But not more precise than Google Maps, which does exactly this.
Oh, do they ? In my (limited) experience I've only seen the "usual lanes plan when there's no construction work" that other satnavs provide too.
But maybe it's not available yet in all cities.
The truth is, even the gold standard for this sort of thing (Google Maps) doesn't have the amount of accuracy required. And can't. These things change too frequently to actually rely on them for that level of detail.
Yup, the guys at Google have often mentioned that indeed. Hence the whole idea of using cameras and teaching the cars to understand their environment.
That would be an improvement, certainly. It's still hard to see how it would be effective enough in practice as a crowdsourced venture. How are they going to convince enough people to outfit their cars with dash cams?
Actually, comma.ai has exactly managed that with chffr : convice enough people to install it so they can have enough data to teach their system.
Their strategy revolve around:
general cheerfulness of the crowd being happy to collaborate on an opensource project
chffr doubling as an actual useful dashcam with full logs of everything, including all that it can gather over the canbus from the odb port
gamifying the whole things with people boasting about their comma points on the forum
a few perks (top contributor could get hardware more easily)
.
If the efforts don't fall into too many different forks / split groups / etc. and if comma.ai doesn't pull an "occulus rift" fuck (e.g.: by suddenly declaring that the trained data will only exclusively be available to car companies that pay a license and to no end-user directly *), comma.ai has a *very good chance* of becoming such a "OpenStreet Map for Autopilots and Self-Driving Car"-like company.
On their blog, they even have documented preliminary experiments with generating a 3D map of the street out of the sensors - using only video and GPS, no LIDAR (chffr works with most android smartphone, it doesn't require a Caterpillar with the built-in lidar).
---
* : I think I've heard somewhere that comma.ai got approached by Tesla exactly for that, and that comma.ai refused and wanted to explicitely stay car-manufacturer-neutral.
Infotainment unit's don't have batteries. Car dashboards don't have batteries.
Technically they all run out of 12v lead batteries (internal combustion engine) or massive lithium battery stacks (electrical vehicle). But...
There isn't a single use case for the i.MX8 series where the SoC is the biggest consumer of power in the system.
...yes indeed, the infotainement basically just feeds out of the 12v instrument bus.
It's the job of other completely different devices to manage the power, and any way the consumption of the infotainment is dwarfed by that of the spark plugs and starter (ICE) or the electric motor (EV), and in both case the battery can be charged using the same electrical motor (respectively by the alternator / by regenerative braking).
At that scale, the power to the infotainment basically comes "for free".
The point of self-driving grade maps is to be precise down to the exact lane.
like "the two left lane are for turning left. the two right lane should be for 'straight ahead', except that the second from the right is currently blocked by construction works".
(i.e.: even more precise than what's available on commercial satnavs).
OSM doesn't go that precisely into details.
For that you need to crowdsource it from video feeds : - the first time the car goes there (either in actual autopilot mode, or in 100% pure manual mode with the camera only working as a glorified dash-cam, like comma.ai's chffr android app), do object recognition on the video feed, notice the street signage, lane arrows and construction work, and beam the info up to the mothership. - the next time one of the car goes there in autopilot mode while connected to the same cloud, the car knows to change to the right most lane, if the satnav's path finding calls for continuing straight ahead.
Google^H Waymo, Tesla, Comma.ai, and countless others are doing this kind of stuff now.
Johnny Mnemonic with Kaenu Reeves is kind of that : it's an adaptation of a Gibson's short story that introduce the universe and a few characters that Gibson will later use for its Sprawl trilogy.
Most of the current smartphone use chipset (mainly Qualcomm) that have the modem integrated into the SoC. That modem (for radio licensing reason) must run a closed proprietary blob, while having full access to the SoC's RAM. Thus you're only an OTA live update away (not even installing an Android upgrade, just sending new bits to your modem to execute), before wire taping law get applied to you and your data start getting siphoned away.
Purism want to make a phone with most of the sensitive part shut away in separate boxes that only speak a standard protocol. i.e.: modem in a separate chip, that only speaks a standard protocol (e.g.: showing up as an ethernet network) with no access RAM. No matter what rogue firmware it runs, such a modem cannot see your data, only sees an ethernet connection (and you're encrypting what goes through that one anyway, unless you stupidly trust the entire internet to be secure).
Thus, for all you concerns, 100% of your system runs opensource auditable code. It's not guaranteed to be secure *yet*, but can eventually be reviewed and secured.
Their target is to make a phone that can run 100% on exclusively free/libre opensource code.
That severly limits which SoC they can choose : - they need a chipset WITHOUT built-in modem, so the modem and its mandatory blob can be pushed out of the main system into an isolated box that only talks a standard protocol (so it doesn't have direct access to RAM. Unlike Qualcomm's chipset, where sometime the modem serves as the northbridge) - they need a chipset with opensource drivers supported by upstream linux kernel.
