danjayh

This is exactly what you do -- lots of spares. If you're big enough, and have a big enough need, you can also design you own hardware ... although even that has limits, since foundries do eventually retire process nodes.

Dead wrong. In sectors that are in a safety critical or mission critical environment, it's not at all unusual to see systems that are decades out of date. We just recently retired the last of our VAX machines, and we still run a large part of our infrastructure for software build on Windows XP machines (airgapped). Our software configuration for these machines is certified, extensively tested, and has heaps of documentation proving that it does exactly what we say it does. Unimaginably large heaps. Moving to new hardware/software is incredibly expensive in some environments (in this case, I'm referring to a build and test platform for DO178B Level A certified software).

Also, replace mentally replace every instance of "museum ready" in this article with "mission critical" (which most of them probably are), and it'll seem a lot less ridiculous that they maintain them.

If you're ever back in Michigan, you should visit the west side ... and also check out lake Michigan. Grand Rapids isn't as big as Detroit, but our economy is growing at a healthy clip and there's a lot of interesting things going on here.

With you all the way.

I have 11 acres, horses, a goat, cats, a dog, and I live 15 minutes from downtown and 17 minutes from work. This is in a mid-sized midwest city (Grand Rapids, MI). While I realize that I could achieve a substantial pay raise on one of the coasts, a similarly sized plot there would cost millions. I paid under 200k 5 years ago with a 1500sf house on it, and it's still probably only worth the mid 200k range after we improved it (cleared 6 acres of trees, planted and fenced a bunch of pasture, made significant improvements to the house). I could not achieve a similar lifestyle on an engineer's salary near a major urban center on a coast ... and I have no desire to move to one :).

Plus, I get to make avionics, which is on my short list of the best jobs ever.

One thing the author totally overlooked - when you have a large percentage of the population that spends an extra 2-6+ years in education, that same large percentage of the population spends 2-6+ less years producing things for people to consume. If those same people aren't more productive as a result of their education, this means that the net productivity of society is lower as a result of the education, while the number of people in the society is the same. End result is that the stuff:people ratio goes down, and poverty naturally increases.

Education has an extremely high cost both in terms of lost productivity and resources consumed to provide it. Misapplied education will throw additional people into poverty, every time. Possibly directly, through onerous student loans, or indirectly by consuming resources that could have been used to produce basic goods for consumption.

He addresses the apple ][ idea. Since his board will be limited to 100s of khz (a PCB has signalling issues that an IC doesn't that will limit the max clock rate) and the apple ][ had HW with dependencies on the CPU being at a certain rate, it will not work just to build an adapter.

Disagree with agriculture being a disaster everywhere. Here in the midwest (Michigan, in my case) we get so much rain that there's rarely, if ever a need to artificially irrigate. Personally, the land that my horse pastures are on is so good that I threw seed on it and it just started growing. I do put a little bit of fertilizer down every few years, but mostly to replace the nutrients that are removed (baling, etc. will eventually deplete the land if you don't). Stuff grows here with very little encouragement.

I wonder what kind of bandwidth they're getting with their car (suv?) full of servers. More seriously, though, wouldn't something like camelcamelcamel be a perfect candidate for AWS or something similar, thereby preventing headaches like physically toting servers hundreds of miles?

Ugh. 'Infinite' vacation policies are starting to infect corporaculture. I work for a megacorp (that I actually rather enjoy working for) that recently went this way, and since my megacorp is a bit of a trend setter, you can expect it to spread. During the explanation meeting, it was made clear that we really aren't supposed to take more vacation than what we used to have, but that we are expected to still actually take about as much vacation as we used to have ... so essentially a no-change there. As far as I can tell, it's mostly about

1) Making it so that employees can't take vacation at times that are particularly bad for a project because they're 'entitled' to it. This was already frowned upon (we always did get to use our vacation eventually, so it wasn't so bad), but now it's in writing -- which is good, rules should be written.

and, more importantly to the megacorp: 2) Eliminating owed vacation from their books & marginally reducing cost. Now if somebody quits or is canned, they're not owed a payout. I'm guessing the average employee will probably also take a couple of days less than they did under the old system, yielding a further savings.

