Become a fan of Slashdot on Facebook


Forgot your password?
Compare cell phone plans using Wirefly's innovative plan comparison tool ×

Comment Re:massive parallel processing=limited application (Score 2) 111

It's an interesting idea, and one I have given a little thought to. ( it would enable a very fault tolerant computer architecture) however, unless you implement highly redundant interconnects/busses, you still have the N-devices fighting for a shared resource problem.

If you make the assertion that all nodes have a private direct connection with all other nodes, and thus eliminate the bottleneck that way, you now have to gracefully decide how to handle a downed private link.

I suppose a hybrid might work. Fully dedicated links, and one shared bus. When dedicated link fails, communicate over the shared bus.

Scaling such a design would become prohibitively costly though. A 200 node design would have orders of magnitude more dedicated links.

The idea I had for playing with this idea, was to use some cheap wired home routers, set up private vlans on the 5 or so ethernet ports each has, then put private patch cables on each port, then put all the Wan ports on a dumb hub.

The local copies of Linux on each system can handle management of local device resources, and a daemon running on each node then handles listening/responding on each interface.

Just what such a thing would be good at doing escapes me though. To be really useful, you would need some way to have nodes specialize, then cooperate, without a central authority.

That way, should we decide to use this network to process live video, one node decodes the input stream, then dispatches portions of the decoded stream to peer devices, who then take the decoded stream and do whatever processing is requested, before sending the processed streams to yet another peer device which assembles the processed stream, then shuttles that to the endpoint node, which reencodes the stream and writes it to the output device. (Or some similarly cellular process)

I suppose this is kinda similar to how a neural colum works, where locally interconnected nets are restricted in the number of true local peers they have, and then communicate collectively to other neUral columns by dedicated interconnects. (Video input source in the above, could be from a camera, but it could also be from another network's output stream.)

The major logical tasks are:
Role selection in the assigned task for each local node.
How to issue instructions to the mesh nodes in a decentralized manner

Depending on how far you wanted to extrapolate this, each mesh node could be treated as a logical unit, where each logical node then is part of another, higher level node of similar topology: each mesh has a direct connection to each other mesh inside its higher order node, and one communal link all nodes can talk on inside that node.

Eg, if I make 5, 5node networks made out of such routers, I need 7 ports on each router. 5 for direct local traffic. 1 for local shared connect, 1 for direct connect to another 5node group. Clever use of subnetting and routing on the shared net would enable there to be a dumb gateway device to allow the shared higher link to function. Each 5node network is connected to every other 5-node network in the scaled up version.

Decisions on how to process incoming data might be tied to which interface received it, or any number of other methods.

Spying on the system state of the whole system should be possible through the shared link infrastructure, though ideally any node you interact with the system with should be a proper peer in it, and nit something sitting on the shared net only.

The drawback of such a design will be signal propogation latency, and keepin all the subnodes, at all levels, synchronized. The human brain uses a support network of astrocytes and glial cells to guide dedicated link physical routing, and to tune propogation delay between neural columns through selective mylienation of trunk bundles.

You could probably fake it with introduced waitstates.

At some point though, the behavior of the whole will revolve around the basic logic baked inside each physical compute unit. Ideally, that is universal, and consistent all over the system. The magic happens based not only on the message sent, but which interface recieved it, etc. The os kernel would have to be baked in at that level. All other results would be totally emergent.

The parallels with neural networks shouldn't be overlooked imo. There's a huge body of work to draw from on that subject.

Comment massive parallel processing=limited applications (Score 2) 111

while being able to leverage that many compute units all a once is quite impressive, most tasks are still serial by nature. computers are not clairvoyant, so cannor know in advance what a branched logic chain will tell them to do for any arbitrary path depth, nor can they perform a computation on data that doesnt exist yet.

thhe benefits of more cores are from parallel execution, not from doing tasks faster. as such, most software is not going to benefit from having access to 8000 more threads.

