Yes the layered xoring approach would be best. But I don't know if you can get error correction at all with a OTP. Ideally there is no math in a OTP just pure randomness and xor.
Yes and no. A carry-on suitcase with 20 3 TB Hard drives ought to do many companies' communications needs. If you are doing more then you will just need more couriers. But you never ever reuse a OTP; if you reuse a OTP it goes from the single best encryption scheme to one of the worst. I took a course where the professor gave us a few strings of OTP encrypted data and it was no problem to crack the encryption. Also I am not some kind of uber crypto mathematician.
-
Will take time. I suspect that companies like Cisco will sigh a breath of relief over the next few months when sales don't plummet. What they won't realizes is that the biggest companies that have no doubt issued directives for an end to end anti-US snooping overhaul will take a while to figure out what needs to be replaced and which products are best. So while these audits and re-architectings take place these companies will continue with business as usual. And even when the plan is deployed I doubt 100,000 employee companies will just toss all their stuff out on Friday to have it all replaced on Monday. They will start with the most critical bits and work their way down the information value chain. So at this point the Cisco type companies will see a slight drop in sales but even still the companies will continue with maintenance contracts to keep their gear going.
But at a certain point you will have an interesting problem. That is that these companies will begin to dump their Cisco gear onto the open market. So along with a sudden drop in sales to key customers you will have a glut of un-trusted gear flood the market.
I use Cisco as an example but you can sub in any American (or American stooge country) networking gear company.
I also expect to see a flourishing of cryptography in various foreign math departments around the world. If I were a Siemens I would be giving fairly large grants to German/Swedish/Norwegian etc math departments to do two things, check for backdoors and to come up with crypto systems that are quite unlike anything that the NSA has recommended.
But switching crypto systems is not as easy as just coming up with something that a bunch of math wizards think is solid. Things like AES crypto is baked right into many modern chip sets at the assembly instruction level. This is why AES based crypto is fantastically fast. So if your new system is different enough yet theoretically computationally equivalent to AES then it will be significantly slower on most chips.
One of the interesting changes that will probably come from this is that people won't trust anything. Thus they will run in 100 different directions. This will be a nightmare for the NSA because even if they can brake every crypto system that comes along they will have to spend the time to break them all.
But there is one system that can't be broken and that is one time pads. You have to physically share the pad but that is not so onerous for most companies as they have trusted employees going from branch to branch all the time. If the border people grab a copy of the OTP then you just toss it in the garbage. Plus one time pads can be layered. So you don't need to trust just one person taking one route.
Ah but the NSA have a huge automated vacuum. For any programmers who have had to import data it is very nice if it came in a nice data format. It is a pain if it doesn't. So a great way to foil the huge vacuum is quite simple. You first wrap your stuff in a bonkers encryption system that your 5 year old might have come up with; this system might be crackable in 5 minutes but that is a highly trained human working very hard for those 5 minutes. Then you wrap your badly encrypted data in a solid and well trusted scheme. So if they haven't broken the trusted system you are golden. But if the common system has been broken then your data needs to be valuable enough for an uber expert to be bothered to pry it open. But if your system is actually half decent then with the resources available your data might be impregnable with the resources available to it. Unless you are a very high value target. But even then you could rotate and modify your encryption systems every week or two. They don't have to be very good, just different.
If I gave a very good programmer the JPEG spec and had them write a import tool from scratch I suspect that most programmers would take some time, make mistakes, and generally not enjoy the experience. So now give the same programmers(who aren't familiar with the inner workings of jpegs) a bunch of jpgs and ask them to build an import tool without the spec. Very hard and a royal pain in the ass. This isn't even encryption.
Now multiply this by hundreds or thousands of companies.
So don't exactly call it security through obscurity; call it security through pain in the ass.
I think that this would be a great business idea. You sell companies bespoke VPN layers to exist underneath the traditional VPN layers. They buy a subscription to the tool which is then upgraded on a random schedule with new and bizarre encryption schemes. But different schemes for all the different companies. The best part is you don't deploy them over the net but in briefcases handcuffed to the guy's wrist. The company's logo would be a tinfoil fedora.
The NSA had in one recruiting thing (I think to students) was that if they figured out how to crack encryption that it would be their patriotic duty to tell the NSA first. After the Snowden revelations that nothing is off the table. Bascially tinfoil hats are in fashion.
Also we have only scratched the surface of Snowden's claims. The pattern seems to be: Government denies, Snowden proves them to be liars, Government denies, Snowden proves them to be liars.
I don't think people outside the US really care if US companies use 10,000 bit quantum spiral elliptical gluon encryption with a half twist of lemon. If the NSA comes to those companies with the Open Sesame court orders then it doesn't matter. This is a massive opportunity for non-US companies to say, "We ignore any pressure from the US." Along with their governments to say, "If a local company gives data to the US government then they go to jail." Put these two together and people will start flocking to their service (assuming it is roughly equal to the US one) so create euromail.eu or whatnot and you've got customers.
Right now is the time to have a marketing shtick where you tell people that you spend all day every day thinking up ways to keep the NSA away from their data.
Also this is the time for Linux to strike. The key is that there are two assumptions being made by most people out there. First is that any US company with closed source software has been strong-armed into leaving a back door. Second is that the NSA have broken any common encryption scheme. So if you use the common ones they might as well be plaintext. But if you are able to use opensource obscure encryption schemes then you stand a chance.
If your organization has a person with the title CISO then your organization is a creaking rusty bucket of stagnation. I had to look up what a CISO was. The description of a CISO is a tiny tiny subset of what the head of IT or the chief Admin should be doing. But giving it the title of CISO makes it sound like the NSA.
A CISO is like in the early 70s when airplanes had engineers in addition to the pilots. In the 60s keeping the engines running and whatnot was hard. But computers could do the work by the 70s so the pilots unions fought to keep them. But this is one of the areas that the newer nimbler airlines cut costs. They only bought airplanes that needed two in the cockpit and eliminated the need for the engineer. But some airlines stuck with them. Pan Am; remember them? The other bloated airlines basically had to go bankrupt one by one to get rid of them. It wasn't that the engineers bankrupted them but that having bloat like an engineer was a symptom of a creaking, rusty bucket of stagnation. I am willing to bet that a high-functioning CISO would be the primary driver of policies that drive the best employees right out of a company "No you can't have an iPad as we haven't reviewed the security implications." "No you can't run apps on your smart-phone until we have reviewed them." "No you can't have your own printer as you might print company secrets."
