ska

15 ms·

151.
by ska 2y ago
I haven't read this new edition, but the previous ones are pretty decent for the why's and hows. If you want to work on your "hairy c++ application" you'll benefit from understanding how "modern c++" it is
152.
by ska 2y ago
> 3 hours in some major That doesn't sound plausible. Got some examples?
153.
by ska 2y ago
does uv still lack lockfiles?
154.
by ska 3y ago
This is sometimes true, sure. And often the older work has more entered the general consciousness than being chased down by searching specific cites. On the other hand, very little is truly new, and recency bias can lead you into all sort
155.
by ska 3y ago
Absolutely, humans are incredibly good at dealing with consistent and predictable latency in a fairly broad range. Dynamic latency, on the other hand ... not so good. I recall a guitarist friend who figured out their playing was going to
156.
by ska 3y ago
No opinion on the specifics of this distinction, but it's worth noting that in research, an awful lot of successful projects have their origins in failed projects of decades ago...
157.
by ska 3y ago
Isn't the article more about algorithmic sorting that you don't have any control over, vs e.g. filters? Or another way of looking at it, the service filtering you rather than you filtering the service. It doesn't seem like y
158.
by ska 3y ago
Sadly I've heard much worse, not in jest. Some academic communities are pretty dysfunctional.
159.
by ska 3y ago
> I'm going to assume that nobody's insurance premiums or deductibles are going down That's exactly backwards. If you have better information than a competitor, you have an effective strategy to steal their lower-risk cus
160.
by ska 3y ago
repair cost goes up, up, up on average -> insurance goes up, up, up. That may only be part of it, but it's definitely part of it.
161.
by ska 3y ago
Risk pooling is fundamental to insurance, but not all pools are the same. The observation is that if you aren't able to discriminate at all or subdivide the pools, the only response is to up the average rate to cover the aggregate risk
162.
by ska 3y ago
> SGI's hardware was cutting-edge and exotic. This was their downfall, trying to scale out adoption with esoteric hardware. I remember being quoted $18k ish for memory upgrade on a O2 or origin, same amount of memory I had just bou
163.
by ska 3y ago
He also mentions the possibility of adversarial collaboration, creating a scenario similar to yours…
164.
by ska 3y ago
I suspect it would work nearly as well on many Americans.
165.
by ska 3y ago
Earthsea was written targeted for children/youth; it’s the same author but in a different register so you may find you enjoy her other stuff more?
166.
by ska 3y ago
This comes up in the article.
167.
by ska 3y ago
Yes this may be the deciding line.
168.
by ska 3y ago
> scan barcode, place item on the weight plate. Repeat. Press pay button, pay with card, whatever, tap/insert. scan barcode, place item on plate mostly works except when barcode is crumpled or something. without barcodes it's a
169.
by ska 3y ago
I've never experienced one reliable enough for this to be true (for me).
170.
by ska 3y ago
> outdated within a few generations (5 years?) Hardware generations are typically closer to a decade than a year. Robotics is moving fast these days, but not that fast.
171.
by ska 3y ago
> I don't know, but I think the big real industrial kind can really really hurt or kill people. Many medium sized arms are quite capable of generating forces that can kill a person.
172.
by ska 3y ago
There are lots of belt driven industrial robots. They have a different set of trade-offs and challenges.
173.
by ska 3y ago
How many degrees of freedom do you actually need, and what reach? Those parameters (plus speed) drive a lot of choices. 1-2k build cost is tiny in this space. I've paid that in individual motors (specialized ones, to be fair).
174.
by ska 3y ago
> cheap, 800x600 cameras with depth estimation ML algos This typically won't be accurate enough for closed loop kinematics, especially at any sort of speed.
175.
by ska 3y ago
Yes, they have systems to learn this way, but it assumes the environment is controlled (always) and the task identical (always). Lots of automation works this way, but it actually limits the applications quite a bit. A robot that can safely
176.
by ska 3y ago
> at there doesn't seem to exist a company yet that mass produces cheap, high quality, reasonably standardized robot arms. These do exist, it's just that "reasonably cheap" is typically low - mid 5 figures for any sor
177.
by ska 3y ago
This is actually pretty common. But getting enough resolution to improve on what you can do with the encoders isn't so easy. The more usual application of multi-camera setups etc. is in path planning and scene understanding, not low l
178.
by ska 3y ago
What you were experiencing was a) probably real and b) due to the conductive properties of the ring , not magnetic. Oversimplification: Moving a conductor in a magnetic field or vice versa indices current in the conductor , resistance in th
179.
by ska 3y ago
Actual accounting software is double entry. You’ll move that invoice through AP, and the balancing transactions will be split by project/type/etc by a combination of accounts and perhaps labels/tags.
180.
by ska 3y ago
That’s a different scenario: could a system be built to eat X’s lunch. The scenario I was calling out is more: Company A is great at X, so clearly that technology could be used to easily be great at Y, and doom company B in 6 months. The pr

More ›