Country analysis of PubPeer annotated articles

Just out of curiosity, after Scihub now an analysis of papers commented at the PubPeer website. Pubpeer is now also screened on a regular basis by Holden Thorp, the chief editor of Science…

Unfortunately I am loosing many records for incomplete or malformed addresses, while some preliminary conclusions can already be made when looking at my world map.

pubpeer.R grey indicates no data, black only a few, red numerous entries.

A further revision will need to include more addresses and also overall research output as a reference.

Continue reading Country analysis of PubPeer annotated articles


CC-BY-NC

AI perpetuating nonsense – the MAD disorder

Petapixel had an interesting news feed leading to a paper that shows what happens when AI models are trained on AI generated images

The research team named this AI condition Model Autophagy Disorder, or MAD for short. Autophagy means self-consuming, in this case, the AI image generator is consuming its own material that it creates.

more seriously

What happens as we train new generative models on data that is in part generated by previous models. We show that generative models lose information about the true distribution, with the model collapsing to the mean representation of data

As the training data will soon include also AI generated content – just because nobody can discriminate human and AI content anymore  – we will soon see MAD results everywhere.


CC-BY-NC

Repressive Toleranz

Aus einem alten, aber längst nicht überholten Beitrag im Deutschlandfunk

Toleranz ist ja in offenen Gesellschaften – in der Theorie – eine Selbstverständlichkeit. […] Vor diesem Hintergrund brandmarkte der Philosoph Herbert Marcuse die „repressive Toleranz“. Oft sei die gefeierte Toleranz in Wahrheit blanke Parteilichkeit im Sinne der Mehrheitsmeinung. So oder so gilt: Die natürliche Grenze jeder Toleranz ist die Intoleranz. In Thomas Manns Zauberberg heißt es: „Toleranz wird zum Verbrechen, wenn sie dem Bösen gilt.“ Und Umberto Eco forderte: „Um tolerant zu sein, muss man die Grenzen dessen festlegen, was nicht mehr tolerierbar ist.“
Als Weichspüler wäre Toleranz in konfliktreicher Zeit in der Tat verschenkt. Wer tolerant sein will, muss die Menschenrechte der Andersdenkenden achten – aber keineswegs jede Überzeugung dulden. Im Gegenteil. Toleranz ist die Lizenz zum zivilisierten Konflikt. Sie ist nicht so sehr ein Wert in sich, sondern die Grundlage, auf der man um Werte streitet.


CC-BY-NC

PubPeer Pearls II

Following part I here are more PubPeer Pearls…

this “ChatGPT”

https://pubpeer.com/publications/CC7BD83B8979D54C5C11F9E3CC61B9

this “XXX Hospital”

https://pubpeer.com/publications/2ACFDF386B4A7703F2A203C667064B#1

this “no data were used”

https://pubpeer.com/publications/E1130B0D1C631F34F40A9CF63CCCCD#1

this “hot environment”

https://doi.org/10.1007/978-3-031-32997-5_7
https://twitter.com/mumumouse/status/1708157290375778631

this “ethics”

https://pubpeer.com/publications/B44F996F0FBD2D2DB60B9EE4D75311

the “mental health”

https://www.pubpeer.com/publications/9351407CB1BAF8BA175220042369CA#7

this “misuse”

https://pubpeer.com/publications/1C2AFBB8147114C1CC37B8F244101A

this “hallucinated

https://pubpeer.com/publications/9A83F00E1BCEBE436797EB829CA46F

this “neon shop”

https://pubpeer.com/publications/61EFC71D08A7338740155D9EF316BA

this “sanctions on Iran”

https://pubpeer.com/publications/F4647FB081B63E5AD30AD3071381F8

this “we collaborated with another group”

https://pubpeer.com/publications/6DA409720C0AB6FB6FA4AA147FB463#2

CC-BY-NC

Übersterblichkeit

Ein Highlight des Wissenschaftsjournalismus in der FAZ zur Übersterblichkeit, nun ja, sie ist variabel