Compressed air storage is only as efficient as the insulation on the storage reservoir. Without insulation, the heat generated by the compression process is lost, and much of the stored energy goes along with it. This drives up the cost of building efficient compressed air storage systems, but there are several projects underway that are attempting to find cost-effective solutions to this problem.

I teach CS classes at a community college when I have the time. Our CC has a 2 year program; having taught classes in the 1st semester and the 4th semester, I have noticed that the 1st semester classes are larger than the 4th semester classes, much larger. This school offers eight sections of the intro programming class, and not more than one section of any top level class (Java, .net, and C++). The top level class sizes are also smaller, and not offered every semester (the intro class is). Net, I'd say that around 3/4 of the students fail out, quit, or change majors.

I've seen students bust their asses to get it. Spend time with tutors, countless hours with the book, with study groups, and other students and still not get it. I've also seen students sleep through most of their classes and pound out not only the lab, but also their entire three week assignment during the lab time.

Anyway, my point is that trying to get a tech degree is far from a sure thing, and for many people, something that they may not be able to finish. It takes a certain kind of thought process to write good software, and while some of it can be taught, I firmly believe that innate ability is equally important. Just like you couldn't teach me to be a good salesman if you had to, some people can't be taught to do tech jobs.

Oh, we still do, they're just in second and third world countries. The author of this article focused on service industries that require a physical presence in the local area ... and did not address outsourcing. I think that one likely outcome of this law will be that people will find creative new ways to move work to lower cost locales.

Wish it had a 'mod' instruction :(

Hell, they'll have half a billion just in the computing & control systems. Having worked to design, build, and certify the avionics systems on three major commercial aiframes, one of which is now flying (787), I am hopeful but beyond skeptical. When a company backed by the government of the world's largest economy (Comac) can't make decent progress on a subsonic airframe built with 20+ year old and well understood technologies, and when leading airframers like Boeing had a helluva tough learning curve building a composite subsonic airframer (787), I'm beyond doubtful that this startup is going to build a supersonic composite airplane before it goes bankrupt. Google '787 wing box' and '787 delamination' if you want to know the kind of pain Boeing went through when trying to build a composite airframe.

If you want to get and idea of the magnitude of the regulatory hurdles in a domain that most here will understand, google DO-178C. The mountains of documentation that you have to produce for the software will push any integrated avionics development effort well into the hundreds of millions, unless they plan on using bog standard off-the-shelf parts, and even then they'll have to deal with the flight control system, which will have to be custom by necessity. For some google food, search on 'wcet analysis', 'structural coverage analysis, 'MCDC analysis', 'requirements traceability', 'stack analysis', etc. There are huge hurdles to leap through to get certified, and that's just for the software.

I hate being a wet blanket :(.

OTOH, if they can get capital in the 5-10+ billion range, poach all the best folks from Boeing, Honeywell, Rockwell Collins, GE, UTC, Gulfstream, etc. etc. etc. (which will involve setting up offices in Seattle, Phoenix, Cedar Rapids, Cincinnati, Grand Rapids, Savannah, the UK, France, etc. -- the aviation scene is a bit older, people have families, and are not so keen to move), then they might actually pull it off. Which would be awesome. And I'd totally be in (if they set up an office in my area, that is).

Sigh. BP can afford to run fiber to their oil rigs, but Comcast can't be bothered to lay it through my neighborhood?

Fascinating article, but it was hard for me to get past 'Immune defence'. I'm sure they meant no offence, but it coloured the rest of the text for me. (If it had been on a .co.uk, I'd have left them alone, scout's honour).

Any theories on why China has a notably less dense distribution?

I taught that class once. It does have some prerequisites, but not a ton. I think it could be taken as early as second year, first semester at that particular school.

Windows before 30 years of updates that were created with a need to maintain backwards compatibility feels so cohesive. I mean, that control panel is just so clean.

