> because you open yourself up for discrimination lawsuits
Is that because when looking to make a possibly subjective judgement on the performance in a test and especially what a side project shows, it then becomes more difficult to prove that the judgement was not instead made because of some protected characteristic of the candidate?
That, and a big chunk of it is kids. In the same way you can't decline to hire someone because they're pregnant or may soon become pregnant (or are the father side of that equation) - requiring side projects is a very thin/loose proxy for "doesn't have kids". It's not a big enough problem to really stop people, but a big company's lawyers will stop it internally.
This is where a scientific approach to measuring opinions and experiences of multiple people would help. Peter Hotez has not demonstrated such an approach and has instead provided a possibly exaggerated and unproven anecdote as an input into discussion on how science is under threat.
He also did not bring up the issue of synthetic polyclonal antibodies, they seem quite relevant to the health of those who lack antibodies, and the transfer of antibodies from those who had developed them through their own COVID exposure or vaccination to those in most need of the antibodies.
I guess he did not even mention or give much thought to other ways that people who need the antibodies could get them, because he's much more interested in promoting vaccines, and treating vaccines as though they are the only option, and those who were on their last breaths could not have had their lives saved through other medical interventions after symptoms developed.
It's been in the works for years. Very good in some ways, a bit rough round the edges though, but it's time I make a substantial demo, docs and marketing website for it. SSR was built into it from the start.
I agree with this statement. Unfortunately, an ineffective (though apparently safe) vaccine got pushed at all cost through various misinformation and confused messaging, probably irreparably damaging the public's trust in vaccines in order to score some political points.
Before covid, nobody would hav considered being against childhood vaccines to be a mainstream or intellectually tenable position. Now those vaccines get lumped in people's minds with the crappy covid vaccine that was forced on everybody.
I'm curious: as you appear to be talking about COVID vaccines, which vaccine do you believe was "ineffective (though apparently safe)", given there are a number of vaccines that were developed in different countries using different methodologies. All the articles I've seen in reputable scientific journals suggest most vaccines, at least those developed in Western countries, were effective. Pointers to articles from scientific journals would be helpful.
It's unclear whether it makes sense to engage with folks who question the effectiveness of COVID vaccines, especially if they are not trained in the medical research/sciences. merely for the practical reason that historically, these discussions/arguments aren't productive, quickly devolve into people calling each other ignorant idiots, and most of the "facts" people cite are really just strongly-held beliefs which are not totally inconsistent with the observed evidence.
In my view it definitely is worth engaging. In order for people to get vaccinated they need to give informed consent. In order for people to give informed consent they need to understand the risks and rewards of getting a vaccine. Arguably what I just said is my strongly held belief and inconsistent with observed evidence, but I'd still like to talk about it without calling anyone an idiot.
When it comes to training, if someone wanted to argue against my position by questioning the credentials of my scientific education perhaps they could be persuasive.
If someone was more interested in taking part in medical research/sciences and also discussions of policy relating to them I could have quite constructive conversations I expect.
While I understand this position (i.e., https://xkcd.com/386/), it (along with an unholy union of the Internet, confirmation bias, pay-per-click, and echo chambers) appears to have contributed to an explosion in reality disconnection that is becoming actually threatening, e.g., there has been a surge in "vaccine hesitancy" for rabies vaccines in pets.
At some point, assertions like "[COVID] vaccines are ineffective" need to be challenged (which? over what time frame? where's the data?) or they become "everybody knows ...".
'ineffective' and 'effective' do not have very strict definitions. Something could be ineffective compared to how effective it was presented as being, such as the recent COVID vaccine. Here is one example of Joe Biden spreading vaccine misinformation: https://www.newsweek.com/joe-biden-2021-video-saying-vaccina...
"[COVID] vaccines are ineffective" is a fair assertion if the standard of something being 'effective' is preventing getting infected at all.
A better way to challenge the validity of that assertion is to make more effective COVID vaccines. In my view, bypassing the stage of creation of spike proteins and more directly getting the cells able to produce the relevant antibodies when needed would be a better vaccine, possibly more effective. It would also likely be safer (as the spike protein itself is dangerous, and the immune response to it can also be dangerous, by the sounds of it when the vaccine has not stayed localised in the arm but moved to the heart), so a vaccine that has solved that issue would do better when looked at in terms of a risk / reward ratio.
"[COVID] vaccines are effective [enough to be worth the risk of side-effects]" is something that would be better addressed through improving the vaccines themselves, possibly through improved public messaging, but I don't think as yet the data is there that supports that in an unequivocal way, and improved vaccines with greatly improved efficacy and safety which are then accurately described would be the best way to get the message accross.