Bei der Wahl dieses Vergleichszeitraums muss man sich aber klarmachen, für welchen Zweck man die Übersterblichkeit berechnet, das heißt, den Einfluss welcher Ereignisse man beobachten will. … Um den Einfluss der Pandemie zu beurteilen, sollte man typische Jahre vor der Pandemie als Referenz nehmen. Doch was ist schon typisch? Hitzewellen, Kälteperioden, neue Grippeviren – all das kann die Sterblichkeit erhöhen. Hier muss man abwägen. Sogar die Länge der Referenzperiode hat Einfluss auf die Übersterblichkeit. Letztes Jahr rechneten Max-Planck-Experten im Fachblatt Population and Development Review vor, dass die Übersterblichkeit in Belgien für das Jahr 2020 mit dem Referenzzeitraum 2010 bis 2019 satte 20 Prozent kleiner ausfällt, als wenn man die Jahre 2017 bis 2019 als Referenz nimmt. In Polen wiederum fiel der Effekt umgekehrt aus.


CC-BY-NC

Falschzitate

“Es gibt Dinge, die so falsch sind, dass nicht einmal das Gegenteil wahr ist.”   G. Krieghofer zu Falschzitaten

1920 schreibt Karl Kraus von Notlügen, “von denen nicht einmal das Gegenteil wahr ist”, aber die  ihm zugeschriebene Verallgemeinerung, “Es gibt Dinge, die so falsch sind, dass nicht einmal das Gegenteil wahr ist”, stammt nicht von Karl Kraus, sondern wurde ihm seit 1976 … unterschoben.

 


CC-BY-NC

Farming and allergy

Hopefully the thread here on “farming and allergy” is now coming to an end. I tried to refute it for many years – last attempt in 2020

the main objection against the farming hypothesis is the interpretation of a negative statistical association as a “protective” effect. Only after thorough exclusion of alternative explanations, this interpretation may be justified.

but  unfortunately missed another paper from the literature that had already been published in 2011, sorry for that.

With environmental exposures aside, at least 2 profound differences between farming and nonfarming families could be a threat to the validity of such investigations. One of these is the long-term selection of traits and genes in favor of the demanding living conditions of farmers because the agricultural lifestyle is often handed down within families. Referring to the “healthy worker effect,” we could expect certain disadvantages to be underrepresented in a farming population, giving rise to the concept of a “healthy farmer effect.”

Also the second point  of Grabenhenrich has never been assessed

The other area expected to be different when comparing farming and nonfarming families constitutes behavioral patterns, lifestyle, and knowledge. For example, “soft factors,” such as health care use, symptom perception, and labeling, as well as access to and interest in health-related information, are most likely to be distributed dissimilarly. This disparity, in turn, could severely affect the response pattern in studies basing their case definition mainly on questionnaires. A temporal shift of these soft factors toward increased cautiousness and awareness has been assumed to contribute to the worldwide increase in symptoms of allergic diseases and, to a lesser extent, to the increase in clinically apparent disease.10, 11 Why should such a change be uniform in all parts of a population? Particular subgroups might be susceptible to catch up more rapidly. This phenomenon could be studied by identifying outliers in otherwise homogenous populations.

in the hope that journalists and politicians will never find it?

Is this “pioneering epidemiology” as judged by MP Dr. Söder?

Promoting a stolen idea (the original was published in German only)  where the source was never cited?

Is it any good science ignoring all objections?

True heroes in epidemiology and public health are Semmelweis, Snow, von Pettenkofer, Doll & Hill, Hesse & Rehn but not a MD who can not even explain an Odds Ratio


CC-BY-NC

Switch off mic during Zoom calls or …

others can use the recording to read what you are typing

This paper presents a practical implementation of a state-of-the-art deep learning model in order to classify laptop keystrokes, using a smartphone integrated microphone. When trained on keystrokes recorded by a nearby phone, the classifier achieved an accuracy of 95%, the highest accuracy seen without the use of a language model.


