Joke aside, that's actually has been Linus' own explanation :
he needs to be frank to people.
I can't use profanity because my Facebook Friends are actual friends, acquaintances, family and stuff.
actual friends, family and stuff are also the people of whom you might not want to hurt feelings and opt to not outright tell everything which goes through your head
("you're ugly as shit", "you clothes/car/whatever is crap", "your idea is stupid and you should be burned in public in the town square for that", etc.)
but where you would restrain yourself
("you've got personality", "well, it's original and has got some charm", "it's a surprising idea").
So again they are the people to which you would "lie" (in a fashion. You're not actively trying to outright deceive them, just not transmitting 100% of the information) which is what this study tries to point out.
Let's be serious here. Three hours? The last long drive I did, my only stops were fuel stops. I was going for about six hours without stopping for anything, and only stopping for fuel and to pee. Stopping every two hours would be incredibly frustrating.
If you want to kill yourself in a car-crash because of attention deficit due to over-tiredness because your concept of a well rested and alert driver is "15 minutes tank-and-pee pause between every block of 6 hours of non-stop driving", be my guest.
But remember :
More seriously :
For all purpose, there isn't a big use case for cars that can drive more 600 km in one go.
Car *can* do, for the convenience of not needing to tank at each stop.
But they are not that much necessary - cause the puny human meat at the driver seat can't reliabily drive safely for extended periods of time.
You either need several puny humans to cycle between. Or you need much longer breaks within these 600km that the car can handle.
are already pretty much enough for most car needs with maybe only a few key exceptions (i.e.: alterning drivers for extremely long trip or people who are really a lot into the "suicide by tree" fad)
This is even more noticeable in the (slightly more) densely populate Europe which is a bit more conscious about its security (higher risk to crash into meaty humans instead of trees when there are more of them around instead of a dust desert), where big centers are close by (very often your travel destination is within a 300km drive anyway), and with a good electrical network with quite some power available (thanks to alpine dams, nukes, or renewable - depending on countries).
Of course, you could argue that all these are example of Nanny-state in our "evil-communist" socialist-leaning European countries, that stay in the way of good drivers and that everyone should be responsible for themselves on the road, and that I'm just brain washed by state propaganda and by government-sponsored bogus research.
 I actually have a military professional driving license. These are the actual number required by local law.
 Though given all the ruckus around Tesla's autopilot, I would guess that things FCAS isn't that popular on your side of the Atlantic.
What range do you think EVs have on a single charge, anyway?
Between 100km and 150km per 20kWh worth of battery charge.
Exact mileage depends on car model.
(e.g.: Tesla use lighter than average material and are designed from the ground up for longer ranges.
Other cars are simply "an electric motor replacing the ICE under the bonnet and batteries bolted wherever there's free place" quick conversion like the VW e-Golf and VW e-Up that VW hastily released in the wake of the diesel scandal, and might have lower mileages).
Also depends on the driver (driving like an aggressive idiot at high speed on the highway, and you'll get a lower range than driving conservatively maybe a bit under the maximum speed limit).
I can drive upwards of 3 hours without a break.
Which is *definitely* not recommanded.
Current recommendations here around in continental Europe is a break each 1 or 2 hours max.
(e.g.: There are big public service campaigns to advise drivers to have at least a quick "turbo-nap" every once in a while when driving long distance)
But let's make the assumption that you are 2 drivers sharing the load, and that you'll switch midway (without charging the car, nor making any break longer than required to change seat - no the best experience, but hey).
With an average-priced EV, that's not even near possible.
Renault Zoe are currently the cheapest e-cars with a decent battery.
(You can even get them for the price of an average priced ICE-car if you decide to rent the battery instead of buying it).
(They are definitely after the same market as Tesla's upcoming model 3, except that Zoes have been on the street for quite some time, and Renault chose the opposite progression from Tesla, release progressively longer range vehicle while staying affordable - instead of long range vehicles while progressively releasing cheaper models)
The latest model has upgraded the battery to 45kWh, which should give you between 200km and 300km of range. (depending on the speed/aggressiveness of driving 130km vs 100km on highway vs. 80km on streets between cities).