Also, would be interested in seeing the orginal Win 95 (before OSR2). It was way way faster than win 98.

I didn't personally downvote you, but I'll give you my best guesses:

1) The percentage of die used for decoding is actually already quite low. Per Anandtech 2014, it was at 10% for x86 and decreasing:

http://www.anandtech.com/show/8776/arm-challinging-intel-in-...

Since there is a floor to the number of transistors needed for decoding, there might not be a whole lot to gain there.

2) Progress has already been made towards updating instruction sets to reflect what CPUs do quickly, well etc. It started with MMX back in the stone age, and has progressed through a plethora of SIMD and media acceleration instructions.

3) Instruction sets are already not designed as if they were supposed to be run directly. Quite to the contrary, they are abstracted -- the instruction set is the API, and the microoperations are the instructions. Designing them as if they were to run directly would mean exposing the micro-ops, which would require backwards compatibility breaking changes to the CPU each generation when there were changes to the micro ops.

4) The current system already pretty well levels energy consumption between competing ISAs. See: http://www.extremetech.com/extreme/188396-the-final-isa-show...

You're describing something that sounds like a bytecode VM, but in essence, that's what modern processors already are. Unfortunately, x86 (at least 32 bit) assembly is pretty unpleasant as an 'API', but ARM and PowerPC are both pretty good.

As far as pointer tagging, that's something that's probably mostly limited by memory bandwidth (it's not a particularly compute heavy thing to do), so unless the HW support came in the form of a dedicated on-chip cache, it probably wouldn't get you very much ... and then, if you're going to the expense of adding a dedicated on-chip cache, it's probably going to be more effective as a general purpose cache -- if the code is accessing the pointer metadata often, it will be in cache, and therefore accelerated.

Not trying to rain on your parade, I like seeing creative ideas, and I have no idea why people are downvoting you. Have an upvote on me :).

EDIT: Also, ARM has NEON.

I see. Many of the countries you listed are actually 'workfare' systems, as this article proposes, and not true 'welfare' systems.

Canada - much of the 'welfare' is paid by the province, not the federal government. The social assistance program is aimed at temporary financial need, and requires people to seek employment:

https://en.wikipedia.org/wiki/Social_programs_in_Canada

http://www.mcss.gov.on.ca/en/mcss/programs/social/

UK: https://www.rt.com/news/uk-welfare-jobless-benefits-535/

Japan - Japanese society actually attempts to shame people into going back into the workforce: http://www.japantimes.co.jp/news/2011/09/25/national/media-n...

I don't have time to go down the rest of the list, but my point is that there are few, if any places where somebody can get a 'free ride' while doing little/no work, and certainly no place that's nice to live in (I'm assuming that since you think workfare is a bad idea, you consider 'good' to be the absence of a work requirement).

I'd take a few of countries that you think are good examples, why, and their median PPP household income and/or examples of the life of a typical person. For example, here are some negative examples:

- USSR:

The USSR had a centrally planned economy, with centrallly planned wages designed to limit differentiation between those at the top of the income scale and those at the bottom. A typical family lived in a ~10x10 single room apartment. Often these apartments did not have kitchens, and never 'bedrooms' (I do mean 'single room.). Even though the workers were 'paid', they often could not buy basic things like food, soap, and toilet paper due to chronic underproduction.

https://en.wikipedia.org/wiki/Communal_apartment

https://en.wikipedia.org/wiki/Economy_of_the_Soviet_Union

http://reason.com/blog/2015/04/06/revealed-the-final-stage-o...

http://articles.philly.com/1989-07-21/news/26131831_1_soap-p...

