People want to make a fully identically functioning system that they can then replicate activity patterns on.
Instead, make use of the way the brain (and subsequently, the consciousness operating inside it) incorporates and induces activity on implanted devices.
Basically, stop with all this scanning shit. Instead, focus on a single, uniform platform that is well known and easy to simulate, then implant the patient with some implants that link the simulation with the still living organic brain. The organic brain will incorporate the functionality of the simulation. In short, go the ship of theseus route. As the brain begins to fail from either injury or old age, it will rely more and more on the simulation hardware, until eventually, it gives up. If you made the right connections, you will have valid activity in the simulation after the death of the organic component. Congrats, you have an upload.
This requires a very robust simulation platform though, which we do not currently have. DARPA is doing some interesting research on simulating neural columns, and the last I heard anything concrete was years ago, so if the project is still active, I can only assume they have gotten much better at their simulation. I understand that Chinese researchers are also working on simulating neuronal columns-- https://www.youtube.com/watch?... (Still, as interesting as these appear, they are noplace near mature enough to have attached to living humans as ancillary networks to support and replace function.)
Again, the issue of "Every organic brain is so vastly different, there is no way we can scan the physical networks for replication!" becomes less important, when you instead say "Let's build a generic, uniform simulation platform that we can then attach, and exploit neuroplasticity of the organic side for deep integration."
Zuck, like many corporate suits, cannot wrap his head around the idea that there are alternatives to the profit motive.
As such, his response is entirely focused there. It basically boils down to:
"There are only two ways that a service that facilitates massive communication world wide can exist: You either pay a subscription fee, or you pay with your eyeball time to advertisers. People have more free time than free cash, so our model makes sense. I am defending this model, because it is the most sensible of the two, and disparage anyone who says that this turns the users of the service into a product we sell to advertisers, especially when the ones making the claim use abusive pricing to enrich themselves at their customer's expense. You should not decide that you love the people hurting you by agreeing with the notion that providing the more scarce asset over the more abundant asset is somehow better for you."
EG, he cannot wrap his head around the idea that there are government supplied utilities, and bulk data delivery can be one of them-- Or that there might be valuable services offered totally for free on ideological grounds. (See for instance, most FOSS development.)
He cannot imagine a world where somebody with the initial outlay for expenses (or a group of people working collectively) could produce something like facebook. Apparently he is not very attentive, or has tunnel vision.
In his world, the people working at the volunteer soup kitchen are there for ulterior motives. ALWAYS.
These devices are essentially a chromebook, just without the custom chromeOS boot loader.
Under the hood they are basically the same kind of beast: eMMC based storage, 4gb of non-upgradable RAM, Celeron or Atom processor, SPI/i2C based bus with keyboard and mouse attached.
If you can get the chromebook cheaper, you are basically getting the same thing, just with a little extra legwork needed. Some chromebooks have NGFF based storage, which you can replace with a fantastically larger storage device. (Most are just eMMC shit though.)
These devices really are made for windows 10, and wont run any microsoft product that is older. (this is due to the use of the SPI/i2C bus architectures for input devices, sound hardware, etc) They basically *NEED* the advanced NTFS compression options offered to be remotely usable, since they are basically crippled by design in terms of storage.
If you use Linux, you will need to use a kernel that supports btrfs, and use the compression. It is also sensible to use an EXT4 volume on the microSD slot (if you have one) that is properly partitioned and formatted for the erase block sizes of that card. (Requires research on the user's part to do correctly. Incorrect formatting and partitioning will kill the card in about 6 months.)
In either case, dont expect these devices to be amazingly wonderful speed demons. Where they excel is being able to run continuously for 8+ hours straight under active use, and being ultraportable.
Darth Pai does not like it when big telecom's interests are threatened. Your tiny satellites, designed to burn up in the atmosphere, pose a significant risk of colliding with established interests. As such, we refused your launch request, then you went over his helmet, and how he he will crush your balls, pitiful startup weaklings.
I have virtualbox installed on my hacked celes (Samsung chromebook 3). It is nothing to write home to mommy about, but it can run another OS fairly well, all things considered.
To be fair though, the Celes has a celeron CPU, instead of the more "atom like" cpus in most other intel chromebooks.
If you want one that is better suited to virtualization/daily driving, you want this guy (especially if you want to upgrade the internal storage to something more reliable than eMMC/microSD)
Just link straight to MrChromebox.tech, since he has the needful to do the firmware portion in the bag right there for nearly all chromebooks in the wild.
All you need to do is remove the write protect screw/jumper, turn on dev mode, then run his script. Booya, bob's your uncle. His firmware has added bonus features, in that it reprograms the chromebook's embedded microcontroller so that the keyboard emulates a PS2 interface, allowing more OSes to run without as much hassle, as well as poking the sound hardware to better approximate an HD Audio Bus device.
Personally, I run GalliumOS as the primary on my Samsung chromebook 3 (celes). I have one of the older ones that only has 2gb of RAM, and I desperately need zram for it to be useful. With how weaksauce it is anyway, nearly any game that 'could' run on it, will also work in WINE, so I dont really need windows.
Seriously, I know editing is hard and all, but can you please tell us-- Is this 4 and a half days, 4 and a half weeks, 4 and a half months, or 4 and a half years?
There is also a very old idea of using memory elements directly to compute results, which is true memputing. (There are few examples of this, because it is costly as an architecture-- but your brain is a pretty good biological example. The same components are used for data storage, as well as data processing.)
Given that such "Computational Ram" devices already exist in the wild, I fail to see why more novel hardware is needed, excepting as a refinement of concept?
I a reminded of Torvald's scathing emails about Intel, their proposed patch sets, and how they pointed toward intel wanting to make future chips "Fast but insecure" by default, and requiring the BIOS or OS to tell the CPU "No bitch, secure mode only please", just so they could continue to claim benchmark scores (naturally, with the anti-spectre and meltdown patches disabled so the chip runs really fast.)
Hopefully these silicon level fixes are *ACTUAL* fixes to the methodology used by the speculative execution implementation of the chip, so that speculative execution still is active, but the chip no longer leaves bits and pieces in the processor cache that can be exploited, and that it does this by default.