Every day I look at my server logs and see all kinds of "attacks" I'm not sure 5 minutes go by without another wp-admin attempt. I suspect that these are probing known easy attacks. So if I were running IT for a company I could make it sound like this was trench warfare WWI style. It would almost be funny to have an air raid siren going off every time one of these attacks came and having a fire pole for the Admins to slide down.
Seeing that these various "attacks" have various goals it can be hard to even define "Penetrated" If their goal is to scrape all the content from a site and they succeed, or if they wanted to get past the spam filters and post something spammy. Then should that be something you alert everyone about? It starts to become clear at a certain point such as the attacker logging in with root access. But what about a spam post that managed to have some active javascript that popped up a casino ad? Or an SQL Injection that resulted in a spam post being approved? SQL Injection is bad but spam is just another day in the office.
Then you have other fuzzies. DDOS attacks that slow things down. DDOS attacks that slow things down that give a competitor an advantage (such as slowing down a brokerages trades stratigically) or a DDOS attack that simply requires a new server (or 10) so you stop caring about it?
Up to a certain point what constitutes a security problem can be very fuzzy. In theory if you have a weak honeypot machine that when attacked cuts the intruders out of the rest of the network you might have an awesome strategy but could be reported(by some auditor looking to get the contract) as having a "weak link" machine that was woefully under-secured and regularly compromised by intruders unknown.
I am going to ignore this story (because it is stupid) and instead comment on Microsoft basically buying Nokia. First a prediction, smartphones are going to be commodity items in 5 years. Yes there will still be some high end units but 5 years from now the fight will not be in features such as screen size, resolution, computing power, etc. It will be in the old features such as durability, battery capacity, cost, and weight. This is where Nokia can excel. Thus if Microsoft can keep from putting MS office on everything it makes then Microsoft has a chance of controlling the OS on a significant proportion of the phones out there.
I have another prediction. MS will think that two wrongs make a right and start giving Surfaces away with their new XBoxes (Limited time offer)
I guess they didn't even think about having that as a slogan. As an engineer you can work to make the world a better place or a worse place. This is a choice that is actively made. Here are 100 people who aren't going to make the right choice. I feel bad for them.
It is all simple math. The graph for LED pricing looks like a double diamond ski slope. The question is will it level out at a level that makes this economic. I suspect that there will be long term downward pressure on LED (or other energy efficient lighting) as more people look to go solar. Every time I go to the hardware store the LEDs are brighter and cheaper.
If I had to guess quite a bit of the present price of LEDs is made up of simple supply and demand. The demand is outstripping the supply resulting in a premium. But capacity is being added every day which may very well overshoot demand. So even if the technology didn't get cheaper after today prices would continue to drop. But quite simply manufacturers are only getting started mass producing large LED bulbs. So I suspect that year on year pricing will keep dropping for a long while yet.
Plus LEDs offer all kinds of interesting options for placement. You could mingle them in with the plants as they don't get hot. Not to mention that selective breeding may result in plants that prefer LED grow lights.
The other complicating factor is when you factor in the whole logistics chain. You can't just compare the cost of production in my all LED system to that of a far away field. You have the vagaries of weather, shipping, market fluctuations, etc. So if you have a LED system that most years makes little money you will have banner years where traditional farmers fail and prices go up. Whereas you will not have failures due to weather or disease. This consistency can be ruinous to your competition. Also you can even eliminate the banner year issue by selling your crops into the distant future. Plus you don't have the issue of seasons thus you get banner prices for much of the year.
So this week LEDs don't work economically, next week maybe they do.
I waited even longer for my XBox 360. When they did that perma ban a zillion people sold their consoles on the used market all at the same time. So I bought one then for around $80 when used consoles sold a couple of weeks before for around $160.
Doubly insulting for Microsoft as it clearly meant that I had zero desire to spend one cent on their Live product.
I bought their Kinect to use in robotics projects.
I don't understand? My premise is that MS is not going to generate much enthusiasm about this device. Thus they are going to have to artificially generate it.
So are you saying that they won't have any problems with people wanting their device and will get genuine hype. Or that they will have disappointing sales and won't resort to underhanded tactics to try to get people hyped up.
Maybe they will hire the same bunch of break dancers as they did for the Surface to inspire people to buy their dud console. My favorite Surface hype misfire was when Oprah tweeted how much she loved her Surface FROM HER IPAD!!!
So my question would be: What is the colour of the sky in Microsoft's Marketing world?
This is a concern with a US manufacturer of a device that is often on, has a camera, has a microphone, and comes from a company that has cooperated with the stazi. If you said this a year ago you would get a tinfoil hat award. Now you just get a "Could be right". So a simple question is: After Sony suffered the hacking scandal would they allow a US agency to spy on their customers opening the door to another Scandal?
Artificial artificial scarcity is going to be the name of this campaign. They don't want real scarcity as that could hurt their probably crap sales. But they want to be able to show the people lining up for days to get one. So I suspect they are going to spin this in some sleazy ways. A few sleazy ideas would be things like having a best buy in a really poor area offer the XBox at 80% off for the first 500 customers. This would then get at least 500 customers lined up for a week or more. You could also have other offers such as contests where you have a 1 in 5 chance of winning either the XBox or a trip or something. Or if worse comes to worse they could just hire actors.
Once they have the scarcity going they will do a huge PR campaign first to promote these 3 week long line-ups and then trying to get the world convinced that there are very few XBoxes in their local retailer. They could then work with the retailers to somehow do something like say 1 store in town has 500 while the other stores only have 50. Which store is it? Then people line up and all find that they were lucky enough to be at the store with 500 as they all had 500.
Lastly we are all going to be reading reviews from Console Times or other made up reviewers saying that this is a "Game Changer" and that while the PS4 is a slight upgrade that the new XBox is revolutionary, a paradigm shift, customer oriented, the only console that will survive.
