(Chess) Life Returning To Normal

[John Upham]

question their dodgy data.

I'm curious to learn of the dodgy data.

Presumably Nick has relevant concrete examples of these.

[Stewart Reuben]

First Saturday Tournaments in Budapest. They take place every month, except January, every year, including 2020 and 2021. APA internationals in rated sections.
Biel, Switzerland is taking place this summer.
FIDE are supporting event this year. This includes Caplin Hastings, that is not finalised, but will probably start 28 or 29 Dec.

[Chris Goodall]

"The roadmap, which has now been published on gov.uk, outlines four steps for easing restrictions. Before proceeding to the next step, the Government will examine the data to assess the impact of previous steps.

This assessment will be based on four tests:

* The vaccine deployment programme continues successfully.
* Evidence shows vaccines are sufficiently effective in reducing hospitalisations and deaths in those vaccinated.
* Infection rates do not risk a surge in hospitalisations which would put unsustainable pressure on the NHS.
* Our assessment of the risks is not fundamentally changed by new Variants of Concern."

In other words, if it's clear that significant numbers of people have refused the vaccine, it's going to make the government think twice about relaxing the rules.

(Not that chess players are the type to be concerned about abstract "rules", but it's just nice to be on the right side of them anyway.)

[David Gilbert]

What is the latest known list of tournaments, nationally or in Europe, starting from mid-May (where many restrictions are expected to be lifted)?
What is the situation in Spain, Germany, France, Czech, etc..?

I only know of:

• 4NCL congress in July
• Golders Green in July
• Northumbria in August (not confirmed)

Don't forget the four ACO Tournaments

11-20 May 2021 Crete The delayed 2020 Open and Seniors (7 sections for each)
17-26 July 2021 Rhodes 2021 Open (7 sections)
1-10 October 2021 Crete 2021 Seniors (7 sections)

The May date appears to be out of the question for English players. We won't be released for holidays or non-essential travel abroad until 17 May 2021 at the earliest. However, the July and October dates do look well set at this time. Currently there's no requirement for visitors to Greece to provide evidence of a vaccination, but since last week the Greek government has been issuing digital vaccination certificates to its own citizens who have had their second dose. It's conceivable that foreign visitors might need to produce some proof in future, or face ten days in quarantine at their own expense instead.

[Paul McKeown]

question their dodgy data.

I'm curious to learn of the dodgy data.

Presumably Nick has relevant concrete examples of these.

He doesn't do evidence. A Brexiter, climate change denier and an anti-vaxxer. Probably disputes quantum mechanics ("bunch of clever clever krauts talking nonsense"), general relativity ("What does Einstein know?") and natural selection ("Charles Darwin, nothing but a middle class liberal laughing at the man in the street"). Probably still supports the Steady State Theory ("proper English astronomy.") He sneers at all expertise except the armchair variety, and has no need of evidence, sticking a finger in the air and spouting any old tabloid prejudice with an ex cathedra certainty is sufficient.

Take the following couple of posts, for example, which did not age well:

Re: Worrying times

Post by NickFaulks » Thu Mar 26, 2020 3:31 pm

I shall be very surprised if it is as high as 0.1%, if properly calculated.

Re: Worrying times

Post by NickFaulks » Fri Mar 27, 2020 7:37 pm

Paul McKeown wrote: ↑
Fri Mar 27, 2020 5:25 pm
But we have had the benefit of a Michael Gove tribute act leading to a defiant statement that the death rate will be less than 0.1% of the infected

I've made a note to return to this when it's over. If there are more than 50,000 excess deaths, you win.

[Chris Goodall]

He sneers at all expertise except the armchair variety, and has no need of evidence, sticking a finger in the air and spouting any old tabloid prejudice with an ex cathedra certainty is sufficient.

When has the Argument From My Opponent Believes Something ever failed?

[JustinHorton]

He doesn't do evidence. A Brexiter, climate change denier and an anti-vaxxer. Probably disputes quantum mechanics ("bunch of clever clever krauts talking nonsense"), general relativity ("What does Einstein know?") and natural selection ("Charles Darwin, nothing but a middle class liberal laughing at the man in the street"). Probably still supports the Steady State Theory ("proper English astronomy.") He sneers at all expertise except the armchair variety, and has no need of evidence, sticking a finger in the air and spouting any old tabloid prejudice with an ex cathedra certainty is sufficient.

I am not clear why you need to set up all these probablies. What do you gain from it except the weakening of your argument?

[NickFaulks]

question their dodgy data.

I'm curious to learn of the dodgy data.

Presumably Nick has relevant concrete examples of these.

Where to start? A good place might be the Imperial College "mathematical model" upon which the Government relied for its decisions a year ago - and possibly still does.

After months of questions and FoI requests the code was finally released, and as anticipated it turned out to be juvenile hackwork. Even then, it had in the interim been handed to professional coders ( at Microsoft, from memory? ) for a couple of weeks and they did their best to make it presentable. Heaven knows what the original version looked like.

[Wadih Khoury]

question their dodgy data.

I'm curious to learn of the dodgy data.

Presumably Nick has relevant concrete examples of these.

Where to start? A good place might be the Imperial College "mathematical model" upon which the Government relied for its decisions a year ago - and possibly still does.

After months of questions and FoI requests the code was finally released, and as anticipated it turned out to be juvenile hackwork. Even then, it had in the interim been handed to professional coders ( at Microsoft, from memory? ) for a couple of weeks and they did their best to make it presentable. Heaven knows what the original version looked like.