Currently, only Freescale i.MX6 fits the bill (Vivante GPU supported by Etnaviv driver), and the Freescale i.MX 8 is their best hope of next chip to be similarly supported.
Yes, it's an old SoC, with low to mid perf, but it's about the only one that fits the bill.
(It might have also been possible with some of the Nvidia Tegra chipsets that are supported by nouveau, but they don't fit the power envelope. Intel's is fully opensourced officially, but doesn't produce anything currently targeting the tablet/smartphone form factor. Qualcomm is completely out of question : even if some are supported by Freedreno, the integrated modem running untrusted proprietary binary firmware, while having full access to RAM is problematic)
Everything else? Nobody is running Linux on a day to day basis.
Yeah, nobody. Except people using smartphone running Android (still a Linux kernel, even if coupled with a weird non-GNU user-space) or nearly any other alternative system (Tizen, Sailfish,... all GNU/Linuxes. iOS (Darwin-like) and BlackBerry (QNX based) are the only exceptions). Except people connecting to a modem/router to go online (Linux has a quasi-monopoly on home routers) Except people using smartTVs, media players, set-top boxes Except people having cars (although QNX is popular on the more critical processors CPUs in a car, the Linux kernel is more popular on the infotainment center) Except people having ebooks (in Europe, the two most popular are Kobo [Linux + embed Qt] and Tolino [Android]) Except people using home NAS to backup their data (Linux is the most popular OS, see Synology for a concrete example) Except people using a USB boot disk from an antivirus maker to scan an infected PC (lots of such recovery sticks actually run a GNU/Linux bootdisk. e.g.: Kaspersky). Except people downloading photo from their camera over WiFi (there are even Wifi-enabled SD-Cards that feature a diminutive embed Linux webdav server) Except people printing document (currently HP has switched to derivative of webOS on their printers) Except people that use a high performance cluster, even for trivial reasons (hint: what did you think Youtube uses to re-compress videos to various formats ?) Except people using SatNav (ever since the original Tomtom device that was Linux powered). etc.
If a gadget has a CPU a little bit more powerful than a micro-controller, and it is not a desktop, chances are very high that it runs Linux.
The "year of Linux on pretty much every other fucking stuff except the desktop" has already passed for quite some time.
Can you expand a bit on how very strong Nokia's position was, the moment that iPhone went on sale?
Here's an overly long rant by an analyst, with lots of details of what went wrong.
I certainly don't remember things that way.
Part of this boils down "In my tiny corner of the world, everybody flocked en masse to the Jesus Phone !" - "Yeah but not all the planet Earth follows what happens to be popular in California, and billions of people can't afford Apple overpriced iGadgets while these billions are still in need of some portable communication tools, and Nokia phones are serving them better than anything".
After the release of iPhone Nokia was basically insanely huge everywhere except in the US (more precisely in the specific sub-market of high range smartphone in the US). In terms of absolute unit shipped or total revenue, that *still* put them ahead of every one else.
Having famous people promote the cars is a sign to me that the cars are not reliable.
The problam is that *you* is not *the general population*.
Us/.ers, given our tendencies, will tend to be over-obsessed with facts, logic, etc. compared to average joe six pack.
On the other hand, random everyday people tend to fall in for quite a lot of social cognitive bias. And if they see a celebrity endorsing something, they'll unconsciously give it more positive attention (there must be something good to it if ${celebrity} endorse it, ${celebrity} must have seen something positive in it).
None of the current crop of assistants (Cortana, Siri, Alexa, OkGoole!, or even just speech engines like Houndify and Dragon Dictate) does run locally except for extremely simple processing (like google detecting locally the "Ok, Google !" stanza, and only starts streaming the audio to the mothership afterwards).
The text commands and audio are transmitted to the company's cloud, and all further processing (full speech recognition when input is audio, then extracting the meaning/intention from the text, taking a decision, and suggesting actions) is entirely done there.
Means that for any kind of assistant to work, their company needs necessarily to have received all of your data (voice stream, chat log, etc.)
And due to the way these thing work (Deep Neural Nets need a big amount of data to train - basically replicating in silico the popular saying that you need to have been doing 10'000 hours of anything to be good at it) they NEED to be centralized on the cloud. It's not possible to have the learning done locally on your smartphone : not only it lacks processing power (for the backpropagation in the neural net) but it also lacks the big masses of data to train on.
So it would NOT be possible to have your very own local copy of Cortana (or at least not in learning mode. Maybe GPU acceleration could at least make possible to simply apply an already trained neural net depending on how big cortana is).
The choice is "an OS that runs the software I have to use" and "an OS that doesn't run the software I have to use."
And in some cases, mainly in "big data" analysis fields, even moreso in bioinformatics, the "OS that runs the software I have to use" tend to be flavors of Unix - so're basically limited to macOS X or Linux.
