I don't know why you repeatedly talk about things you know nothing about: https://arxiv.org/help/withdraw

Did you actually read the page you are linking to? Nothing can be deleted from arXiv.

Also: "It is not appropriate to withdraw a paper because it is being updated. Instead you could submit a replacement."

Why should he do a full withdrawal in order to correct a petty point that has no material implications for the contents of the paper? He could even correct the phrase to "published on Github with no references to existing literature". That would certainly be unassailable.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Next

Return to Sci.Physics.Foundations

Who is online

Users browsing this forum: No registered users and 87 guests

cron
CodeCogs - An Open Source Scientific Library