ashwinpp

I'm looking into reducing loneliness and providing companionship for aging parents with an AI powered talking companion. If you have faced any issues with providing companionship to aging parents, I would love to hear more about it!

Note that most registrars participating in EAP or landrush charge a non-refundable fee, which is major chunk of the price increase from day 7 to day 1. It doesn't seem that there is "no cost" to participating in this period if one doesn't get the domain name.

Apart from all the questions about the camera, resolution and culpability, we need to re-think what it means to be training and evaluation of autonomous cars with respect to the rare incidents.

Training the cars based on data collected from roads is heavily biased towards incident free conditions. This does not give any training or feedback on the rare occasions such as these. If there was a learning algorithm deciding what to do (assuming hand coded rules are brittle and hence one would want to learn handling these scenarios) then it perhaps has no training data.

Evaluating the cars based on incidents per million is fine but doesn't tell anything about how the incidents would have been handled if they had happened. There is no incentive for the learning algorithm to slow down the car to prevent fatality, if all it cares about is an incident happening and not the severity.

One possible solution, autonomous cars are trained in real life simulations (using realistic lighting conditions, dummies and what not) to be able to handle the rare incidents and they are also required to pass regulatory testing in similarly realistic conditions to test for their behavior in rare incidents, before they are allowed to drive on actual roads.

As far as analogies go, I'll rescind my analogy based on the difference you have stated. However, it does not negate the point I was trying to make, that the problem lies not with feedback but with willful ignorance of it.

Since, I didn't realize you were the same person, here's my response for the other comment (https://news.ycombinator.com/item?id=15942520) you made. I think I understand what you mean by saying that what appears unwanted to one party might appear playful and wanted to another party based on their mindset. But then what guarantee is there that if the victim escalates the issue, that the escalation isn't considered playful either. Note that escalations can come in increasingly severe degrees and can also have equally "playful" connotations from the perpetrators perspective. If a victim tells her close girlfriends about the sexual assault, to the perpetrator it can sound as "I made such love that she couldn't stop talking about it". If a perpetrator has violated many victims, who have tried to warn their female colleagues, it can sound as "I'm so good that, some women want me for themselves and are spreading rumors about me to prevent others from getting close to me". I feel vile coming up with legitimate sounding sentences such as these, but I believe putting the onus on victims to give feedback need not make the perpetrators realize that boundaries have been crossed.

In my opinion there is an objective measure of whether a person is welcome or unwelcome. Now perpetrators might have a different viewpoint about it in which case no matter what the victims do to escalate the issue, it need not necessarily mean feedback for the perpetrator. On the other hand from the victim's perspective there is a cost to be paid in terms of engaging with the perpetrator apart from other costs which you already mentioned.

Finally, a few perpetrators might have such a removed world view that nothing below public shaming opens their eyes to the harassment, which is the only time we come to know about it. We might assume that no feedback was given by victims, but it might just be that the feedback wasn't nearly enough.

To summarize, the victims do not have nearly enough incentive to give feedback and perhaps many did and the harassment didn't stop until public shaming. It is definitely not the case of victims not having enough integrity to overcome costs to of giving feedback.

I think there are two orthogonal issues in the discussion which I would like to untangle. Some of the proposed solutions have been about maintaining a "social distance" (https://news.ycombinator.com/item?id=15939009) to "reduce the odds" of sexual attraction [Pencesque] (https://news.ycombinator.com/item?id=15940041) and consequently sexual harassment at the workplace. Another comment asked about whether pool parties at a conference make sense (https://news.ycombinator.com/item?id=15937938)

The orthogonal issues are

1) Non-consensual unwanted advances

2) Curbing avenues for sexual attraction

I call them orthogonal because,

- A person with enough avenues for sexual attraction at workplace or otherwise and choose not to do anything non-consentual and to recognize the people he/she feels attracted towards as equal humans even if they are his/her superiors or juniors.