The boot loader looks for an unsigned kernel and initrd on a specific partition, formatted as FAT32, with a specific file name.
You can bake your own and put it on the drive, and the mycloud will boot that image and initrd without complaints.
In the community pages, we have been working on a straight up clean debian for quite some time. There are instructions on how to configure and compile your own kernel from the stock device tree.
Considering that I own one of these devices, AND I participate frequently on the WD community group for this device, I can confirm that the base package is identical between consumer and midsize business class offerings. I can *ALSO* confirm that we have proposed workable patches on many numerous occasions, in every "Leave feedback" location WD makes available. (Protip, THEY IGNORE US.)
At least on the older Gen1 consumer mycloud units, the web interface and the/etc/passwd file were hosted by a real, writable filesystem that could handle persistent changes, meaning that the information we share in the community pages could be used to correct the security vulnerabilities if you wanted to take the matter into your own hands. Many people did this.
On the Gen2 however, WD decided that the user being able to modify the root file system persistently was just not something they felt comfortable with. It is a ram-backed root file system from an initial ramdisk, into which a cramfs container gets automatically mounted at a defined mount point. The cramfs container contains the web UI, and all the major system binaries./etc/passwd and pals are all obliterated on every boot, because they live in the initial ramdisk image.
That said, the hardware itself is *NOT* that bad. Just the horror-show WD offers software wise. (STAY THE FUCK AWAY FROM WD-SYNC.) There is a russian frequent contributor to the community site who has created instructions for a "From scratch" pure debian deployment on the Mycloud (both gen1 and gen2) units. This completely replaces the WD software with fully FOSS software, and gives the user full control over the unit.
Many people in the community board are Americans, like myself. We have proposed many solutions and fixes that we have locally implemented and given local testing to. We are not afraid to propose solutions, or to do things ourselves. The problem is that big companies do not feel comfortable testing things (Its a cost center! How horrible!!), and would rather just take a generic canned product, slap their name on it, and run. That is hardly an American thing.
The issue with H1B labor is that H1B software people often tend to do the same thing. They will take the homebrewed things people like myself make, (which have **NOT*** been sufficiently tested to mass deploy!!!!!!), slap their names on them and run with them, and their bosses, doing the same exact thing, are just peachy keen with this.
As for your assertion that we should make our own startups--- Do you have any idea how much the industry is stacked against new blood entering the market? Are you fucking delusional? Do you think that there are no people in the US that are just straight up makers, builders, and engineers who do shit for fun, who would love to make neat products available to people? The major reasons why we dont have many people doing that are two-fold; 1) Our erstwhile government frowns on individual thought and self-empowerment of the citizenry, and actively promotes a narrative that if you see your neighbor making something suspicious, he is probably an islamic terrorist making some kind of improvised explosive to blow your kids up on the bus with, or some crazy shit. The mainstream press eats that shit up like candy because it is over the top, and our culture is conditioned to soak it up like a sponge. There is a damned near moral panic against people doing neat stuff in garages these days. 2) Big corporations dont like new products entering the market, so they lobby to require "You must be this big to play" hurdles thrown in. Now, not only do you need to have a good idea and a working prototype, you need to have your entire product vetted for intellectual property form other vendors, even if you have no idea those vendors even exist, (and in many cases, are simply patent trolls!), which means hiring a small team of lawyers--- You also need investment capital to meet stringent manufacturing and materials use regulations, pay for init
The thing is, much like the clothing patterns I mentioned, the ability to make your own (the core tenet of "freedom" espoused by RMS as needed for GNU is the freedom to modify, not to be free from pricetags.) and to distribute those changes is a thing that only appeals to enthusiasts. In the case of clothing, that is people who enjoy playing with textiles, or doing neat things with garments.
In the case of software, that is your typical computer enthusiast. Exactly the people that the masses ARE NOT.
This does not prevent the masses from using product that is freely redistributable and alterable, (and just not exercising their freedom to do said actions themselves, instead relying on those that have such an impulse), but the realities of being practical and efficient with resources and resource availability means that for the same reasons mass manufactured clothing is popular, proprietary large-development-cycle software (with its non-free paradigms) will dominate, because the freedom espoused has no value to them. (Again, they never exercise that freedom, even when it is right there, because it does not interest them.)
The people that really care about free software do so because they are enthusiasts. The average person does not care about that, and just wants something that works with minimal costs (either financial, or in terms of invested time or mental effort.)
I think you just chose to see what you wanted to see in my statement, rather than what I actually said. Linux (and GNU software as a whole) appeal to enthusiasts, because the freedoms that are cornerstones of their development process are of value to enthusiasts. Non-enthusiasts are not interested. Software made by enthusiasts, for enthusiasts, will be poor fits for non-enthusiasts, especially when more simplified offerings exist.
The problem, is that Linux appeals to people who are computer enthusiasts-- people who LOVE computers, because they are simply amazing things, and they want to get the most out of that purchase.
Most people are not like that. They want a computer to do a very short list of things, and want one that will never slow down, break, or get infected with something. For most people, that thing is "I need the internet, facebook, and stuff for work/school." The less they have to actually know about computers, or how computers work (EG, the more "Magic box" like they are) the happier these people are.
Linux dares to expose its internals, and worse yet, DEMANDS that you learn about how it works underneath in order to use it effectively. That is why it has never, and likely will never, take off as a mainstream desktop.
Apple and Microsoft have created the "Shiny plastic experience", and people love it. Linux might as well say "Batteries not included, setup time 6 hours, major assembly required" on the box.
Asking why Linux is not a mainstream desktop environment is like asking why McCalls clothing patterns are not the dominant source of apparel in the market. Sure, you can customize the clothing however you want, and you can modify the patterns to your hearts content--- But dammit, you gotta get the cloth, cut it, sew it together, and all that shit. Why bother when you just want a fashionable new sport top, eh? People would rather spend the money on something somebody else already put together-- VIOLA-- OSX and pals. Shiny plastic. No work.
Linux needs to stop chasing this fantasy where everyone stops being lazy gits and becomes excited computer enthusiasts. They need to understand that they are a niche market, and do that niche very well. Last I checked, that was the Unix philosophy anyway.
