This question was answered by Armin Ronacher three days ago.[0]
"Why for instance is it not a good idea to write a library in Go? The reason for this is that Go for needs quite a heavy runtime that does garbage collection and provides a scheduler for it's coroutines."
This was a badly chosen example indeed: in many countries lawyers are obliged to continue studying and completing tests throughout the career. In fact, one of Holland's most prominent lawyers got suspended for not obtaining enough study points for an extended period of time.
Anyway, in a sense I agree with the author. To stay sharp programmers need to get out of their comfort zone. There's multiple ways to do that; side projects is one of them. If you don't get out of your comfort zone, you run the risk of becoming like a pilot who forgot how to land or take off, because you have only been flying mid air for so such a long period of time.
I've abandoned purchases as well. When I know I might be missing out on a coupon, I go hunt for it. Most of the time I don't find the coupon I'm looking for, but I do stumble across (coupons for) other products, which might offer a better deal. I end up purchasing those, or not purchasing anything anymore.
Instinctively I agree with the author: when a customer has effectively made a purchase decision, a webshop shouldn't divert him to random pages on the internet, unless there is a mechanism in place that leads to extra sales. Showing a coupon box during one of the final phases of the purchasing process is a way of diverting customers to random pages on the internet.
Compensating for the movement in the way you mention it shouldn't be too hard I guess.
I'm curious if they will need to compensate for the speed differential between the source and the target. Can anyone explain if the following will cause any problems for this technology? If the source and target move away from eachother quickly, the target will observe an increased wavelength, and vice versa. Apparently this is called Redshift [http://en.wikipedia.org/wiki/Redshift] and technically there is some similarity to the Doppler effect.
The Doppler effect (of which redshift is a special case) already has to be taken into account with radio communications (which is EM radiation just like light). In fact, all data from the Huygens Titan lander was almost lost because the software onboard the Cassini probe, on which Huygens piggybacked, couldn't have locked onto the lander's carrier wave due to too steep a Doppler shift. This was fixed by altering the Titan approach trajectory so that the Huygens' relative velocity didn't grow too high.
The Doppler effect will be present, but will not cause any problems that aren't already present. It exists in current radio communications - light is no different in that regard as it's all just electromagnetic waves.
Would that matter if the information weren't encoded using the wavelength? Eg, imagine that they toggle the light very quickly, Morse-code style, to encode messages. If the frequency changes slightly, no big deal.
EDIT: I suppose this effect could make the speed of the toggling appear faster or slower, but that could be adjusted for.
I'm not fond of declaring her mentally ill based on the offenses she has committed. Clearly, she isn't living an ordinary life, but plenty of ordinary people have used drugs, drank alcohol, trespassed and stole during the course of their lives. Prostitution might be a different case, but prostitution ultimately also is a choice. I can even imagine her shoplifting because she wasn't able to obtain the necessary money through different means: nobody would hire her, and whenever she prostitutes she gets arrested. Also, it appears that she doesn't rip out her veins when she isn't incarcerated - otherwise she would be dead by now. Furthermore it is not uncommon for "normal" people who get arrested to be put on suicide watch. The tendency for self affliction perhaps isn't so uncommon for the imprisoned.
I think that if the government would stop disturbing her in her daily life, a Pareto optimum would be reached. Naturally we have the moral obligation to kindly offer help with her problems - rehab, etc. - but shouldn't have the right to force her into anything as long as she isn't a menace to society.
What I'm trying to say is that the system now is turning her into a menace to society. Stop it!
It's sad that in my country, where prisons are currently run by the government, there exist plans for privatization. With privatization there's almost no benefits, and a lot of opportunities for trouble; it the Dutch government feels the urge to repeat the mistakes of others.
The concern isn't the privatization[1], it is the profit motive of the contract does not align with the a just system. If the contract was based on potential cells as opposed to the actual number[2], then bribing to get more profit couldn't happen. The minute any groups profit incentive doesn't match your goal is going to give you problems.
1) do some reading on prison worker unions and you'll see the perversity isn't restricted to private enterprise
2) we will pay you a fixed X for your Y number of cell facility
Not going to disagree with the intent, but that would make it very worth the effort to get as many children who shouldn't be in prison incarcerated since they likely wouldn't be prone to going back. In fact, it might lead to no prison time for hard cases since that would affect he prison's profits.