That's definitely in near the 3 hours of your example (and by now, both drivers of our assumption should get a nap, or at least make a long break - enough to put quite some additionnal range back into the battery using standard 50kW chargers)
For a car that cost in the general ballpark figure of ~30k USD (not some 100k+ USD Tesla Model S super car).
And all of the above aren't made up numbers, but my actual experience with Zoes.
They are available at the local car-sharing company (though not the more recent 45kWh battery), and I've already driven quite a lot of trip with them.
I can easy get 100km when I drive aggressively or 150km when much more conservative.
The current drawback I see, is that Renault doesn't have collision avoidances option available on their smaller cars like the Zoe.
(unlike VW where - like lots of european constructors - for the last several years even the lowest entry-level model like Up comes with a LIDAR [a.k.a. "City Safety"] in standard configurations,
or unlike all the noise that Tesla is making around their "Autopilot" since a couple of years ago).
If I understand it right, it's a GNU/Linux distro without a Linux kernel on top of a compatibility layer on Windows, right?
So "GNU/Windows NT Kernel" is better than "Linux" - That actually one of the rare few occastion a typical "GNU/Linux" distro gets used without the Linux kernel part.
But because "Linux" has brand recognition, it's still used.
(*): there's no separate compatibility layer (unlike things like Cygwin which are a user-mode compatibility layer that translates POSIX API-calls into Win32 calls - and thus enables soure compatibility).
The NT-Kernel has a bizare peculiarity : it can export several different ABI's to usermode software - it has different "personnalities".
- Win32 is just *one* of the set of ABI available.
- A long time ago, that made it possible to run OS/2 software on Windows NT.
- A little bit less longer time ago, Windows NT also had a "Unix" personality.
- Now WSL is actually the NT kernel exhibiting a small subset of the ABI featured by the linux kernel - about the bare minimum to get a few basic user-mode software (e,.g.: the "GNU" part of "GNU/Linux") run unmodified.
These are straight ABI available from the NT-Kernel, not a mere Linux-to-Win32 API conversion like Cygwin.
- Among other defaults Win32 has a poor multi-processing (forking is expensive). Cygwin application have to rely on that poorer cousin in order to provide multi-processing to POSIX.
- The recent kernels of Windows NT intoduced pico-thread which are very cheap, weren't available in the Win32 API back when introduced, but where exposed through the "Linux-lite" API that is WSL in order to make a usefull multiprocessing.
On the other hand WSL is far from complete. There is tons of stuff that you can do on your GNU/Linux that you can't do with WSL (e.g.: filesystem drivers)
Then you could use either ReactOS in your VM, or run Wine straight in your userspace.
And again there are also companies supporting *that*.
(e.g.: CrossOver pays developers)
So *there is* company-sponsored efforts to be able to run windows programs in a GNU/LInux or Android/Linux environment.
Trolley buses are - unfortunately - only widespread in the former soviet union and its client states
CH, here. The country has mostly been neutral during cold war and is far from being a client state.
But bigger cities here love trolley and trams too.
Electricity is easily available (thanks to alpine dams)
And city centers are rather densely populated - and thus the network of bus stops is also dense (you don't need hundreds of km of wire just to link 2 bus stops)
(for some reason soviet government seriously loved trolley buses, they have even built a trolley bus line in Afghanistan, back then they were there)
I would say that electric motors are simpler, smaller, easier to install into a vehicle. And are easy to ship around.
Whereas ICE are more a custom job that is vehicle specific.
Thus it's much easier for a Sovietic planned economy to make a 5-years plan to build a huge mega factory in one client state (e.g.: Bulgaria) and ship motors and install them into bus through the whole communist world.
Unload truck onto train, ship, unload train onto truck, deliver?
Drive truck onto flatbed railcar, ship, drive truck off railcar, deliver?
For trucks that are separate tractor trailer, just the trailer on the flatbed.
The law doesn't specify anything.
