pacbard

I want to say it's almost there with 4.0. It seems that they re-implementend most features.

For me, this is still the roadblock:

> More features, including support for volume mounts from the host, are planned for Podman v4.1, so stay tuned for more updates.

I do not run many self-contained containers and I usually end up mounting a local folder in the container, so I'll have to wait a little bit longer to fully transition to Podman.

Also, it doesn't have a native interface like docker desktop and it works mostly through the terminal. Last time I checked, there are some external tools that add a Docker Desktop app like on top of podman.

Meta-analysis is an application of idea #4 (Bayesian Multilevel Models) in the article.

What makes meta-analysis special within a multilevel framework is that you know the level 1 variance. This creates a special case of a generalized multilevel model where you leverage your knowledge of L1 mean and variance (from each individual study's results) to estimate the possible mean and variance of the population effect.

The population mean and variance is usually presented in funnel plots where you can see the expected distribution of effect sizes/point estimates given a sample size/standard error.

Researchers have also started to plot actual point estimates from published papers in this plot, showing that most of the published results are "inside the funnel", a result that is usually cited as evidence of publication bias. In other words, the missing studies end up in researchers' file drawers instead of being published somewhere.

p-hacking is a research "dark pattern" where a researcher fits several similar models and reports only the one that has the significant p-value for the relationship of interest.

This strategy is possible because p-values are themselves stochastic and a researcher will find one significant p-value for every 20 models that they run (at least on average).

p-hacking could also refer to pushing a p-value close to the significant cut-off (usually 0.05) by modifying the statistical model slightly until the desired result is achieved. This process usually involves the inclusion of control variables that are not really related to the outcome but that will change the standard errors/p-values.

Another way to p-hack is to drop specific observations until the desired p-value is reached. This process usually involves removing participants from a sample for a seemingly legitimate reason until the desired p-value is achieved. Usually identifying and eliminating a few high leverage observations is enough to change the significance level of a point estimate.

Multiple strategies to address p-hacking have been proposed and discussed. One of the most popular ones is pre-registration of research designs and models. The idea here is that a researcher would publish their research design and models before conducting the experiment and they will report only the results from the pre-registered models. This process eliminates the "fishing expedition" nature of p-hacking.

Other strategies involve better research designs that are not sensitive to model respecification. These are usually experimental and quasi-experimental methods that leverage an external source of variation (external to both the researcher and the studied system, like random assignment to conditions) to isolate the relationship between two variables.

Depending on the type of chestnut, you can "fry" them in a pan with the skin still on. I think that that's the "roasted chestnuts by the fire" thing in the Christmas song. All you have to do is to cut a small slit in the skin (I was told as a kid that otherwise a chestnut would explode but I don't know how true that is as I have never seen one explode myself). After you have scored them, you can just "fry" the chestnuts in a pan on a stove or an open flame. The traditional roasting pans were made of sheet metal and had holes in them to let the heat and flame in to speed up the cooking. I have found that a cast-iron pan cooks them well enough even if it's slower. You know that the chestnuts are cooked when the skin becomes dry and it breaks off easily from the chestnuts. If they are undercooked, you will still see the "fuzzy" second skin sticking to them and they will taste very bitter. Cooked ones will be golden in color and their skins (both the outside one and the fuzzy one) will just peel off. My favorite thing is to give peeled chestnuts to someone you care about while you eat them. They will appreciate it because peeling chestnuts inevitably burns your fingers.

Keep in mind that roasting chestnuts is appropriate only for some types of chestnuts. the roasting kind have a "square" bottom with two long, parallel sides. Wild chestnuts (which usually are smaller and have a bulgy side) do not roast well at all. Instead, you will want to boil them in salt until they are soft. To eat them, you bite and split them in the middle with your teeth and "suck" out the pulp. The pulp is somewhat sweet and not bad. They are good for a snack as they are easy to carry around and last a few days. Roasted chestnuts are only good while warm and they become very hard to eat after they cool off.

If you can make (or find) some chestnut flour, there are recipes for chestnut cake, chestnut pancakes, or fried chestnut dumplings. All of those taste very unique and weird (I never got used to it) and are way worse than anything "modern" made with regular white flour. I really got to eat them when visiting "old school", cheap relatives for Christmas. I heard that people used to make chestnut bread but I never had that.

It does in Italy: https://it.wikipedia.org/wiki/Equo_compenso_in_Italia#Tariff...