I was not aware of this topic, thank you for raising it. I therefore did some research and found the following summary:
https://www.nature.com/articles/d41586-020-01685-y

But the main findings is that the code was reproducible, and was working for all intent on its goal. However, it was poor by programming standards, which is to be expected since these are researchers, not programmers doing the coding. In short, the model did work as expected (which makes sense given that we have more than 100k deaths even with 2 lockdowns. Without any intervention, it would not have been surprising to have a much higher toll).

Some excerpts:

When a cleaned-up version was released at the end of April, software engineers disparaged its quality and said the simulation needed to be repeated by others
....
Media articles cast further doubt on the Imperial work by reporting online comments suggesting that other scientists had experienced problems rerunning the code. Nature has now ascertained that these were taken out of context: they related to work done with the Imperial group to ensure that the publicly released code ran correctly in every possible computing environment.
...
Ferguson — who didn’t comment on the criticisms at the time — agrees that the simulation didn’t use current best-practice coding methods, because it had to be adapted from a model created more than a decade ago to simulate an influenza pandemic. There was no time to generate new simulations of the same complexity from scratch, he says, but the team has used more modern coding approaches in its other work. However, none of the criticisms of the code affects the mathematics or science of the simulation, he says.
...
The politicized debate around the Imperial code demonstrates some of the reasons that scientists might still hesitate to openly release the code underlying their work, researchers say: academic programs often have shortcomings that software engineers can pick at
...

British science advisers, however, asked multiple teams to model the emerging pandemic, and they produced results similar to Imperial’s.

....

As part of this effort, they checked that the public and original code reliably produced the same findings from the same inputs.

The RAMP group’s work included a separate effort to test the robustness of the simulation by trying to break it under various operating conditions, says Graeme Ackland, a physicist at the University of Edinburgh, UK. The team involved, including software specialists at Edinburgh and at Europe’s particle-physics laboratory CERN, near Geneva, Switzerland, posted comments on GitHub as they went. It was these comments that newspaper articles erroneously quoted as casting doubt on whether the code could be reproduced.

So thankfully all was working as it should, and the worries did not impact substantially the nature of the models.

[jholyhead]

As a one time computer science researcher and current software engineer I can attempt to put forward a few informed statements re the criticisms of the Imperial model.

1. All research code is garbage from a software engineering perspective
2. Models like this are ultimately maths, implemented in code. It is the maths that is important, not the code, and it is the maths (along with data/params) that makes a study reproducible or not, not the code.
3. Writing portable software is not a skill taught to researchers, not is it something that would be particularly useful on a day to day basis.
4. Models represent only the best approximation of how a system will behave given the data and underlying assumptions available at the time. All models are wrong, but some are less wrong that others. The Imperial model was far less wrong than most.

[David Sedgwick]

He doesn't do evidence. A Brexiter, climate change denier and an anti-vaxxer. Probably disputes quantum mechanics ("bunch of clever clever krauts talking nonsense"), general relativity ("What does Einstein know?") and natural selection ("Charles Darwin, nothing but a middle class liberal laughing at the man in the street"). Probably still supports the Steady State Theory ("proper English astronomy.") He sneers at all expertise except the armchair variety, and has no need of evidence, sticking a finger in the air and spouting any old tabloid prejudice with an ex cathedra certainty is sufficient.

No cause was ever advanced by being gratuitously unpleasant.

I am not clear why you need to set up all these probablies. What do you gain from it except the weakening of your argument?

Quite so, Justin.

[NickFaulks]

1. All research code is garbage from a software engineering perspective

Of course. That is one reason why decisions of massive significance to the whole country should not be based on research code.

When people asked to know more about the model, did the authorities reply "you can't, because the code is garbage"? No they didn't, and I doubt that they even began to understand that it was garbage - it had been sold to them as the gold standard, a claim it would not have occurred to them to challenge.

[jholyhead]

1. All research code is garbage from a software engineering perspective

Of course. That is one reason why decisions of massive significance to the whole country should not be based on research code.

When people asked to know more about the model, did the authorities reply "you can't, because the code is garbage"? No they didn't, and I doubt that they even began to understand that it was garbage - it had been sold to them as the gold standard, a claim it would not have occurred to them to challenge.

I refer you to point 2. Those decisions were based on maths not on code. The code was used to generate some plots and to make playing with variables convenient, but it is the underlying maths that is important. Now, it's possible that they wrote the code, never checked that it worked or that it conformed to the underlying equations and just slapped some PNGs into their Latex file and went for a beer, but if you're going to assume bad faith without reason, then I'd steer clear of bridges, tall buildings and aircraft, as they all rely on modelling far more complex than the Imperial model you're worried about.

[Roger Lancaster]

I thought Nick's preference for taking rat poison rather than a Covid-19 vaccine was somewhat extreme, and I'm not convinced he meant it literally, but some of the verbal assault on him seems well over the top. It seems to me plain stupid to accept every computer-generated forecast as totally reliable. If they were then, to take a simple example, one would expect different forecasts of election results to produce the same projected outcome. That's not to denigrate mathematical modelling, which has a sound theoretical basis in statistics, but to say that it doesn't always get the right answer.

[Chris Goodall]

I thought Nick's preference for taking rat poison rather than a Covid-19 vaccine was somewhat extreme, and I'm not convinced he meant it literally, but some of the verbal assault on him seems well over the top. It seems to me plain stupid to accept every computer-generated forecast as totally reliable. If they were then, to take a simple example, one would expect different forecasts of election results to produce the same projected outcome. That's not to denigrate mathematical modelling, which has a sound theoretical basis in statistics, but to say that it doesn't always get the right answer.

The best science - or the best mathematical modelling - is that which makes the quality of the conclusions most closely approximate the quality of the data.

Give bad data to a bad model and you might get the right answer by sheer luck. Give bad data to a good model and you're guaranteed the wrong answer.