woeirua

33 ms·

571.
by woeirua 5y ago
Really? You don’t have a problem with an app that causes 1% of teens that use it to develop suicidal thoughts? By the way, according to the leaked study these teens directly attributed their suicidal ideation to Instagram.
572.
by woeirua 5y ago
No recommendation algorithms for certain classes of websites. News feeds must be sorted in chronological order only. Users may selectively filter their feeds if they choose. No infinite scrolling.
573.
by woeirua 5y ago
I think a lot of people are choosing to ignore that a lot of companies have done things in the past that were not illegal at the time of action. However, those actions were later decided to be made illegal because the behavior was deemed
574.
by woeirua 5y ago
You are replying to a direct quote from FBs article. I don’t personally think that 1 person is too many, but clearly 1% of their user base is a lot of people.
575.
by woeirua 5y ago
First the article directly contradicts you on the number of people sampled: “ ² According to the raw unweighted data, we surveyed 1,296 teens in the US and 1,309 teens in the UK and asked them if they experienced a range of feelings or expe
576.
by woeirua 5y ago
Just in case anyone else is wondering recent studies have said that up to 80% of nineteen year old women have body image issues. So maybe it’s not 1/3 of all teenage girls that are having their body image issues worsened by Instagram b
577.
by woeirua 5y ago
It’s simple math, if even 1% of the 1% who have suicidal thoughts as a result of Instagram are successful in a suicide attempt then based on the size of their user base that would mean thousands of people die from suicides at least partly c
578.
by woeirua 5y ago
This is the same argument that people who have bullied others into committing suicide have used to try and avoid responsibility. The courts disagree with you.
579.
by woeirua 5y ago
If 1% of the users of your app start having suicidal thoughts as a direct result of using your app, then guess what: your app is going to be responsible for a significant number of suicides. No matter how FB spins this, it’s just disgusting
580.
by woeirua 5y ago
And it was true, until cheap LI ion batteries became commercially available because they have the required power density to make it viable. From there it was just a matter of the economies of scale.
581.
by woeirua 5y ago
Maybe you should RTFA before commenting. The people who did the detailed analysis in the article have extensive experience doing exactly this thing and even using the most optimistic projections still found that it's just not feasible.
582.
by woeirua 5y ago
The article makes a bunch of points about economics, and those may be true now, but there are many products widely available today that were at one point considered impossible to produce at significant scale. That said, I think the article
583.
by woeirua 5y ago
There's a difference between the natural progression of technology, manufacturing capability, and economic and efficiency improvements brought about by large scale manufacturing. The path to cheap solar panels was relatively straightfo
584.
by woeirua 5y ago
Colorado trail or segments of the Continental Divide trail.
585.
by woeirua 5y ago
A couple questions I have after skimming the paper, so forgive me if they were answered somewhere in the 54 page manuscript: They mention in A.3 that they explicitly reject dynamically generated training worlds/games that collide with
586.
by woeirua 5y ago
By that definition wouldn't a regular RNN or LSTM also possess awareness?
587.
by woeirua 5y ago
Because the "best" electric car company in the world is barely eeking out a profit from selling a million cars a year, and even then only through regulatory credits, Bitcoin shenanigans, and outright fraud with FSD. That said, I a
588.
by woeirua 5y ago
The overwhelming majority of households use their cars such that one essentially is used only for commuting and running errands on the weekend. The second car, can quite easily be electric given the state of today's technology, and tha
589.
by woeirua 5y ago
In the US at least the USGS builds floodplain predictions for most of the major streams/rivers using this exact method. The problem is that those maps typically underestimate the amount of water that falls during these extreme events.
590.
by woeirua 5y ago
Congrats on getting out into this space. I suspect you will be very successful. Have you considered training a model to use the quarterly reports to identify companies likely to see a significant upside or downside over the next N months? S
591.
by woeirua 5y ago
No, it's not. In the full length video, there were 3 manual interventions in less than 30 minutes of drive time. If you're counting interventions per second, then sure that's 99.9%, but if you're counting safety critical
592.
by woeirua 5y ago
I'm not nearly as worried about fracking when it comes to PFAS/PFOAs as I am about airports. It is well documented, that airports (especially military ones) have used PFAS/PFOAs as firefighting agents for years and years. If
593.
by woeirua 5y ago
I'm not worried about programs like Copilot, in the same way that I'm not worried about outsourcing. If price was the sole factor in software engineering then all jobs would have already been shipped overseas. Instead, an awful lo
594.
by woeirua 5y ago
It's just bizarre to me that people are focused on building homes more efficiently, when the overwhelming majority of the cost of housing in most areas is the cost of the land, not the actual cost of building the house. How do we kn
595.
by woeirua 5y ago
LOL. I get that bashing the US is popular on this site, but this comment is really detached from reality. The US landed its own rover on Mars not that long ago, and they're not even covering it in the domestic press. Why would anyone e
596.
by woeirua 5y ago
Can someone explain how Julia MT got a super-linear improvement in performance by using 4 cores + HT vs 1 core? Fortran’s MT performance is more along the lines of what I would expect.
597.
by woeirua 5y ago
They usually can, but not necessarily under strict time constraints and the pressure associated with a whiteboard coding exercise.
598.
by woeirua 5y ago
If RL is enough, then there is no physically realizable way to actually train a RL based GAI in the near future. RL based learning requires evaluating the outcome of millions or billions of scenarios over time in order to optimize the netwo
599.
by woeirua 5y ago
Many employment contracts expressly forbid you from accepting other employment without prior authorization from the company. If his employment contracts contain that provision, then he would definitely be in breach. But... damages would p
600.
by woeirua 5y ago
Which explains why China has spent hundreds of billions of dollars trying to catch the US tech advantage? Please. This is comically false.

More ›