This is probably related to teenagers enrolling in college not only for an education but also for the amenities that come with college life (think of greek life, moving away from parents, college sports, freedom to explore your identity, etc.) The pandemic has put a stop on most of all non-academic stuff, making enrolling in college less appealing to this group of students.

It will be interesting to see a follow-up analysis that parses out enrollment behavior by subgroup (e.g., by SAT/ACT score) as it will be easier to understand who is choosing not to enroll in college.

Another follow-up could be to see which institutions are losing students. It is known that college enrollment is counter-cyclical to the economy and that enrollment declines at community colleges and open access universities when people can get a job right out of high school.

The reviewer shows a setting for it. By default, each track loops on its own. You can toggle a setting to link tracks and make them play together (I forget how it’s called).

I wonder if it's because of the quirky Swiss military setup where evertyone is basically enlisted as a reservist. I would imagine that the local village will have monthly drills and that they have to organize them somehow. Using WhatsApp/Telegram/Signal might be useful for them as they might already use them for other purposes and everyone has it installed.

Source: https://en.wikipedia.org/wiki/Conscription_in_Switzerland

The Italian retail sector (i.e., small local shops) is ripe for disruption. The shops pretty much just buy from wholesalers and provide some local inventory in the shop. This usually comes with a high markup. Competition is highly regulated (e.g., sales can only happen during a certain time window each year, most places have a cap on the maximum discount, returns are close to non-existent). I am not sure what other value they add to the retail experience beyond maybe having a relationship with the shopkeeper and going in to shoot the shit while buying stuff.

I'm not surprised that something like Amazon is able to disrupt the system. Even my parents have started buying stuff on Amazon instead of going to the store and they barely use a credit card.

I'm also not surprised that the retail guild/cartel is pushing back through legal means. It will be interesting to see what will happen. My guess is that the Italian consumer will end up losing at the end and will have to pay more for stuff from Amazon.

> wonder how he convinced the team to drop the "for free" academic HPC/Slurm

Probably because they are working with publicly available data so they don't need to worry about HIPAA/restricted data compliance. One of the main reasons for using the HPC/Slurm cluster is that IT (or the data compliance office) guarantees compliance with restricted data.

This paper shows the problem with interrupted time series analyses.

Mainly, (1) there is an external shock that happened in 2020 that interrupts the historical trend in the data. This shock is probably the covid pandemic, but the time series can't really disentangle how covid impacts death rates (i.e., is it the vaccine? Is it an undiagnosed infection? Is it putting off care because hospitals are full? Is it stress? Is it something else?). In other words, interrupted time series don't really isolate the effect beyond the general shock.

(2) It is possible that a similar pattern is present in a different subpopulation that hasn't experienced the same shock. Unfortunately, I cannot think of anyone that hasn't experienced in some way or another the covid pandemic that the authors can use as a possible counterfactual for these analyses. Without that, the interrupted time series could just be an example of an omitted variable that correlated with both the external shock and the outcome and that it is the actual "root cause" of the outcome. For example, it could be that covid lead athletes practice less and become more sedentary. When they resumed their training regimens in the second half of 2020, they were out of shape leading to heart problems. I'm not a heart disease researcher so I don't know if this happens, but it could be.

I 100% agree with you that math is real and that there is real benefit from learning it. With that out of the way,

> It's treated as some kind of tournament to get into college

and

> But most people admit that they forgot all of their math immediately upon finishing school.

in those 2 quotes, you summarize the current issues with school mathematics. There is no real benefit in learning school math as most people will just use up to 6th/7th grade math in their daily work. Even professions that "do math" daily (like carpenters) have, most of the time, figured out ways to do it "quickly" without really doing math (see, for example, the diamonds on a measuring tape and how they are used to solve the Pythagorean theorem).

That leaves school math as "the great sorter" of students. Students that "can" take advanced math classes and use those as leverage in college admissions. To see this process in motion, look up parents' concerns with California's new math sequence and how those usually involve college readiness (as in, college admission advantage) over career readiness (as in, calculus II is needed for whatever profession my kid will do after high school).

I had a friend living at Munger at Umich and visited a few times.

The vibe of the building is similar to an hotel. As you enter, there is a foyer with elevators. The upstairs floors have windowless corridors where you enter the suites.

Each suite has a long corridor with doors to the rooms (4 on each side, if I remember correctly). Each room is on the "inside" of the building, so they don't have windows, but they have private bathrooms.

The living quarters have a shared double kitchen (two fridges, two cooking ranges, etc) and a double living room (two couches and two tvs). This room also has a wall of windows so there is plenty of sunlight there.

The suite I visited was coed and had graduate students from different schools living together. It was definitely an interesting vibe.