- A person with few avenues for sexual attraction can choose to be non-consentual and delegitimize the humanity of his/her victims despite the lack of abundance of opportunity.

Mixing up these issues ends up confusing infidelity and sexual harassment. A person who seeks and engages in consensual sex outside his/her marriage is guilty of infidelity but not guilty of sexual harassment. A person who makes unwanted advances towards a married partner is guilty of sexual harassment irrespective of the fact that the victim is a married partner.

Note that I would categorize anything that makes the other person uncomfortable as an unwanted sexual advance. If a person talks about his/her pornography fetishes, but the other isn't interested, it is unwanted. If a person grabs the other other by the waist, while the other does not like it and tries to wriggle free, it is unwanted. If a person asks a subordinate for drinks and the subordinate is unwilling but pressured into accompanying, it is unwanted. Being unwanted is a clear signal for a person to not engage in a certain activity in the future and the victim should not be expected to create a ruckus in order to reject unwanted advances.

I don't think the problem lies with feedback. I think it lies with willful ignorance of it.

I believe that most people in group 1, are in fact aware of the problem of their behaviour and are willingly ignorant of it. This is because if they made any unwanted advances against women, even if the women didn't tell them to stop or publicly humiliate them, they surely ignored or tried to get away from the situation. At the very least the victim did not enjoy it and that is feedback enough.

Now the question boils down to whether a perpetrator considers the victim to be a human with a standing equal to his/her own. So as far as I can tell, it is the general problem of dehumanizing victims. The blame does not lie with Type 2 people who did not feel "comfortable" enough to publicly speak up.

Take any crime which doesn't take in public view, let's say child abuse. Is it the responsibility of the victim to give feedback to the perpetrator, or is it the responsibility of the perpetrator to view the child as a human being?

PS - I made a very similar comment down the thread to another commenter as well.

I like nuanced comments and you make cogent arguments. However, I believe that most people in group 1, are in fact aware of the problem of their behaviour and are willingly ignorant of it. This is because if they made any unwanted advances against women, even if the women didn't tell them to stop or publicly humiliate them, they surely ignored or tried to get away from the situation. At the very least the victim did not enjoy it and that is feedback enough. Anywhere in the course of history where there was a stronger party forcing themselves on a weaker party, the pain of the weaker party is visible in plain sight and ignored by the stronger party.

Now the question boils down to whether a perpetrator considers the victim to be a human with a standing equal to his/her own. The answer to that question has been a resounding no, whether it was slavery or war crimes or sexual assault.

To summarize, lack of feedback cannot be the reason, willful ignorance and dehumanization perhaps is and that is something the victim cannot be tasked with rectifying.

Curious about why this article went missing from the first few pages in a few minutes it took me to read it. Is it by design or intervention and if intervention then the rationale for it.

I found the article to be underscoring in detail the modus operandi of sexual harassment in the film industry (and workspaces in general) and adds a fair amount of interesting thoughts to my mind about the functioning of society and inherent oppression within.

I have a counterpoint for discussion against taking the flu vaccine based on the facts we know, and a two independent assumptions I make.

We know that influenza is different from other viruses because 1) There are many strains of influenza, it has a high mutation rate 2) Flu vaccines are based on a forecast of likely strains and typically cover only at most 50% of the strains in a given year [1]

Now let us make the following assumption - Actual flu infection boosts immunity after recovery from the infection compared to vaccination (perhaps an analogy would be how muscle recovery happens after microtears after a strength training workout with heavier weights).

In which case, one might conclude - Since a stronger immune response is great against a fast mutating virus and around 50% of strains are not covered by vaccines anyways, that the benefit of boosting immune response against future infections (beyond the current year) outweighs cost of getting infected and is a better choice than getting immunized against 50% of the strains with increased chances of getting flu due to the other 50% anyways

Assumption 2 - Getting exposed, but not infected by flu, increases the number of strains one's body is immune to. Something sort of like vaccination, but without taking the vaccines themselves.