(With only very recently Windows starting to be able to run these same software, thanks to WSL).
If we don't use Windows 10, what other OS can we use instead? {...} Linux isn't really an option, especially with so many distros including systemd. I hate to say this, but I've found Windows 10 to boot more reliably for me than the versions of Debian and Ubuntu that use systemd! {...} We use Windows 10 because it's the only practical option.
On the other hand, *I* find Linux to be less of a nightmare whenever I upgrade storage (replace HDD with SSD, replace old busted optical bay with a bay holding an extra SSD, rebalance BTRFS on the fly to RAID1 across the 2 SSDs, etc.), compared to Windows 10 (Whenever you move windows partitions around. Or switch between BIOS and UEFI modes. Or replace DOS-partition scheme with GPT : Windows becomes unbootable and you must yet again get the recovery disk. Or you must do the above action using specially crafted tools by the manufacturer which often are... Linux bootdisks, actually).
So as the saying goes, "Your Mileage Might Vary".
In my case, I use openSUSE's Tumbleweed Linux disto because it's a very practical option for me.
Its ridiculous that Android phone I bought a year ago will never get a security update.
Part of this isn't due to Android, but to hardware manufacturers - drivers aren't avaible / updated.
The current tendencies for chipset manufacturers is to fork whatever Linux kernel version happens to be the base of the Android Letter of the day. Then slap some binary drivers on it, and call it a day and never ever touch it again.
Hardware manufacturer come, and to be faster to market, basically just re-adapt an existing board design from the chipset manufacturer, and quickly botch some android user space on top of the above mentioned kernel. Once they sold the smartphones to retailer they abandon it and move to the next model.
By they time you want an update for your phone, the phone's manufacturer might not even exist anymore or they might have abandonned it long ago. Even if they wanted to make updates, there would by then the problems of getting a newer kernel + userland drivers set - but the chipset manufacturer has completely abandoned it.
Google might be happily still providing newer versions of android and fixes (currently all the way back to Android KitKat), it will take some tedious work by the people of LineageOS (formely CyanogenMod) to build an image you can actually use......as long as your phone will actually authorize you to flash it.
Moving to another OS isn't going to fix these troubles : you'll still be bound to the same binary drivers (running thanks to libhybris adaptation layer, because you want an actual GNU/Linux OS instead of the weird Android user space and driver API).
Case in point, the original Jolla 1 smartphone by Jolla Oy. It runs Sailfish OS (a descendent of Nokia's Meamo/Meego, and cousin of Samsung's Tizen). As of 2017, the os itself is still getting the same upgrades as all the other devices officially supported by Jolla Oy (currently 2.1.1, with 2.1.2 coming out soonish). But you're still stuck running on the Jolla whatever Linux kernel (3.4.xx) Qualcomm happened to fork back when they developped the drivers for the onboards Snapdragon 400. And thus the provided android application compatibility (Aliendalvik by Myriad) is limited to Jellybean, not Kitkat like on the other devices supported by Jolla.
The only exception are a few chipsets by Intel (official upstream drivers in kernel - but they exitet the smartphone market), by Qualcomm (some of their GPU can work with Freedreno driver, if you're lucky) and a couple of chipsets by Freescale (some of their Vivante GPUs are supported by Etnaviv driver, or support could be comming soon. That's part of the reasons why they got picked up by Purism for their Librem smartphone). But none of the sexier more powerful chipset is currently supported well enough by opensource drivers. Thus you're still stuck with manufacturer-provided, outdated "android" linux kernel and drivers.
Instead, they let RIM eat their lunch, then Apple, then Google. All the while kinda half-assing multiple doomed attempts in what is reminiscent of a shakespearean tragedy.
And managed to burn Nokia in the process (who were in a very strong position before Stephan Elop and Microsoft happened to them).
There exists a non-zero chance the reason we don't see the galaxy teeming with extraterrestrial life is that an advanced civilization simply comes along and destroys all other civilizations once they make themselves known. They do this simply because...why put your own civilization at risk on the chance the new civilization is or becomes malevolent? {..} Then 200 years later the Sun is obliterated.
It could also go the other way around. Any civilisation that has is that much advance, that high on the kardashev scale, absolutely *MUST* evolve to be pacifist and non-aggressive. Otherwise they would have had a too high risks of destroying themselves in the process of advancing (it's one thing to be aggressive, when the worst that could happen is that your primitive tribes club each other to death while other tribe in the neighbouring valley don't even notice. It's an entirely other thing to have trigger-happy rulers when you possess enough power to obliterate your whole solar system).
The apparent lack of advanced civilisation is then simply a consequence of not all of them learning fast enough to de-escalate violent situation, and too many of these civilisation taking the wrong turn on their historical version of the Cuban missile crisis - having opted instead for the nuclear option.
