As an operator you're always making a tradeoff between the "negatives" of a license and the "positives" enabled by the actual software.
So for one example, the recent (bait and) switch of Elasticsearch to SSPL effectively means that for many actual real-world users of Elasticsearch, the benefits of staying on the official Elastic branch has ceased to be worth the costs of the license when your alternative is running FOSS Elasticsearch/Opensearch.
So to answer your actual question - presumably as an operator you want to use the AGPL-licensed code more than you want to avoid those same negatives of AGPL.
> There’s nothing silly about a novel virus having greater abilities to overcome barriers when compared to other pathogens that humans have had years decades and centuries to develop defenses again.
This isn't really quite accurate though. Before SARS-CoV-2 ever emerged, our immune systems already had defenses against it. This is because of the already-extant circulating hCoVs.
This is why something like 80% of blood samples taken before SARS-CoV-2 emerged showed T-cell cross-reactivity. From the perspective of our immune system, it was never a "novel" virus. It was novel to scientists and politicians but not to the human immune system.
> Seems like the simplest answer is just "actions that have reduced the spread of one disease have also reduced the spread of another, that's historically less widespread already."
This is not the simplest answer, and the evidence that the measures have really slowed spread is extremely low except in places like New Zealand and Australia which are small islands in Oceania.
---
> A bias towards testing for Covid wouldn't explain fewer cases of the flu unless those cases of the flu were coming back as Covid falsely instead of so further diagnosis was stopped.
Exactly this. I'm partial to the viral interference hypothesis, but what you don't seem to realize is that if you get infected for SARS-CoV-2 and recover in 7-14 days, you will still test PCR+ for months after. This goes into the widespread mistuning of the cycle threshold. Case in point: They tested George Floyd's corpse for COVID-19 and he was PCR+, despite having recovered from COVID-19 a couple months before. The test hit on the remnant viral debris from his long-gone infection.
> This is not the simplest answer, and the evidence that the measures have really slowed spread is extremely low except in places like New Zealand and Australia which are small islands in Oceania.
Have you ever heard of this small Asian country called CHINA?
The study isolated SARS-CoV-2 samples from every positive case, sequenced genome of virus, and tracked the mutation patterns. So, that will avoid a lot of the errors that improper qPCR usage can result in.
First things first, age and viral susceptibility:
> Of the 564 children under the age of 10 years in the targeted testing group, 38 (6.7%) tested positive, in contrast to positive test result in 1183 of 8635 persons who were 10 years of age or older (13.7%). In analyses involving participants up to 20 years of age, we observed a gradual increase with older age in the percentage who tested positive (Fig. S5).
> Children under 10 are less likely to get infected than adults and if they get infected, they are less likely to get seriously ill. What is interesting is that even if children do get infected, they are less likely to transmit the disease to others than adults. We have not found a single instance of a child infecting parents.
> There is an amazing diversity in the way in which we react to the virus.
---
(I recognize I only presented one study here - there's only so much time in the day :P - but the other high-quality studies I've seen confirm this. Mechanistically it makes sense if you look at the enormous T-cell cross-reactivity in those age groups. And BTW, data on school closures for Influenza (which children seem to transmit much more readily) showed that school closures were ineffective anyway because they would just spread it more outside of school)
Interesting that this was getting downvoted. Obviously relates to what the poster above was saying about it being "politically incorrect."
We may disagree on what the science implies, but downvoting someone for sharing research means that people are refusing to listen to anything that disagrees with their preconceived notions.
(I guess perhaps the implicit assumption is that anyone pushing articles that show that kids are at less risk, or that kids don't spread Covid as well, must also be pushing for all kids to be running free and socializing?)
Edit: And now this comment is getting downvoted. Folks who disagree, why not speak up about what's wrong with the study, instead of just downvoting?
To be fair, that implicit assumption is pretty strong. There is an absurd bias to push kids back in schools so parents can get back to work.
Still, your points stand that there are some good studies showing kids are not necessarily the major spreaders. It is somewhat surprising to me, but it is data.