For this reason I am opposed to the efforts of Poettering and Pals. Dont dumb down Linux for the masses. There are plenty of shiny plastic offerings out there. There aren't a lot of highly mature offerings for enthusiasts.
The majority of soaring house costs are from opportunity and scarcity costs.
EG, "Good jobs in the city--- City is mostly big business buildings, no housing. What little housing there is, is very high demand-- so HUGE FUCKING PRICES" Couple that with absurd city zoning, and you have spiraling costs.
As jobs get more and more concentrated into tinier and tinier geographic areas, this trend will continue. Throw in high wage disparities, and you end up with very strange things happening indeed.
It has little to do with the actual costs of house manufacture in terms of man hours, or lumber costs. It has much more to do with the fact that the likes of Elon Musk (and pals) is offering 100mil for a chunk of property, because he really, really wants it, driving up property values, and increasing scarcity because once he buys it for his new expanded campus, he never lets it go, and that property is essentially off the market forever.
What you have is housing insufficiency. (or more specifically, insufficient housing that is within an acceptable distance of suitable employment.)
You need one of two things to happen: Big employers need to decentralize their location (Yeah, Apple and Google? Having ONE FUCKING HUGE CAMPUS is bad for people in Cupertino and the rest of the valley, ok? Break that shit up-- That's what VPNs are for, M'kay? The same goes for other big name employers driving up housing through scarcity.) so that the impact of their demands on the local real-estate market is not magnified.
City planners need to collectively grow some balls, and mandate that there be sufficient reserve housing available to accommodate the demolition of previously built high density structures while new ones are constructed, and to subsidize this so that costs are kept affordable. (which means telling Apple and Google to go fuck themselves when they want to expand their campus *AGAIN* by gobbling up that nearby housing block.)
That might sound very California centric, but like many things, California does things (and has consequences) before anyone else does, and unless you *LIKE* the housing situation there, now is the time to nix that shit.
1) This approach just puts all the eggs into one easily taxed basket. 2) Does NOTHING to combat last mile access issues, which is the real thorn in the side here (The same people that control the last mile, and thus prevent competition in their blocks, are the very same people behind wanting to murder net neutrality. They will just deny you access to this service over their network/refuse a peer relationship with them/charge you a shitload of money if they detect packets for this network originating from your home.) 3) As others have pointed out, this does nothing to "take the profit out". What WOULD take the profit out, would be competing infrastructure and cessation of franchise monopolies, which is exactly why Pai is working so hard to stop states and municipalities from doing exactly that, by saying they cannot enact their own legislation to impose net neutrality anyway within their borders.
No, you want to stop this shit, you need to get Congress to fucking do something other than fellate the ISPs.
Low memory is also significantly less important on a UEFI system, because it boots straight into protected mode. Eventually, Intel will completely do away with trappings like v86 mode and pals, because they wont really be needed or useful, and will just be gobbling up die space.
What complicates intel's master plan, is that DOS (especially since the freedos project is very mature and has no licensing fees) is a very approachable target for many applications even in the modern era (Many things, from airport metal detectors to vinyl cutters, to industrial robots and pals), and that requires BIOS to operate. That you do not need to lug around a huge OS stack (DOS lives comfortably in less than 1mb of RAM), and dont have to contend with hundreds of multitasking processes (So your single task-oriented solution does not end up competing for resources or hardware events, because it is operating at realtime instead of time slices or having to wait for spin locks to disengage, etc) makes DOS a very approachable platform even today.
Intel just does not like that. It sees UEFI and their management processor security device model being the future in modern computing, and much like AMD, probably will only give up the keys to the management engine's castle after the vandals storm the place. (Meaning CoreBoot and pals will have to find ways to smash down the custom minix's doors and take over by force to overcome the designed security features of the processor, and hand them over to proper user control.) This is because the premise of the technology defacto asserts that the end user is not capable of being trusted with the security of the platform, and that only trusted persons or entities (orgs) can be vested with that responsibility. (This is at odds with GNU's philosophy.) Intel has many deep-pocketed orgs demanding this level of digital lordship, (microsoft *AND* apple being among the big ones), so the money is in giving the big pocketed groups what they want, which is mutually exclusive to projects like coreboot.
A primer cap filled with crushed glass and lead azide has a little "hammer" inside it that compresses the mixture inducing a small detonation. That detonation then detonates the powder in the shell.
In a fire cracker, a burning fuse extends into a sealed container (usually made of paper and clay)- when the inner explosive mix is ignited, it produces lots of gas. This increases the pressure inside the firecracker until the rest of the explosive detonates from the rapid increase in that pressure. (When it does NOT do that, it shoots off like a bottle rocket.)
Using your jargon, the black cat still detonates. It just does so using a sacrificial pressure vessel to achieve the detonation pressure.
Since we are discussing items that do indeed detonate when there is sufficient pressure (both the black cat, and the shaped explosive), we are still comparing apples to apples.
This then goes to explosive force per gram of material. This comes from a number of factors: Rate of combustion (since it needs to combust faster than the pressure wave that initiates the reaction to sustain that compression wave), and amount of expansion of the end products being the main ones. Black powder burns very slowly, but still detonates when combusted in an enclosed chamber. (If it did not, you couldn't make guns with it.) It lacks much power as an explosive agent, because uncombusted material falls behind the detonation front, and does not contribute meaningfully to the pressure wave. The rate of combustion is very important in that respect; Uncombusted explosive that does not burn faster than the speed of sound cannot apply force to the shock front. In a gun, that waste is taken up by the barrel; Combustion continues after the chamber in the barrel, and gases continue to exert pressure on the bullet as it leaves the gun. Most black powder munitions are subsonic, and this is a major factor in why. In a fire cracker, it adds pretty sparks; the primary detonation ends when the paper wrapper ruptures.