Also, if alien life behaves anything like earth life, and they somehow manage to get here, we will all likely get to go off planet - probably via some high powered energy beam weapon.
On the other hand, if any contacted civilisation would have been that aggressive, chances are high that they will have had plenty of opportunities to blow themselves all up in the process of getting to that point on the Kardashev scale. Think of their historical equivalent of the Cuban missile taking the wrong (destructive non pacifist) turn. - Now think of it, but in the context of an advanced enough civilisation with enough technology and power to be capable of interstellar travel. A "wrong turn" of crisis would have vastly devastating consequences on their existence.
Any civilisation more advanced than us is *bound* to be more pacifist than us, otherwise they would certainly not have been able to survive up to that point. It would be very unlikely for such an advanced civilisation to be nefarious to the point of taking opportunity to send "viruses" to us, because such a nefarious civilisation that also has access to such advanced technology would have had plenty of opportunity to obliterate it self before.
Our history is filled with critical point where we had way too much destructive power, and had the deciding powers being any much crazier, we would have destroyed our civilisation. Think about the cuban missile crisis. Even think at the undergoing North Korean crisis. If any of the involved powers was any bit more trigger happy, we would have not be here around anymore.
To paraphrase the movie catch phrase : with great power comes great responsibility. As we advance in technological power, we need to also advance in responsibility or risk not existing anymore.
By the time any civilisation has significantly enough risen on the Kardashev scale to be capable of interstellar contacts, it must have becom enough violence/cruelty-averse to not have been blown itself in the process of that rise.
Excel spreadsheet {...} So nowadays I end up relying on various modules far more often than I use perl's own built-in formatting capabilities.
Story of my life.
We arrived at the conclusion that Excel Spreadsheets (with drop-down selectors) is the best ever interface that our lab technician would dream of for the LIMS of our high throughput sequencer.
(Ended up using XLSXWriter and co, so that the LIMS could spit out Excel Spreadsheet and accept re-uploads of them).
and use it as if it was just a more convenient awk. Which has no useful capabilities for reporting except an END clause.
Actually, like many other languages (including shell, including Perl, including PHP and even including Python where they are first-class citizens), AWK has support for C-style formating (printf, etc.) and thus allow some form of formatting.
Of course it's not as beautifully integrate into the core language as perlform, and isn't visually clear (As in "%-10s" vs "@") and neither as versatile (you're going to abuse a lot of function calls and passing variables and hashes around to achieve a poor man's replacement of declaring formats). But still it has some limited formating capabilities.
On the other hand, in the modern days (where terminal is seldom the final output of a program, and instead JSON message sent to some Javascript WebApp is increasingly the norm), this kind of formatting is becoming less useful.
(Just like PHP's ability to interweave static blocks (e.g.: of HTML) with dynamic code isn't making much sense with modern nearly entirely dynamically generated websites).
But if you dig into *why* these prophecies tended to self-fulfil, things would be clearer.
Star Trek is a fiction show by writters. Their only limitation is their imagination, and they tended to imagine what would be cool to have no matter how realistic (pocket computer) or not (warp drive). The show writter "simply" put form to some of the more technical human desires.
Then some of the gadget ended up appearing in real life, because : - lots of engineers agreed or were already thinking that, indeed, it would be could. - in those fields the technology would eventually catch up.
On the other hand, I you would have ask *market* specialist, investors, etc. during the same time instead of asking Sci-Fi writers you would have gotten something along the lines of : "More big iron ! Invest in IBM now ! " and "Teletype to connect to the big iron in each middle-upper class home ! Invest in Bell !"
Back in 2007, the Slashdot consensus was that the iPhone was lame, had pointless features, and was doomed to failure since nobody wanted to use an iPod as a phone.
On the other hand, another slice of the geek population has had been having PDAs such as Psion and Palm for a around a decade. They already had pocket computers, with very rich ecosystems of application. And some already used to connect their PDAs to networks thank to nascent Wifi or by pairing (Bluetooth or even IrDA) with GPRS-capable dumb phones.
(That was my case : Palm IIIc, then Tungsten T3. Pairing with IrDA, then Bluetooth to an Ericsson T39, then a Wifi SDIO card)
To that crowd, an iPhone was pretty lame and useless. - the only new feature was the multi-touch capacitive screen (most PDA used resistive single touch). - it lacked everything else that made PDA great.
It wasn't such a huge innovation that we weren't able to grasp it. It was a huge step backward.
I didn't want a smartphone, because I already had way much better when the original iPhone launched.
Slashdot Mirror
But not more precise than Google Maps, which does exactly this.
Oh, do they ? In my (limited) experience I've only seen the "usual lanes plan when there's no construction work" that other satnavs provide too.
But maybe it's not available yet in all cities.