Absolutely agree with you. It's a general problem with using any performance metric as a sole-focus driver for optimization. You'll find the optimizations may drive behavior beyond where the metric has direct value. That's a truism if you're talking about performance of prison contracts, computer systems, factory throughput, or even capitalism as a whole.
And really, that's what a well-running oversight function should be doing in the case of prison management, adjusting metrics to drive the system toward good whole-society outcomes. Public vs private prisons theoretically doesn't matter, except that introducing private prisons introduces a dynamic where corruption of the oversight function and metrics is used for optimization of profits.
> Public vs private prisons theoretically doesn't matter, except that introducing private prisons introduces a dynamic where corruption of the oversight function and metrics is used for optimization of profits.
Read up on some of the campaigning done by the prison worker's union in California and you will see public has its own profit motives that are just as bad.
"Privatization" where the government contracts to third parties, is not privatization. When the decision is centrally-managed, and the revenue raised by force (taxes), you haven't addressed the central problem. So calling it privatization is misdirection. It's a confusion of terms that is meant to keep people arguing in circles.
Actual privatization removes centralized decision-making. We should be able to opt-out of governments and opt-in to alternative sovereigns. The fact that governments are allowed to claim authority on the basis of geography is why they're still in business, and it's why these messed up situations occur.
How would you describe the central problem you're referring to? I think the main issue here is that "the system" allows for the possibility of an unjust justice system due to commercial interests.
Everybody agrees that what has happened was illegal and should not have happened, and that it is the sovereign's responsibility to create a system in which the risks of such events are minimized, simply because the consequences can be so far stretching. This can be done by ensuring financial independence of judges; eliminating the role of parties who benefit from outcomes that are not necessarily aligned with society's benefits.
I don't yet see how decentralizing decision making reduces this risk.
The central problem is any entity that is allowed to raise revenue by force. Which is the definition of a State.
> the possibility of an unjust justice system due to commercial interests.
There's no such thing as a non-commercial interest. Money is not the only thing that is directly in demand. Power is valuable. Regardless of how the government labels its activities, it has things people want; labeling those things non-commercial is just marketing. Trying to plug the holes with regulations and endless "reforms" is a distraction meant to saturate your time.
> ensuring financial independence of judges; eliminating the role of parties who benefit
"Oversight" is a hand-wavy solution to a never-ending problem. The feedback loop is too long.
> how decentralizing decision making reduces this risk
Competition provides alternatives. Lack of competition reduces incentives for providing great service. The state has no competition, and it raises revenue by force, it's silly to expect a good outcome from that.
I am a proponent of strong government but I agree about your point about this fake privatization which most of the time works the same way as before but with more overhead and corruption. It is just switching a government agency to a company filling exactly the same role in the same way.
If the privatization plan had no incentive for increasing occupancy, then this problem would not occur. Regardless of whether things are government run or privately run the key is oversight by independent bodies that have as little incentive to be biased in one way or the other.
Privatized prisons will always have an incentive to increase occupancy because they will always benefit by having more prisons and providing more jobs.
The problem with having an independent body provide oversight is that in the worst case it's just another level of corruption. Even in the best case they have an incentive to keep the prison system larger because it means more work (aka money) investigating and auditing.
I'm not a big fan of the government, but I think prisons are one thing the government should do itself.
True - you would have to set rules in place that would prevent any kind of incentive for increased occupancy, and then create government bodies that enforce these rules. Also, you would have to set rules in place for many kinds of standards for living conditions for inmates, and set up bodies that enforce those as well.
The thing is that when commercial incentives come into play, companies will go out of their way to find ways to increase profits. Privatization essentially starts a race to the bottom; a race that can only be slowed down by regulatory institutions. In the end, the company always wins and society loses. Banks have showed us that.
What also bothered me was that I'm not sure if the conclusion of the research is correct. At least, from the fact that people who were asked to memorize more, we can't deduce that willpower and cognitive processing draw from the same pool of resources. The conclusion could be correct (and I confess that I haven't read the paper), but there is one other obvious reason that may be possible.