The second feature is where there is a big difference between flash powder and C4. Flash powder detonates, and burns faster than the speed of sound, so the expansion of combustion adds to the shockwave. It also does not need an enclosing container to supply the pressure to initiate detonation in most cases. (a blasting cap inserted into the explosive without a housing is usually sufficient.) Both flash powder and C4 technically meet that requirement (but the loose powder nature of blasting powder typically requires a container of some sort, just not involved in the pressure to detonate. An open topped jar is fine.), but C4 produces much more volume of end product gases than does the flash powder, meaning that the force it puts on the pressure wave as it combusts is much greater.
A typical consumer fire cracker's design is the same as one made from black powder, even if it contains flash powder instead. A fuse leads down into a paper barrel that has clay plugs at the ends. A burning ember begins combustion of the explosive, which rapidly produces gases faster than they can be expelled through the hole in the clay plug, a pressure wave forms, and the rest of the explosive detonates. The vast majority of the mass of the device is the paper and clay.
C4 is an entirely different beast. You can light it on fire, and it will likely just go out. Instead, it uses a detonation cap of some kind to supply the needed pressure wave to initiate the combustion; again, it combusts faster than the speed of sound, so that combustion effectively adds to the pressure wave. It produces a large volume of gas product from that combustion, adding high energy to the shock front.
This comes back to my original post, which was a refutation of your "basically a firecracker" argument.
C4 explosive-- 1) Does not require a heavy pressure vessel to activate (a small percussion hammer cap will do.) 2) Burns significantly faster than black powder (and still faster than
Flash powder is not exclusively used in typical bangpops. Black powder is indeed often used.
However, for the sake of argument, let's use flash powder. It indeed does burn significantly faster-- 25000 feet per second. (Not too shabby, and in the same general ballpark as C4. However, a significant fraction of a fire cracker is paper and clay containment. You dont need that with a plastic explosive, meaning you get more actual explosive for the payload. There is maybe a gram of flash powder inside a black cat, if that.)
However, there is a significant caveat. Again, this kind of explosive detonates uniformly. The explosion goes in all directions.
A shaped explosion focuses the majority of the concussive force of the explosion on a very small area. Still apples and oranges.
As for your nonsense rambling there at the end, I am sorry, I cant fix your stupidness. Comparing burn rates, and giving a "foo times more explosive force" is very much comparing apples to apples. Specifically, it is comparing burn rates to burn rates. Want to know the magic you are missing there smart ass? An object "Detonates" when the burn rate exceeds the speed of sound (which causes a pressure wave to form.) Black powder burns at subsonic speeds, and thus does not detonate. (It relies on an external compression sleeve to focus the gases created by combustion to focus that energy. In a gun, this is the combination of the cartridge shell and the barrel. In a fire cracker, it is the paper body and clay plugs at the ends. This is important, because of the aside I mentioned prior-- most of the mass of the black cat is paper and clay. With a high explosive, you dont need that. With a shaped high explosive, that force can be focused on a very tiny (say, 2cm dia) area. Two very important bits you of course, speciously omitted in any of your posts, while insisting on being a douche bag and splitting hairs.)
If after correcting your absurdity as I did above, you are still too stupid to realize that you are fucking wrong, I am sorry. Go fuck yourself, I'm done with you,
There is a big difference between a black cat, and a shaped high explosive charge.
Take for instance, semtex. This is a commercially available plastic explosive used for demolition. (and frequently used by terrorists.)
250 grams of it is enough to destroy an in-flight airplane, if properly placed.
The premise of the video is that a human skull is pretty thin, and not evolved to stop a shaped explosive's concussion wave. If one uses something like this, they can blow half your skull off with just a few grams of material, pretty much exactly like in the video.
A shaped charge explosive works by having a special void in the explosive material on the surface that is to be favored for blast-wave creation. This provides a high velocity path of least resistance, through which combustion products of the explosion will favor being expelled, and giving the explosion a preferred direction for energy delivery. (This is very different from a fire cracker, which explodes basically uniformly.)
Considering that just about any high explosive is many times more powerful per gram than the black powder found inside the black cat mentioned by the grandparent, and are capable of producing shaped shock fronts on detonation, I basically call bullshit on grandparent's dismissal. For reference, military grade C4 plastic explosive detonates with a combustion rate 29,000 feet per second. Black powder? Between 600 and 1400 feet per second. Literally, just replacing that "black cat" with the same weight of C4, increases the explosive force 20 times, at best, and 48 times at worst.
Apples and oranges sir. Your black cat is not even in the same class as the material they are suggesting could be inside these drones.
This is very easy to do if you have some savvy. The software has been around for a very long time. The hardware is cheap.
Hardware: TPLink OpenWRT compatible wifi router with USB port. USB Hub USB hard disk USB sound hardware. [already existing home stereo(s)] Stereo patch cables
Software: OpenWRT with suitable sound modules, the MPD (Media Player Daemon), netjukebox, and apache web server (with php) packages.
Really, ANY minimalist linux box with USB ports and a network stack would work, but these wifi routers are cheap. Super cheap.
You only need one of these (if you install more than one) to have storage attached, if you enable NFS shares, and configure the others to connect. If you have a NAS, you can just use that most likely.
I personally have no love for chromeOS, and was happy to remove it as soon as possible.
Arch is very much for power users, but if you just want a GNU/Linux deploy free of most hassle, Gallium already detects and configures itself properly for most chromebook hardware our there. Just be aware that it is an ubuntu variant. (actually, an xubuntu variant.)
Legacy boot mode gives a very minimal BIOS implementation, and does not do much to simulate ps2 keyboards or mice, and most chromebooks use an SPI or I2C connected keyboard/mouse. Most normal distros are not geared to detect and use such hardware, so manually configuring them is a PITA.
Slashdot Mirror
Approach is wrong.
People want to make a fully identically functioning system that they can then replicate activity patterns on.
Instead, make use of the way the brain (and subsequently, the consciousness operating inside it) incorporates and induces activity on implanted devices.
Basically, stop with all this scanning shit. Instead, focus on a single, uniform platform that is well known and easy to simulate, then implant the patient with some implants that link the simulation with the still living organic brain. The organic brain will incorporate the functionality of the simulation. In short, go the ship of theseus route. As the brain begins to fail from either injury or old age, it will rely more and more on the simulation hardware, until eventually, it gives up. If you made the right connections, you will have valid activity in the simulation after the death of the organic component. Congrats, you have an upload.