Comment Re:It's all true (Score 1) 75

Nonviolent revolution (a la, civil disobedience in India) has at least one reported instance of success.

However, the US electorate lacks a suitable cultural stereotype upon which to sustain nonviolence, in the face of government totalitarianism. Eg, when people start getting arrested and destroyed financially, people will either suddenly forget their morals, or will turn violent.

You asked for an alternative, the problem is that it does not look plausible for US culture.

Comment Re: Criminal (Score 4, Insightful) 301

How delightfully backwards!

The first step in addressing a problem, is acceptance that there is in fact a problem to address.

Rightly pointing out that all current choices are shit, is therefore the correct and proper direction toward addressing that problem, eg, by not voting for more political choices that are shit.

Hilariously, your rebuttle is whimsically absurd! Denial of the problem, will somehow result in correction of the problem-- and drawing attention to the fact that there is in fact a problem purpetuates the problem.

What really needs to happen is for the US to grow a pair and demand no confidence vote powers. That way when presented with the choice of 70 year old rancid tuna and 50+ year old trolls with verbal diarrhea, we can return the offered plate to the kitchen and demand they do it right this time.

Arguing bitterly over the pros and cons of rancid tuna over shitspeaking oompahloompahs does nothing to chastise the kitchen. Next time they wI'll serve you a festering blue waffle with whipped cream and shriveled oil baron salomi with a side of polluted tap water.

Unless you want to keep getting served shit, I suggest you address the problem of being served shit.

Comment Re:Pile it on.. (Score 3, Informative) 301

Saggy pants is still not racist, per se.

It is related to prison culture, and ghetto culture. The association with a specific racial group is a consequence of over expression of incidence in criminality and poverty in that racial group. Jumping straight to the racecard is itself racist; it defacto implies that the saggy pants, and the culture behind it, are intrinsically linked to that racial group. It is equally as onerous as saying "all black people are criminally minded."

Instead, saggy pants is just saggy pants. Complaints about the glorification of prominent displays of saggy pants is not racist, per the above reasoning. Counter-claims that such complaints are racism are themselves what is racist.

You were born with a powerful organ, refined through millions of years of evolution, for the primary functions of advanced abstract and critical thinking. Please use it responsibly.

Comment Re:Mobile Web (Score 1) 79


But then our phones would probably explode under the pressure of fullscreen autostarating flash ads.

That and embedded java applets, fully interactive canvas elements, and millions of cross site activity trace pings.

Throw in some android specific attacks, and some other kinds of fun we don't know about yet, and it could be loads more entertaining to browse on a mobile device.

(I can hardly wait for the aggressive AR adverts of the near future.)

Comment Re:Easy. (Score 3, Interesting) 110

From ths perspective of people who like to watch, as the nsa does, telling the landlord about where all the best places to peep so they can be fixed, is counter intuitive.

That the same building techniques are used in thier own house, and that other people who like to watch can peep on them through them, is not seriously considered.

Instead, only the loss of really good ways to peep is what is considered. If the method of peeping is likely to be discovered, or the architecture behind the means of peeping changes such that the approach becomes less valuable, the peeping Tom may delude himself into thinking that he is doing a service to society and the landlord by pointing out how that peeping may happen. (See for instance, methods used to remotely observe what is displayed on a crt monitor by monitoring the em spectrum for telltale radio artifacts-- who uses crts these days?)

The nsa is sick, and likes to watch. The very idea that they would feel they should stopper up the holes they look through, or alert people that they are looking at them through them, is counter to their fetish. The very idea is absurd to them. Only somebody that sees by accident, and is disgusted by having seen, has motive to see to it that no such seeing ever happens again, apart of course, from somebody catching somebody peeping on them, and discovering the hole that way.

From the perspective of the nsa, if they have eno ugh places to peep through, you can putty up holes all day, and they can wack off to watching you do it. Telling you where all the holes are stops that from happening. They want to watch you. Not keep you safe from being peeped on.

Comment Re:Easy. (Score 4, Interesting) 110

I have.