But the worst is going to be in "Votable" forums like the Slashdots and Reddits; where they nodoubt have an army of voters seeking out to kill the bad reviews and promote the shill reviews. A simple example of this would be in any MS related Slashdot posting the anti MS positions are voted down initially but after a while they resurface and then become the norm. My guess is that the PR firms have a limited number of Karma points at any time and try to steer the mood at the beginning but then run out of ammunition.
I'll add one more level of indoors. How about underground? That is to put the plants in a completely controlled environment underground. You can then control for absolutely everything. You could even then have solar on the surface getting the power. LED lights are in a price freefall and last for just about forever. Plus with careful study you can have the LED lights produce only those frequencies that are of use to the plants. Other factors such as temperature, humidity, and nutrients can also be perfectly controlled. Also by going underground even the air quality can be uber controlled preventing fungus and insect pests from entering.
Plus many plants can be cajoled into doing desired things under highly controlled conditions. For some plants days can be shortened to 16 hours; other plants like huge temperature swings from night to day; and many plants will produce deeper colours and flavours when given the occasional UV sunburn.
Then you get other interesting systems such as aquaculture(with fish) that allows you to scientifically have a fairly organic system that also produces some protein.
Once you go indoors the world food trade system could be thrown into chaos as what happens if we start to grow organic, high quality, cheap coffee (bananas, coconuts, chocolate, etc) that normal come from countries that are presently struggling to get by. The reality of our system is that two things that grocery stores prize above many things are consistency of supply and price. But if you can deliver a "local" good that is organically grown (at least without pesticides and herbicides and with fairly natural fertilizers and no GMOs) then you are golden.
But then you start to get other interesting economic things happening (some good, some bad, some terrible). If you look at the average household budget a fair chunk is for food. For most places food is largely imported. Thus a good chunk of any given areas spending leaves that area. With localized production this keeps money circulating locally for longer. For areas that will continue to export goods from their regions (say oil, minerals, manufacturing, fish, etc) this is excellent news. As the trade flow will become even more imbalanced. For areas that traditionally were have nots and didn't export much and imported food it will help stem the bleed. But for areas that primarily exported food this is a disaster; A total disaster.
But you are right. Putting farming in close to the cities (Market gardens) with nearly perfect closed non polluting systems will be quite cool. I can even make an interesting robotic suggestion. With robot cars, most buildings won't need all their underground parking. So you swap out a level or two of parking for a farm. Seeing that most plants only reach a few feet up you might get two or three layers of plants per parking story. So if 20% of the buildings in a city dedicated 2 levels of parking on which 3 layers were planted. You would have a hydroponic farming area equal to 120% of those buildings footprints. So in theory you could have Manhattan farming at a higher productivity than the Central Valley of California.
You have hit at the exact problem with all robotics where modern robotics will eat all low skilled jobs. It is a cultural problem not a technological or economic one. Some societies will become feudal with a small few owning everything and the great unwashed masses completely left out of the economic game and on some kind of punitive welfare.
But some societies will know that they are all about their people. One guess is that concepts like Minimum Basic Wages (different from minimum wage) and high income taxes will shift the focus from production and capitalism (which is easy with robotics and thus shouldn't be greatly rewarded) to consumption and fairness.
I am not talking about communism for if you look at the defective planned economy of the Soviet Union where they focused on production and things still sucked. The idea is that you focus on simple things that encourage consumption and equality and then let people figure the rest out themselves. But most societies focus on the magic term GDP and with robots that number can be very very high even with extreme unemployment. Thus it is a terrible standard to measure a happy economic situation.
But the stupidest societies of all will ban or fight robotic production.
I agree with our points completely; especially the points about teaching math earlier. There are great math courses (from the Great Courses company along with some from the Annenburg Foundation and MathDVD) that are awesome. I can't see why you couldn't start the kids with these videos in early grade 7 and be done basic calculus by grade 9.
There are efforts to create open textbooks but these don't have commissioned salesmen. When you are looking at the textbook budget for a state it is huge (10's-100's of millions). The technique that seems to be working to prevent these open source textbooks is to have convinced the various school boards to group buy their books. That is they buy 200 different books at once instead of say a single tender for just the math textbook. Plus these textbooks come with all sorts of kickbacks along the lines of "training seminars" it is nice when these seminars are in Hawaii.
And this is where the power of computers can work in education. For it is not just Teacher B who is bad but possibly B E and L which means that it becomes almost impossible for a human to sort out the data. But for a computer not only would it be easy but the computer could even shuffle the students/teachers so as to create the ideal statistical model for detecting lousy teachers. Also this would allow for targeted tutoring. If student A was doing very well in math years 1-6 and then hits a terrible teacher in 7 that is detected through poor grades in 8 you could say, "That student is performing below expectations and with tutoring can be restored."
The downside for a system like this is that it mathematically exposes bad teachers and it isn't very glamorous. Handing out iPads is glamorous.
I did some work analyzing the standardized test results for 3 states. The three states used student IDs that were 9-12 digits. Yet the data I was given had students with 3 digit ID and many IDs with letters. The students' names often had Tildes in them or were just "Geo(&YNT". This was the official state data. My job was to see if I could use some form of ML or other analysis to track students from one year to the next. Basically this was impossible. I could look at the probability that Zach Mill~r ID 123456789 was the same as Ach Miller ID 1234S6I8H seeing that the second one was a grade ahead in the same school and was the best match. But do that with all the students and you lose all kinds of data such as dropouts, new students, students who move from school to school (even though they kept the same ID in the state)
My conclusion was that they didn't want to track anything and that with fuzzy data you could model it to fit what you wanted. You could say that students didn't drop out if you just could make one student into another. Basically any claims you made would not be auditable as you would have started off by legitimately bringing opinions in to the data at the first step. So how could they prove that you tailored your opinions to produce the desired results?