This requires a very robust simulation platform though, which we do not currently have. DARPA is doing some interesting research on simulating neural columns, and the last I heard anything concrete was years ago, so if the project is still active, I can only assume they have gotten much better at their simulation. I understand that Chinese researchers are also working on simulating neuronal columns--
https://www.youtube.com/watch?... (Still, as interesting as these appear, they are noplace near mature enough to have attached to living humans as ancillary networks to support and replace function.)
Again, the issue of "Every organic brain is so vastly different, there is no way we can scan the physical networks for replication!" becomes less important, when you instead say "Let's build a generic, uniform simulation platform that we can then attach, and exploit neuroplasticity of the organic side for deep integration."
Zuck, like many corporate suits, cannot wrap his head around the idea that there are alternatives to the profit motive.
As such, his response is entirely focused there. It basically boils down to:
"There are only two ways that a service that facilitates massive communication world wide can exist: You either pay a subscription fee, or you pay with your eyeball time to advertisers. People have more free time than free cash, so our model makes sense. I am defending this model, because it is the most sensible of the two, and disparage anyone who says that this turns the users of the service into a product we sell to advertisers, especially when the ones making the claim use abusive pricing to enrich themselves at their customer's expense. You should not decide that you love the people hurting you by agreeing with the notion that providing the more scarce asset over the more abundant asset is somehow better for you."
EG, he cannot wrap his head around the idea that there are government supplied utilities, and bulk data delivery can be one of them-- Or that there might be valuable services offered totally for free on ideological grounds. (See for instance, most FOSS development.)
He cannot imagine a world where somebody with the initial outlay for expenses (or a group of people working collectively) could produce something like facebook. Apparently he is not very attentive, or has tunnel vision.
In his world, the people working at the volunteer soup kitchen are there for ulterior motives. ALWAYS.
These devices are essentially a chromebook, just without the custom chromeOS boot loader.
Under the hood they are basically the same kind of beast: eMMC based storage, 4gb of non-upgradable RAM, Celeron or Atom processor, SPI/i2C based bus with keyboard and mouse attached.
If you can get the chromebook cheaper, you are basically getting the same thing, just with a little extra legwork needed. Some chromebooks have NGFF based storage, which you can replace with a fantastically larger storage device. (Most are just eMMC shit though.)
These devices really are made for windows 10, and wont run any microsoft product that is older. (this is due to the use of the SPI/i2C bus architectures for input devices, sound hardware, etc) They basically *NEED* the advanced NTFS compression options offered to be remotely usable, since they are basically crippled by design in terms of storage.
If you use Linux, you will need to use a kernel that supports btrfs, and use the compression. It is also sensible to use an EXT4 volume on the microSD slot (if you have one) that is properly partitioned and formatted for the erase block sizes of that card. (Requires research on the user's part to do correctly. Incorrect formatting and partitioning will kill the card in about 6 months.)
In either case, dont expect these devices to be amazingly wonderful speed demons. Where they excel is being able to run continuously for 8+ hours straight under active use, and being ultraportable.
1) You can connect to an SMB share while not in dev mode. You can push your data off at this time.
2) Activate Dev mode. (this DOES wipe the system and start it fresh).
3) In Dev Mode, install MrChromebox's UEFI bios. This completely replaces the bootloader.
4) Install GNU/Linux
YES, this voids the warranty. I remind you that it is absurd to complain about this, as Google is only going to support chromeos anywyay.
https://www.youtube.com/watch?...
Darth Pai does not like it when big telecom's interests are threatened. Your tiny satellites, designed to burn up in the atmosphere, pose a significant risk of colliding with established interests. As such, we refused your launch request, then you went over his helmet, and how he he will crush your balls, pitiful startup weaklings.
I have virtualbox installed on my hacked celes (Samsung chromebook 3). It is nothing to write home to mommy about, but it can run another OS fairly well, all things considered.
To be fair though, the Celes has a celeron CPU, instead of the more "atom like" cpus in most other intel chromebooks.
If you want one that is better suited to virtualization/daily driving, you want this guy (especially if you want to upgrade the internal storage to something more reliable than eMMC/microSD)
https://www.amazon.com/dp/B015...
Features an i3 processor, with NGFF SSD socket.
Just link straight to MrChromebox.tech, since he has the needful to do the firmware portion in the bag right there for nearly all chromebooks in the wild.
https://mrchromebox.tech/#devi...
https://mrchromebox.tech/#fwsc...
All you need to do is remove the write protect screw/jumper, turn on dev mode, then run his script. Booya, bob's your uncle. His firmware has added bonus features, in that it reprograms the chromebook's embedded microcontroller so that the keyboard emulates a PS2 interface, allowing more OSes to run without as much hassle, as well as poking the sound hardware to better approximate an HD Audio Bus device.
Personally, I run GalliumOS as the primary on my Samsung chromebook 3 (celes). I have one of the older ones that only has 2gb of RAM, and I desperately need zram for it to be useful. With how weaksauce it is anyway, nearly any game that 'could' run on it, will also work in WINE, so I dont really need windows.
Nevermind-- I misread. Tiny assed phone displays, messin' with my myopia.
I see it is one and half hours. Meh.
Seriously, I know editing is hard and all, but can you please tell us-- Is this 4 and a half days, 4 and a half weeks, 4 and a half months, or 4 and a half years?
Such things include "Computational Ram"
https://en.wikipedia.org/wiki/...
There is also a very old idea of using memory elements directly to compute results, which is true memputing. (There are few examples of this, because it is costly as an architecture-- but your brain is a pretty good biological example. The same components are used for data storage, as well as data processing.)
Given that such "Computational Ram" devices already exist in the wild, I fail to see why more novel hardware is needed, excepting as a refinement of concept?
I a reminded of Torvald's scathing emails about Intel, their proposed patch sets, and how they pointed toward intel wanting to make future chips "Fast but insecure" by default, and requiring the BIOS or OS to tell the CPU "No bitch, secure mode only please", just so they could continue to claim benchmark scores (naturally, with the anti-spectre and meltdown patches disabled so the chip runs really fast.)