If I do physical exercise, I have an easier time allowing myself to eat some chocolate. If I work a long day, I have an easier time allowing myself to sit down and watch TV for a while. If I solve a difficult puzzle, I have an easier time allowing myself to do something fun.
I like to believe that this is not the result of my lacking willpower after cognitive processing or physical exercise, but of a moral justification that it related to the quid pro quo principle: if I do something good, I have deserved the right to do something bad.
What you described is exactly how lack of will power looks like.
You find rationalization, an argument why it is OK to do something you otherwise would prefer not to do (eat cake, avoid work-out...).
I've always had to pay attention to my weight and the difficult part is persuading yourself not to make an exception no matter how compelling the argument for it looks like (and "deserve" is one of the more difficult ones).
It assumes that cake to all participants is decadent and requires willpower. As a college student in my 20s, no amount of food was off limits and therefore I wouldn't have batted an eye at choosing the cake.
I read the results as more of "I completed a hard task therefore I feel it's appropriate choosing the greater reward."
If that were true, i don't see any reason why people would ever choose the lesser reward, since both were presented in an equal fashion. They chose fruit because they consider it the healthier alternative, not because it's a more appropriate reward. It really is the case that it's all about our finite capacity of willpower, and how we rationalize it away.
Since realizing a few years ago that willpower is finite and must be periodically recharged i've remapped the way i go about things to remove the possibility for negative choices or to make the right choice require less willpower. For example, when i commit source code the server runs a jshint syntax check and prevents my commit if there are issues. I have no choice but to make all my code pass jshint checks. Another example: recently i noticed i spent a lot of time playing a game that i considered a negative use of my time. In a single limited moment of willpower i deleted it and my savegames, so that it would cost a great deal of effort to get back to where i was. The easiest path now is not to play that game. If only i could do the same thing for my internet addiction :)
The problem I have is that 'willpower' is completely subjective and arbitrary. Exerting your willpower is not the same as me exerting mine. Also, your definition of a willpower choice might not be the same to me.
Like I said, in my 20s, food was meaningless. Calories/fat/cholesterol weren't even something I considered. I wasn't overweight due to my high metabolism, so the choice between fruit or cake would be arbitrary. It wouldn't depend on using my brain, it would depend on any number of other factors. To put it simply: if I felt like the fruit, I'd eat the fruit, otherwise cake.
You felt that playing a specific game was a waste of your time and you felt you'd have difficulty in quitting so you made it significantly hard to start again. Someone else might be able to just shut it down, leave it all installed, never go back to it, and, maybe, never even think about it again.
So, unless you know that a person has some reservations about eating cake, it's hard to say whether choosing cake over fruit is a willpower decision. There's an even worse example in the article:
Spend hours at work on a tricky design problem? You’re more likely to stop at Burger King on the drive home.
Once again, it makes the assumption that stopping off at Burger King is somehow taboo to the person and because they blew a ton of cognitive cycles their willpower is blown. And all this completely inferred by a flawed premise (or at least flawed in how it is presented to us in this blog post).
Let me give a reverse example. I chew my nails, sometimes very badly, and this can get to the point of the skin around the nail as well. It can be pretty painful. I've tried just about everything to quit, yet I've been doing this for at least early middle school. I find that if I can keep myself busy enough, either through work or non-stressful activities, I do not chew my nails (or at least it's minimized. My willpower is low when my cognitive functions are in excess - the exact opposite of what this research and accompanying blog assumes. That is one reason I find it to be somewhat questionable.
For me this isn't a lack of willpower, it's creating my own incentivization system so that I can get more accomplished. If I get more work done than usual and then reward myself, this creates a positive connection in my mind. Sometimes it is hard to stay motivated and by having a carrot teasing me on I can achieve more of my goals.
I agree, the conclusion drawn is certainly plausible, but there are other possibilities. Does the use of the brain in somewhat complex tasks deplete blood sugar, in a way which the subconscious mind realizes cake would be a more suitable remedy (fast injection of sugar and/or caffeine)? I think various trials are necessary using different "healthy"/"unhealthy" snacks would be necessary to confirm the author's conclusion. For instance, are the results the same for veggies vs. Doritos?
That said, willpower is certainly limited, and effort must be rewarded or it will become harder and harder to press on. Yes, cognitive tasks can be draining, but they can also be rewarding and invigorating, so it's not as simple as drawing from a limited willpower/cognition well.