- Cuba

Cuba's economy is dominated by state-run enterprises, with 76% public sector employment. In Cuba, employees in some skilled professions that are not favored by the state, like medical doctors, work in slave-like conditions. GDP per capita is only $10k PPP, less than 1/5th that of the USA. References:

http://www.wsj.com/articles/mary-ogrady-cubas-slave-trade-in...

https://en.wikipedia.org/wiki/Economy_of_Cuba

https://en.wikipedia.org/wiki/Economy_of_the_United_States

- China

In the mid 1990s, China's state-owned economy was teetering on bankruptcy. As an experiment, China began to allow select cities to ignore the old rules on enterprise management, wages, guaranteed employment, etc. By the late 90's, these cities had spawned some of the largest appliance manufacturers in the world, and China began to implement these policies more broadly. Since the reforms, China's GDP has risen tenfold, with productivity increases accounting for nearly half of the rise. During the reform period, PPP per capita incomes have grown at ~6.6% annually.

http://www.economist.com/node/21528262

https://en.wikipedia.org/wiki/Chinese_economic_reform

That is what I would consider two well-documented examples countries with large social apparatuses that resulted in disaster, and one that significantly improved the standard of living for everyone by scrapping their old guarantees. I was hoping that you could provide me with something similar to support your views.

> Yeah, a year or two. That is a much different situation than the poor are in. And this is the problem...you think the situation is the same. So do a lot of other well meaning people. They think that the poor are not making the decisions you would make so they must be stupid or not capable of reason somehow. But the fact is, facing hunger in a year is much, much different than facing hunger next week. Having some sort of built up human capital, be it a degree or work experience or connections or whatever is much different than having none of those things to fall back on. Having connections, be it family or friends, that can help you out in a pinch is much different than everyone you know balancing on the razor's edge too. Missing a payment on your car in a month or two if you lose your job and making uncomfortable decisions is much different than not being able to eat or put gas in your car and leads to much different calculations. It is the myth that we are all in the same boat that allows us to vilify those stupid poor people who can't just do what needs to be done to get themselves out of their situation.

You can make 20k flipping burgers at McDonalds. 20k is enough to build up a safety cushion. I'm a big fan of Dave Ramsey's, I listen to his show, and couples with a family income of ~40k call in literally every day having paid down ALL of their debt and built up a 3-6 month emergency fund (cash!). If they are willing to indebt themselves deeply, they have the same 'year or two' that I'd have if I indebted myself deeply. I have friends who have accomplished this, on 40k/year. Granted, the cost of living in the midwest is fairly low, but wages are also lower than elsewhere (our local McDonalds pays $9.50/hr starting, Wal-Mart is in the same ballpark).

> Perhaps they can, but at what cost?

I know exactly what it feels like. We were not in a good financial situation when we got married. Beans & rice, rice & beans. But we dug ourselves out and bought a modest house. No, we didn't go to movies, on vacations, or out to eat; we didn't have high end smartphones (even 'good' -- mine had a cracked screen), nice cars (we drove rustbuckets), remotely up-to-date PCs, or decent clothes (I wore my clothes to the point that they began to shred). I regularly caulked our brittle siding and tarred our aging roof to keep the water out, bondo'd our cars, and fixed their myriad mechanical problems with shitty, nearly free Harbor Freight tools. So yes, they can, and at a cost that I know and that I've paid myself.

>I've seen people making 100k+ spend so foolishly that they are effectively 'pushing for nothing' (zero or negative net assets).

Me too. But you mistake 'spending on consumption (enjoyment of their time)' with working for nothing. That's not nothing -- they are getting value for their money and time. That's not saving but I am talking about being in a situation where you have to transact with someone or lose everything. Very, very different.

Not different at all. If that fool loses his job, he'll lose every last thing he has, because he doesn't own it and the bank will take it back. Yes, his path to homelessness was more comfortable, but he'll end up in the same place.

> threatening another with imprisonment if they don't remit a significant portion of their income to that class -- are you OK with that?

* Yes, because if they are like your friends above and they truly get themselves into trouble, they'd have that guarantee as well.*

If they don't want that guarantee (I don't), should they have option to opt-out? Personally, I find the idea of forcibly taking from one person to give to another abhorrent. Especially when you will send the person to prison if the resist the taking, and to their grave if they resist the prison. I love charities, I give to charities, and I think they're a great solution. I'm proud to live in an area with extremely high charitable giving (Grand Rapids, MI), and I love to be a part of that. I do not love being forced into it, and if a government wants to implement something like that, it should be done at a more local level (certainly no higher than the state), so that the people effected by the decision have a greater voice in it.