Hopefully these silicon level fixes are *ACTUAL* fixes to the methodology used by the speculative execution implementation of the chip, so that speculative execution still is active, but the chip no longer leaves bits and pieces in the processor cache that can be exploited, and that it does this by default.
Hopefully.
You TOTALLY CAN do that on the MyCloud.
The boot loader looks for an unsigned kernel and initrd on a specific partition, formatted as FAT32, with a specific file name.
You can bake your own and put it on the drive, and the mycloud will boot that image and initrd without complaints.
In the community pages, we have been working on a straight up clean debian for quite some time. There are instructions on how to configure and compile your own kernel from the stock device tree.
Considering that I own one of these devices, AND I participate frequently on the WD community group for this device, I can confirm that the base package is identical between consumer and midsize business class offerings. I can *ALSO* confirm that we have proposed workable patches on many numerous occasions, in every "Leave feedback" location WD makes available. (Protip, THEY IGNORE US.)
At least on the older Gen1 consumer mycloud units, the web interface and the /etc/passwd file were hosted by a real, writable filesystem that could handle persistent changes, meaning that the information we share in the community pages could be used to correct the security vulnerabilities if you wanted to take the matter into your own hands. Many people did this.
On the Gen2 however, WD decided that the user being able to modify the root file system persistently was just not something they felt comfortable with. It is a ram-backed root file system from an initial ramdisk, into which a cramfs container gets automatically mounted at a defined mount point. The cramfs container contains the web UI, and all the major system binaries. /etc/passwd and pals are all obliterated on every boot, because they live in the initial ramdisk image.
That said, the hardware itself is *NOT* that bad. Just the horror-show WD offers software wise. (STAY THE FUCK AWAY FROM WD-SYNC.) There is a russian frequent contributor to the community site who has created instructions for a "From scratch" pure debian deployment on the Mycloud (both gen1 and gen2) units. This completely replaces the WD software with fully FOSS software, and gives the user full control over the unit.
Many people in the community board are Americans, like myself. We have proposed many solutions and fixes that we have locally implemented and given local testing to. We are not afraid to propose solutions, or to do things ourselves. The problem is that big companies do not feel comfortable testing things (Its a cost center! How horrible!!), and would rather just take a generic canned product, slap their name on it, and run. That is hardly an American thing.
The issue with H1B labor is that H1B software people often tend to do the same thing. They will take the homebrewed things people like myself make, (which have **NOT*** been sufficiently tested to mass deploy!!!!!!), slap their names on them and run with them, and their bosses, doing the same exact thing, are just peachy keen with this.
As for your assertion that we should make our own startups--- Do you have any idea how much the industry is stacked against new blood entering the market? Are you fucking delusional? Do you think that there are no people in the US that are just straight up makers, builders, and engineers who do shit for fun, who would love to make neat products available to people? The major reasons why we dont have many people doing that are two-fold; 1) Our erstwhile government frowns on individual thought and self-empowerment of the citizenry, and actively promotes a narrative that if you see your neighbor making something suspicious, he is probably an islamic terrorist making some kind of improvised explosive to blow your kids up on the bus with, or some crazy shit. The mainstream press eats that shit up like candy because it is over the top, and our culture is conditioned to soak it up like a sponge. There is a damned near moral panic against people doing neat stuff in garages these days. 2) Big corporations dont like new products entering the market, so they lobby to require "You must be this big to play" hurdles thrown in. Now, not only do you need to have a good idea and a working prototype, you need to have your entire product vetted for intellectual property form other vendors, even if you have no idea those vendors even exist, (and in many cases, are simply patent trolls!), which means hiring a small team of lawyers--- You also need investment capital to meet stringent manufacturing and materials use regulations, pay for init
Well, with the internet shut off, they had to send the angry missive SOMEHOW! ;P
The thing is, much like the clothing patterns I mentioned, the ability to make your own (the core tenet of "freedom" espoused by RMS as needed for GNU is the freedom to modify, not to be free from pricetags.) and to distribute those changes is a thing that only appeals to enthusiasts. In the case of clothing, that is people who enjoy playing with textiles, or doing neat things with garments.
In the case of software, that is your typical computer enthusiast. Exactly the people that the masses ARE NOT.
This does not prevent the masses from using product that is freely redistributable and alterable, (and just not exercising their freedom to do said actions themselves, instead relying on those that have such an impulse), but the realities of being practical and efficient with resources and resource availability means that for the same reasons mass manufactured clothing is popular, proprietary large-development-cycle software (with its non-free paradigms) will dominate, because the freedom espoused has no value to them. (Again, they never exercise that freedom, even when it is right there, because it does not interest them.)
The people that really care about free software do so because they are enthusiasts. The average person does not care about that, and just wants something that works with minimal costs (either financial, or in terms of invested time or mental effort.)
I think you just chose to see what you wanted to see in my statement, rather than what I actually said. Linux (and GNU software as a whole) appeal to enthusiasts, because the freedoms that are cornerstones of their development process are of value to enthusiasts. Non-enthusiasts are not interested. Software made by enthusiasts, for enthusiasts, will be poor fits for non-enthusiasts, especially when more simplified offerings exist.
The problem, is that Linux appeals to people who are computer enthusiasts-- people who LOVE computers, because they are simply amazing things, and they want to get the most out of that purchase.
Most people are not like that. They want a computer to do a very short list of things, and want one that will never slow down, break, or get infected with something. For most people, that thing is "I need the internet, facebook, and stuff for work/school." The less they have to actually know about computers, or how computers work (EG, the more "Magic box" like they are) the happier these people are.
Linux dares to expose its internals, and worse yet, DEMANDS that you learn about how it works underneath in order to use it effectively. That is why it has never, and likely will never, take off as a mainstream desktop.
Apple and Microsoft have created the "Shiny plastic experience", and people love it. Linux might as well say "Batteries not included, setup time 6 hours, major assembly required" on the box.
Asking why Linux is not a mainstream desktop environment is like asking why McCalls clothing patterns are not the dominant source of apparel in the market. Sure, you can customize the clothing however you want, and you can modify the patterns to your hearts content--- But dammit, you gotta get the cloth, cut it, sew it together, and all that shit. Why bother when you just want a fashionable new sport top, eh? People would rather spend the money on something somebody else already put together-- VIOLA-- OSX and pals. Shiny plastic. No work.
