atleta

12 ms·

91.
by atleta 4y ago
The problem with comparing the cost of different energy production means is that most of these have huge externalities. After all, this is why we're still burning hydrocarbons, because they seem cheap compared to anything else, excep
92.
by atleta 4y ago
I actually find the model in the article pretty convincing despite agreeing with you on that it's a profession for people who like to learn knew things and that university/grad school should teach more generic, theoretical knowled
93.
by atleta 4y ago
It's not so much a desire for stability in general rather than the depreciation. I found that piece in the post rather convincing. I don't think most people have a problem with learning new skills in this field. But having to thr
94.
by atleta 4y ago
Which is pretty rare and the reason you know they are wrong is because science proves them wrong - later. The thing most people don't get is that if you want a pick viable strategy, then it's not enough to simply criticize one of
95.
by atleta 4y ago
I'm not making it out to be anything and I'm definitely not claiming that I know the solution. If anything, I was contemplating that the problem is more complex than it seems: it doesn't matter how you measure self-awarness,
96.
by atleta 4y ago
Regardless of whether this guy is right or wrong, this brings up an interesting angle: we know that we won't be able to (we are not able to) distinguish self-conscious and non-self-conscious entities with a 100% accuracy. Both because
97.
by atleta 4y ago
>> It's enough to show the errors of the current explanation > Not if you want to argue in favor of a specific alternative. Sure, but this is not what you actually said in your original comment. You said "if you want to c
98.
by atleta 4y ago
You don't need to identify who is an expert and who is not by yourself: experts (scientists) compete with and cross validate each-other. I.e. experts themselves will self-select who qualifies as experts. And since there are a lot of th
99.
by atleta 4y ago
I mostly agree, but I think you are confounding two things here. The actual argument doesn't have to contain a better alternative than the current best explanation. It's enough to show the errors of the current explanation - as lo
100.
by atleta 4y ago
Well, they weren't really proposing solutions , Ivermectin and Hydroxychloroquine were really just suggestions, candidates. I.e. ideas for what might work. Now in both both cases there were no reasons why anyone would think they cou
101.
by atleta 4y ago
Mainly the SMS thing. As far as I can remember, when I reinstalled it on my new phone (about a year ago) I just had to verify my phone number via the SMS they sent. Also, as I have already been using the Google Auth app, I was somewhat igno
102.
by atleta 4y ago
Looking at one everyone say, using backup codes and asking friends to store them (preferably on a USB drive in an encrypted password store) can be a good solution. I don't think most people need to worry about criminals (or the state)
103.
by atleta 4y ago
Well, Google Authenticator uses a standard, well documented algorithm (TOTP) and there is really no reason to touch it unless they discover a security issue, it becomes incompatible with the latest version of android or there is a pressing
104.
by atleta 4y ago
Google authenticator will also backup your keys to the google cloud so installing it on a new device will just work. You'll, of course, need to remember your google account password and not set it up with authenticator as the second fa
105.
by atleta 4y ago
Protecting your password db with a physical key sounds pretty stupid. I was always wary of 2FA for exactly this reason. It's something that you can lose or damage. Actually, it's pretty common to kill your most used 2nd factor, yo
106.
by atleta 4y ago
I'm still looking for the note taking app that can replace Tomboy. Not because Tomboy is great (it isn't), but because it's actually a note taking app. Most apps seem to be this huge convoluted editors, with a UI resembling a
107.
by atleta 4y ago
I've been doing this for well over a decade and while I had similar experiences sometimes, I don't see how this was a mistake by any means. Yep, not many companies sell or leak your email, but some do. And let's not forget
108.
by atleta 4y ago
Yep, it's indeed a trivial idea. I'd say I probably seen it mentioned in the explanation of convnets: they compress the images more and more as you go deeper in the network (i.e. they extract the features). Compression+decompressi
109.
by atleta 4y ago
E.g.: "Neural Image Compression and Explanation" https://arxiv.org/abs/1908.08988
110.
by atleta 4y ago
Well, matlab didn't pivot into that esoteric toolbox of numerical methods for engineering tasks. It has always been that. That's probably the very reason ML researchers picked it up in the first place, because everyone was using i
111.
by atleta 4y ago
This is definitely how it will play out. Whoever creates AGI first (and second). After all, they are investing a lot of money and resources and a "super human" AGI is very likely worth an unimaginably lot. Also, given the processi
112.
by atleta 4y ago
It depends on how you define "improve". I wouldn't call a runaway AI an improvement - from the users' perspective. E.g. if you think about the Chernobyl power plant accident, when they tried to stop the reactor by lower
113.
by atleta 4y ago
Symbian did everything better than iOS and most things better than android. At least it could do a lot more than any of these. Symbian had 'push' email (i.e. email that would just arrive, like it does now), whereas at least with
114.
by atleta 4y ago
He was not allowed to share the stuff (and the information) he had even with us. But yep, super crazy. That was the time before if was obvious that you do want to give access to all code being written to all of your developers. (And eventua
115.
by atleta 4y ago
I worked in Budapest and didn't have connection with the Mountain View site, but I remember some of my direct colleagues talking about it. I think one or maybe two of them did visit it too, so I guess they did work with people from ove
116.
by atleta 4y ago
C++ strings might be problematic, but the solution that EPOC came up with and then Symbian inherited was terribly bad. Symbian didn't really just have a single string type (as I'm sure you remember), but a set of different types f
117.
by atleta 4y ago
I wanted to try it but for some reason it doesn't accept Lex Fridman's channel as a valid URL: https://www.youtube.com/c/lexfridman
118.
by atleta 4y ago
Maybe. But by then (2011 IIRC) it was way too late, sales had been going South, so one wouldn't need too much of a genius to figure out that Symbian was dead. The problem was that in the mid 2000s things were going just too good and
119.
by atleta 4y ago
Quite a few of those hours were wasted and contributed to Nokia's demise. They should have switched away from Symbian waaay sooner. I think clinging to it is a classic case of the sunken cost fallacy. By ~2005-2006 it was obvious, at l
120.
by atleta 4y ago
I used to work for Nokia Research during the mid 2000's. I remember how hard it was to get the source code for this back then as internal employees... One of my colleagues was working on a research prototype that needed tighter integra

More ›