The truth is, even the gold standard for this sort of thing (Google Maps) doesn't have the amount of accuracy required. And can't. These things change too frequently to actually rely on them for that level of detail.
Yup, the guys at Google have often mentioned that indeed.
Hence the whole idea of using cameras and teaching the cars to understand their environment.
That would be an improvement, certainly. It's still hard to see how it would be effective enough in practice as a crowdsourced venture. How are they going to convince enough people to outfit their cars with dash cams?
Actually, comma.ai has exactly managed that with chffr : convice enough people to install it so they can have enough data to teach their system.
Their strategy revolve around :
.
If the efforts don't fall into too many different forks / split groups / etc. and if comma.ai doesn't pull an "occulus rift" fuck (e.g.: by suddenly declaring that the trained data will only exclusively be available to car companies that pay a license and to no end-user directly *), comma.ai has a *very good chance* of becoming such a "OpenStreet Map for Autopilots and Self-Driving Car"-like company.
On their blog, they even have documented preliminary experiments with generating a 3D map of the street out of the sensors - using only video and GPS, no LIDAR (chffr works with most android smartphone, it doesn't require a Caterpillar with the built-in lidar).
---
* : I think I've heard somewhere that comma.ai got approached by Tesla exactly for that, and that comma.ai refused and wanted to explicitely stay car-manufacturer-neutral.
Infotainment unit's don't have batteries. Car dashboards don't have batteries.
Technically they all run out of 12v lead batteries (internal combustion engine) or massive lithium battery stacks (electrical vehicle). But...
There isn't a single use case for the i.MX8 series where the SoC is the biggest consumer of power in the system.
...yes indeed, the infotainement basically just feeds out of the 12v instrument bus.
It's the job of other completely different devices to manage the power, and any way the consumption of the infotainment is dwarfed by that of the spark plugs and starter (ICE) or the electric motor (EV), and in both case the battery can be charged using the same electrical motor (respectively by the alternator / by regenerative braking).
At that scale, the power to the infotainment basically comes "for free".
Upgrade the universe to 64-bits already!
But then I'll need to reboot the universe, and I still have a few unsaved tabs.
The point of self-driving grade maps is to be precise down to the exact lane.
like "the two left lane are for turning left. the two right lane should be for 'straight ahead', except that the second from the right is currently blocked by construction works".
(i.e.: even more precise than what's available on commercial satnavs).
OSM doesn't go that precisely into details.
For that you need to crowdsource it from video feeds :
- the first time the car goes there (either in actual autopilot mode, or in 100% pure manual mode with the camera only working as a glorified dash-cam, like comma.ai's chffr android app), do object recognition on the video feed, notice the street signage, lane arrows and construction work, and beam the info up to the mothership.
- the next time one of the car goes there in autopilot mode while connected to the same cloud, the car knows to change to the right most lane, if the satnav's path finding calls for continuing straight ahead.
Google^H Waymo, Tesla, Comma.ai, and countless others are doing this kind of stuff now.
as does comma.ai
(With the small difference that their software is opensource,
and their hardware is not vendor specific)
Johnny Mnemonic with Kaenu Reeves is kind of that :
it's an adaptation of a Gibson's short story that introduce the universe and a few characters that Gibson will later use for its Sprawl trilogy.
Their main arguments are around the chipset :
Most of the current smartphone use chipset (mainly Qualcomm) that have the modem integrated into the SoC.
That modem (for radio licensing reason) must run a closed proprietary blob, while having full access to the SoC's RAM.
Thus you're only an OTA live update away (not even installing an Android upgrade, just sending new bits to your modem to execute), before wire taping law get applied to you and your data start getting siphoned away.
Purism want to make a phone with most of the sensitive part shut away in separate boxes that only speak a standard protocol. i.e.: modem in a separate chip, that only speaks a standard protocol (e.g.: showing up as an ethernet network) with no access RAM. No matter what rogue firmware it runs, such a modem cannot see your data, only sees an ethernet connection (and you're encrypting what goes through that one anyway, unless you stupidly trust the entire internet to be secure).
Thus, for all you concerns, 100% of your system runs opensource auditable code. It's not guaranteed to be secure *yet*, but can eventually be reviewed and secured.
Their target is to make a phone that can run 100% on exclusively free/libre opensource code.
That severly limits which SoC they can choose :
- they need a chipset WITHOUT built-in modem, so the modem and its mandatory blob can be pushed out of the main system into an isolated box that only talks a standard protocol (so it doesn't have direct access to RAM. Unlike Qualcomm's chipset, where sometime the modem serves as the northbridge)
- they need a chipset with opensource drivers supported by upstream linux kernel.
Currently, only Freescale i.MX6 fits the bill (Vivante GPU supported by Etnaviv driver), and the Freescale i.MX 8 is their best hope of next chip to be similarly supported.