Linux needs to stop chasing this fantasy where everyone stops being lazy gits and becomes excited computer enthusiasts. They need to understand that they are a niche market, and do that niche very well. Last I checked, that was the Unix philosophy anyway.
For this reason I am opposed to the efforts of Poettering and Pals. Dont dumb down Linux for the masses. There are plenty of shiny plastic offerings out there. There aren't a lot of highly mature offerings for enthusiasts.
The majority of soaring house costs are from opportunity and scarcity costs.
EG, "Good jobs in the city--- City is mostly big business buildings, no housing. What little housing there is, is very high demand-- so HUGE FUCKING PRICES" Couple that with absurd city zoning, and you have spiraling costs.
As jobs get more and more concentrated into tinier and tinier geographic areas, this trend will continue. Throw in high wage disparities, and you end up with very strange things happening indeed.
It has little to do with the actual costs of house manufacture in terms of man hours, or lumber costs. It has much more to do with the fact that the likes of Elon Musk (and pals) is offering 100mil for a chunk of property, because he really, really wants it, driving up property values, and increasing scarcity because once he buys it for his new expanded campus, he never lets it go, and that property is essentially off the market forever.
What you have is housing insufficiency. (or more specifically, insufficient housing that is within an acceptable distance of suitable employment.)
You need one of two things to happen:
Big employers need to decentralize their location (Yeah, Apple and Google? Having ONE FUCKING HUGE CAMPUS is bad for people in Cupertino and the rest of the valley, ok? Break that shit up-- That's what VPNs are for, M'kay? The same goes for other big name employers driving up housing through scarcity.) so that the impact of their demands on the local real-estate market is not magnified.
City planners need to collectively grow some balls, and mandate that there be sufficient reserve housing available to accommodate the demolition of previously built high density structures while new ones are constructed, and to subsidize this so that costs are kept affordable. (which means telling Apple and Google to go fuck themselves when they want to expand their campus *AGAIN* by gobbling up that nearby housing block.)
That might sound very California centric, but like many things, California does things (and has consequences) before anyone else does, and unless you *LIKE* the housing situation there, now is the time to nix that shit.
No, really, Just No.
1) This approach just puts all the eggs into one easily taxed basket.
2) Does NOTHING to combat last mile access issues, which is the real thorn in the side here (The same people that control the last mile, and thus prevent competition in their blocks, are the very same people behind wanting to murder net neutrality. They will just deny you access to this service over their network/refuse a peer relationship with them/charge you a shitload of money if they detect packets for this network originating from your home.)
3) As others have pointed out, this does nothing to "take the profit out". What WOULD take the profit out, would be competing infrastructure and cessation of franchise monopolies, which is exactly why Pai is working so hard to stop states and municipalities from doing exactly that, by saying they cannot enact their own legislation to impose net neutrality anyway within their borders.
No, you want to stop this shit, you need to get Congress to fucking do something other than fellate the ISPs.
Especially since you can construct something much smaller, with COTS parts, for less than half that.
(psst... hacked zsun + USB battery pack. Other than N, and maybe really loud antennas, it can do anything this thing does. Total price: retail ~35$)
Low memory is also significantly less important on a UEFI system, because it boots straight into protected mode. Eventually, Intel will completely do away with trappings like v86 mode and pals, because they wont really be needed or useful, and will just be gobbling up die space.
What complicates intel's master plan, is that DOS (especially since the freedos project is very mature and has no licensing fees) is a very approachable target for many applications even in the modern era (Many things, from airport metal detectors to vinyl cutters, to industrial robots and pals), and that requires BIOS to operate. That you do not need to lug around a huge OS stack (DOS lives comfortably in less than 1mb of RAM), and dont have to contend with hundreds of multitasking processes (So your single task-oriented solution does not end up competing for resources or hardware events, because it is operating at realtime instead of time slices or having to wait for spin locks to disengage, etc) makes DOS a very approachable platform even today.
Intel just does not like that. It sees UEFI and their management processor security device model being the future in modern computing, and much like AMD, probably will only give up the keys to the management engine's castle after the vandals storm the place. (Meaning CoreBoot and pals will have to find ways to smash down the custom minix's doors and take over by force to overcome the designed security features of the processor, and hand them over to proper user control.) This is because the premise of the technology defacto asserts that the end user is not capable of being trusted with the security of the platform, and that only trusted persons or entities (orgs) can be vested with that responsibility. (This is at odds with GNU's philosophy.) Intel has many deep-pocketed orgs demanding this level of digital lordship, (microsoft *AND* apple being among the big ones), so the money is in giving the big pocketed groups what they want, which is mutually exclusive to projects like coreboot.
You are aware of how a bullet gets fired, right?
A primer cap filled with crushed glass and lead azide has a little "hammer" inside it that compresses the mixture inducing a small detonation. That detonation then detonates the powder in the shell.
In a fire cracker, a burning fuse extends into a sealed container (usually made of paper and clay)- when the inner explosive mix is ignited, it produces lots of gas. This increases the pressure inside the firecracker until the rest of the explosive detonates from the rapid increase in that pressure. (When it does NOT do that, it shoots off like a bottle rocket.)
Using your jargon, the black cat still detonates. It just does so using a sacrificial pressure vessel to achieve the detonation pressure.
Since we are discussing items that do indeed detonate when there is sufficient pressure (both the black cat, and the shaped explosive), we are still comparing apples to apples.
This then goes to explosive force per gram of material. This comes from a number of factors: Rate of combustion (since it needs to combust faster than the pressure wave that initiates the reaction to sustain that compression wave), and amount of expansion of the end products being the main ones. Black powder burns very slowly, but still detonates when combusted in an enclosed chamber. (If it did not, you couldn't make guns with it.) It lacks much power as an explosive agent, because uncombusted material falls behind the detonation front, and does not contribute meaningfully to the pressure wave. The rate of combustion is very important in that respect; Uncombusted explosive that does not burn faster than the speed of sound cannot apply force to the shock front. In a gun, that waste is taken up by the barrel; Combustion continues after the chamber in the barrel, and gases continue to exert pressure on the bullet as it leaves the gun. Most black powder munitions are subsonic, and this is a major factor in why. In a fire cracker, it adds pretty sparks; the primary detonation ends when the paper wrapper ruptures.