Note in citizen 4, the first phase of the nsa's activity against Snowden was in sussing out his aberrent behavior, and surveiling his girlfriend, family, and Hawaii place of residence. This is what happens in phase 1) of my short list. A list of persons of interest is produced using psych details, and active monitoring starts. Connections maps are created. Points of surveillance are established, and monitoring priority increases. Phase one ideally (for the nsa) ends with apprehension of their leak, but the process does not end there.

After sussing out the entry point of the leaker, the companion network is either dismantled, or subtly repurposed for cointel.
False intel is fed to the group. If the false intel causes the foriegn agency to suspect compromise, it sends the message to that foriegn agency that their action was detected, and that thier methods are not valid any longer. If the foriegn agency fails to change the operational behavior of the cell, then it may become beneficial to plant a double agent. This double agent can then cause the foriegn power to change its policies or public activities, through contaminated or misleading intelligence, created specifically for this purpose.

That they can conduct such a profoundly invasive phase one investigation using literally any internet connected, or broadcast capable device, along with your financial data, and the information about you provided by your so called friends on social media, is the primary thrust behind snowden's leaks. What the NSA will do, and why they will do it is not going to change. The leaks from snowden concerned the how and the what.

Comment Easy. (Score 5, Insightful) 110

1) there will be a witch hunt.
The nsa will investigate its own employees against its already existing psych profile sheets to see who is the most likely to have been motivated to steal the data. Then they will set up an internal emtrapment scenario to catch the leaker red handed. They will then be charged with federal espionage, and put into prison.

2) the same investigation will sift out accomplices and contacts. The trap will not be sprung until positive id has been made on all members of the cell.

3) the nsa will not directly move against the other members of the cell. Instead it will monitor, and selectively leak false intel to this cell, making it ineffectual, or worse, countereffectual to the foriegn government operating it.

4) if deemed useful to do so, the cell will be infiltrated with a new "insider", who will actually be collecting and analyzing the cells instructions to better predict and respond to the foriegn power's activities.

Really, this is not hard.

Comment Can the WiFi do monitor mode? (Score 1) 124

I ask, because these tiny, fully integrated and inexpensive devices look like very inexpensive, and easily concealed WPA password crackers, or network listening devices to me.

The one with the sdcard slot could be made to do quite a few things, if you don't mind wearing out the card using it for swap space. A combination of zram and sdcard spillover swap, some sensors, and an ext data partition on the card would let this thing do quite a lot, such as sticking it on the back of a USB printer to make the printer network enabled, to running an electronic door control, camera monitoring (using vfl capable cameras), making WiFi driven toy cars or drones, and quite a few others.

For many things, you don't need a whole lot of power. I have used an old router as a pvpgn server for some time, in fact. Works fine.

For a lot of possible projects, physical size of the package and power draw are the obstacles, not the memory size or the processor.

Comment Welding hoods shut (Score 4, Interesting) 128

The issue with efuses is like welding the hood shut, so you cannot service the engine.

It forces "physical damage" to become necessary to gain the access you should have been able to get anyway. In the case of the welded hood: you have to cut through the welds. In the case of efuses protected boot loaders: the boot loader enforces strong crypto against custom kernels, preventing boot of devices without the magic number baked into them, and if you flash a new boot loader, bam, efuses blown.

When the hardware that gets damaged is little more than a "warranty void" sticker, just in digital form, the oems are stretching things pretty thin.

I would love to see them be told that they cannot do these kinds of things. Sadly, that is not how the world works today.

Comment Re:16gb ssd (Score 1) 55

You have to be kidding.

Unless it is USB 3.0, that port will saturate far faster than the sata2 port on the minnowboard.

An actual nvme based M.2 ssd put on would smoke both suggestions. If the device is running its own transactional database for object identification, having access times that fast becomes more than just a nice thing to have-- something i mention because of the suggested use case for the board, from intel themselves.

Slashdot Top Deals

RAM wasn't built in a day.