HNHacker News
TopNewBestAskShowJobs

hcknwscommenter

1,715 karma · joined November 7, 2017

submissionscomments
hcknwscommenter··on Japan HP accidentally deleted 77TB data in Kyoto U. supercomputing system
Or that is what he used to justify a free (likely business class) flight to JP.
hcknwscommenter··on Down the Rabbit Hole: The world of estranged parents' forums (2015)
I mean it kinda is the same panacea it was back then no? As in, it's not a panacea, but its a ton better than no job and just sucking off your parents largesse like a parasite.
hcknwscommenter··on Pfizer and Biontech provide update on Omicron variant
I'm not splitting hairs at all. The point that infection with one serotype elicits lifelong immunity to that serotype only, is not original antigenic sin (OAS) at all, much less lifelong OAS. To make sure we are discussing the same thing, OAS (to me) refers to the effect that infection with (or vaccination against) one serotype makes it difficult/impossible to obtain any useful immunity to a second serotype. Lifelong, means that this effect is very long-lasting (years). If dengue researchers routinely talk about such, then you should be able to produce one paper that demonstrates it. Certainly Halstead does not although it contains an interesting reference to and discussion of the earlier work by Eisen, suggesting OAS lasting 7 months in rabbits. However, my larger point is this: a strong enough OAS effect to actually blunt vaccine effectiveness is rare. In fact, other than Dengue and RSV, have we really every seen proof of it at all, never mind whether it is long-lasting? For example, Influenza vaccines infections do show a slight OAS effect, but vaccination still works very well.
hcknwscommenter··on Harvard won’t require SAT or ACT through 2026 as test-optional push grows
This is a weird comment in my opinion. MO says regular colleges are good too. She doesn't say "just as good". Of course going to Harvard can (not necessarily does) provide certain advantages. She didn't deny that.
hcknwscommenter··on Omicron evades Moderna vaccine too, study suggests, but boosters help
> When the spike changes sufficiently, there is no mechanism by which the vaccine works

this is not entirely correct T cell response uses a vast array of much smaller epitopes as compared to the antibody response. These much smaller and much more numerous epitopes almost certainly cannot be entirely mutated away while maintaining function.

hcknwscommenter··on Pfizer and Biontech provide update on Omicron variant
your link only reports on primary and secondary dengue. There is no analysis of how long of a gap there was between primary and secondary, and whether that length of time matters. I don't see any evidence whatsoever in that link of a "long-lasting" original antigenic sin.
hcknwscommenter··on Pfizer and Biontech provide update on Omicron variant
Okay. I admit that my characterization was sloppy/incorrect. Thank you for pointing that out. The paper is still not strong evidence though. To whit:

"the higher hospitalization rate in the 32,000-person analysis was based on just eight hospitalizations in a vaccinated group and one in a previously infected group. And the 13-fold increased risk of infection in the same analysis was based on just 238 infections in the vaccinated population, less than 1.5% of the more than 16,000 people, versus 19 reinfections among a similar number of people who once had SARS-CoV-2."

These numbers are very small for distinguishing between the effectiveness of two different and effective immunization methods (natural and vaccine). By now, we should have more data to support if the effect size is that strong. Do we?

hcknwscommenter··on Pfizer and Biontech provide update on Omicron variant
Except all available safety and efficacy data shows that those who have been infected but not vaccinated should get vaccinated.
hcknwscommenter··on Pfizer and Biontech provide update on Omicron variant
No there is not strong evidence. The Israeli study highlighted previous infection versus vaccinated but not (yet) infected. Note the "previously infected" category did not exclude the vaccinated, but the vaccinated category did exclude the previously infected. The only solid conclusion to make from that is: that even if you have been previously infected, get a vaccine.
hcknwscommenter··on Pfizer and Biontech provide update on Omicron variant
Long-lasting original antigenic sin (e.g., an immune system trapped for years (or forever) in a sub-optimal state by a vaccination event) is a very unlikely scenario. I don't think we have ever seen something like that for any other virus? So, it's no surprise you haven't found much data. I would also suggest you critically examine whatever data you have found that supposedly backs up this concern of yours.
hcknwscommenter··on Pfizer and Biontech provide update on Omicron variant
I briefly skimmed the article at the medrxiv link, but it does not seem to assay neutralizing antibody titer. Rather the medrxiv link just assays "reactivity." I suppose it would be great if the booster increased reactivity against all variants. But it is not so surprising that a booster furthers the selection of higher affinity antibodies, which tend to be more specifically adapted to the antigen. However, at the end of the day, the Pfizer data shows that either this decrease in reactivity is not enough to overcome the recovery in titer elicited by the booster or that the decrease in reactivity is not a general phenomenon. Or, hopefully not the case, we will start to see this decrease in reactivity effect as we look at larger sample sizes. I think there is a ton of data available to Pfizer about booster titers/effectiveness, so I would be surprised to see larger sample sizes change the conclusions.
hcknwscommenter··on The Next ‘Lab Leak’
umm. a paucity. So like, you admit that betacoronaviruses naturally do indeed sometimes have furin cleavage sites?