I was recently picking blueberries at a u-pick. This is easily the best year I have ever seen. Literally the bushes were breaking under the weight of the blueberries. You could eat the berries off the bushes like corn on the cob. The problem is that most berry areas are having a similar banner year along with there being a huge amount of berries planted. All this has resulted in a price crash. This crash has made it borderline uneconomic to harvest the berries. But if you had a robotic harvester this changes the pricing quite a bit. Once you have purchased the machine the price to run it should be very low and the amortized costs are there regardless if you run the machine or not. Thus you can harvest the berries even in banner years. Another option is to also plant excessive crops of different types and then focus your harvesting on the most profitable crops in any given year.
It is my firm belief that robotic agriculture will change the entirety of how we produce food. A few simple examples of changes that few people discuss would be the terrain that is used for harvesting. Two of the key advantages of flat land for grains is that the crop will develop consistently across large areas and thus when harvested be of a predictable quality when turned into bread and whatnot. The other is that it is far easier to build the massive harvesting machines if they don't have to contend with any variations in the terrain. The goal of the massive machines is to vastly increase the ability of a single human to do a huge amount of work.
But with robotic planting, tending, and harvesting you don't need to "multiply" the work of a single human. Thus the robots can be fairly small. Also the robots can adjust the feeding of the plants so to grow a fairly consistent crop in inconsistent terrain. Then in the end when it comes time to harvest. The robot can methodically harvest at the perfect moment for any given plant (repeatedly bypassing those not ready) plus it can methodically sort even down the single grain.
Another advantage is where the cost of the entire cycle of agriculture can be so low that you could robotically convert marginal land into low producing land and still produce food at a very low cost. The return on quality land would be higher but by being able to cheaply bring marginal land into production it will form a scenario of relentless competition thus holding down prices. Plus once again due to the nature of robot economics once marginal land was in production the cost of continued production would be very low. This could also be carefully factored into the logistics calculations where a less efficient production is competitive where it might reduce some other cost such as shipping.
This last factor might result in it being cheaper to produce greenhouses and then produce goods year-round much closer to the point of consumption rather than shipping them half way around the world.
Also robotics can be used inefficient ways such as massively processing marginal land making it quite productive. Normally this is a time eating process that is not worth it. But if you can leave some robots cooking away in a forest for a few years and come back to find nutrient rich terra pretta then again the economics change.
What I can't foresee is which direction agriculture will take. I have a feeling it will be mega massive monster farming companies with very few employees that depopulate the rural farm communities. But at the same time the low barriers to entry might mean that many people will jump in the moment a competitive opportunity is perceived. Personally where food is such a fundamental part of living (right there after clean water) that I don't believe that any small group of companies should be allowed to concentrate ownership of any nation's food production. If they get it wrong, or play evil games, massive numbers of people could suffer.
One prediction that I will solidly make is that there will be very very very very few people employed in agriculture in 20-50 years.
His point was that he wasn't in daycare and didn't need to learn their song in order to work well with others and produce awesome code. So he told them that if they were going to treat him like he was 5 he would act like he was 5. Plus he knew that they would not bend to him and just buy him out. The other two who sold stuck it out and I think went a little bit insane.
Very often every system in education becomes hijacked by some interest group. Textbooks are a great example. Looking through my daughters' very expensive textbooks I can see that the science and math textbooks were written by non mathematician/scientists. One of my favorite questions went something like Jamal has 5 candies that are 5 different flavours; how can he distribute them among his 5 friends? Write all the ways. WHAT? Or just the usual questions that are missing some element such as you have a triangle that is 2 units on the bottom side and 3 units high. How long is the remaining side? But there is no picture of the triangle. Is this a right-triangle. Are they talking about the hypotenuse? And then one of the best. A grade 10 math textbook with a section on parabolas. My daughter was assigned the usual questions 1-20 at the end of the chapter. I don't quite remember how to find the vertex or some such so I leaf through the textbook to find out how. All it does is define the parabola and give some examples of how they can be used for things like flashlight reflectors. But absolutely no math involving the parabolas. None. Lots of parabola questions but no math. This was not some kind of workbook but a textbook where they had just been sloppy.
Then there is the technology. They are so lost. So so so lost. They have just grasp at technology. The usual result is that they buy big systems where moodle would be fine. But at no point do they really leverage the technology much. A great example is both of my daughters' schools have robocalls to tell me about things like vaccinations, school trips, etc. This is very annoying in that the calls usually waste most of the call telling me things that I don't care about. The worst part is that the critical bits are at the end. So I hear about things like congratulations to some student for winning a sack race in Kalamazoo and then in the end learn that some critical form needs to be turned in by 9am the next morning. Hello please use at least email. Maybe a website? The 20th century is calling and wants their robocaller back! I wonder how much they pay for this service?
But there is a wonderfully effective way to use computers in education. You look at student's marks. You then look at the pattern of the marks as the student's pass through various teachers. I am not talking about standardized tests but just comparing the marks of various students in the same classrooms. The key being that you can see that when a batch of students hits a truly great or terrible teacher that their marks will thrive or suffer for years to come. Bad teachers are like boulders in the stream; they result in much turbulence and waves far beyond their position in the time stream. Both of my daughters hit the same terrible math teacher. I tutored both of them past this disaster of a teacher but many of their co-students may have lost any hope at a career in STEM as their grade 10 math would then suck with little time left to recover to the point where they could leave HS with a good mark in Pre-cal let alone Calculus.
Magical. I think this comes from some misfiring in some people's brains of their "Fairness" circuit. I have been here for 20 years so it is not "fair" that the new guy who runs circles around me gets more money, in fact, it is not "fair" that they even let him run circles around me. They should put limits on how many customers he can call per day. The horrible problem is that in these situations it is often the guy who is envious who is also BBQ buddies with someone else who can set policy. So suddenly things like your IBM story happen.
A happier story was a guy I know was part owner of a company bought out by a large old company. Part of the deal was that he would work there as a "Senior Software Architect" so they bring him in with a group of new employees where he has to learn the company's song. He walks up to the HR guy and tells him, "I won't learn your stupid song, to make it clear, I won't do anything that isn't directly involved in being a Senior Software Architect." Long story short they tried to bully him but in the end just bought out his year. There was one moment in the process where he put his fingers in his ears and said over and over, "I'm not listening".
I am certain that in the above company that when they brought in a new CEO that he didn't have to learn the company song.
