> The effect in Java is like every pointer (or object reference or whatever you want to call > them) were shared_ptr (except better).
Except that the GC implies that there is no destructors, and you never know whether finalizers are called, and if they do, when. That alone barred Java from the neat RAII technique that C++ programmers are so accustomed to.
Re:Can someone please explain to me...
on
GPLv2 Vs. GPLv3
·
· Score: 1
The problem is that people in FSF believe (and I think the belief is correct) that most people writing GPL code does not think that this is their intention.
Re:Gah GPLv3 is total bullshit
on
GPLv2 Vs. GPLv3
·
· Score: 1
Why did you think that there had been nobody crazy enough to try challenging GPL? See
Re:Detect Modified Software?
on
GPLv2 Vs. GPLv3
·
· Score: 1
Easy: have a ROM with hard-coded public key that it will use. Boot from the ROM. Immediately read the entire flash RAM with its signature. Verify it using the public key in the ROM, and continue if and only if it checks correctly. With a public key system, nothing in your setup reveals the private key. Unless the user can install another ROM, he is stuck with whatever signed by the vendor's private key, which only the vendor possess. There are quite a few things in the world doing similar things (see FON).
> i wonder, could you sue someone for writing bad code for a project based on > a GPLv3 because they were unable to explain how it actually works? i mean, > if you wrote the code bad enough (but still had it somehow work) you'd have > a DRM scheme more powerful than AACS...
Then the program writers are depending of the assumption that they have better ability than the collective ability of everybody who tries to read the source code. In my understanding, this might succeed only if you are writing a niche system that there are few who are in any way interested in it.
Did you actually read GPLv3? It quite explicitly requires that everybody distribute the software will not just also give out the source code when asked, but also tells how a modification of the software can be installed to the hardware that the software was distributed in (which means it is not legal to use a digital signature to prevent running modified binaries). It does not just prevent one to use "legal means", but one of the "technical means" to achieve DRM. Well... you can do DRM with GPLv3 code, but expect people to modify it and edit it out, install it to his own hardware to make sure it works, and post a patch to the world to tell how to do it in their hardware as well. And according to the license, you must provide a way to do all these. Of course, as long as the software distributed is not the one doing DRM, it doesn't matter, so it is still possible to do DRM (say, in hardware), but not in GPLv3 protected code.
> So I was saying that for small to medium scale projects pretending that there is no centralized server, just > people's repository is stupid.
I don't think everybody has to use every single feature of every tool or that is "stupid". If you are in a "small" project that you are the only user of it, you can of course have a single repository in your own hard disk. If you are in a project that you are the only developer of it, you can of course have a single repository in the server of your company and have every user checkout from there. It doesn't make use of the distributed feature, but that give you the flexibility to grow to a point where there are multiple developers and there is a QA team who want to have their own repositories.
> For large scale projects it can work, like it does for Linux, but then you have a > dedicated core team that is necessary in judging what goes in and what stays out.
If you want to "simulate" a CVS or Subversion repository, you can always have a "merge robot" in the repository that everybody uses, which tries to pull the change anybody send to it, and if successful, it is committed. It doesn't work in practice. Not because you cannot write the code, but because there are a lot of crap that people are sending in. The problem you mentioned is not a problem of the repository. It is the problem that every big enough project cannot "simply trust" the code that any programmer sends in.
> It doesn't matter if you call it > people x's tree or commit right to the central repository. That is the same thing. The terminology Linus uses is > annoying because it lies. Not Linus, but the terminology.
If it is not for that ability of the individual, perhaps junior, programmers to have write access to their own repository without affect the others and thus can save their partial changes, and if it is not for that ability of the individual programmers to safely merge code from multiple branches at will without worrying that some code in these branches are actually themselves merged from the same branch and thus must not be merged twice again, etc, then you're right. But because of those abilities, I do think that they are perfectly entitled to say "it is different" and "it is distributed, others are centralized".
When every of us have 1, 2 or 4 core/CPUs doing all the tasks that need to be done in the computer from the kernel to the desktop environment to the virus scanner to the servers to the BT clients to the video player to the real application, who need parallel programming? Give every of us 1024 core/CPUs, and every CPU-bound program will be parallel. But that's not a day that everybody see as about to come.
> China and India will both be producing 5-10 times more emissions than they are today, > and today China and India are producing almost as much as the US. They aren't covered > by this agreement at all.
I don't know where you get the figures making you think they are emitting as much as US, which doesn't seem very plausible for me. But even if this is true, it doesn't mean that they will be "producing 5-10 times more emissions than they are today". Instead, their economy is much smaller than those in US at this point of time, and if they still make the same amount of emissions, that means they have far more capacity than US to *reduce* their emission. All it takes is to export them the low emission technology. The target is not to make money over it, instead it is to make sure emission is low. Then it doesn't matter for the export. If no company in the US or Europe will trade with any company having a large amount of emission, then the companies in China or India will be forced to use much lower emission equipment even if they don't sign any treaty, since US and Europe are the biggest trade partners with China and India. But the prerequisite is that US and Europe must themselves be energy efficient, must be able to pass laws to make such trading impossible, and have low emission equipments to export!
