This is Turing Test

This was Alan Turing who broke the Engima code with this machine

The three laws of robotics: 1. Robot may not injure humans or allow harm 2. Must obey orders issued to it by humans unless such orders can harm fellow humans 3. Robot must protect its own existence unless laws 1 and 2

Important for installing npm -g whatever

mkdir ~/.npm-global

npm config set-prefix '~/.npm-global'

export PATH=~/.npm-global/bin:$PATH

(Basically add to the path)

source ~/.bashrc

Functional mental model

Both double vaccination and negative test within 72 hours are not fool-proof that a traveller will not end up with a positive test after arriving at the destination. Breakthrough infections are possible.

The authors discussed the implications of moving away from elimination or hard elimination strategy to a softer containment strategy. At present, elimination strategy is about zero tolerance towards new cases, not so much as tha the total number of cases in the country will be 0. This is also impossible given the emergence of delta variant. The authors argue that to do that, the country first needs to vaccinate every eligible individual. This will shift the "risk perception" of people such that even if deaths occur and infections continue, there will be less anxiety and urgency to act in a way that is strict and hard, rather, covid19 related hospitalisations and deaths would be viewed as "unavoidable" but risks associated with normal life, much as we think about these things in case of lnfluenza. They also argue that once universal vaccination for covid19 is in place, then the government, in order to sustain a zero tolerance policy towards death from covid19 need to implement three things:

• Electronic device based contact tracing and tracking
• Issuing vaccine passports to the fully vaccinated and providing facilities to vaccine passport holders that will not allowed to the non-passport holders
• Mass testing of people for covid19 using saliva and other tests (such as rapid antigen testing but they have not mentioned that either)

Beyond this, they think that the health system and relationship of the government with public health and businesses need to be "overhauled" to some extent. Their advices include:

• Delay an urgent health reform
• create a closer partnership with the businesses such that along the lines of biosecurity and primary industries in a way that businesses are registered that will have implemented minimising covid19
• Increase workforce capacity so that as the country is dependent on overseas health care workers, fast track residency applications and retrain or arrange for training of existing workforces and retired but capable workers
• They have not mentioned but reasonable to understand that this is about minimising attrition of health workforce
• Develop purpose built MIQs rather than depend on ad-hoc MIQ facilities
• Develop a specific Pandemic controlling agency

The authors make several assumptions but not quite clear from what they have written how they address them, although several of their desiderata and suggestions will have bearing on those assumptions. First of all, this is now established that covid19 is largely airborne, and therefore, ventilation and masking have a more prominent role in non-pharmacological intervention that what was considered earlier. But in order to enforce or enable this, individuals need to change their behaviours about precautionary practices and businesses will need to adjust their practices such as ventilation on shop floor and how many people to allow; restaurants may consider to open more open spaced. Second, while they have considered endemicity of covid19, this is open to debate. Endemicity of covid19 implies that the authors have assumed that COVID19 will have reinfections; so far, there is little evidence of re-infections as a dominant mode of transmission of this virus. Besides, as it is now known that extant vaccines do not confer "sterilising immunity", the discussions around "booster vaccinations" are in order. If so, the question of an endemic infection and thereby the premise that there will be a "shift in public risk tolerance" needs careful consideration. Third, the authors have not discussed the potential issues around "long covid", and the need to study the implications of long covid. Albeit, we may be in the throe of the birth of a new speciality in public health and medicine devoted to the study of the causes and consequences of Covid. What the implications of long COVID19 will be is open to speculations.

What was the rationale for accrual of 130 cases has not been explained here. Would be nice to have that explanation. When the study parameters (95% confidence interval, 60% vaccine efficacy, 1% baseline attack rate) were applied to the WHO calculator obtained from here, the sample size turned out to be 86000, and when 80% confidence interval, 60% vaccine efficacy, and 1% baseline rate among unvaccinated) was used to assess sample size, it turned out to be 21000; in any case, the study then would be underpowered unless the rationale of 130 case accrual is made.<br> Why was not the WHO sample size estimator used?