CC-BY-NC

Wir müssen lernen, überrascht zu sein

„Ich würde über Individuen sagen, dass ein Individuum stirbt, wenn es aufhört, überrascht zu sein. Ich bin jeden Morgen aufs Neue überrascht, wenn die Sonne scheint. Wenn ich eine böse Tat sehe, bin ich nicht indifferent. Ich gewöhne mich nicht an die Gewalt, der ich begegne, ich bin immer noch überrascht. Deshalb bin ich dagegen, deshalb kann ich ihr meine Hoffnung entgegensetzen. Wir müssen lernen überrascht zu sein, nicht uns anzupassen.“

Abraham Joshua Heschel,  genaue Quelle nicht bekannt


CC-BY-NC

What to do when you disagree with your PI

There was a nice thread at X (former Twitter) about interview questions and how to respond. Here are three  question that are answered by Aakash Gupta along with ChatGPT responses.

 

How would you handle a situation where you strongly disagree with the higher-ups’ decision?

  1. I tried to understand their point of view
  2. Then I weighed everything on first principles
  3. I spoke up once more
  4. But ultimately disagreed & committed

As an AI language model, I don’t have personal feelings or opinions, but I can provide you with some advice on how to handle a situation where you strongly disagree with higher-ups’ decisions:

  1. Remain calm and composed: It’s essential to keep your emotions in check when expressing disagreement. Approach the situation with a professional and respectful demeanor.
  2. Understand the rationale: Take the time to understand the reasoning behind the decision. There might be factors or information that you are not aware of, which could influence their choice.
  3. Gather evidence and prepare your argument: If you still believe the decision is not the best course of action, collect data and evidence that support your viewpoint. Having a well-reasoned argument will help you present your case effectively.
  4. Choose an appropriate time and place: Find a suitable moment to discuss your concerns with the higher-ups. Avoid doing so in public or during high-stress situations.
  5. Be open to discussion: Be receptive to their perspective as well. A constructive dialogue can lead to better understanding and, potentially, a more acceptable resolution.
  6. Propose alternative solutions: Instead of just pointing out the problems, suggest alternative approaches that you believe might yield better results.
  7. Consider escalation: If the decision impacts the company’s values, ethics, or legality, and you can’t find common ground, consider escalating the matter through proper channels within the organization.
  8. Accept the decision gracefully (if necessary): In some cases, despite your efforts, the higher-ups may still proceed with their decision. It’s essential to respect their authority and demonstrate professionalism by accepting the outcome gracefully.
  9. Focus on the bigger picture: Remember that organizations function as a team, and decisions may be made based on broader considerations. While you may disagree with a specific decision, maintaining a positive attitude and focusing on the organization’s overall success is crucial.
  10. Learn from the experience: Use the situation as an opportunity for growth. Reflect on how you handled the disagreement and consider what you can do differently in the future.

Ultimately, how you handle the situation will depend on the specific circumstances and your organization’s culture. Always prioritize maintaining a respectful and professional demeanor, even when expressing strong disagreement.

 
Continue reading What to do when you disagree with your PI


CC-BY-NC

Stack Overflow importance declining

newsletter.devmoh.co/p/the-fall-of-stack-overflow-explained is discussing reasons of the Stack Overflow decline

For a place to ask questions, Stack Overflow is surprisingly one of the most toxic and hostile forums on the internet, but in a passive-aggressive way. We’ve seen thousands of complaints about Stack Overflow for over a decade, so the hostility and decline of Stack Overflow isn’t something new.

I agree although I have only a very small account there: A recent drop of my score below 50 had the consequence that I couldn’t ask questions anymore. Funny enough, the score jumped back without any interaction.

Screenshot

CC-BY-NC

Complex Email Search

Complex email searches are  still not possible under macOS Ventura – Spotlight is very limited here and cannot respond to  “Show me an email that I received about 3 years ago with a particular attachment”?

Using an email plugin this is however possible.

Screenshot Email Search

Houdah Spot (38€) may be  life saving here, look for the free trial.


CC-BY-NC