I have no opinion on DRASTIC.

hcknwscommenter··on The Next ‘Lab Leak’
Really? I'd like to hear more about this. What exactly looks so "suspicious" that it can't be explained by typical recombination events that are known to occur naturally in RNA viral genomes?
hcknwscommenter··on The Next ‘Lab Leak’
You have two #1's, and 2 #2's, so it's a bit hard to respond to any of those clearly.

I will first address the second #1, and note that your point is very hard to understand here. COVID-19 has been found in multiple places "in the wild." It's been found in racoon dogs, which were known to be sold at the Wuhan wet market. It even seems to be endemic in certain geographic areas of the U.S. deer population now, which makes me think this thing is never going away.

Regarding your point #3, I think 100s of miles for an animal reservoir of a closely related virus, where the animal is capable of flight, is not so far?

Regarding your point #4, I just don't get it. Yes, we don't know who patient zero is. Are you trying to imply that patient zero is being obfuscated by the government for some nefarious reason?

Regarding your first point #1, as a (former) protein engineer, I can assure you that viral engineering is just not a thing. It is not currently possible according to the brightest minds to just engineer a virus to have any particular infectious biological behavior in a host. Generally, we can engineer viruses to knockout core functions and thereby weaken them or render them non-replicative or non-infective. Sometimes we can engineer viruses to change their tropism (the cells they infect). But taking say a bat virus and combining with a different virus (SARS?), adding a few key mutations, and ending up with a weapon or something is just science fiction.

Could we use directed evolution? Maybe, but you can't just apply some sort of selective pressure in a cell culture that is applicable to actual human infectivity/morbidity/etc.

IMO The only "lab leak" hypothesis that makes any sense at all, was that someone just **ked up, the dice rolled horribly badly (like the worst outcome out of 10^64 possible outcomes), and it all went pear shaped. But it's a lot more likely an infected racoon dog at the wet market scratched or breathed on someone.

hcknwscommenter··on The Next ‘Lab Leak’
FYI, the wastewater story you refer to was clearly nonsense and has not withstood any serious scrutiny. But, I agree with your larger point.
hcknwscommenter··on WHO asks South African startup to replicate Moderna's mRNA vaccine
Stage A must be provably useful, provably non-obvious, and provably new.
hcknwscommenter··on WHO asks South African startup to replicate Moderna's mRNA vaccine
The very act of providing the product to the market place is a sufficient disclosure of "how to make the thing at all" for anyone with a modest level of skill in this field. Moreover, there are plenty of detailed descriptions in Moderna's patent filings and other publicly available sources on how to make the thing at all. The issue is very much one of scale and I would wager that Moderna doesn't quite have the expertise either to reproduce their manufacturing pipeline at will. I have have personally witnessed a massive project failure in moving a therapeutic molecule from one manufacturing site to another, presumed identically equipped, manufacturing site. These things are not easy. It may be that this South African startup actually cannot use much of the information Moderna has on how to manufacture at scale because it only reproduces in Moderna's hands at Moderna's sites. This would be the expected outcome really.
hcknwscommenter··on Impact of circulating SARS-CoV-2 variants on mRNA vaccine-induced immunity
We have over a decade of data on this for coronaviruses.
hcknwscommenter··on Impact of circulating SARS-CoV-2 variants on mRNA vaccine-induced immunity
We are pretty much in agreement. My point was that the OPs statement that there is no point in getting a vaccine if you are not going to get COVID was nonsense.
hcknwscommenter··on Impact of circulating SARS-CoV-2 variants on mRNA vaccine-induced immunity
Over a long enough timeline, the likelihood of an unvaccinated person catching COVID is essentially 100%.
hcknwscommenter··on Impact of circulating SARS-CoV-2 variants on mRNA vaccine-induced immunity
Immediate vaccination does not inhibit long term protection. If the vaccine efficacy wanes to a significant degree (which is not a given for each individual), then you are still in a better place than having never been vaccinated. The safest way to get covid is as a vaccinated person. Not the other way round. There is no "tak[ing] one for the team" here.
hcknwscommenter··on Laser for control of mosquitos, weeds and pests
That's not UV light
hcknwscommenter··on Lead time for new Intel NIC orders is quoted around 52 weeks
More profound misunderstanding. Influenza tests exists and are deployed at scale. In the US, I assure you that anyone in the hospital with a high fever of unknown origin is tested for several influenza variants. No one just "assumes" covid. That is total and complete nonsense. We have clear data on influenza hospitalizations (edit: and deaths) and the numbers are way way down.
hcknwscommenter··on Lead time for new Intel NIC orders is quoted around 52 weeks
Your last paragraph demonstrates a number of profound and fundamental misunderstandings.