This means that by design the trial would MISS asymptomatic COVID19 cases that nevertheless make up a substantial proportion (about 60% of covid cases). But this was mitigated by taking monthly swabs for Category 2 participants.

This opens up the possibility of bias in the analysis if the same statistician was involved in the data analysis of the efficacy. How was that mitigated?

Why was the decision of per-protocol taken if the study was blinded to the participants and the study personnel BUT unblinded to the analysts?

That still leaves open the role of unblinded data analysts. Why were they not blinded as well or at least made agnostic as to the status of the allocation. How was allocation concealment achieved and why was this not described here?

How did they mitigate self-report bias?

??What does blinding mean here? Double blinded or single blinded?

Again, wild speculation. Popularity is not necessarily as a result of tendency to evoke an emotional response but perhaps the fact that they play a powerful position in society or "famous" in the first place. Merely rousing emotional responses in others would not lead to centrality of anyone's position in a network, will it?

How can online networks with long ties increase disease burden UNLESS the implication is that, they eventually translate to physical contact?

What is the source of this information?

kaplan meier curve

Very useful

They are saying that presenc of antibody is not necessarily the same thing as actual exposure to the virus.

I do not understand the problem I suppose. The professor has taught a class and given lectures. Anyone could listen to those lectures and write down the ideas obtained from the professor in his or her own terms. If copyright is a matter of expression and not ideas, does it not follow that the professor enabled new copyrights to be reserved by the student who attended the lecture, "listened" or "watched" the professor's lectures, perhaps recorded it on a device and then "re-created" in her own way? What else matters in education and teaching if you cannot expect your student to "remember" and "apply" the knowledge you have accrued and then distribute to the students who pay money to come to your class and be graded by you as a professor? Seems the basis of the lawsuit is that another company, cannot profit off the same "words uttered by the professor" on the same topic. Even if that is the case, it makes little sense to me, as the exactness of the specific expression patterns matter little compared with the content of the thought. For disciplines like science (which the professor seems to teach), the truth can be expressed in so many ways.

But what is "normal"?

Five hindrances:

• Sensual urge (sex, food, craving)
• Anger (Rage, dissatisfaction, ...)
• Sloth-torpor (sluggishness, mainly mental fatigue before physical fatigue)
• Excitement (Overactivity, unable to stop, ...)
• Doubt (Do not have confidence, no faith in the course of action, ...)

Seven factors:

• Mindfulness
• Investigation
• Joy
• Energy
• Tranquility
• Concentration
• Equipoise

matches the reference, and on the forward strand ,: matches the reference genome but on the reverse strand

First variant in the genome was found at ...

header, each version slightly different specifications What programs are used in generating the file

Take line wrapping off

b = binary

This is where they go in:

The selection process was quite complex. It assumed:

• particiipants had access to the Internet and had valid email addresses that they accessed
• participants were able to access URLs on a web browser that would work with their site design
• participants were literate enough to read and understand the background of the study and consent process
• Participants had relatively mild depression
• the participants possessed phones
• the participants were wiling to talk to the researchers and were willing to be interviewed (this had the possibility of selection bias)

This is randomized controlled trial and it is the correct type of intervention research design for this question

The purpose of the trial was to study the effectiveness of a stand-alone, unguided, internet-based mindfulness treatment programme for anxiety

Suggests that the funding body had no role to play in the outcomes of this study

Some conflict of interest

Discuss how you will convert these questions relevant to your own field of study.

This is the "first" of a series of articles where David Sackett defined Evidence Based Medicine. If you are from Public Health, how does what he says apply to you?

In this paper, Trisha Greenhalgh has argued that patient care is multidimensional. As a result, if someone tries to measure the extent to which practice is evidence-based, he/she cannot arrive at a true estimate unless all dimensions are captured. Merely focusing on whether one used the results of a randomised controlled trial or meta analysis to prescribe medicine/therapy is not sufficient by itself to justify that evidence based practice was achieved and evidence based practice is not binary.

