einarfd

When I got my Tesla a couple of years ago. I ended up getting it with FSD, not because I had much faith in Tesla pulling off FSD, but that was the only way to get active cruise control. The lane follow and active cruise control that my HW2 hardware supports are nice, and I use them a lot. But anyone that thinks that, that is in the vicinity of self driving are naive. They are definitely parts of what you would need for self driving, but only small parts at best. Maybe things are better with HW3, and I'll probably switch to HW3 this summer. But I have less faith in Tesla pulling off self driving for my car now, than I had when I bought it.

The transport authority in Oslo (Ruter#), has been running self driving services for a few years now and there is some info about it at https://ruter.no/en/about-ruter/reports-projects-plans/auton... . I don't live in a part of town they've been experimenting in. But what I've seen indicates that what they are doing is running slow moving, small self driving busses in areas where it would not be cost efficient to use a bus with a driver. Since they are slow driving, the self driving problem is a lot easier, you don't have to see that far off, and stopping is faster so you don't need to model the future far out to be safe. In addition low speed collision are safer, which makes failures safer aswell.

Yes I'm definitely sure he is speaking for the hospital, and that he was speaking at the hospitals press conference. This has been all over the news here in Norway, every news outlet treat this as legit, and from the hospital / government.

Do remember that this expert is the person the premier Norwegian hospital picked as their spokesman for their press conference on this. So in essence what he says is the opinion of the Norwegian health care system, not just his own, about these cases. This isn't some random expert, giving an of the cuff opinion.

There is actually a commercial product for killing lice on farmed salmon with lasers (https://www.stingray.no/delousing-with-laser/?lang=en), but I have to admit that I don't know how well it works. So at least there are related products. Probably easier to solve the lice on salmon problem though.

I've got one of the Tesla Model 3 cars, which is afaik. the model which Tesla has gone farthest with the touch screen, and before I had that I drove a lot of different cars.

After driving Model 3 for a couple of years, I have to say that the touch screen is not a usability problem. The controls I need to have quick access to are not on the touch screen, and the ones that are on that screen, are controls I would have to take my eyes of the road and look for anyway.

I will have to grant that they have gone a bit overboard with putting controls in the touch interface, and I still don't get why the glove box opening button is in the touch UI, but that isn't a problem by any means.

I was a bit skeptical of the Model 3 UI, before I got it. But for me it is clearly a step up in usability from the "a million little buttons with weird symbols that who knows does what", and while that view of traditional car interfaces might be a bit unfair. So is all the "any touch screen UI in a car is bad", sentiment going around. Touch interfaces can work well in a car, as long as you don't need to use it for the drive task, but only for auxiliary tasks that don't need, and shouldn't be split seconds tasks.

My understanding is that it isn't the loan that is banned but charging interest. A fairly simple workaround would then be to charge a monthly fee instead, and you would then end up with something that in practice is more or less the same.

The Norwegian branch of Circle K seems to be interested, and they have put up a charging points at some of their locations.

I've never tried the charging point at Circle K though, they might be usable or they might not, at least they look like they are looked after and not physical broken.

But after trying some of the other brand of chargers, I've largely given up on using any non Tesla fast charger. I will use non Tesla ones in a pinch. But if I do, my expectation is worse user experience, slower charging and more expensive, which does seem like somewhat of a bad deal.

I have a Model 3 where the rear window developed cracks and the integrated heating in it stopped working. Tesla exchanged it under warranty, and I haven't had any problems with the new one. I'm not an expert, but my impressions is that these things happens from time to time. Safety glass doesn't like to be under tension and if you mess up the mounting of it, and do that, it can crack or break.

Implementing session handling as URL parameters makes no difference from a legal perspective. It will be covered by the law excatly the same as if one used cookies. It's the tracking the law is about, not the implementation of it. Your just picking a worse technical solution for no legal gain.

It's named in tribute after a guy called Rocky McGaugh. I don't think they want to change it.

No vaccine is ever 100%, you will see that some vaccinated people do get sick. In addition you have people that can't be vaccinated, these are still at risk. But if you vaccinated enough people, the successful vaccinated will act as a kind of firewall and hinder the sickness from effectively spreading.

> Either your goal is to model reality and real life, or your goal is to model your idealization of what you feel real life should be.

> If you pick option #2 then your model does not reflect real life.