First, herd immunity does not benefit from more bodies per se. You are portraying it as some sort of wall, where taking members out of the wall reduces the effectiveness. It is not like that at all. The proportion of previously exposed individuals required for herd immunity effects goes down as the frequency of potential spreading interactions goes down.

Second, there is no correlation between lockdown and lower vitamin D.

Third, we have clear data showing a profound reduction in the spread of other respiratory viruses (RSV, rhino, influenza). So, we have had many fewer deaths during lockdown that in previous years. A win, and yet you characterize it as a loss. Odd.

hcknwscommenter··on Silicon Valley stirs over proposal to end popular QSBS tax break
We are talking about a US tax proposal. I'll bite, what is country B you are talking about that is next door and has better services and infrastructure and lower taxes? Heck, remove the "next door" requirement you put in there and I still think your statement is indefensible.
hcknwscommenter··on Silicon Valley stirs over proposal to end popular QSBS tax break
By the same logic, if you every drive or walk or bike on a road or sidewalk, you are taking what does not belong to you. If you ever eat food that is healthy and nutritious because the government has tested and regulated food production, you are taking what doesn't belong to you.

If you live in your home safe and secure from crime . . . Taxes pay for basic infrastructure, defense, education, and regulation. These things that taxes pay for are good for us and you use them every day. You want them for free apparently. Do you seriously think you would be better off having to procure all those benefits yourself as an individual?

hcknwscommenter··on Ivermectin: * Must show pic of you and your horse *
I mean, anything is possible. Death rays from mars could strike me down tomorrow at 10 am. But it is certainly unlikely that a given antiviral would be effective for long. Tamiflu is very much widely available, but it is not really deployed (administered) at scale. At peak, and now down >50%, tamiflu sales were about 1B. Even if tamiflu were $1 a dose (it's not), we are talking about a very very small portion of the global influenza patients having taken it. So yes I think that, if tamiflu were a strong protection against severe flu (it's not) and it were deployed at scale (never has never was), its usefulness would not last long.
hcknwscommenter··on Ivermectin: * Must show pic of you and your horse *
IF ivermectin were an actually useful treatment (huge IF), and it were actually deployed at scale, you would see an immediate rise in ivermectin resistant virus. Antivirals are very different from vaccines. They hit one target at one place in on specific binding mode, and it is easy to mutate so that the antiviral doesn't work. A vaccine produces a different repertoire of antibodies in every single person. No single mutation or even group of mutations affects the immune response in the same way from one person to another. It's bad luck that delta got so "lucky" in its enhanced transmissibility, but the vaccines are still effective against delta. An antiviral will not be effective against the next variant that comes after the antiviral is launched. Vaccines are the answer. If we need a slightly tweaked new booster for delta or lambda or whatever is next, I'm getting that and staying away from ivermectin et al.
hcknwscommenter··on Vaccination blunts, but does not defeat Delta
Of course it was realistic. China basically achieved it (I am not a China fan or defender), but the rest of the world failed. The whole world achieved it for the first SARS. However, that ship sailed a long time ago. Probably before the first surge in Italy of COVID-19.
hcknwscommenter··on Vaccination blunts, but does not defeat Delta
Vaccines are very different from antibiotics. Antibiotics are single molecules that bind a very specific target. A single mutation of a single protein can render an antibiotic useless. Vaccines induce the immune system to generate a repertoire of antibodies, each of which binds in a different way and each of which is then selected and "evolved" (hypermutation) for increased efficacy. And each individual person creates a different starting and ending repertoire of antibodies. A single mutation, or even a large group of mutations, does not so easily side step all this diversity of antibody types/binding modes. Moreover, we have evidence that the idea that vaccines increase the likelihood of variants is just wrong. We do not see that effect for mumps, polio, smallpox, or any other virus. Moreover, delta appears to have arisen in India, which at the time had vaccination rates very close to zero. Again, providing strong evidence that vaccines are not applying selective pressure for more virulent strains.
← PreviousPage 6 of 26Next →