A great meta analysis tutorial

Some neat pointers

A classic example of eminence versus evidence debate. Use it for classroom teaching. Eminent doctors are rejecting the claims of an evidence generated using case control study

This is not supported.

fatty acid metabolism

hypoglycin A and mcpg toxicity. Overnight no feed and litchi consumtion (preferably unripe?) What is the significance of routine practice of not washing the food before eating and encephalopathy?

LItichi consumption with overnight fasting

What are the possible reasons for abnormal fatty acid metabolism

mahmud shabistari Read the dude's poems.

Cancer research trust

nitrate levels are high, a good opportunity for environmental health risk assessment and conduct epidemioogical studies.

Follow up on this with Alistair and see if you can get a grant for prostrate cancer with the cancer study

Once knowledge disappears into code, how do we retrieve it?

A fantastic piece of writing.

Fascinating story

useful data set

quantitative text data analysis from Berkeley. Useful.

Great list

HEIC A new file format to view jpg and mp4 files. Need an heic converter to jpg

No, he is not a muslim

Awesome tutorial

very well written

skip grams

use globs if you can. Look up glob https://en.wikipedia.org/wiki/Glob_%28programming%29

stop_words package called stop_words snowball stopwords package

polysemes homophones

part of speech tagger

A collection of texts. Corpus = collection of documents

Lecture notes from kenneth benoit

Who would call him visionary?

In the battle between Hindu supremacists and elite egalitarian, 'minority-exploitative' political forces, supremacists won. A new, ''Modified' India is on the wings. Not for everyone.

This will go from bad to worse

Fiction + trolls + pliant news media = Modi returns. This will go from bad to worse.

This was not just his narrative; it was the narrative of everyone in his coterie.

The two India's India and Bharat now come to haunt us. This is the rise of an apparently disenfranchised India. Will it augur well for all of us?

Unequal and inequitable

Consolidation of Hindu supremacist regime.

This will get worse over the next five years!

and more to come

Yet he keeps winning. Why and how?

Sounds more like incumbency

Great person!

isaac study on the prevalence of food allergy among children.

Here it asks to use it from jupyter_notebook_config.py

Which is why you will not see a single profile and if you have two brave browsers installed on two devices, they will not sync. OK.

How to render

provides advanced tricks and tips for jupyter notebooks

• Remember
• Understand
• Apply
• Analyse
• Evaluate
• Create

A document for writing in markdown

The above code solved the problem. The problem was that, every time I opened Brave browser, it showed an error message, unsupported command-line flag: -no-sandbox

Good summary

Follow the instructions. There are two ways to go with this: 1) git clone the browser-extension repo, then cd into src/ folder and then go to the injector.js file and change the settings manually taking from the diff file OR 2) use git apply to change the injector.js file 3) Or use git apply as the

set the permissions for entire google.com

ipfs.io documentation

Read about cli-fi here.

An excellent short essay on clifi

Watch this video: https://www.youtube.com/watch?v=7nIMa7Fq0eo

Very interesting perspective!

Very important! do not miss it!

ISAAC study found food allergy among children in Australia was about 10%, and similar prevalence expected in NZ, and all ethnicity are similar in this respect. No separate estimate for NZ is available

important point.

Strategies that help in moving information from short term to long term memory:

• chunking (tying disparate facts together)
• concept mapping (done best before lessons begin)
• repeated formative assessments
• faded examples All times, avoid cognitive load

Guided practice is where we do:

• We work, our learners in the workshop work
• We use frequent formative evaluations every 5-10 minutes using multiple choice questions and the MCQs are so created that they can tap misconceptions, not just fact checking.
• We hold student's hand and guide
• We do this so that repeated practice will help to store information from short term memory to longer term memory
• Minimalist guidance actually increases the burden of learning new tools

Great tip for desining a lesson

Steps of drawing concept maps:

1. List all concepts
2. Put most abstract concepts on top
3. Place all specific concepts at bottom
4. Link them using verbs
5. The more links the better

This is the principle of chunking. In chunking you relate or connect the disparate pieces of information by forming connections among them, and then that forms the story or the narrative. this is like the 'attic' of Sherlock Holmes.