But it seems to me that for logistic reasons, the 3rd option you mention is the most popular :
the trailer is a standardized contrainer that can be moved from the truck to the train without needing to lose time for unloading/reloading the merchandise.
For cost reasons, any consumer electronic device will usually be implemented on the cheapest possible SoC with the bare minimum of unused resources (eMMC, RAM, MIPS etc.)
Nowadays ROM chips with a custom firmware burned in cost too much (add maybe a 0.5$ per unit), firmware is flashed on built-in EEPROM inside the SoC.
In theory, most of modern-day widgets have field-upgradeable firmware.
In practice, no company bothers to do the necessary work, specially since by the time the firmware upgrade is necessary, the device has already been sold and the money has been earned. There is no big immediate advantage in providing upgrade. It's an eventual long term advantage for the end-user, but by then the end-user can only regret having spent the money on the gadget.
(e.g.: Not exactly wireless DAB radio, but out of all the wireless bluetooth speakers I've seen, only Logitech/Ultimate Ears bothers to make regular updates that actually add new features. None of the DAB radio I've owned has ever bothered releasing a firmware upgrade, even if some did advertise the possibility)
For lots of music usage (again, anything beyond the handheld DAB/FM receiver*) the "bare minimum SoC" is already quite powerful.
e.g.: bigger multi-media device need AAC decoding capabilities (to play music from USB sticks / MP3 players in USB-Storage mode, etc.) - That's about what is needed to move a DAB-enabled device to DAB+.
e.g.: In Vehicle Infotainment have a fuckton more processing power (Some high end device are the equivalent of a big over-powered tablet / a small netbook), that's way more than enough for playing DAB+ (or even support OPUS).
*: the handhelds tend to be a single chip with a hardware DAB receiver piping its data straigh into a MP2/MP3 hardware core. There's no real CPU.
This kind chip is designed to be used as a DAB solution for media devices.
(e.g.: combine it with a CD player and a few such other parts, and you can make a cheap all-in-one audio device)
But the micro-controller on this SoC can run a firmware that gives it limited stand-alone properties: it can handle a few bare simple menus and can be wired straight to an LCD with a couple of buttons.
Thus, this kind of chip can be also useful to make dead-cheap handheld radios.
(example of such radio: Revo Pico+ - though it wasn't sold at a cheap price back then)
I happen to live in a populated city that brags about having a ton of electric/hydrogen fueled vehicles (Porto, Portugal),
Long range public transportation:
We have an extensive train network (all electric, thus mostly hydro-electric and nuke powered, with a little bit of solar and wind sprinkled in) (Thank the *Alps* for nearly perfectly clean hydroelectric - unlike tropical hydroelectric which tends to be giant glorified swamps)
It covers most of the territory except for remote less populated area (and as they are less populated, the long-range public transportation using busses hardly makes a dent in the total energy tally)
Most big cities have a dense network of tramway and trolley buses (aerial electric power delivery makes much more sense in a densely packed area) also sometime metro/subs for some cities.
They are also joined by (diesel) bus. But the electric propotion of short-range transportation is quite significant and hardly just for the show.
Private companies in public transportation / ride sharing:
Most taxi fleets in big cities tend to rely on hybrid vehicle (lower gaz consumption makes operations cheaper)
the rise of Uber (mostly privately owned car with classical ICE drives) is actually a step backward environmentally. (But as taking transportation instead of driving a car around is better anyway, the end tally might not be bad).
Private companies car sharing:
The main car sharing operator in Switzerland (mobility) operates a mixed fleet featuring ICE (mostly), hybrid (fewer) and electric vehicle (only a few, usually available at sharing stations where high electrical power is available : eg.: parking near trainstation usually feature 1 or 2 Renault Zoé. But other places feature them too. Random example : EPFL institute).
From that point of view we are less ecologically advanced than france, where the dominating car sharing companies tend to have all-electric fleets (e.g.: Autolib in Paris).
Though there are smaller CH player with electric fleets (e.g.: ElectrEasy)
So globally, in Switzerland, the role played by electricity in public transportation (specially by public company like national trains and city public transportation) is really significant.