And yes, I wholeheartedly agree with her sentiment regarding valuing users' cognitive resources, and appreciated how she expressed it. If everyone had such good intentions and proper perspective, the world would be a far better place. Thanks for sharing yourself with us Kathy.
Same here. I'm going to read this book and hope to learn more about functional programming and the Haskell language, while (hopefully) not being bothered with the kind of trivial examples that you commonly encounter in tutorials.
You can consider this problem in two ways. Either you see the need for increasing the patent lifetime - at least for certain drugs - or you consider this a weakness of a privatized pharmaceutical industry, and take it as an argument in favor of placing the government in charge of pharmaceutical developments.
It is time for software companies to unite. Feds can't just continue roaming around, asking companies for their users' password hashes and other things.
In the current state, some big companies have the means to fight such requests, some big companies are very willing to cooperate, and small companies rarely have the means to go into a legal battle.
Because of the current fragmentation and secrecy surrounding feds' requests with software companies, users do not have the possibility of knowing what they're in for with which company. Also, the divide and conquer tactics used by the Feds really allow them to extract much more information than what would otherwise be the case. Ideally there should be a union for software companies, which makes agreements with the feds concerning their access rights; agreements which then apply to all members of the union.
Currently I have two rules of thumb: 1) for critical services, avoid companies located or significantly involved in the US or UK and 2) at all costs, stay away from Microsoft.
This is more apt advice. At Hacker News we are swimming in a sea of start-ups, who are constantly evangelising the 'cloud' (it's often their bread and butter). But if you genuinely want privacy then keep it local and locked down. Stick to mainstream open source products and keep things as simple as possible.
Well ... What if I launched a cloud based startup that used homomorphic encryption on your data? I mean, it'll be another 10 years before we get the encryption overhead below a factor of a million, but at least we won't be able to give away your data in any useful form...
I've been saying the same thing for a while. Many companies need to form some sort of alliance against government censorship and surveillance, not just in US, but globally. One company alone, even one as big as Google, can't stand up to a government like the Chinese one. 100 big American companies that are vital for their economy, might be able to do it.
Ditto. I'm of the notion that the government can't put the entire company in jail. Could you imagine if Google, Microsoft, and Yahoo were effectively put out of business for these decisions? The repercussions would be devastating. Even placing a lot of key officials from these companies in jail would have lasting effects.
To me this is the prime definition of "too big to fail". It would only require a small percentage of these companies uniting "for the greater good" to produce meaningful results. Not cowing to the NSA is not treason in this instance so I can't even possibly understand why complying with "laws that aren't on any books so are they really laws?" has any positive merit.
Yes, but they won't do it until it hurts their bottom line. If I owned a European cloud business of any kind I would be heavily advertising to the US market right now. When customers start leaving major US internet companies because they no longer feel that they adequately protect their data and privacy, things will change.
IMO the very nature of the centralization of power works against individual rights, one of which is (arguably) privacy. As corporations grow they tend to lose a sense of the customer as a means, and instead choose which type of customer they need in order to maximize profits (or other goals).
Corporations (or any large centralized power base) will optimize for the most exploitable customer or user base, culturing this base if possible. To help broaden a target user base corporations need strong centralized governments more than they need even sizable (but less "culturable") segments of their market base.
Upshot: mature corporations (political parties / religions / etc) will not typically stand up to a centralized government on behalf of a rights-demanding fraction of their market... indeed, typically, they will do the opposite.
And let's face it, it's not like corporate America needs the government's help to abuse you based on your private information. Not in a day and age where you can be denied a job because of your credit history or kicked off your insurance because of your health records. People throw off tons of data, and companies have been working for decades to figure out how to use it to screw you.
Well that sounds great if the large companies fight the good fight. If they don't, you have a very large, unaccountable companies, able to fight the governments to get [lower taxes/lower wages/monopolies]
Do we think that services like Mint are handing over all our financial data to the government (making it easy for them to have a picture of your entire finances)?
If so, are there any viable, offline alternatives?
Never mind Mint, if your bank accounts are in the US your financial data is already available for inspection by the IRS, DHS, and probably many other three letter agencies. I'd wager that Mint, not being a bank, has far less an obligation to hand over your financial data than the banks you have accounts with.