These deep learning models built by corporations are not scientific models, they are engineering solutions, built to solve problems. Reflecting the real world is only useful if it furthers what the company wants to solve. If they for example remove swear words from their training set, that will make them a less accurate model of the world, but make them more useful for building solutions. But it is probably a trade of they would be happy with. We've also seen example of risk scoring application for felons that seem to end up doing racial profiling, because that is what the data seem to indicate makes sense. But that's deeply problematic and runs counter to laws in some places and seem ethically problematic (https://www.theverge.com/2020/6/24/21301465/ai-machine-learn...).

> Bias is by definition the way the model returns results that don't match the real world and real life, in frequency and in proportion.

Getting a good data set without bias is hard, even if you crawl the whole internet like Google does. Not everything is on the internet and there are systematic drivers that make some part of the human condition over represented (English, science, the views of the affluent and educated), and some under presented (small languages, the discourse of people behind the Chinese great firewall, the poor). So just getting a ginormous data set does not fix bias.

> If your goal is to use your model to manipulate and control society based on your own personal criteria, by manipulating it to return results that distort the real world and real life, then call it something else, because bias is not it.

Positive bias is absolutely something that we use, and while it might seem sinister it does not have to be. The example I'm most familiar with a facial recognition technology. Most groups building that ends up with a model that is better at some groups than others. Asian research groups often end up with models that does well with asians and worse with whites, while European groups usually end up with the reverse. In some sense these results do reflect the reality of these groups, most people in Europe is white and most people Asian is asians, so that you training sets ends up like that is not surprising. But no one is happy with these kind of results, and everyone wants to fix that.

To bring it back to speech and text models, let's say you are building a customer service solutions incorporating a deep learning model, the reality might be that you current customer service representatives treat blacks (or people who use "black" dialects), worse than people who sound white. An accurate model built on this data set will then also do that. But is that acceptable? I hope most companies would want to fix that, and be fine with adding some positive bias in their solution.

> I think this sort of veiled personal attack resorting to baseless extrapolation is not a productive line of public discourse.

I thought that my comment was fitting to the tone of the text I replied too, but fair enough, maybe I shouldn't have included that paragraph.

> You should take a step back and think about a) the field of study, b) how the main argument completely ignores the basis of said technical field and what it studies, c) how the argument being created lies on the idea that a self-annointed elite should have the right to manipulate the public to force it to fall in place with it's goals and desires.

It is unclear to me what this means, is it arguing that studying bias in AI and specifically deep learning is not germane? Is it arguing that it is not acceptable to institute policies and action to create a level playing field for minorities and repressed group because that would be social engineering, and forcing the common man to do the elites bidding? If that is what it means, I have to disagree, I find both of those things worthwhile.

>-- Has she calculated the oposite alternatives? People driving to libraries, to search for something?

Driving to the library, that is the alternative to training these models, really?

> -- Ok, basically she is pissed as AI is picking normal language, that people use, and not using the vocabulary that is in vogue on certain political circles. Basically censure, and forced speech.

If you had a model trained on a large corpus of data from the pre civil war southern American states, it would have been deeply racist, and would even view black people as possible property. If you had one that was trained on data from the 1950 it would be less racist but still problematic viewed by people from today. Is there really something special with today, that removes these kind of concerns with a model trained with current data?

> I think social engineering b.s. should be kept as far away as possible from science. This is turning true ai research into a masquerade to push certain political agendas.

It seems to me that it would be impossible to do any social science research, and specifically any research on racism. With this kind of attitude.

Some of the concerns brought up in the paper seems less consequential than others. Especially the pollution one seems weak to me, that doesn't make it false, and fair enough that is was brought up. I find the racism issue a lot more problematic, and is something I've run up to working on deep learning solutions myself. Even if it worked fine for my group, and most of our customers, that is definitely not fun, and something practitioners should consider when building these things.

If you can drill down to identifiable users. Then I have a hard time seeing how this is not a blatant GDPR violation. The biggest GDPR fine so far has been over workplace monitoring. So brave of Microsoft to roll out something like this, this might end up being a costly move on their part.

It's nice that the 737 Max has now been recertified by the FAA. But to be honest, after the FAA's handling of the original certification of the Max, I don't trust them to get it right now. So as long as it hasn't been certified by any of the other air safety agencies around the world. The 737 Max is a no go for me.

It looks like what Apple did was take their cellphone Arm processors. Do the minimal changes to get it to work in a laptop, and jammed that into laptops they already had. Which seems to me, to be the right strategy, but you do then end up with some weird limitations. I expect these issues to be sorted in the future generations of Apple silicon laptops.

At least they've caught the leaker, https://en.wikipedia.org/wiki/Natalie_Edwards, and is in the process of punishing her.