There is an absurd bias to push kids back in schools so parents can get back to work.
Why is that absurd? Millions of Americans (mostly women) have had to work less or leave their jobs entirely in order to take care of their children, which is a major hit to both their finances and careers. It's a huge cost.
It is somewhat surprising to me
It shouldn't be if the media had been doing their jobs, but due to either incompetence or bias they haven't. People really don't seem to realize that schools have been open in large parts of the US (and many other countries) for months, and it hasn't led to mass infections of students or teachers or parents.
I'm far enough on the liberal end that a living income makes sense to me. Raising your kids is a job. And it is of value to society that one can do that well. Juggling low paying jobs is not beneficial.
I think it’s entirely reasonable to believe that individual kids are more resistant to spreading COVID, but closing schools is a powerful intervention. Schools have _lots_ of kids in them, in close proximity, and _every_ child lives in a household with at least one, but on average more than one, other person.
This is not the only study implying closing schools was a sensible step:
I'm going to lift one part of your quoted material, because this should get more eyeballs:
> We have not found a single instance of a child infecting parents.
That is a really strong claim. It also goes against common logic.
Children are more likely to be asymptomatic. Fine. They tend to resist the virus better, and avoid severe cases even when they do get it. Yup, all good so far. They are less likely to infect others around them. Still makes sense.[ß]
What doesn't make sense is that transmission probabilities in the list are all above zero. From a purely mathematical perspective, transmission probability of "infected child -> parent" should not be zero. I am not stupid enough to dispute scientific finds, but I strongly suspect there are more factors in play.
Also the cynic in me notes that "infected child -> parent" is NOT the same as "infected child -> adult".
ß: Recent news indicates that the latest variants do spread more aggressively among children and teens, and are more likely to show up with symptoms in them. I haven't seen anything about increased mortality among the same groups, though.
As a parent of a 6 year old and 8 year old, I'm thankful for this cutoff. Many studies define child as <= 18 or < 20, and I'm not sure how relevant they are for our particular situation.
For context, in India there's about 27,000 deaths per day in a normal year (napkin math: 7.344 deaths per 1000 from https://www.macrotrends.net/countries/IND/india/death-rate, multiplied by the indian population size of 1.390 billion, divided by 365 days per year). Currently there are around 2700 deaths being attributed to COVID-19 per day.
I am much more concerned about poverty, starvation, death, and decreased non-COVID vaccination rates, than I am about COVID itself, for a country like India. SARS-CoV-2 primarily kills the very old and the very unhealthy, particularly diseases of modernity (this is not a value judgement but rather a statement of fact).
The indian life expectancy is somewhere around 68 years. Compare that to 78.7 in the US (https://www.cdc.gov/nchs/fastats/life-expectancy.htm). We've seen in the US how we likely caused iatrogenic harm in places like New York by practicing early invasive ventilation. We've seen (although we don't have the true long-term data yet) the impacts of unnecessarily suspending elective surgeries (such as was done in my state of California for a month despite a total availability of ICU capacity, and the fact that even if there weren't capacity many outpatient surgeries are more valuable than an extra COVID-bed-day)
> It estimates that there have been 228,000 additional deaths of children under five in these six countries due to crucial services, ranging from nutrition benefits to immunisation, being halted.
> It says the number of children being treated for severe malnutrition fell by more than 80% in Bangladesh and Nepal, and immunisation among children dropped by 35% and 65% in India and Pakistan respectively.
> The report also says that child mortality rose the highest in India in 2020 - up by 15.4% - followed by Bangladesh at 13%. Sri Lanka saw the sharpest increase in maternal deaths - 21.5% followed by Pakistan's 21.3%.
It should go without saying that a 15.4% increase in child mortality is NOT a result of the SARS-CoV-2 virus, which spares children. It's a result of missed medical appointments, decreased non-COVID vaccinations, economic disruption, and the general environment of largely unnecessary and maladaptive fear and anxiety that the populace has collectively been exposed to.