For the purpose of modelling, let us also make the assumptions that all strains arise and propagate independently. That is the strains not covered by vaccination expose the same number of people unaffected by herd immunity, while the strains covered by the vaccination face herd immunity.

Now consider the strains covered by the vaccine, whose spread is reduced due to herd immunity. It is great for everyone vaccinated against the strains, but for people not vaccinated against it, it prevents exposure from and hence immunity. This might be a case where the utility of vaccination decreases for non-vaccinated individuals instead of increasing (herd immunity) as a function of the fraction of vaccinated population. In other words, the more people vaccinated for flu, the less likely an non-vaccinated person is to get exposed, the less likely is he/she to develop immunity towards those strains naturally. Thus this would prevent population level immunity from arising, where it would have arisen naturally.

Note that in this scenario, one might question that in the natural course of events, a certain percentage of population will develop immunity anyways and hence become equivalent to the scenario where some part of the population is vaccinated. However there are three major differences. First is that there is a certain delay between getting infected, spreading to others and gaining immunity. Thus for the same percentage of immune population, there might be larger population exposed in the natural course of events. Second is that many individuals might get infected, transmit and immunized to a strain without getting infected, thus such an individual gets all the benefits of vaccination, but might help others get exposed and immunized to a strain. And third that vaccination is more aligned with socio-economic factors while infection less so.

To conclude, if any of the two assumptions I state are true, then we should perhaps not blindly apply the same logic that applies to other viruses for which we have vaccinations.

[1] https://news.ycombinator.com/item?id=15562258

Thanks. Should have googled.

However, upon closer inspection, it seems like there was a wave of research during 1960s which looked at Copper ion concentration between 0.02 - 10g/L

In the 2000s there was another wave of research looking at antimicrobial properties, however sampling a few of them, it does seem like they were seen as growing surfaces rather than containers for water purification.

So, I guess it was observed for a long time, scientifically studied fairly recently, and not in particular whether it was viable in a practical setting?

and were proven long ago.

Can you back up your claim?

Anonymity of poster doesn't render the poster's experience unworthy of thought. If it did, I would be ignoring this comment from a throwaway account.

Cartoonish and poorly written are subjective judgements.

These experiences might seem implausible to you because they might never be happening when you're looking. In light of other, non-anonymous and fact-checked accounts of Susan's experience, I wouldn't think it is implausible.

All-in-all, while I agree in principle that one should apply fair judgement before believing someone's anonymous story, I do not think it warrants incredulity, especially and ironically from an anon account. If you believe that Amy should not have posted anonymously, I find it laughable that you use an anonymous account fearing flagging and down-voting in a low stakes environment fearing the "groupthink brigade".

* There is no proof in your statement that currrently 9X% of their users have fast internet connection.

* Even so, it would simply be an artefact of them never bothering to optimize for people with slower internet connections. If they did optimize, a lot of their traffic would come from the so called "lowest common denominator" just as the blog post says it did for Google.

* Finally I find the term "lowest common denominator" misrepresentative because it implies that optimization needs to cater to the slowest connection on earth, which is clearly not the case. If the average speed is above 90% of internet connections (as per Akamai report cited in the blog post) then the distribution of internet speeds is clearly skewed and there's value trying to even cater for median if not the minimum speed.

As the comments point out, apart from immediate use to consumers, it can also help understand underlying temporal trends and perhaps with some data analysis suggest healthier substitutes based on user tastes.

However, as part of a course project in using Convolutional nets for image recognition, I was wondering, what if, we could simply take a picture of a rack in a grocery store and augment the image with the overall "healthiness" heatmap.

And now to the meat of the problem, data. Is there a readily available dataset on which the image recognition can be trained? Note that each product will need to have a images taken from a variety of angles and lighting conditions. There exist methods to automatically generate alternate sets of images, but they are not perfect and we would still need data.