Slashdot Mirror
Yes the layered xoring approach would be best. But I don't know if you can get error correction at all with a OTP. Ideally there is no math in a OTP just pure randomness and xor.
Yes and no. A carry-on suitcase with 20 3 TB Hard drives ought to do many companies' communications needs. If you are doing more then you will just need more couriers. But you never ever reuse a OTP; if you reuse a OTP it goes from the single best encryption scheme to one of the worst. I took a course where the professor gave us a few strings of OTP encrypted data and it was no problem to crack the encryption. Also I am not some kind of uber crypto mathematician. -
Will take time. I suspect that companies like Cisco will sigh a breath of relief over the next few months when sales don't plummet. What they won't realizes is that the biggest companies that have no doubt issued directives for an end to end anti-US snooping overhaul will take a while to figure out what needs to be replaced and which products are best. So while these audits and re-architectings take place these companies will continue with business as usual. And even when the plan is deployed I doubt 100,000 employee companies will just toss all their stuff out on Friday to have it all replaced on Monday. They will start with the most critical bits and work their way down the information value chain. So at this point the Cisco type companies will see a slight drop in sales but even still the companies will continue with maintenance contracts to keep their gear going.
But at a certain point you will have an interesting problem. That is that these companies will begin to dump their Cisco gear onto the open market. So along with a sudden drop in sales to key customers you will have a glut of un-trusted gear flood the market.
I use Cisco as an example but you can sub in any American (or American stooge country) networking gear company.
I also expect to see a flourishing of cryptography in various foreign math departments around the world. If I were a Siemens I would be giving fairly large grants to German/Swedish/Norwegian etc math departments to do two things, check for backdoors and to come up with crypto systems that are quite unlike anything that the NSA has recommended.
But switching crypto systems is not as easy as just coming up with something that a bunch of math wizards think is solid. Things like AES crypto is baked right into many modern chip sets at the assembly instruction level. This is why AES based crypto is fantastically fast. So if your new system is different enough yet theoretically computationally equivalent to AES then it will be significantly slower on most chips.
One of the interesting changes that will probably come from this is that people won't trust anything. Thus they will run in 100 different directions. This will be a nightmare for the NSA because even if they can brake every crypto system that comes along they will have to spend the time to break them all.
But there is one system that can't be broken and that is one time pads. You have to physically share the pad but that is not so onerous for most companies as they have trusted employees going from branch to branch all the time. If the border people grab a copy of the OTP then you just toss it in the garbage. Plus one time pads can be layered. So you don't need to trust just one person taking one route.
Ah but the NSA have a huge automated vacuum. For any programmers who have had to import data it is very nice if it came in a nice data format. It is a pain if it doesn't. So a great way to foil the huge vacuum is quite simple. You first wrap your stuff in a bonkers encryption system that your 5 year old might have come up with; this system might be crackable in 5 minutes but that is a highly trained human working very hard for those 5 minutes. Then you wrap your badly encrypted data in a solid and well trusted scheme. So if they haven't broken the trusted system you are golden. But if the common system has been broken then your data needs to be valuable enough for an uber expert to be bothered to pry it open. But if your system is actually half decent then with the resources available your data might be impregnable with the resources available to it. Unless you are a very high value target. But even then you could rotate and modify your encryption systems every week or two. They don't have to be very good, just different.
If I gave a very good programmer the JPEG spec and had them write a import tool from scratch I suspect that most programmers would take some time, make mistakes, and generally not enjoy the experience. So now give the same programmers(who aren't familiar with the inner workings of jpegs) a bunch of jpgs and ask them to build an import tool without the spec. Very hard and a royal pain in the ass. This isn't even encryption.
Now multiply this by hundreds or thousands of companies.
So don't exactly call it security through obscurity; call it security through pain in the ass.
I think that this would be a great business idea. You sell companies bespoke VPN layers to exist underneath the traditional VPN layers. They buy a subscription to the tool which is then upgraded on a random schedule with new and bizarre encryption schemes. But different schemes for all the different companies. The best part is you don't deploy them over the net but in briefcases handcuffed to the guy's wrist. The company's logo would be a tinfoil fedora.
The NSA had in one recruiting thing (I think to students) was that if they figured out how to crack encryption that it would be their patriotic duty to tell the NSA first. After the Snowden revelations that nothing is off the table. Bascially tinfoil hats are in fashion.
Also we have only scratched the surface of Snowden's claims. The pattern seems to be: Government denies, Snowden proves them to be liars, Government denies, Snowden proves them to be liars.
I don't think people outside the US really care if US companies use 10,000 bit quantum spiral elliptical gluon encryption with a half twist of lemon. If the NSA comes to those companies with the Open Sesame court orders then it doesn't matter. This is a massive opportunity for non-US companies to say, "We ignore any pressure from the US." Along with their governments to say, "If a local company gives data to the US government then they go to jail." Put these two together and people will start flocking to their service (assuming it is roughly equal to the US one) so create euromail.eu or whatnot and you've got customers.
Right now is the time to have a marketing shtick where you tell people that you spend all day every day thinking up ways to keep the NSA away from their data.
Also this is the time for Linux to strike. The key is that there are two assumptions being made by most people out there. First is that any US company with closed source software has been strong-armed into leaving a back door. Second is that the NSA have broken any common encryption scheme. So if you use the common ones they might as well be plaintext. But if you are able to use opensource obscure encryption schemes then you stand a chance.
If your organization has a person with the title CISO then your organization is a creaking rusty bucket of stagnation. I had to look up what a CISO was. The description of a CISO is a tiny tiny subset of what the head of IT or the chief Admin should be doing. But giving it the title of CISO makes it sound like the NSA.
A CISO is like in the early 70s when airplanes had engineers in addition to the pilots. In the 60s keeping the engines running and whatnot was hard. But computers could do the work by the 70s so the pilots unions fought to keep them. But this is one of the areas that the newer nimbler airlines cut costs. They only bought airplanes that needed two in the cockpit and eliminated the need for the engineer. But some airlines stuck with them. Pan Am; remember them? The other bloated airlines basically had to go bankrupt one by one to get rid of them. It wasn't that the engineers bankrupted them but that having bloat like an engineer was a symptom of a creaking, rusty bucket of stagnation. I am willing to bet that a high-functioning CISO would be the primary driver of policies that drive the best employees right out of a company "No you can't have an iPad as we haven't reviewed the security implications." "No you can't run apps on your smart-phone until we have reviewed them." "No you can't have your own printer as you might print company secrets."
