anon946

At 25:15: "More than 8 million words can be stored and sent back by each disk pack".

Each disk looks the size of a vinyl record, and there are a whole stack of them in a disk pack.

Also, a sustainable model, even at large universities, would need to have a source for funding the up-front work. Most obvious source would be to tax the prize money. Long-term, I'm not sure what the end-game would look like, but I have a lot of doubts.

There are lots of problems with the way research is currently funded. But replacing the current system completely, or mostly, with prizes has an issue that I've not seen discussed, and that is the social/organizational aspects of it. In order to fund the work (that might lead to winning), a pool of money will be needed. Where does that pool come from? One way would be tax the winners. If, say, 1 in 20 efforts wins, this tax will need to be quite high. If too high, however, then you've essentially negated the "prize" aspect of it.

Then, within the institution, there will need to be internal competitions as to see which proposal(s) gets the funds to go after a prize.

I am not a lawyer, but it seems that intentionally and falsely shouting fire to cause panic would not have any kind of blanket 1A protection. From https://www.washingtonpost.com/news/volokh-conspiracy/wp/201...:

> And in fact the line from Justice Holmes in Schenck v. United States is “The most stringent protection of free speech would not protect a man in falsely shouting fire in a theatre and causing a panic.” That “falsely” is what’s doing the work, both in Justice Holmes’s hypothetical, and in how such a false shout would be treated by First Amendment law today. Knowingly false statements of fact are often constitutionally unprotected — consider, for instance, libel, fraud, perjury, and false light invasion of privacy. That would presumably apply to knowing falsehoods that cause a panic.

Also see: https://www.law.cornell.edu/uscode/text/18/1038.

That said, this doesn't change your broader point that it seems that most people (myself included) are not very clear on 1A exceptions.

It's obvious that Zoom is listening, because if I try to speak when I'm muted, it tells me that I'm muted.

As a professor, this completely resonates with me. For example, I take attendance and make it 5% of the grade. Then I give 5 free days and am generous with absences due to whatever. Why? Because it's a nudge for many students to get them to come to class, which makes them stay engaged, and ultimately get a better grade.

(The other reason I take attendance is so that I can recognize at least most of them by mid-semester, so can call on them by name when they raise their hand.)

And I'm often torn with taking points off for submitting work late. On one hand, why should it matter exactly when they submitted the work, if it's good work? On the other hand, I know that if I just said that there's no late penalty, some significant fraction of the students would wait till the end of the semester, then realize that they haven't been keeping up, then create headaches for everyone involved, including themselves.

Looks like about $10K.

https://www.cell.com/open-access

I would be very much interested in seeing the IRB submission/application that was submitted for this study. I wonder whether or not it was mischaracterized to the IRB, or written in such a way as to diminish the problematic aspects.

Can't speak for others, but I rarely, if ever, look at the table of contents paragraph. The reason is that I find it faster and more convenient to simply scan the section headings, since most CS papers are relatively short.

The biggest issue is the remote part, not the part-time part. I'm a professor, and myself and others in the dept. have remote, part-time PhD students. However, they have finished their coursework, which must be done in person for the most part. The coursework could be done part-time, though.

I am also in CS. My impression is that the situation in CS is different than that in other fields (better). The only publications in commercial venues I have have been with collaborators from areas outside of CS, in journals in their areas.

I can't say whether or not they are worth it, but they provide some degree of branding, mainly to "adjacent" communities. If I see a paper in IEEE something, I have some degree of assurance that there is a minimum quality level, even if I'm not that familiar with the area.

Also, at least for IEEE, they provide some help with running the conference, publishing the proceedings, etc.

Whether or not they provide value commensurate with their cost is a separate question.

Definitely no malice. You could start an open-access journal and claim that it will be as selective and prestigious as Nature. But where are your initial papers going to come from? Can you get enough people to submit their Nature-level paper to your journal?

Yes, the problem is one of collective action. A single, pre-tenure faculty member would be taking a significant risk to publish only in open-access venues. Thus, they don't, and the cycle continues.

This material is aimed at those who will be developing/implementing HPC software. HPC techniques are also used for ML approaches, since deep learning also requires a massive of computation, and benefits greatly from massive parallelization. GPUs are common in HPC, for DL and also other computations.

That said, there certainly is a great interest in the scientific computing community in ML/DL.

Investigation found that the incidence was no higher than for unvaccinated individuals (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2870557/):

>By 2001, with over 1·4 million Lyme vaccine doses distributed in the United States the VAERS database included 905 reports of mild self-limited reactions and 59 reports of arthritis associated with vaccination [29]. The arthritis incidence in the patients receiving Lyme vaccine occurred at the same rate as the background in unvaccinated individuals. In addition, the data did not show a temporal spike in arthritis diagnoses after the second and third vaccine dose expected for an immune-mediated phenomenon. The FDA found no suggestion that the Lyme vaccine caused harm to its recipients.

Depends on how you define "consumer", but hybrid drones can fly a long time: https://www.quaternium.com/ .

I just use a fixed count, like 5j5j5j or whatever to get me close, then jjj or whatever.

If the question were about full validated parsing of HTML with a regex, then I'd agree that "You can't do that" might be part of a valid answer. But finding tags is not doing a full validating parse.

Note that the set of valid C programs is not a context-free language. Yet it's common to use a context free-based approach to parsing. You just add additional code to handle the context-sensitive aspects (such as a symbol table).

Giving a talk is normal for many academic and "scientist" level positions. Being interrupted with questions is also normal, and I'm personally completely fine with it. The only annoying thing is that the company canceled at the last minute, but that's probably because they had multiple candidates "in-flight" at the same time.

That email was normal to me for small audience talks. Also, as a professor, I encourage students to ask questions during lecture (my class sizes are small enough so that it works fine).

Short term, yes. But when costs go up, they know that they can then later use it justify a proportional increase in premiums.

The US HC system basically runs as a "skim" operation. Since skim is fundamentally percentage based, the more money that flows through the system, the greater the absolute amount that is skimmed. Thus, no one has incentive, in the long run, to reduce costs. Short term, insurance companies might benefit from cutting costs, but they know that ultimately, when costs go up, they can justify proportional premium increases, which increase their skim in absolute value.

More here: https://ieeexplore.ieee.org/abstract/document/1188746

From what I could see in the table of contents, this is at a different level. Basically, this teaches you the concepts used to implement something like Vulkan. They probably implement a software Z-buffer, while a graphics card will do much of it in hardware.

To be honest, it's fairly specialized knowledge, though fundamental.

Should the long read lengths allow error correction to work well if there is sufficient coverage?

If the goal is to lower Rt to below 1, however, it still may be acceptable. See:

https://advances.sciencemag.org/content/early/2020/11/20/sci... https://www.medrxiv.org/content/10.1101/2020.06.22.20136309v... (preprint of above)

Large scale testing is likely aimed at public health, where the goal is to lower Rt < 1. Missing some positives is tolerable. Also, the antigen tests are more sensitive when the person is most likely to be contagious.

There are also fast molecular tests, using RT-LAMP.

https://www.lucirahealth.com/

Also when thinking about accuracy, I think it's helpful to distinguish between testing for public health, and testing for medical diagnosis/treatment. Antigen tests tend to be cheaper and have better sensitivity when the person is most likely to be contagious, for example, which, likely make them more suited for public health-oriented frequent, mass-testing.

The checksum should be checked by your computer. So somehow the packet is being repackaged with the correct checksum, but for the wrong data. In other words, when your computer checks the checksum, it matches. Another possibility is that somehow only errors that result in the same checksum are being generated.