Someone (including the President of the US) asserting that any vaccine is 100% effective and will prevent someone from "getting infected at all" simply means that individual does not understand how vaccines and infectious diseases caused by viruses work.
Indeed. My point is that this sets a standard for what is considered 'effective' and if it falls short of that standard, those who say it's 'ineffective' by that unrealistic standard, that is still correct.
Something can be ineffective compared to something else. Such as a COVID vaccine being ineffective compared to some expectations or representations of it.
The article was both about attacks on scientists and attacks on science itself.
My view disagrees with that of Peter Hotez, as I think that science should be vigorously attacked, especially by trying to disprove all kinds of scientific things, while also trying to prove things that contradict science. I expect that after such attacks, science would be strengthened rather than destroyed though.
No, in my view Youtube can not do whatever they like with this kind of thing, at least as far as my potential outrage is concerned. If Youtube gives a false reason (to me as well as others) about the reason for demonetising someone then I have a problem with that.
When some content is banned from Youtube, it's got positives and negatives. Like when Alex Jones was banned, I was annoyed that I could no longer watch Alex Jones on Youtube if I ever wanted to, but more than that glad that he'd never appear in my autoplay or recommended videos. While I think there is some truth that YouTube can do as it likes, people talking about what their rules are, complaining about them, lobbying Youtube even, is all fair too. A fair complaint would be that the user does not get enough control over what gets recommended. If enough people are talking about that issue, it could motivate Youtube or a competitor to provide that kind of control, as it would be a signal that it would attract an audience to that platform and keep them engaged if recommendation control was a major concern of theirs.
Also, in some circumstances I could be quite annoyed with Youtube for not demonetising or banning some content. It could be something I don't want to watch personally, or more likely something I feel disgusted by such as Elsagate type scandals where the 'protect the children' type argument or instinct in my opinion or feelings override free speech concerns.
People criticising what Youtube does and talking about what a video hosting website would ideally do helps to create the conceptual foundations for the ideal video hosting website, and which Youtube and anyone else who reads the comments can use.
Also, discussing how such a system works produces what would be considered 'prior art' when it comes to patents.
> People criticising what Youtube does and talking about what a video hosting website would ideally do helps to create the conceptual foundations for the ideal video hosting website.
The only problem is that your ideal video hosting website doesn't work. You won't:
a) Get enough users because most people want moderation.
b) Raise enough revenue because most advertisers want moderation.
c) Be able to legally operate because most legislators want moderation.
I never said that my ideal video hosting website would lack moderation.
Ideally I would have control (which I can delegate) over moderation rather than someone I disagreed with, and not have to spend much time or effort on moderation either.
Training and/or fine tuning my own moderation AI would be a useful feature.
After you get fired/retire/die someone else will take over your role. They will have their own ideals. Will they follow your ideals? Maybe. History says not likely.
Communism never succeeds past the first (maybe second) generation before corruption takes root. Same with obtusely vague terms of service and privacy policies. Google circa 2003, awesome. Google circa 2023, not so sure anymore
So what your saying is you want it so I can come to your house/business and talk shit and you have no recourse of kicking the person out, or severing a contract with them?
If you want to deal with Youtube, break them in a way that promotes competition instead of having the largest ad company also owning the largest video company. Trying to otherwise restrict their rights has many other bad outcomes for all businesses and individuals.
The OpenBLAS package was missing on ARM, along with some other dependencies I needed for compilation.
At the end of the day, even with many tweaks and custom compilation flags, the instance was averaging below 1 token/sec as a Kobold Horde host, which is below the threshold to even be allowed as a llm host.
Here is the problem with that: Apple would have a public-facing online search engine. The article addresses the advantages for Apple of having Safari users use Apple's search engine, but does not mention what the strategy would be for non-Apple users. Would they make a full-fledged search engine that optimises advertising for users who are not using Apple devices?
That's not to say it's not a good idea - a lot would depend on the quality of the search that Apple would be able to provide.
Why would Apple need to worry about non-Apple users?
Ultimately they could offer the service to them but they may find having Apple Search exclusive to their devices as a positive. It would allow them to offer many unique features such as integrating local and web results seamlessly (amongst many other ideas).
Also, and Apple knows this more than anyone, Apple users are the most capable in their ability to spend money. So although Safari users would be only 20% of the global search market, it's likely the most lucrative 20%.