Yes, it's an old SoC, with low to mid perf, but it's about the only one that fits the bill.
(It might have also been possible with some of the Nvidia Tegra chipsets that are supported by nouveau, but they don't fit the power envelope.
Intel's is fully opensourced officially, but doesn't produce anything currently targeting the tablet/smartphone form factor.
Qualcomm is completely out of question : even if some are supported by Freedreno, the integrated modem running untrusted proprietary binary firmware, while having full access to RAM is problematic)
Everything else? Nobody is running Linux on a day to day basis.
Yeah, nobody. ... all GNU/Linuxes. iOS (Darwin-like) and BlackBerry (QNX based) are the only exceptions).
Except people using smartphone running Android (still a Linux kernel, even if coupled with a weird non-GNU user-space) or nearly any other alternative system (Tizen, Sailfish,
Except people connecting to a modem/router to go online (Linux has a quasi-monopoly on home routers)
Except people using smartTVs, media players, set-top boxes
Except people having cars (although QNX is popular on the more critical processors CPUs in a car, the Linux kernel is more popular on the infotainment center)
Except people having ebooks (in Europe, the two most popular are Kobo [Linux + embed Qt] and Tolino [Android])
Except people using home NAS to backup their data (Linux is the most popular OS, see Synology for a concrete example)
Except people using a USB boot disk from an antivirus maker to scan an infected PC (lots of such recovery sticks actually run a GNU/Linux bootdisk. e.g.: Kaspersky).
Except people downloading photo from their camera over WiFi (there are even Wifi-enabled SD-Cards that feature a diminutive embed Linux webdav server)
Except people printing document (currently HP has switched to derivative of webOS on their printers)
Except people that use a high performance cluster, even for trivial reasons (hint: what did you think Youtube uses to re-compress videos to various formats ?)
Except people using SatNav (ever since the original Tomtom device that was Linux powered).
etc.
If a gadget has a CPU a little bit more powerful than a micro-controller, and it is not a desktop, chances are very high that it runs Linux.
The "year of Linux on pretty much every other fucking stuff except the desktop" has already passed for quite some time.
Dragon Dictate only works locally on desktops, and for some markets only.
There's currently no local Dragon Dictate for embed device.
There's currently no local medical dragon dictate, it only exists as a cloud-only product.
Can you expand a bit on how very strong Nokia's position was, the moment that iPhone went on sale?
Here's an overly long rant by an analyst, with lots of details of what went wrong.
I certainly don't remember things that way.
Part of this boils down "In my tiny corner of the world, everybody flocked en masse to the Jesus Phone !" - "Yeah but not all the planet Earth follows what happens to be popular in California, and billions of people can't afford Apple overpriced iGadgets while these billions are still in need of some portable communication tools, and Nokia phones are serving them better than anything".
After the release of iPhone Nokia was basically insanely huge everywhere except in the US (more precisely in the specific sub-market of high range smartphone in the US).
In terms of absolute unit shipped or total revenue, that *still* put them ahead of every one else.
Having famous people promote the cars is a sign to me that the cars are not reliable.
The problam is that *you* is not *the general population*.
Us /.ers, given our tendencies, will tend to be over-obsessed with facts, logic, etc. compared to average joe six pack.
On the other hand, random everyday people tend to fall in for quite a lot of social cognitive bias. And if they see a celebrity endorsing something, they'll unconsciously give it more positive attention (there must be something good to it if ${celebrity} endorse it, ${celebrity} must have seen something positive in it).
None of the current crop of assistants (Cortana, Siri, Alexa, OkGoole!, or even just speech engines like Houndify and Dragon Dictate) does run locally except for extremely simple processing (like google detecting locally the "Ok, Google !" stanza, and only starts streaming the audio to the mothership afterwards).
The text commands and audio are transmitted to the company's cloud, and all further processing (full speech recognition when input is audio, then extracting the meaning/intention from the text, taking a decision, and suggesting actions) is entirely done there.
Means that for any kind of assistant to work, their company needs necessarily to have received all of your data (voice stream, chat log, etc.)
And due to the way these thing work (Deep Neural Nets need a big amount of data to train - basically replicating in silico the popular saying that you need to have been doing 10'000 hours of anything to be good at it) they NEED to be centralized on the cloud.
It's not possible to have the learning done locally on your smartphone : not only it lacks processing power (for the backpropagation in the neural net) but it also lacks the big masses of data to train on.
So it would NOT be possible to have your very own local copy of Cortana
(or at least not in learning mode. Maybe GPU acceleration could at least make possible to simply apply an already trained neural net depending on how big cortana is).
The choice is "an OS that runs the software I have to use" and "an OS that doesn't run the software I have to use."
And in some cases, mainly in "big data" analysis fields, even moreso in bioinformatics, the "OS that runs the software I have to use" tend to be flavors of Unix - so're basically limited to macOS X or Linux.