The second feature is where there is a big difference between flash powder and C4. Flash powder detonates, and burns faster than the speed of sound, so the expansion of combustion adds to the shockwave. It also does not need an enclosing container to supply the pressure to initiate detonation in most cases. (a blasting cap inserted into the explosive without a housing is usually sufficient.) Both flash powder and C4 technically meet that requirement (but the loose powder nature of blasting powder typically requires a container of some sort, just not involved in the pressure to detonate. An open topped jar is fine.), but C4 produces much more volume of end product gases than does the flash powder, meaning that the force it puts on the pressure wave as it combusts is much greater.
A typical consumer fire cracker's design is the same as one made from black powder, even if it contains flash powder instead. A fuse leads down into a paper barrel that has clay plugs at the ends. A burning ember begins combustion of the explosive, which rapidly produces gases faster than they can be expelled through the hole in the clay plug, a pressure wave forms, and the rest of the explosive detonates. The vast majority of the mass of the device is the paper and clay.
C4 is an entirely different beast. You can light it on fire, and it will likely just go out. Instead, it uses a detonation cap of some kind to supply the needed pressure wave to initiate the combustion; again, it combusts faster than the speed of sound, so that combustion effectively adds to the pressure wave. It produces a large volume of gas product from that combustion, adding high energy to the shock front.
This comes back to my original post, which was a refutation of your "basically a firecracker" argument.
C4 explosive--
1) Does not require a heavy pressure vessel to activate (a small percussion hammer cap will do.)
2) Burns significantly faster than black powder (and still faster than
Flash powder is not exclusively used in typical bangpops. Black powder is indeed often used.
However, for the sake of argument, let's use flash powder. It indeed does burn significantly faster-- 25000 feet per second. (Not too shabby, and in the same general ballpark as C4. However, a significant fraction of a fire cracker is paper and clay containment. You dont need that with a plastic explosive, meaning you get more actual explosive for the payload. There is maybe a gram of flash powder inside a black cat, if that.)
However, there is a significant caveat. Again, this kind of explosive detonates uniformly. The explosion goes in all directions.
A shaped explosion focuses the majority of the concussive force of the explosion on a very small area. Still apples and oranges.
As for your nonsense rambling there at the end, I am sorry, I cant fix your stupidness. Comparing burn rates, and giving a "foo times more explosive force" is very much comparing apples to apples. Specifically, it is comparing burn rates to burn rates. Want to know the magic you are missing there smart ass? An object "Detonates" when the burn rate exceeds the speed of sound (which causes a pressure wave to form.) Black powder burns at subsonic speeds, and thus does not detonate. (It relies on an external compression sleeve to focus the gases created by combustion to focus that energy. In a gun, this is the combination of the cartridge shell and the barrel. In a fire cracker, it is the paper body and clay plugs at the ends. This is important, because of the aside I mentioned prior-- most of the mass of the black cat is paper and clay. With a high explosive, you dont need that. With a shaped high explosive, that force can be focused on a very tiny (say, 2cm dia) area. Two very important bits you of course, speciously omitted in any of your posts, while insisting on being a douche bag and splitting hairs.)
If after correcting your absurdity as I did above, you are still too stupid to realize that you are fucking wrong, I am sorry. Go fuck yourself, I'm done with you,
There is a big difference between a black cat, and a shaped high explosive charge.
Take for instance, semtex. This is a commercially available plastic explosive used for demolition. (and frequently used by terrorists.)
250 grams of it is enough to destroy an in-flight airplane, if properly placed.
The premise of the video is that a human skull is pretty thin, and not evolved to stop a shaped explosive's concussion wave. If one uses something like this, they can blow half your skull off with just a few grams of material, pretty much exactly like in the video.
A shaped charge explosive works by having a special void in the explosive material on the surface that is to be favored for blast-wave creation. This provides a high velocity path of least resistance, through which combustion products of the explosion will favor being expelled, and giving the explosion a preferred direction for energy delivery. (This is very different from a fire cracker, which explodes basically uniformly.)
Considering that just about any high explosive is many times more powerful per gram than the black powder found inside the black cat mentioned by the grandparent, and are capable of producing shaped shock fronts on detonation, I basically call bullshit on grandparent's dismissal. For reference, military grade C4 plastic explosive detonates with a combustion rate 29,000 feet per second. Black powder? Between 600 and 1400 feet per second. Literally, just replacing that "black cat" with the same weight of C4, increases the explosive force 20 times, at best, and 48 times at worst.
Apples and oranges sir. Your black cat is not even in the same class as the material they are suggesting could be inside these drones.
This is very easy to do if you have some savvy. The software has been around for a very long time. The hardware is cheap.
Hardware:
TPLink OpenWRT compatible wifi router with USB port.
USB Hub
USB hard disk
USB sound hardware.
[already existing home stereo(s)]
Stereo patch cables
Software:
OpenWRT with suitable sound modules, the MPD (Media Player Daemon), netjukebox, and apache web server (with php) packages.
Really, ANY minimalist linux box with USB ports and a network stack would work, but these wifi routers are cheap. Super cheap.
You only need one of these (if you install more than one) to have storage attached, if you enable NFS shares, and configure the others to connect. If you have a NAS, you can just use that most likely.
Essentially, yes, I do believe so.
I personally have no love for chromeOS, and was happy to remove it as soon as possible.
Arch is very much for power users, but if you just want a GNU/Linux deploy free of most hassle, Gallium already detects and configures itself properly for most chromebook hardware our there. Just be aware that it is an ubuntu variant. (actually, an xubuntu variant.)
Legacy boot mode gives a very minimal BIOS implementation, and does not do much to simulate ps2 keyboards or mice, and most chromebooks use an SPI or I2C connected keyboard/mouse. Most normal distros are not geared to detect and use such hardware, so manually configuring them is a PITA.