skgoa

19 ms·

151.
by skgoa 8y ago
Surprisingly often the consultants are not hired to bring entirely new knowledge. Instead they are hired to make certain people in the org listen to and accept a message they would have rejected, had it come from the inside.
152.
by skgoa 8y ago
The quickest growing software cluster in Europe is Dresden, though.
153.
by skgoa 9y ago
I agree with you. You express precisely what I believe is going wrong with Tesla/Autopilot right now.
154.
by skgoa 9y ago
I'm talking about active level 2 systems, which the Autopilot is an example of. A level 2 system has to be monitored 100% of the time, because it can not be trusted to warn the driver and fall back to manual automatically in every situ
155.
by skgoa 9y ago
> Actual usage is really the only way anyone will know if the models are trained appropriately to handle most/all situations they will encounter in the real world. Because testing stuff before throwing it on the market isn't a
156.
by skgoa 9y ago
> This is why, as the head of Google's self driving effort once said, partial self driving "assistance" is inherently unsafe. Sorry, but this is the exact wrong conclusion to make from the available data. Partially automa
157.
by skgoa 9y ago
Surely you would want internet at a charging location just so you can monitor it remotely.
158.
by skgoa 9y ago
It is absolutely necessary. Windows will happily downloads tens of GB worth of updates (or, you know, a full OS that MS is trying to push on you) without the user being able to do anything about it.
159.
by skgoa 9y ago
> This thinking doesn't appreciate that once the Model 3 manufacturing system is paid for most of that capital expenditure will cease. That's not how this works. You can't just stop investing in capital. Especially not whe
160.
by skgoa 9y ago
> Then, by 2021 or so, we'll look back at a company making over a million electric cars per year at high margin Yeah, sure. Why not say that it is "most likely" that they will have a colony on Mars by that time, too? >
161.
by skgoa 9y ago
(in the one quarter they made a huge, unsustainable effort to push production rate as high as possible)
162.
by skgoa 9y ago
that's the ironic outcome of their misguided attempt to just throw automation at every step in the process. They have ended up with the least automated car factory in the world.
163.
by skgoa 9y ago
That bullshit story being pushed by this bullshitter for years does not constitute evidence. There was no way for him to send data to any of the safety critical systems.
164.
by skgoa 9y ago
This is being looked into right now by Here, VW etc. Their plan is to for all cars that use the Here map backend (All Merc, BMW, VAG cars plus probably others who join down the line) to send updates automatically. MobilEye (now owned by Int
165.
by skgoa 9y ago
> I've mentioned that before. You build a map of what's around the vehicle. If it were that easy, everyone would have solved it already. Detecting distances, objects etc. from several varied sensors is exactly how you build th
166.
by skgoa 9y ago
> 1) Conform to a standard set of protocols for how sensors provide data to a self-driving software system. Doesn't exist yet, though.
167.
by skgoa 9y ago
The core of this debate is the matter of what intelligence actually is. We don't really know, but the most common definition I have heard/read is that intelligence is the ability to generate new solutions when presented with previ
168.
by skgoa 9y ago
It is exactly how it works. There ~40-115 different computers in a modern car that often have completely walled-off modules running on them. Most 'models' you will find in a car's software aren't even generated through m
169.
by skgoa 9y ago
> So if a normal Mercedez has a collision avoidance system that automatically brakes, there is no reason why an AI based system can't be built on top of that and the collision avoidance system automatically braking without intervent
170.
by skgoa 9y ago
But his conclusions turn out to be flat-out wrong.
171.
by skgoa 9y ago
Yeah, some of these comments here seem very naive.
172.
by skgoa 9y ago
The big success story is MobilEye, which was bought out by Intel recently.
173.
by skgoa 9y ago
Wow, this explains why the clock on my gf's oven had deviated so much after years of keeping perfectly in sync with all the other clocks on our devices.
174.
by skgoa 9y ago
Also, it only applies to a tiny minority of "tech" sub-fields. A lot of "tech industry" employees are nowhere close to getting those salaries.
175.
by skgoa 9y ago
DF itself already computes the prosperity of your fortress for you. This score and the sheer number of (non-enemy) dwarfs are the way you progress through the game. So a reward function based on these two inputs, plus maybe something like &
176.
by skgoa 9y ago
But with deletionism an unknown (yet significant) number of pages are still crap. One of the sad aspects of wikipedia is that their idea of "maintenance" enshrines crap and their idea of "sources" makes it incredibly dif
177.
by skgoa 9y ago
The former is enough, but having the latter makes it much easier.
178.
by skgoa 9y ago
You might want to read the article...
179.
by skgoa 9y ago
FB's chief avenue to success was that they turned the chicken-and-egg problem of social networks (they are only helpful when lots of people your know are on the same network) into a shining example of the strength of the network effect
180.
by skgoa 9y ago
That's not all that practical. Modern cars are rolling datacenters. Most have 50-100 computers in them that are on several different networks. Features are distributed over several nodes ("ECUs") and everything is statically

More ›