Every day I look at my server logs and see all kinds of "attacks" I'm not sure 5 minutes go by without another wp-admin attempt. I suspect that these are probing known easy attacks. So if I were running IT for a company I could make it sound like this was trench warfare WWI style. It would almost be funny to have an air raid siren going off every time one of these attacks came and having a fire pole for the Admins to slide down.
Seeing that these various "attacks" have various goals it can be hard to even define "Penetrated" If their goal is to scrape all the content from a site and they succeed, or if they wanted to get past the spam filters and post something spammy. Then should that be something you alert everyone about? It starts to become clear at a certain point such as the attacker logging in with root access. But what about a spam post that managed to have some active javascript that popped up a casino ad? Or an SQL Injection that resulted in a spam post being approved? SQL Injection is bad but spam is just another day in the office.
Then you have other fuzzies. DDOS attacks that slow things down. DDOS attacks that slow things down that give a competitor an advantage (such as slowing down a brokerages trades stratigically) or a DDOS attack that simply requires a new server (or 10) so you stop caring about it?
Up to a certain point what constitutes a security problem can be very fuzzy. In theory if you have a weak honeypot machine that when attacked cuts the intruders out of the rest of the network you might have an awesome strategy but could be reported(by some auditor looking to get the contract) as having a "weak link" machine that was woefully under-secured and regularly compromised by intruders unknown.
I am going to ignore this story (because it is stupid) and instead comment on Microsoft basically buying Nokia. First a prediction, smartphones are going to be commodity items in 5 years. Yes there will still be some high end units but 5 years from now the fight will not be in features such as screen size, resolution, computing power, etc. It will be in the old features such as durability, battery capacity, cost, and weight. This is where Nokia can excel. Thus if Microsoft can keep from putting MS office on everything it makes then Microsoft has a chance of controlling the OS on a significant proportion of the phones out there.
I have another prediction. MS will think that two wrongs make a right and start giving Surfaces away with their new XBoxes (Limited time offer)
I guess they didn't even think about having that as a slogan. As an engineer you can work to make the world a better place or a worse place. This is a choice that is actively made. Here are 100 people who aren't going to make the right choice. I feel bad for them.
It is all simple math. The graph for LED pricing looks like a double diamond ski slope. The question is will it level out at a level that makes this economic. I suspect that there will be long term downward pressure on LED (or other energy efficient lighting) as more people look to go solar. Every time I go to the hardware store the LEDs are brighter and cheaper.
If I had to guess quite a bit of the present price of LEDs is made up of simple supply and demand. The demand is outstripping the supply resulting in a premium. But capacity is being added every day which may very well overshoot demand. So even if the technology didn't get cheaper after today prices would continue to drop. But quite simply manufacturers are only getting started mass producing large LED bulbs. So I suspect that year on year pricing will keep dropping for a long while yet.
Plus LEDs offer all kinds of interesting options for placement. You could mingle them in with the plants as they don't get hot. Not to mention that selective breeding may result in plants that prefer LED grow lights.
The other complicating factor is when you factor in the whole logistics chain. You can't just compare the cost of production in my all LED system to that of a far away field. You have the vagaries of weather, shipping, market fluctuations, etc. So if you have a LED system that most years makes little money you will have banner years where traditional farmers fail and prices go up. Whereas you will not have failures due to weather or disease. This consistency can be ruinous to your competition. Also you can even eliminate the banner year issue by selling your crops into the distant future. Plus you don't have the issue of seasons thus you get banner prices for much of the year. So this week LEDs don't work economically, next week maybe they do.
I waited even longer for my XBox 360. When they did that perma ban a zillion people sold their consoles on the used market all at the same time. So I bought one then for around $80 when used consoles sold a couple of weeks before for around $160.
Doubly insulting for Microsoft as it clearly meant that I had zero desire to spend one cent on their Live product.
I bought their Kinect to use in robotics projects.
I don't understand? My premise is that MS is not going to generate much enthusiasm about this device. Thus they are going to have to artificially generate it.
So are you saying that they won't have any problems with people wanting their device and will get genuine hype. Or that they will have disappointing sales and won't resort to underhanded tactics to try to get people hyped up.
Maybe they will hire the same bunch of break dancers as they did for the Surface to inspire people to buy their dud console. My favorite Surface hype misfire was when Oprah tweeted how much she loved her Surface FROM HER IPAD!!!
So my question would be: What is the colour of the sky in Microsoft's Marketing world?
This is a concern with a US manufacturer of a device that is often on, has a camera, has a microphone, and comes from a company that has cooperated with the stazi. If you said this a year ago you would get a tinfoil hat award. Now you just get a "Could be right". So a simple question is: After Sony suffered the hacking scandal would they allow a US agency to spy on their customers opening the door to another Scandal?
Artificial artificial scarcity is going to be the name of this campaign. They don't want real scarcity as that could hurt their probably crap sales. But they want to be able to show the people lining up for days to get one. So I suspect they are going to spin this in some sleazy ways. A few sleazy ideas would be things like having a best buy in a really poor area offer the XBox at 80% off for the first 500 customers. This would then get at least 500 customers lined up for a week or more. You could also have other offers such as contests where you have a 1 in 5 chance of winning either the XBox or a trip or something. Or if worse comes to worse they could just hire actors.
Once they have the scarcity going they will do a huge PR campaign first to promote these 3 week long line-ups and then trying to get the world convinced that there are very few XBoxes in their local retailer. They could then work with the retailers to somehow do something like say 1 store in town has 500 while the other stores only have 50. Which store is it? Then people line up and all find that they were lucky enough to be at the store with 500 as they all had 500.
Lastly we are all going to be reading reviews from Console Times or other made up reviewers saying that this is a "Game Changer" and that while the PS4 is a slight upgrade that the new XBox is revolutionary, a paradigm shift, customer oriented, the only console that will survive.