So even if curtailing dirty money in our financial system does not seem to be on the agenda. They are doing something about reporting about it.

A bit over ten years ago I was in my mid thirties, and working in a company that was acquired by one of our industry giants.

After the acquisition was finalized, all our products where put on the back burner, and we where all inn in on integrating our production as a feature in their offering for their next release.

Unfortunately for me, the part of the organization I was in, was minuscule compared to the size it would have been in the mothership. Which meant that the expectations of what my team could accomplish, was wildly inaccurate.

So for the year I stayed on. Every workday, no matter how hard I worked, the stack of unfinished task was always higher at the end, than at the beginning of the day. I was so stressed out during this, that if I wanted to fall asleep in the evening, I either had to drink or go for a run.

After a year of this, I managed to get a new job, and leaving was the best decision I've ever made. But getting a new job while struggling is hard, and I was lucky to pull it of, in hindsight, I should probably have started looking way earlier, before I was as stressed out as I was at the end.

Now ten years later, I'm doing fine, and the whole ordeal is just a bad memory. But I do know that one of my colleagues from an other team, ended up burnt out and spent at least a year on sick leave before getting somewhat on their feet again.

So my recommendation to those that are struggling. Take action now, rather than later. For me, leaving worked, but I'm sure that it isn't the only solution for everyone. If you think that there is nothing you can do, that sounds wrong, and you should look again. Maybe talking it out with someone else, either a mental health professional, or a friend would help figuring out the next step? I did not do that, and that is something I regret, I think that would have helped me handle the whole thing better.

A lot of people complain about Zoom fatigue, and while on one hand I'm fairly happy that it is a competitors offering that has ended up with being a part of that moniker, and not the product I'm working on, it is somewhat unfair to blame Zoom or the software stack for that. The main culprit for messing up meetings are the cameras and the microphones on laptops. I'm not sure if there are any vendors at all, including Apple which is the laptop vendor I use, which has god quality camera and microphones on their kit. I'm lucky enough to work in a group where everyone has dedicated video conferencing kit, and this makes a big difference. The quality gap between someone on system with proper camera and microphones, and someone calling in on laptop is huge. I get that not everyone can go out and buy dedicated kit, and maybe the dedicated kit don't work with you video conferencing vendor. But getting the people you work with on a decent USB camera with inbuilt microphones, is something everyone should have a look at. The chances are high it will make your days more pleasant.

When It comes to five year olds doing schooling over video conferencing. I'm far from an expert on kids, I don't have kids and I don't want kids. But it does sound to me like a fools errand, and I'm super happy I'm not going to participate in that nightmare.

I wonder how the TikTok userbase in the US is going to react to this? This could galvanized teens to action against the Trump administration. On the other hand, they could just ignore it aswell.

Al Jazeera is reporting that according to the president in Lebanon, it was almost three thousand tons of ammonium nitrate that blew up https://www.aljazeera.com/news/2020/08/huge-explosion-rocks-... .

It's more the opposite, "Intelligent Design" seems to be faith dressed up as science. While this would be more faith that doesn't contradict science.

The description of Reverend Sydney Smith in the article seem to indicate that he was an intellectual during the enlightenment period. Given that, I would presume that the phrase "rational religion" would mean religion interpretations that would not be in conflict with science and rationality as understood at the time.

Nothing in Europe north of Belgium and the Netherlands made the cut. It is kind of weird that there isn't several lists for Europe. When there is for several other geographic regions.

Singapore seem to be doing well with a solution that doesn't store everything on the governments servers.

Why not follow their lead?

My government (Norway), seems to be going (link in Norwegian https://www.simula.no/news/digital-smittesporing-apper-i-and... ), for a solution where they keep all the data on their servers, and privacy be damned. The upside they argue for with this, seems to boil down to them then having a lot more data to look at.

Having an app to help tracking infections, is something I would install. But what my government seem to be going for, no way. Way to intrusive and all encompassing.

I haven't found Simula, which is building the Norwegian app, on the list of partner of PEP-PT. So this is probably a saner initiative.

That's not really Boeing or the airlines call. If the relevant certification authorities refuse to certify it, then it isn't going to fly no matter how much Boeing or the airlines want it too.

It is going to be interesting to see how it goes. That it has taken this long to fix, points to it being, at a minimum a lot harder to fix that first suspected, and it might even be unfixable given the laws and regulation around airtravel.

Move to somewhere they speak it and force yourself to use it. Even if you could get away with using a language you know better. Accepting that you would get it wrong, and keep trying even if you sound stupid and it feels hard would also help.