As any dorm, there are other amenities that are shared with everyone like a patio on the roof, a gym, rec room, and study spaces. Also, all the rooms are furnished, so it might be a good set up for someone that doesn't have furniture or doesn't want to worry about it.

When compared to other grad student housing options (Northwood 1-3 for single students or a shared room in Northwood 4-5), Munger is definitely an improvement (mostly it's location and the fact that is a new construction, to be honest) but it is a little bit more expensive than NW offerings.

I don't get what you mean by front-facing website. Do you mean accessible by the public? Isn't that the point of websites that allows the public to look up professional licenses? I would imagine that there are similar websites for all other licensed professionals.

BTW, California has a similar website that allows employers to put in SSN and DOB to lookup a teacher's license: https://www.ctc.ca.gov/commission/lookup

It seems that Missouri had a similar setup that required a teacher's last name and the last 4 SSN digits.

Most states allow you to lookup teacher licenses using last name and they have a “secure” version of the same site for employers and employees to update data. Most of the secure sites use a combination of last names, dob, and/or SSNs to authenticate teachers.

> drawing sweeping conclusions from a single specific experiment

It seems that you are bringing up the main issue with modern scientific production. Research studies that do not have sweeping/popular/weird conclusions do not get press and the ones that do get coverage. That creates a perverse incentive for researchers to either do research in "popular" topics and to have new insights/conclusions even when these are not really supported by the results.

It is not an accident that you are reading about this specific study in vice rather than the 100s of other incrementalist studies about depression that got published over the past year.

This is the paper I had in mind: https://journals.sagepub.com/doi/full/10.1177/23328584156221...

> Students of color are underrepresented in gifted programs relative to White students, but the reasons for this underrepresentation are poorly understood. We investigate the predictors of gifted assignment using nationally representative, longitudinal data on elementary students. We document that even among students with high standardized test scores, Black students are less likely to be assigned to gifted services in both math and reading, a pattern that persists when controlling for other background factors, such as health and socioeconomic status, and characteristics of classrooms and schools. We then investigate the role of teacher discretion, leveraging research from political science suggesting that clients of government services from traditionally underrepresented groups benefit from diversity in the providers of those services, including teachers. Even after conditioning on test scores and other factors, Black students indeed are referred to gifted programs, particularly in reading, at significantly lower rates when taught by non-Black teachers, a concerning result given the relatively low incidence of assignment to own-race teachers among Black students.

The context for the use of "racially segregated" is probably within the current anti-racist movement (see, for example, Kendi's work). This movement has moved away from labeling racist a policy that was implemented with racist intent (like segregated schools during Jim Crow) towards labeling racist a policy that has unequal outcomes based on race, regardless of policymakers' intent.

If you start from that perspective, it naturally follows to label G&T education as segregated if we observe that one group being overrepresented in it and another group being underrepresented.

Of note here, it's the assumption that the distribution of gifted and talented eligibility attributes is uniformly distributed among different groups of students and that somehow the system is not picking up eligibility signals for specific students. There is some emerging evidence that this is the case for Black students, as the identification gap between Black and White students closes when the G&T assessor is Black (I think that Grissom has a recent paper on this). If we reject the uniform distribution assumption, we end up assuming that the G&T eligibility traits are unevenly distributed among students, which I believe is one of the longest lasting beliefs of racist thought in education.

It could even be argued that the barricade around scientific research is what kept the epistemic "superiority" of academia and academics. Opening up the tools to review the scientific literature gives people the power of "doing their own research" without having the tools and skills to decide whether a paper they have found on whatever topic is good research. It's possible that this democratization process has opened up the scientific knowledge to the masses but it is also possible that access to scientific literature doesn't really help without the training required to make sense of it.

uBlock Origin has been broken on macOS since Safari 13 / macOS Catalina [1].

If they ever get the port to work on a newer version of Safari, it could work on iOS.

[1]: https://github.com/el1t/uBlock-Safari/issues/158

While not necessarily being “woke”, any research tradition that identifies with post-positivism [1] tends to reject the traditional scientific method and proposes new ways to reach “scientific truth”.

There are many post-positivist research traditions in many fields that you will be able to look up and read more about if you are interested in specific citations and authors that do this kind of work.