In a workshop setting repeatedly ask: "what questions do you have?"

Fluid here signifies an easy flow or flexibility.

This is knowledge graphs. Nodes and edges connect disparate pieces of information

Just memorising facts do not matter. You need to have a vault in the mind where you keep pieces of information and then forge links between them.

Several teaching strategies were discussed. Start with reverse design and start with the goals in mind and tailor your teaching to that end. Then, use formative assessments

Reverse-design setting is where you start with the goals and objectives and tailor your teaching to meet these goals. See more ...

See book summary and annotation here

What does the word 'real understanding' mean here? We need to contextualise all information we learn to our own lived experiences. Abstract thoughts are not enough.

Key notes from understanding by design

Notes are here

The summary article about growth and fixed mindsets written from the perspective of high school student teaching but it seems that this can be used for all levels of students and individuals.

A book by Carol Dweck on mindsets.

Interesting

annotations here: Article in pdf

Some more notes about concept maps:

• Each concept is a noun or adjective
• Each connector is a verb
• More general concepts are arranged on the top of the page (so we have a top down or TD arrangement for graphing)
• Highly specific terms are placed in the bottom ## Software for concept mapping:
• Paper pencil (most convenient)
• cmaptools, free and excellent tools
• freemind (mind mapping as well but you can use for concept mapping) See a large list here

Here is a 'coggle' that shows how concept maps work (a very simple concept map just two nodes and one edge)

where you create mental models by linking various pieces of information together. A concept map or a mind map is useful. See also this annotation

check also this hacker noon article

A discussion on the definition of what constitutes a knowledge graph. Looks like a knowledge graph is where different pieces of information are connected to each other based on their ontology.

I set Firefox to open PDF files. When I tried to open PDFs, Firefox repeatedly opened empty tabs. I had to open preferences and click 'preview pdfs in firefox' rather than open pdfs in firefox.

Hack Education Archives

Teaching Machines

What could this be?

Cross link with direct linking

Intriguing. Needs closer reading till I figure out a way where I can export the annotations on individual passages and notes to a text document. So, I'd keep integrating hypothesis and zotero together and write papers using Authorea or Jupyter notebooks using a bibtex formatted bibliography management system.

For this to happen, start with:

1. Open an article in the web browser (Google Chrome or Firefox) that can support pdf.js or epub.js or html if that is available
2. Open hypothesis app, and start annotating at passages (as in this annotation)
3. Add a consistent tag (see the other how_to tagged articles)
4. Then search all related hypothesis annotations and notes on pages using those tags and organise knowledge that way.

This is a great tool for using hypothes.is in the classroom. Select article, ask focused questions, prepare your own answers and let students use this in real time in the classroom.

Overview of data repositories.

This is an excellent use case for teaching and research, section by section, segment by segment. As an extension, this can lead to a new form of scholarly documentation, where mashup and commentary can itself become the basis of peer reviewed contribution.

https://scholar.google.co.nz/scholar?hl=en&as_sdt=0%2C5&q=polygenic+risk+scores+tutorial&btnG=

test

http://bst.sagepub.com/cgi/content/abstract/24/3/188

So, this page lays out the basic idea behind hypothes.is. The page note is an overarching summary, or can add another layer of information of related pages, annotate 'weblinkography', 'bibliography' if it is an article, a set of equations or perhaps even point to something like codes or jupyter notebooks that can flow from a 'page' or a 'document' level. Yet embedded within that document could be other commentaries that are marginal notes. What this system needs is a seamless way to extract these notes per user basis, so that I can collate a set of links strewn across the web with identical tags. For example, if have tagged this note along with other marginal notes I created with tags 'howto', and 'hypothesis_how_to'. These were created as I tried to get to understand this system. If I were able to now extract these various pieces of texts to a single document that would contain a link to this parent document, then I could use this to teach or let others know how to use this system as well, and for my future notes. In theory, the notes system could be used for a thought logging system as one were to read a 'paper' or a webpage, or add comments. Then these comments could be either emailed or 'shared' in some way that would make it useful to write a script (say a python script using Beautiful Soup or somehting) to scrape all comments to a theme based set of plain text, which in turn could be arranged to develop into an annotated bibliography with very little effort. The effort would be to read and annotate on the fly. it could also be used to curate knowledge across the web and bring everything into one place, in this case, hypothes.is user page.