(With only very recently Windows starting to be able to run these same software, thanks to WSL).
If we don't use Windows 10, what other OS can we use instead? {...} Linux isn't really an option, especially with so many distros including systemd. I hate to say this, but I've found Windows 10 to boot more reliably for me than the versions of Debian and Ubuntu that use systemd! {...} We use Windows 10 because it's the only practical option.
On the other hand, *I* find Linux to be less of a nightmare whenever I upgrade storage (replace HDD with SSD, replace old busted optical bay with a bay holding an extra SSD, rebalance BTRFS on the fly to RAID1 across the 2 SSDs, etc.), compared to Windows 10 (Whenever you move windows partitions around. Or switch between BIOS and UEFI modes. Or replace DOS-partition scheme with GPT : Windows becomes unbootable and you must yet again get the recovery disk. Or you must do the above action using specially crafted tools by the manufacturer which often are... Linux bootdisks, actually).
So as the saying goes, "Your Mileage Might Vary".
In my case, I use openSUSE's Tumbleweed Linux disto because it's a very practical option for me.
Other users might have even other experiences.
Its ridiculous that Android phone I bought a year ago will never get a security update.
Part of this isn't due to Android, but to hardware manufacturers - drivers aren't avaible / updated.
The current tendencies for chipset manufacturers is to fork whatever Linux kernel version happens to be the base of the Android Letter of the day.
Then slap some binary drivers on it, and call it a day and never ever touch it again.
Hardware manufacturer come, and to be faster to market, basically just re-adapt an existing board design from the chipset manufacturer, and quickly botch some android user space on top of the above mentioned kernel. Once they sold the smartphones to retailer they abandon it and move to the next model.
By they time you want an update for your phone, the phone's manufacturer might not even exist anymore or they might have abandonned it long ago. Even if they wanted to make updates, there would by then the problems of getting a newer kernel + userland drivers set - but the chipset manufacturer has completely abandoned it.
Google might be happily still providing newer versions of android and fixes (currently all the way back to Android KitKat), it will take some tedious work by the people of LineageOS (formely CyanogenMod) to build an image you can actually use... ...as long as your phone will actually authorize you to flash it.
Moving to another OS isn't going to fix these troubles : you'll still be bound to the same binary drivers (running thanks to libhybris adaptation layer, because you want an actual GNU/Linux OS instead of the weird Android user space and driver API).
Case in point, the original Jolla 1 smartphone by Jolla Oy. It runs Sailfish OS (a descendent of Nokia's Meamo/Meego, and cousin of Samsung's Tizen).
As of 2017, the os itself is still getting the same upgrades as all the other devices officially supported by Jolla Oy (currently 2.1.1, with 2.1.2 coming out soonish).
But you're still stuck running on the Jolla whatever Linux kernel (3.4.xx) Qualcomm happened to fork back when they developped the drivers for the onboards Snapdragon 400.
And thus the provided android application compatibility (Aliendalvik by Myriad) is limited to Jellybean, not Kitkat like on the other devices supported by Jolla.
The only exception are a few chipsets by Intel (official upstream drivers in kernel - but they exitet the smartphone market), by Qualcomm (some of their GPU can work with Freedreno driver, if you're lucky) and a couple of chipsets by Freescale (some of their Vivante GPUs are supported by Etnaviv driver, or support could be comming soon. That's part of the reasons why they got picked up by Purism for their Librem smartphone).
But none of the sexier more powerful chipset is currently supported well enough by opensource drivers. Thus you're still stuck with manufacturer-provided, outdated "android" linux kernel and drivers.
And this while still providing for legacy devices, even like their first Jolla smartphone.
(2.1.2 would be out for it any moment now)
Instead, they let RIM eat their lunch, then Apple, then Google. All the while kinda half-assing multiple doomed attempts in what is reminiscent of a shakespearean tragedy.
And managed to burn Nokia in the process (who were in a very strong position before Stephan Elop and Microsoft happened to them).
There exists a non-zero chance the reason we don't see the galaxy teeming with extraterrestrial life is that an advanced civilization simply comes along and destroys all other civilizations once they make themselves known. They do this simply because...why put your own civilization at risk on the chance the new civilization is or becomes malevolent? {..} Then 200 years later the Sun is obliterated.
It could also go the other way around.
Any civilisation that has is that much advance, that high on the kardashev scale, absolutely *MUST* evolve to be pacifist and non-aggressive.
Otherwise they would have had a too high risks of destroying themselves in the process of advancing (it's one thing to be aggressive, when the worst that could happen is that your primitive tribes club each other to death while other tribe in the neighbouring valley don't even notice. It's an entirely other thing to have trigger-happy rulers when you possess enough power to obliterate your whole solar system).