Also, regarding merchandise :
Switzerland is peculiar in that transport of merchandise *across* the country is *forbidden in trucks*.
Trucks can be used to deliver merchandise to/from and within cities.
But if you want to transport merchandise long distance or across the country, it's mandatory to load it on trains.
When driving on the highway, you're going to see way much less trucks compared to other European countries (e.g.: Italy, France, Spain...)
Last but not least a few interesting corner case :
I few touristic cities (mostly in the mountains like Zermatt and Saas Fee) have completely banned ICE engines within the city (with a few exceptions like firefighters, ambulances)
Thus nearly the whole fleet is small electric glof-cart-like cars and taxis.
Fun to see (even if completely insignificant statistically to the rest of the country).
Fixed the title for you.
And I think that's genuinely the point of this:
it would be possible to run a whole cluster of compute nodes with VEGA GPUs,
and have all the data within a single unified address space across the whole cluster.
Just throw in a few IOMMU to handle access rights, and a fabric like Infiniband, or some PCI-E based one.
For bonus point have the storage it self being memory mapped non-volatile RAM (X-Point, etc.)
(But then you DO run out of address space - clusters tend to have data in the peta-byte range).
Even if some pages emit ultrasound - others will play sound and remind me to always mute.
And the default behaviour in firefox is to display a small "speaker" icon next the title of any tab that plays audio.
You can cut the audio off simply by clicking on the icon.
On android, the non-focused tabs don't even play audio by default (it's not possible to listen to music in a background tab).
Even if some PCs emits ultrasound, who will leave a mic on and run receiving sw? Not me, for sure.If this gets popular, muting the mic will be standard . .
The thing is : YOU might not be in control of the mic (that does the recording).
The whole point is locating YOUR laptop. So by definition, the mic that is doing the recording is on some other hardware.
- That could be hardware purposefully left to record.
- That could be the smartphone of some other user who's a lot less carefully than you (has a mic and location service both available, and currently abused by some random JS ad)
- That could be the government. (In some phone, the portion of the radio chipset that is not under your personal control is able to record audio and position. Happens in some Qualcom SoC, where the radio chips works as a kind of "north bridge" to the rest.
If mandated correctly, the information services of a country could remotely start to eavesdrop on whatever the phone is hearing around - as long as the radio chip is within range of a cell tower).
Only a tiny minirity nerds were ever wanking their micropeens over modular phones.
Meanwhile, Fairphone 2 is still selling well.
The key point is that, through their "environmental-friendly" policy, they have managed to find a way to make modules also appealing to a larger audience
(modules means easier to repair a phone by swapping a broken component instead of throwing it away whole)
and not only to the 3 geeks that are salivating about the idea of creating their own custom module to tap into the available internal USB pogopins.
Thus they have found a way to make modularity commercially viable and even trendy (by being eco-friendly).
We were just as early adopters, but in an effort to give as many as possible the finger it will be exclusively DAB+. So if you bought a DAB radio it has both been born and died in less time than most FM radios have lived.
Depends on the type of radio receiver.
DAB and DAB+ are nearly exactly the same.
Same modulation, same DAB emmiter data stream, same list of radio channel, same menu of available stations.
Only once you select a station inside the list available at the emitter, if it's a DAB one, you'll get an MP2-compressed audio stream with some old data correction scheme, if it's a DAB+ one, the audio stream will be AAC with Reed-Solomon error correction.
If you radio is a very low power one, with a single chip that handles everything in hardware (a portable hand held radio that should work with a few AA batteries), yup the SoC won't work with DAB+ because it only has a MP2/3 audio decompression hardware acceleration.
If it's anything bigger that can also play MP3/AAC/OGG files from a USB stick or SD-Card, then it's already has all the capability necessary to play a DAB+ station from the list available at your emitter.
It should be only a software/firmware update. If it's not availble, it's only the manufacturer's fault of being lazy / wanting you to rebuy new hardware.
Life is a whim of several billion cells to be you for a while.