But the worst is going to be in "Votable" forums like the Slashdots and Reddits; where they nodoubt have an army of voters seeking out to kill the bad reviews and promote the shill reviews. A simple example of this would be in any MS related Slashdot posting the anti MS positions are voted down initially but after a while they resurface and then become the norm. My guess is that the PR firms have a limited number of Karma points at any time and try to steer the mood at the beginning but then run out of ammunition.
In base rand(73);
I'll add one more level of indoors. How about underground? That is to put the plants in a completely controlled environment underground. You can then control for absolutely everything. You could even then have solar on the surface getting the power. LED lights are in a price freefall and last for just about forever. Plus with careful study you can have the LED lights produce only those frequencies that are of use to the plants. Other factors such as temperature, humidity, and nutrients can also be perfectly controlled. Also by going underground even the air quality can be uber controlled preventing fungus and insect pests from entering.
Plus many plants can be cajoled into doing desired things under highly controlled conditions. For some plants days can be shortened to 16 hours; other plants like huge temperature swings from night to day; and many plants will produce deeper colours and flavours when given the occasional UV sunburn.
Then you get other interesting systems such as aquaculture(with fish) that allows you to scientifically have a fairly organic system that also produces some protein.
Once you go indoors the world food trade system could be thrown into chaos as what happens if we start to grow organic, high quality, cheap coffee (bananas, coconuts, chocolate, etc) that normal come from countries that are presently struggling to get by. The reality of our system is that two things that grocery stores prize above many things are consistency of supply and price. But if you can deliver a "local" good that is organically grown (at least without pesticides and herbicides and with fairly natural fertilizers and no GMOs) then you are golden.
But then you start to get other interesting economic things happening (some good, some bad, some terrible). If you look at the average household budget a fair chunk is for food. For most places food is largely imported. Thus a good chunk of any given areas spending leaves that area. With localized production this keeps money circulating locally for longer. For areas that will continue to export goods from their regions (say oil, minerals, manufacturing, fish, etc) this is excellent news. As the trade flow will become even more imbalanced. For areas that traditionally were have nots and didn't export much and imported food it will help stem the bleed. But for areas that primarily exported food this is a disaster; A total disaster.
But you are right. Putting farming in close to the cities (Market gardens) with nearly perfect closed non polluting systems will be quite cool. I can even make an interesting robotic suggestion. With robot cars, most buildings won't need all their underground parking. So you swap out a level or two of parking for a farm. Seeing that most plants only reach a few feet up you might get two or three layers of plants per parking story. So if 20% of the buildings in a city dedicated 2 levels of parking on which 3 layers were planted. You would have a hydroponic farming area equal to 120% of those buildings footprints. So in theory you could have Manhattan farming at a higher productivity than the Central Valley of California.
You have hit at the exact problem with all robotics where modern robotics will eat all low skilled jobs. It is a cultural problem not a technological or economic one. Some societies will become feudal with a small few owning everything and the great unwashed masses completely left out of the economic game and on some kind of punitive welfare.
But some societies will know that they are all about their people. One guess is that concepts like Minimum Basic Wages (different from minimum wage) and high income taxes will shift the focus from production and capitalism (which is easy with robotics and thus shouldn't be greatly rewarded) to consumption and fairness.
I am not talking about communism for if you look at the defective planned economy of the Soviet Union where they focused on production and things still sucked. The idea is that you focus on simple things that encourage consumption and equality and then let people figure the rest out themselves. But most societies focus on the magic term GDP and with robots that number can be very very high even with extreme unemployment. Thus it is a terrible standard to measure a happy economic situation.
But the stupidest societies of all will ban or fight robotic production.
I agree with our points completely; especially the points about teaching math earlier. There are great math courses (from the Great Courses company along with some from the Annenburg Foundation and MathDVD) that are awesome. I can't see why you couldn't start the kids with these videos in early grade 7 and be done basic calculus by grade 9.
There are efforts to create open textbooks but these don't have commissioned salesmen. When you are looking at the textbook budget for a state it is huge (10's-100's of millions). The technique that seems to be working to prevent these open source textbooks is to have convinced the various school boards to group buy their books. That is they buy 200 different books at once instead of say a single tender for just the math textbook. Plus these textbooks come with all sorts of kickbacks along the lines of "training seminars" it is nice when these seminars are in Hawaii.
And this is where the power of computers can work in education. For it is not just Teacher B who is bad but possibly B E and L which means that it becomes almost impossible for a human to sort out the data. But for a computer not only would it be easy but the computer could even shuffle the students/teachers so as to create the ideal statistical model for detecting lousy teachers. Also this would allow for targeted tutoring. If student A was doing very well in math years 1-6 and then hits a terrible teacher in 7 that is detected through poor grades in 8 you could say, "That student is performing below expectations and with tutoring can be restored."
The downside for a system like this is that it mathematically exposes bad teachers and it isn't very glamorous. Handing out iPads is glamorous.
I did some work analyzing the standardized test results for 3 states. The three states used student IDs that were 9-12 digits. Yet the data I was given had students with 3 digit ID and many IDs with letters. The students' names often had Tildes in them or were just "Geo(&YNT". This was the official state data. My job was to see if I could use some form of ML or other analysis to track students from one year to the next. Basically this was impossible. I could look at the probability that Zach Mill~r ID 123456789 was the same as Ach Miller ID 1234S6I8H seeing that the second one was a grade ahead in the same school and was the best match. But do that with all the students and you lose all kinds of data such as dropouts, new students, students who move from school to school (even though they kept the same ID in the state)
My conclusion was that they didn't want to track anything and that with fuzzy data you could model it to fit what you wanted. You could say that students didn't drop out if you just could make one student into another. Basically any claims you made would not be auditable as you would have started off by legitimately bringing opinions in to the data at the first step. So how could they prove that you tailored your opinions to produce the desired results?