This is more like a summary or a critique or a succinct account of a page that gets annotated. it could be a video site as well, say a youtube site. Or an audio site. It could provide additional information to a page or a related page.

This is step one. Set up a standard set of tags. These tags will then classify resources. Let's say I want to classify all studies on polygenic risk scores on the web. These documents come in the form of HTML, or PDF. If on the top of that, you'd also like to tag something like tutorial, or software to do the job, then write those tags as well when you come across a resource and tag it. So, a list of tags to learn about polygenic risk scores could be something like:

• polygenic risk score
• polygenic_risk_score
• tutorial
• utility
• why_how So, all resources that pertain to polygenic risk scores can now be divided into a set of main document types: some that discuss tutorials and how tos, others discuss the utilisation values and debates. Later, these could be reassembled.

Test

Test

This paper provides a tutorial and a comparison between risk modelling for colorectal cancer

I cannot understand what is meant by open gray?

What is AMSTAR assessment? What is the URL where I can find the information?

What is meant by systematic meta-review

Not 13, but 3 variables!

Another point worth mention, you must have beamer package installed in your texmf tree or if you use linux or macOS, install beamer separately. None of these commands or options will work if beamer itself is not installed. These tutorials are useless if they do not mention these things.

This information is wrong or outdated. The way to insert a default org-exporter is to do M-x-org-export-default and select default when they present for options.

Too complicated!

Good paper

What do you think?

Need to add a code that can produce boxplot for a single variable and state that the code will be

data %>%
 ggplot() +
 geom_boxplot(aes(x = "", y = continuous_variable))

Useful tip to use reveal.js on github

link is broken, the correct link is https://linuxbrew.sh/

A big listicle with plenty of misinformation about prices of the tools, and outdates or broken links. These people do not maintain what they write. Practically none of the tools are good, except for freeplane/freemind to some extent, and I'd say wisemapping.

There is a need for developing better concept/mind mapping tools for regular use.

A good and free tool true, but you cannot connect the nodes and you cannot label the connectors. So, while a good mind mapping tool, practically useless for concept map as well.

Free, fast and easy, but the interface is clumsy. It does not allow for linking of nodes, and it does not allow for labelling of arcs. Useless but a nice diagramming tool.

It is nice and has free options, and for educators. Can be "tweaked" to produce concept maps, but clumsy.

Good tool but NOT free.

Looks good, but too expensive for 20 bucks a month. Useless.

A nice visual editor that works in html but clumsy. Not worth the six dollars a month they are asking. But can be useful for concept mapping, not so much as mind mapping

Anything but. The site did not open after five minutes and I gave up.

Broken link, useless

Not quite. Very ugly and not at all user friendly.

No, it is expensive, there is no free option.

Select short texts >> pass around the class >> ask students to annotate

Theory of planned behaviour by Ajzen; veridical here means truthful. Three factors:

1. What is my attitude to the behaviour?
2. What do others think or I think others think towards my behaviour (normative belief)?
3. How much control I think or I believe I have towards my behaviour or what factors either make it easy or make it difficult for me to conduct my behaviour?

These will determine my intention to actually act my behaviour, and then intention precedes my actual conduct.

A very important note for writing the introduction of a journal article, and I think it is useful for any article for that matter. Essentially four or five points:

• Start with a compelling question, story, data, quote
• Layout the map of the rest of the paper right there in the introduction leading to the next section
• Spell out the main argument in the introduction section, this helps to lay out the rest of the map of the paper together
• Use simple language.
• Use one of the two approaches or horses for courses - either write a rough sketch in the beginning and then rewrite, or write the perfect story right from the beginning. I like the former approach.