The US already has complete financial surveillance over all US financial activity that isn't a cash trade, and they've been expanding it around the globe aggressively. For decades.
In this day and age where everyone does everything through credit cards, everyone already has all your financial data. Certainly the government does, and the credit card companies hand information out like candy.
For what it's worth, I'm working on a Mint competitor of sorts (that takes advantage of Machine Learning to automatically help you save. It will be based in Australia, not the US, and the basic app will be released as open source for personal self hosting.
I don't think any country is safe at this point. TNO (trust no one) is the only solution. Your cloud provider should have no ability to hand over your data because they can't decrypt it themselves. For example, Lastpass has an architecture where the passwords are encrypted and decrypted on the client, the server never sees anything but pseudorandom noise, and you can audit their browser addon to verify this. You can, with careful design, build many - if not most - cloud services in this way.
That is exactly what we are trying to do. The problem is that is somewhat at odds with machine learning in practice, but I have some ideas in the space.
Just an FYI, a Mint competitor, Wesabe, went out of business some time ago. When they did, they open sourced their software. Not sure of the state it's in. Maybe you can find some good things in their bank interface/scraping code...
After reading rlvesco7's comment, I immediately deleted my Mint account. It always made me uncomfortable to have all my financial data sitting in the cloud. I stopped actively using it a few months ago.
An open source service sounds interesting, but I don't think I'll ever be willing to post all my financial data to a web service again. It would be great to have a locally installed application that could keep track of all those accounts. Having some algorithms run to help me save would be great, but it would take some demonstrated assurances to get me to provide even anonymous data for the machine learning process.
This is interesting. Machine learning requires data. Will the app be sending back up anonymized data that then gets used to help the app make better decisions? Or how else will you make your app smart like Mint? Can't wait to see what you're doing.
The former is what we are doing, but coupling it with some basic statistical financial methods (and some stuff I've come up with myself!) that you can rely on if you'd rather not send the information. That's also what the cheapest plan for our hosted version relies on.
The entire premise is personal finance software that learns your habits to make it easier to use :)
The government having access to your financial data is a prerequisite for a functioning tax system. If you are audited, the IRS has the right to look inside your bank accounts.
The policies (or location) of online budgeting tools are entirely irrelevant. Hiding financial data from the government involves well established trades dating to long before the internet (or PRISM): money laundering and tax evasion.
Check out GNUCash. I've been using it for about a year. Entering all your stuff is tedious, but it's open source and integrates with some banks (also import from Quicken and CSV).
I could never sign up for Mint. I see the value in it, but providing a private business with a view into all my financial accounts just seems like a huge mistake.
Other than banks, creditors, and credit-score companies? I imagine there are far more businesses with access to view our financial accounts than we realize day-to-day.
It's not just that, but they use information in your accounts to recommend other companies' products. Today I got a notification that I was paying more than average for car insurance, with a link to a competing product.
text files, paper statements and your file cabinet. There are various open source check register and book keeping packages. GnuCash is complete but possibly overkill for some people.
What part of "privacy is dead" do people not understand? You think multi-core processors in your iPhone or android is to make the calls more clear? How many people here can say they haven't integrated a "smart" Apple or Google phone into their lifestyle? Someone with more HN love should post that poll.
Everyone needs to reconsider their worldview and a few important definitions they hold. One of those is privacy.
My definition of privacy: anything I relay to ANY one person is no longer private. What's the old saying about three people keeping a secret? Information wants to be free and privacy is not its natural state. It's always been this way, but the physical barriers to diffusion have been completely decimated in the past two decades.
This is not a mere blip in a long term trend, it is fundamental, IMO.
That being said, I believe there are new values we can all embrace to make the most of the state of the human experience today. Perhaps someone should can a thread on Internet values for the 21st century and beyond.
PS - Anyone ever wonder how MSFT got an anti-trust pass in the US, but not in the EU?
"Why for instance is it not a good idea to write a library in Go? The reason for this is that Go for needs quite a heavy runtime that does garbage collection and provides a scheduler for it's coroutines."
[0] http://lucumr.pocoo.org/2013/8/18/beautiful-native-libraries...