The apparent lack of advanced civilisation is then simply a consequence of not all of them learning fast enough to de-escalate violent situation, and too many of these civilisation taking the wrong turn on their historical version of the Cuban missile crisis - having opted instead for the nuclear option.
Also, if alien life behaves anything like earth life, and they somehow manage to get here, we will all likely get to go off planet - probably via some high powered energy beam weapon.
On the other hand, if any contacted civilisation would have been that aggressive, chances are high that they will have had plenty of opportunities to blow themselves all up in the process of getting to that point on the Kardashev scale.
Think of their historical equivalent of the Cuban missile taking the wrong (destructive non pacifist) turn. - Now think of it, but in the context of an advanced enough civilisation with enough technology and power to be capable of interstellar travel.
A "wrong turn" of crisis would have vastly devastating consequences on their existence.
Or you could imagine the other way around.
Any civilisation more advanced than us is *bound* to be more pacifist than us, otherwise they would certainly not have been able to survive up to that point.
It would be very unlikely for such an advanced civilisation to be nefarious to the point of taking opportunity to send "viruses" to us, because such a nefarious civilisation that also has access to such advanced technology would have had plenty of opportunity to obliterate it self before.
Our history is filled with critical point where we had way too much destructive power, and had the deciding powers being any much crazier, we would have destroyed our civilisation.
Think about the cuban missile crisis.
Even think at the undergoing North Korean crisis.
If any of the involved powers was any bit more trigger happy, we would have not be here around anymore.
To paraphrase the movie catch phrase : with great power comes great responsibility. As we advance in technological power, we need to also advance in responsibility or risk not existing anymore.
By the time any civilisation has significantly enough risen on the Kardashev scale to be capable of interstellar contacts, it must have becom enough violence/cruelty-averse to not have been blown itself in the process of that rise.
Excel spreadsheet {...} So nowadays I end up relying on various modules far more often than I use perl's own built-in formatting capabilities.
Story of my life.
We arrived at the conclusion that Excel Spreadsheets (with drop-down selectors) is the best ever interface that our lab technician would dream of for the LIMS of our high throughput sequencer.
(Ended up using XLSXWriter and co, so that the LIMS could spit out Excel Spreadsheet and accept re-uploads of them).
and use it as if it was just a more convenient awk. Which has no useful capabilities for reporting except an END clause.
Actually, like many other languages (including shell, including Perl, including PHP and even including Python where they are first-class citizens), AWK has support for C-style formating (printf, etc.) and thus allow some form of formatting.
Of course it's not as beautifully integrate into the core language as perlform, and isn't visually clear (As in "%-10s" vs "@") and neither as versatile (you're going to abuse a lot of function calls and passing variables and hashes around to achieve a poor man's replacement of declaring formats).
But still it has some limited formating capabilities.
On the other hand, in the modern days (where terminal is seldom the final output of a program, and instead JSON message sent to some Javascript WebApp is increasingly the norm), this kind of formatting is becoming less useful.
(Just like PHP's ability to interweave static blocks (e.g.: of HTML) with dynamic code isn't making much sense with modern nearly entirely dynamically generated websites).
Self fulfilling prophecy, mostly.
But if you dig into *why* these prophecies tended to self-fulfil, things would be clearer.
Star Trek is a fiction show by writters. Their only limitation is their imagination, and they tended to imagine what would be cool to have no matter how realistic (pocket computer) or not (warp drive).
The show writter "simply" put form to some of the more technical human desires.
Then some of the gadget ended up appearing in real life, because :
- lots of engineers agreed or were already thinking that, indeed, it would be could.
- in those fields the technology would eventually catch up.
On the other hand, I you would have ask *market* specialist, investors, etc. during the same time instead of asking Sci-Fi writers you would have gotten something along the lines of :
"More big iron ! Invest in IBM now ! " and "Teletype to connect to the big iron in each middle-upper class home ! Invest in Bell !"
Back in 2007, the Slashdot consensus was that the iPhone was lame, had pointless features, and was doomed to failure since nobody wanted to use an iPod as a phone.
On the other hand, another slice of the geek population has had been having PDAs such as Psion and Palm for a around a decade.
They already had pocket computers, with very rich ecosystems of application.
And some already used to connect their PDAs to networks thank to nascent Wifi or by pairing (Bluetooth or even IrDA) with GPRS-capable dumb phones.
(That was my case : Palm IIIc, then Tungsten T3. Pairing with IrDA, then Bluetooth to an Ericsson T39, then a Wifi SDIO card)
To that crowd, an iPhone was pretty lame and useless.
- the only new feature was the multi-touch capacitive screen (most PDA used resistive single touch).
- it lacked everything else that made PDA great.
It wasn't such a huge innovation that we weren't able to grasp it.
It was a huge step backward.
I didn't want a smartphone, because I already had way much better when the original iPhone launched.