>professionals are only partially paid based on the profit they generate. Partially, in this case, makes all of the difference. > labor demand vs. labor supply, just like everyone else's

Except the poor. Sure their labor is more plentiful and therefore cheaper, but cost of that labor is depressed because __their bargaining position, their best alternative to a negotiated agreement, is negative__. That is a big weight on the scale.

Not sure what you're trying to say here: __ .... __.

> it is actually asking them to work for exactly what their labor is worth Again, it isn't because of the 'or else' chip on the scale. Besides, should we accept jobs in our society that don't pay enough to support the maintenance of the capital it uses? We don't offer to subsidize the cost of ink for a printing press or gas for a logistics company, why should we subsidize the cost of human labor?

No, we should not subsidize the cost of the labor. Exactly my point. Many professionals work 60+ hours/week, and even at crap jobs like McDonalds, 60+ hr/week pays $30k+, which is ample income to support a person.

Once again, I ask for hard statistics and facts, with references. Data, my friend, show me the data.

> Providing a base minimal level of living for each person tends to produce nice societies. What people with money get in return for supporting the deadbeats is definitely worth the, not large, cost.

Hate to do this to you, but if you were on the other side of the issue someone would have asked, and I honestly haven't seen data on it. Do you have any stats/facts/references to back this up? To be blunt, I'm just old enough that I remember societies that tried this, and in their implementations it drove everybody's standard of living toward what we would consider poverty, so I am expressing a bit of doubt here. "They pretend to pay us, and we pretend to work" (old Soviet political joke).

> This forum is full of people that will not starve if they don't work -- at least not for a while. But the vast majority are doing what they can to create value in the world.

Not immediately, but eventually. Also depends greatly on the person. Plenty of people making 100k+ work paycheck to paycheck, and will be in the exact position a poor person would be in (relying on credit, missing payments) immediately after losing their work. Put simply, I'm calling you on this: most people in this forum do work under threat of either losing housing/food or relying on some safety net without a job. I make really good money, and without work I'd probably be screwed in a year or two, assuming I'm willing to borrow heavily first.

> Are they just better human beings than the poor who must be threatened in order to create value? Or have we built enough capital of one type or another to make the transaction of our time and effort and application of our capital worth it? An improvement of our lot?

None of the above. Working for food, shelter, and transportation, just like everyone else.

> If all of us more fortunate creatures were told that we'd never turn a profit for our efforts, how many of us would put in that effort? I would wager not many (a situation that plagued most of those socialist systems you mentioned). The poor are put in a situation where they are asked to work but never profit. They must work to limit their loss. They must work to merely not die.

Once again, calling you on this. Unlike the socialist systems, in our system even the poor can accumulate savings and wealth, albeit at a slower rate. After I graduated school, I lived cheaply enough that I could have accumulated savings on a McDonald's wage, if I had needed to. It's all a matter of personal choices, fiscal discipline, and fiscal education. Maybe we should focus on adding appropriate personal finance education to school curriculum as a way to help people get away from the cycle of debt.

> It is little wonder that we have people that prefer welfare and don't do much to better their position because the alternative is to spend their time and effort pushing for -- nothing.

I've seen people making 100k+ spend so foolishly that they are effectively 'pushing for nothing' (zero or negative net assets). I've seen people making 20k/ea spend so wisely that they can buy a house, a car, have a child, and save. Hardly 'nothing.'

> What I would push for is not Socialism -- which is where everybody owns the means of production which is dragging everyone down into that same pitiable state of never being able harvest your own profit for your own efforts -- but merely giving the poor a guarantee that they will not die or be in pain -- enough food, housing, medical care to live.

Hey, if I could have all of those things guaranteed, even I might consider sitting around exercising my (low-cost) hobbies all day. Not beneficial to society. Also, guaranteeing those things to one class of people means threatening another with imprisonment if they don't remit a significant portion of their income to that class -- are you OK with that?