[1]: https://en.m.wikipedia.org/wiki/Postpositivism

>the risk of developing symptomatic COVID-19 was 27 times higher among the vaccinated, and the risk of hospitalization eight times higher

The denominator in those odds ratios seem to fishy to me. Comparing people who were previously infected with COVID and vaccine recipients does some weird conditioning on getting COVID, so it should hide risks of developing symptomatic COVID and hospitalization for people who got COVID previously. It seems to me that the risks should be "cumulative" for people who were previously infected (i.e., the denominator should include symptomatic infections and hospitalizations for previous COVID variants and not just delta) rather than "zeroing" those risks for this group and just focus on delta re-infection. The decision to only focus on delta infections will bias up the OR estimates for vaccine (in)effectiveness, so we should read the results of this paper as the possible upper bound of the ORs.

Also, the study compares the risk of delta infection, meaning that people cannot really just say "I won't get vaccinated because natural immunity is better" as the variants that provided that natural immunity are not "dominant" anymore, meaning that the opportunity to acquire natural immunity against delta is probably close to zero at this time.

Also also, isn't it something that is known that natural immunity is better than vaccine-acquired immunity? The trade-off is that natural immunity has the "side effect" of actually having to fight off and survive the illness?

I guess they offer similar subscription services but it seems to me (at least) that they serve completely different niches. For example, if a content creator says "check out my OnlyFans for subscriber-only content" versus "check out my Patreon for subscriber-only content", I would expect that they are selling completely different things.

Another parallel could be comparing HomeDepot and HobbyLobby. Both stores sell you things for DIY stuff, but they target different kinds of DIYers (even if they could sell some of the same items).

(Most of) All the things that fly do it because of lift. Lift acts like piece of string that keeps up the flying object. The point where this string is attached to the object is called the center of lift. This study shows that the center of lift for a flat round object used as a frisbee is towards the front of the disk. This results in the frisbee flying at an angle instead of flat because lift is pulling its nose up. This results in the disk eventually back spinning, making it bad at flying (think of it as the lift as yanking the front of the disk and causing a tailspin).

I wouldn't really trust correlational studies that link hospital capacity and mortality rate during covid. You want to look for causal studies that try to isolate variation in hospital capacity using some sort of instrument to get to a causal estimate of the increase in capacity on mortality. A lot was going on during the covid pandemic that was probably correlating with hospital overcapacity that will bias the estimates from a correlational study.

Out of curiosity, which mechanism are you thinking about that would lead you to the conclusion that hospital capacity doesn't impact mortality? The story that I have heard the most is that hospitals start to turn away people if they don't have beds from them, leading to either delay in care or no care at all. Both delay or no care lead to higher mortality. I cannot really think of a story where an over-capacity hospital leads to the same or reduced mortality.

If you believe that hospital closures lead to hospital overcapacity or delays in care, you can find some literature that shows that closing hospitals leads to an increase in mortality in the community [1].

1: https://www.nber.org/papers/w26182

From a quick google scholar search, it seems that there is a trade-off between heating/cooling costs between summer and winter:

> The results indicate that the energy savings ratios of the rooms with the sedum-tray garden roof and with the white roof were 25.0% and 20.5%, respectively, as compared with the black-roofed room, in the summer; by contrast, the energy savings ratios were −9.9% and −2.7%, respectively, in the winter.

It seems that covering the building with a green garden is better than painting the roof white, but the heating costs were higher in the winter for both designs.

I guess that changing the roof could work in climates where the winter is not too cold.

On a side note, I wonder if it would be better to work on better insulation (both roof and sidings) rather than re-roofing a house.

Note: I am not a civil engineer so I have no clue if the paper is good or not.

[1]: https://doi.org/10.1016/j.enbuild.2017.09.091

I wonder if the stress comes from how a police officer dies on the job rather than the absolute death rate of the profession.

Lumber harvesting is the most dangerous profession in the US but I imagine that most deaths would be described as "accidents". For example, someone cuts a leg with a chainsaw, a tree snaps in a weird way, someone gets pushed off a cliff.

Police officers, on the other hand, usually due to violence while interacting with other people.

In other words, a lumberjack, roofer, etc could think that they won't die today on the job because they are being careful, because there are safety protocols set by OHSA, etc. The same might not be true for police officers because there is little that they can do to address how others will interact with them.

In a sense, police officers might process their job more closely to how a soldier process their job than a lumberjack/roofer.