A coding camp is the opposite of hustling. It's called "bootcamp" for a reason - it's there to impose some structure for those who need it.
Also bootcamp is very overhyped. There's some great people that come out of them, but I think mainly the hype is the fact that college is overrated (this is a cliche opinion I know) so bootcamps are there to try to exploit some of the delta that the myopic focus on college degrees creates. But many boot camps are super myopic in their own ways.
Right on. I'm self taught before bootcamps were a thing. I learned by doing by launching a bunch of startup prototypes. Then I ran out of money so I started interviewing, went 1 for 1 and got a good 6 figure Rails job at a big corporation (skipping junior level), all with less than 1 year experience.
Many people getting into coding I talk to think this path is too hard and that a bootcamp somehow looks good on a resume and is therefore safe.
The same arguments can be made for startup accelerators. You really don't need one if you're a hustler who can build businesses. Thus they also attract some percentage of people who want a "safe" approach for their resume.
> Difference is that I know person that got COVID in November and now has issues with memory.
> I don't know a single person that has that after flue and cold - and those are with us much longer.
Look into ME/CFS, whose existence is still contested (or rather I should say, whether it's a physical or psychogenic illness is disputed). I know you're just speaking anecdotally but just wanted to mention that post-viral issues (fatigue, memory, etc) absolutely does happen.
> As for viruses mutations etc. do we for sure know that this one is not man made?
At most SARS-2 is the result of extensive gain of function research on https://en.wikipedia.org/wiki/RaTG13. I don't know any credible individual that thinks it's fully artificial / manmade (and to be clear, my definition of "credible" is not the "anyone who agrees with the WHO/CDC and nobody else" definition that the establishment relies on).
Personally I think it's more likely that SARS-2 was GoF'd into existence rather than was a purely natural zoonotic leap, whereas I think the probability that it was fully artificial is almost zero.
The fact that the vaccines create just a single spike protein and the real virus creates much more is actually one of the issues.
First: does the rate at which the cells are made to artificially produce spike protein follow a different curve than the rate at which SARS-2 would? i.e. could mRNA vaccination cause a much more aggressive "inflammatory cliff", thus the huge percentage of "mild" adverse reactions (mild meaning, you feel like death for a day but end up fine with no detectable long-term issues)? It's possible.
And switching to efficacy, while personally I think resistance to the spike protein alone will be sufficient, because SARS-2 does not have the same ability to mutate/evolve the way Influenza does (for example, I can't imagine SARS-2 evolving away from the spike protein), it's very possible that the diverse epitopes produced by real SARS-2 infection give a much more robust and enduring immunity.
Your point about efficacy is not currently born out by the data, AFAIK. Seems like re-infection after COVID is much higher than infection after vaccine.
(Not the one you responded to) I know what an mRNA vaccine is and the only clearcut answer is that "yes, obviously the long-tail risks of mRNA vaccines could outpace the long-tail risk of COVID-19". The statement is also true when you say "...COVID-19 could outpace...mRNA vaccines" as well, if that's not clear.
BTW the "long tail" of COVID-19 in children is totally unproven and the whole "long haulers" phenomenom is likely (a) a small part normal post-viral fatigue which we see with basically any virus, and (b) mostly psychosomatic/psychogenic illness.
That uncertainty is exactly the point. It's possible that the cure could be worse than the disease, but we have every reason to believe that it won't be in the long-term, and that in fact the long-term prospects from COVID-19 are probably much worse than any of the vaccines. And in the short-term the question is decided quite clearly.
So when faced with the choice of vaccine or virus, it should be a foregone conclusion.
So for one example, the recent (bait and) switch of Elasticsearch to SSPL effectively means that for many actual real-world users of Elasticsearch, the benefits of staying on the official Elastic branch has ceased to be worth the costs of the license when your alternative is running FOSS Elasticsearch/Opensearch.
So to answer your actual question - presumably as an operator you want to use the AGPL-licensed code more than you want to avoid those same negatives of AGPL.