> Those that need their labor will then have to entice them to work. They have to look at how much their labor is worth to them and try and pay them based on how much profit they can wring from the value they can create -- like lawyers are paid, engineers, management, skilled workers, professionals...literally everyone else.

Trust me, professionals are only partially paid based on the profit they generate. Yes, if they were not generating profit their jobs would be eliminated, just like many lower paid jobs are automated when faced with forced wage increases. However, their salaries are set as a function of labor demand vs. labor supply, just like everyone else's. I teach SW development classes at a local college as a side-job, and I can tell you that people struggle with professional programs vs. other majors. My wife is completing a second degree, an associates in Nursing. She has commented repeatedly how she's had to work way harder to get this nursing degree than her 4-year bachelors, which was not a hard science degree (these days, nursing definitely is a hard science degree). These fields have high pay because the number of people willing and able to get into them is low, and the demand for them is high.

> It's not crazy, it is how 3/4s of the population in the US conducts business. It is just an extension of that strong bargaining position in the free market to those in danger of having to work for less than their time is worth for no hope of bettering their position just to merely not die. It may mean that I have to take it on a bit of faith, but I trust that 1/4 at the bottom will work to better their position just like the rest of us if given the chance. I don't think we're any better, when it comes down to it.

For the reasons stated above, I disagree with the proposition that a lack of a social safety net that doesn't require work as a precondition puts them in a position with "no hope of bettering their position just to merely not die." Asking them to do work to receive benefits (as long as we don't specify what work we're asking them to do) is not asking them to work for less than they're worth, it is actually asking them to work for exactly what their labor is worth, with the added benefit of a bonus above that value being taken from another person and given to them in the form of a social safety net.

Actually, now that I think about it, since they're receiving this benefit, we're asking them to work for more than their labor is worth, and it's actually the person paying for the benefit that we're asking to work for less than they're worth ... and we're asking the person paying for this benefit to do it under the threat of imprisonment if they decide not to pay.

> Personally, I have more faith in a third alternative: a virtual instruction set. Like bytecode, though not managed like Java, and not meant to be interpreted or JIT compiled either. It could run on the bare metal, or be translated into something that is —like the Mill CPU. That way you can keep the illusion of having a single instruction set, while sacrificing virtually nothing. Moreover, future CPUs don't even need backward compatibility, as long as you can translate (and optimise!) the virtual assembly for them.

I don't think he really gets how CPUs work -- it's already that way, and has been for quite a while. The published instruction sets are the 'virtual' instructions, and translation of these instructions (be they ARM, x86, or PowerPC) is baked into each CPU's microcode. We actually have no visibility into the 'real' instructions that CPUs execute ('micro-ops'), because they're proprietary.

If the vast majority were employed, then the proposal to require able-bodied individuals to work as a precondition for receiving welfare should only affect a small minority of people who have chosen to live off the system. Why is that ridiculous?

> Are you willing to create value for other humans at less than your time is worth? Because that is what you are asking them to do. If it was worth their time we wouldn't have to threaten them with pain or death to incentivise them.

Friend, hunger pain and risk of homelessness are what motivate many people to drag themselves to work. Without that threat, many people choose to be unproductive, and it makes society as a whole poor. If you take too heavily from those who are productive to enable this, it dicencentivises them. This has been demonstrated repeatedly -- in the USSR, in China's old system, in Cuba, in the DPRK, in Saudi Arabia (who, with less oil money, are having serious problems), and even in the Internet's beloved Denmark ... literally over and over and over.

(ref for Denmark, since I know that will be one of the more contentious entries on my list: http://www.businessinsider.com/free-universities-and-no-stud... )

Asking them to take what someone else is willing to give them for their work, without the threat of imprisonment from the government to motivate the giving, is not asking them to work for less than they're worth. Quite the contrary, it's asking them to work for an amount that exactly reflects the value other humans assign to their labor.