I dug up SFUSD board policy [1] around failing notifications. The policy reads:

"Whenever it becomes evident to a teacher that a student is in danger of failing a course, the teacher shall arrange a conference with the student's parent/guardian or send the parent/guardian a written report. The refusal of the parent to attend the conference, or to respond to the written report, shall not preclude failing the pupil at the end of the grading period. (Education Code 49067)"

The policy clearly states that the student has to be "in danger of failing a course" and not "guaranteed to fail a course". But, if a teacher knows that they might fail a student, they must notify parents by some reasonable deadline (again, it seems that OP's school sets it by week 9 of the semester or by the mid-semester mark). Without this notification, they might not be able to fail a student because they weren't in compliance with the CA ed code and district policy. This becomes even more of an issue when HS graduation is concerned.

I don't know if the forms are new or not. The policy was last updated in 2017 so they are at least 4 years old. What what is worth, I filled out similar forms when I was a teacher 10 years ago in another part of CA, so I suspect that they have been around for longer than that (and the forms were mass produced on carbon paper slips that seemed printed in the 70s). I am also wondering if the word "guaranteed" came up in response to COVID-related changes to grading policies and the district having implemented a credit/no credit grading system.

Links:

[1] Board policy 5121 cf. 6154 para. 4: https://go.boarddocs.com/ca/sfusd/Board.nsf/goto?open&id=AKG...

> With half the term remaining, teachers of seniors received a notification that they would not be allowed to fail students unless they filled out a form right then and there declaring that the student was certain to receive an F.

I think that OP is misconstruing the reasons behind this notification requirement. This notification has to do more with a "cover your ass" policy than lowering expectations for seniors.

The California Education code [1] requires school districts to develop procedures to notify parents of a failing grade. All the CA districts that I worked for had some procedures in place for when and how notify families of failing grades in response to this law. From a quick search, it seems that SFUSD policies 4.2.5 [2] and 4.2.6 apply in this case. The policy clearly states that teachers have to notify parents of grades either 1 or 2 times during the semester (i.e., mid-semester report cards-if you went to school in CA you will remember receiving those).

It seems that OP's school was on a 9-week reporting period, requiring teachers to notify parents of grades mid-semester and that their school added a requirement of filling out an additional form outlining that the student was in danger of failing the class specifically for seniors (probably because most families stopped looking at report cards for seniors). This form is required to be in compliance with the CA ed code as parents have the right to appeal failing grades and not being notified is probably an enough reason for the district to change a failing grade to a passing one in case of a lawsuit/complaint.

I would imagine that failing a class is more "high stakes" for seniors than other students, so they have more paperwork involved for seniors if you want to fail them. Families might be more "litigious" if a student ends up failing a class required for graduation. It is simpler to complain to the board/suing that having to repeat a school year.

The bottom line is that the CA ed code gives teachers final say in grades and even the superintendent cannot change a grade without the teacher's consent. On the other hand, the CA ed code gives some rights to parents to appeal grades and has some notifications guidelines so they shouldn't be surprised of failing grades. OP will be able to fail as many seniors as they want, but they just have to notify their families that they are in danger of failing the class at some point before the end of the semester. At the end of the day, if they have enough evidence for failing a student (that would stand up during a public board meeting/lawsuit), notifying the student's family 9 weeks before the end of the semester shouldn't be too much of a burden.

Links:

[1]: https://leginfo.legislature.ca.gov/faces/codes_displaySectio.... [2]: ]https://www.sfusd.edu/services/know-your-rights/student-fami...

I saw a chart once that made the point that most research studies can be classified along two axes: rigorousness of methods and popularity of results. Most published papers have either high rigor/low popularity and low rigor/high popularity. The trade-off is in the fact that highly rigorous studies only allow for narrow, unexciting results, while popular studies with flashy results will have to compromise on their rigorousness. This is not really a rule, but it is an interesting way to see research and the editorial/peer-review process.

The example discussed in OP seems to fall in the category of low rigor/high popularity. I am not 100% on my history of psych research, but it seems to me that the stereotype threat was all the rage in the late 90s following the publication of Steele and Aronson (1995). OP study seems to follow a similar experimental setup as S&A with a new group of people (Asian-American women).

As far as meta-science is concerned, I think that it remains mostly a part of philosophy (as in epistemology) and the focus of a few (senior?) scholars in each field. There is really no space to publish meta-scientific papers that "shake up" the field and call out established researchers, as editors that publish those pieces could come under similar criticism for their work. I think that it is not an accident that the discussion of the replication crisis in psychology started from blog posts and other non-academic avenues and then found its way to more "established" publications in the field (again, if I remember the context of those conversations).

I really wish that the review process was open. It would be interesting to see the reviewers comments to this specific paper and how the editor decided to pick up and engage with them. All those conversations are usually locked up in some editorial management system and are seldom made public. I don't know if we can really have open science without having open peer reviews.