I was recently picking blueberries at a u-pick. This is easily the best year I have ever seen. Literally the bushes were breaking under the weight of the blueberries. You could eat the berries off the bushes like corn on the cob. The problem is that most berry areas are having a similar banner year along with there being a huge amount of berries planted. All this has resulted in a price crash. This crash has made it borderline uneconomic to harvest the berries. But if you had a robotic harvester this changes the pricing quite a bit. Once you have purchased the machine the price to run it should be very low and the amortized costs are there regardless if you run the machine or not. Thus you can harvest the berries even in banner years. Another option is to also plant excessive crops of different types and then focus your harvesting on the most profitable crops in any given year.
It is my firm belief that robotic agriculture will change the entirety of how we produce food. A few simple examples of changes that few people discuss would be the terrain that is used for harvesting. Two of the key advantages of flat land for grains is that the crop will develop consistently across large areas and thus when harvested be of a predictable quality when turned into bread and whatnot. The other is that it is far easier to build the massive harvesting machines if they don't have to contend with any variations in the terrain. The goal of the massive machines is to vastly increase the ability of a single human to do a huge amount of work.
But with robotic planting, tending, and harvesting you don't need to "multiply" the work of a single human. Thus the robots can be fairly small. Also the robots can adjust the feeding of the plants so to grow a fairly consistent crop in inconsistent terrain. Then in the end when it comes time to harvest. The robot can methodically harvest at the perfect moment for any given plant (repeatedly bypassing those not ready) plus it can methodically sort even down the single grain.
Another advantage is where the cost of the entire cycle of agriculture can be so low that you could robotically convert marginal land into low producing land and still produce food at a very low cost. The return on quality land would be higher but by being able to cheaply bring marginal land into production it will form a scenario of relentless competition thus holding down prices. Plus once again due to the nature of robot economics once marginal land was in production the cost of continued production would be very low. This could also be carefully factored into the logistics calculations where a less efficient production is competitive where it might reduce some other cost such as shipping.
This last factor might result in it being cheaper to produce greenhouses and then produce goods year-round much closer to the point of consumption rather than shipping them half way around the world.
Also robotics can be used inefficient ways such as massively processing marginal land making it quite productive. Normally this is a time eating process that is not worth it. But if you can leave some robots cooking away in a forest for a few years and come back to find nutrient rich terra pretta then again the economics change.
What I can't foresee is which direction agriculture will take. I have a feeling it will be mega massive monster farming companies with very few employees that depopulate the rural farm communities. But at the same time the low barriers to entry might mean that many people will jump in the moment a competitive opportunity is perceived. Personally where food is such a fundamental part of living (right there after clean water) that I don't believe that any small group of companies should be allowed to concentrate ownership of any nation's food production. If they get it wrong, or play evil games, massive numbers of people could suffer.
One prediction that I will solidly make is that there will be very very very very few people employed in agriculture in 20-50 years.
His point was that he wasn't in daycare and didn't need to learn their song in order to work well with others and produce awesome code. So he told them that if they were going to treat him like he was 5 he would act like he was 5. Plus he knew that they would not bend to him and just buy him out. The other two who sold stuck it out and I think went a little bit insane.
Very often every system in education becomes hijacked by some interest group. Textbooks are a great example. Looking through my daughters' very expensive textbooks I can see that the science and math textbooks were written by non mathematician/scientists. One of my favorite questions went something like Jamal has 5 candies that are 5 different flavours; how can he distribute them among his 5 friends? Write all the ways. WHAT? Or just the usual questions that are missing some element such as you have a triangle that is 2 units on the bottom side and 3 units high. How long is the remaining side? But there is no picture of the triangle. Is this a right-triangle. Are they talking about the hypotenuse? And then one of the best. A grade 10 math textbook with a section on parabolas. My daughter was assigned the usual questions 1-20 at the end of the chapter. I don't quite remember how to find the vertex or some such so I leaf through the textbook to find out how. All it does is define the parabola and give some examples of how they can be used for things like flashlight reflectors. But absolutely no math involving the parabolas. None. Lots of parabola questions but no math. This was not some kind of workbook but a textbook where they had just been sloppy.
Then there is the technology. They are so lost. So so so lost. They have just grasp at technology. The usual result is that they buy big systems where moodle would be fine. But at no point do they really leverage the technology much. A great example is both of my daughters' schools have robocalls to tell me about things like vaccinations, school trips, etc. This is very annoying in that the calls usually waste most of the call telling me things that I don't care about. The worst part is that the critical bits are at the end. So I hear about things like congratulations to some student for winning a sack race in Kalamazoo and then in the end learn that some critical form needs to be turned in by 9am the next morning. Hello please use at least email. Maybe a website? The 20th century is calling and wants their robocaller back! I wonder how much they pay for this service?
But there is a wonderfully effective way to use computers in education. You look at student's marks. You then look at the pattern of the marks as the student's pass through various teachers. I am not talking about standardized tests but just comparing the marks of various students in the same classrooms. The key being that you can see that when a batch of students hits a truly great or terrible teacher that their marks will thrive or suffer for years to come. Bad teachers are like boulders in the stream; they result in much turbulence and waves far beyond their position in the time stream. Both of my daughters hit the same terrible math teacher. I tutored both of them past this disaster of a teacher but many of their co-students may have lost any hope at a career in STEM as their grade 10 math would then suck with little time left to recover to the point where they could leave HS with a good mark in Pre-cal let alone Calculus.
Magical. I think this comes from some misfiring in some people's brains of their "Fairness" circuit. I have been here for 20 years so it is not "fair" that the new guy who runs circles around me gets more money, in fact, it is not "fair" that they even let him run circles around me. They should put limits on how many customers he can call per day. The horrible problem is that in these situations it is often the guy who is envious who is also BBQ buddies with someone else who can set policy. So suddenly things like your IBM story happen.
A happier story was a guy I know was part owner of a company bought out by a large old company. Part of the deal was that he would work there as a "Senior Software Architect" so they bring him in with a group of new employees where he has to learn the company's song. He walks up to the HR guy and tells him, "I won't learn your stupid song, to make it clear, I won't do anything that isn't directly involved in being a Senior Software Architect." Long story short they tried to bully him but in the end just bought out his year. There was one moment in the process where he put his fingers in his ears and said over and over, "I'm not listening".
I am certain that in the above company that when they brought in a new CEO that he didn't have to learn the company song.