I think so. All you do is to modify the intensity of the 4 sub-pixels to have carefully chosen relative intensity, so that the first one is of intensity 1 in each of its levels, the second intensity 64, the third intensity 64*64, and the last intensity 64*64*64. Of course, you will need to calibrate each of the pixels very exactly, so that, e.g., the 100th level of the brightest pixel of intensity 100*64*64*64, rather than 100*64*64*64+1. You also have to mix the sub-pixels good enough not to create really bad artifacts. You also have to make sure that you create bright enough pixels so that quantum effects do not run in. I wonder whether anybody in the industry will try such a thing, perhaps you rock every of us by doing so.
> Time to learn binary my friend. A 6 bit binary number can only represent the > values 0 through 63.
Time to learn counting my friend. Counting from 0 through 63 in steps of 0.25 (because each pixel is created by 4 sub-pixels), you get 0, 0.25, 0.5, 0.75,..., 62.75, 63; which are 253 different levels of intensity.
I still don't quite understand why the hell somebody would like to use scripts in their office documents, especially if they are so capable. Documents are things people send and read without even thinking about, with no expectation that there can be any harm. Why program must be attached to those, rather than some extensions that is independent of the document, which people can only install when they can be expected to know the possible harm?
March 2007 issue of Scientific American has an article "Diesels come Clean", suggesting that a number of models of SUVs are being manufactured to changed that (with improved Diesel and engine design). So things might change in the (not so distant) future.
My experience is just the opposite: most articles of Wikipedia is quite readable, at least in the introduction. There are knowledges that not everybody can receive, so of course there are parts of many articles that are beyond many people. But my expectation is that most of the time, those people has no business for those parts anyway.
Let's look at what TFA points out (disclaimer: my biology studies ends on secondary 5, equivalent to grade 11 in US systems). For epigenetics, it says:
Epigenetics is the study of changes in gene regulation that occur without a change in DNA sequence (genotype). When a cell undergoes such an epigenetic change, it is the phenotype of the cell which is affected.
Epigenetic events during embryo development lead to differentiation of the fetal cells. The process fetal development with the differentiation of cells and organs is called epigenesis.
In biology, while the subject of genetics focuses on how organisms can inherit traits by inheriting genes from their parent(s), which encode information for cell function as sequences of DNA, epigenetics is sometimes used to refer to additional methods of biological inheritance that do not directly relate to the inheritance of collections of genes, or soft inheritance.
Huh? What's the problem? The single first paragraph effectively says everything that his cited alternative tells except the example. I don't know what is phenotype, but now I learn that it is something about "gene characteristics other than DNA". The second sentence is hard for me, but clearly it is about some mechanisms that give rise to the behaviour, so I blame myself for not knowing enough biology rather than the article. The third paragraph supplements the first that sometimes things outside the gene can be considered epigenetics as well. I guess a normal person who has any training of technical reading should not have problem getting this far.
Now let's turn to the next, Fluid mechanics. Wiki has it this way:
Fluid mechanics is the subdiscipline of continuum mechanics that studies fluids, that is, liquids and gases. It can be further subdivided into fluid statics, the study of fluids at rest, and fluid dynamics, the study of fluids in motion. Modern applications use the computational approach to develop solutions to fluid mechanics problems; the discipline concerned with this is the CFD, Computational Fluid Dynamics.
Exactly one term I don't quite understand, "continuum mechanics". If I just ignore this little problem of terms, the remainder doesn't seem to have a problem to understand. That term is a link, so I click on it, getting this at the beginning:
Continuum mechanics is a branch of physics (specifically mechanics) that deals with continuous matter, including both solids and fluids (i.e., liquids and gases).
The fact that matter is made of atoms and that it commonly has some sort of heterogeneous microstructure is ignored in the simplifying approximation that physical quantities, such as energy and momentum, can be handled in the infinitesimal limit.
Okay, intuitive and I need nothing more. Understanding completed. My guess is that if somebody has problem doing just that (click on a link to find out more about something he doesn't completely understand), he should be having problem surfing web at all.
Now the last one, "Mitochondrial DNA". Its entry looks like this:
Mitochondrial DNA (mtDNA) is DNA that is located in mitochondria. This is in contrast to most DNA of eukaryotic organisms, which is found in the nucleus. Nuclear and mtDNA are thought to be of separate evolutionary origin, with the mtDNA being derived from bacteria that were engulfed by early precursors of eukaryotic cells....
I find it quite difficult to understand. But then, why I'd like to know anything about "Mitochondrial DNS" if I know nothing about "mitochondria"? It seems to be that clicking on that link will give me some
What a billiant example of Slashdotters who don't know the topics he post on. Please point your browser at the following and read the first two paragraph:
What do you think about it? 0.9? It's less than 1, meaning at least 1 in 10 couples never give birth to any children in their whole life span, the number is likely much more since there are couples giving birth to more than one! According to your theory, now Hong Kong should start criminalizing all porns, so that the people has to "do" it rather than "watch" it, and accidentally give birth to children? (According to recent survey, around 40% of people in Hong Kong who did give birth to children *regret* about it!)
> One problem here that I see, and which I can't see any commenting on, is that this > is Hong Kong.
But this is the actual problem. It is in HK. Hong Kong is probably the most conservative cities among all modern cities concerning public expression of any content slightly remotely related to sex, with the only exception being academic studies. It has nothing to do with returning to China, its people are very conservative. In fact, I consider it even more conservative than mainland China.
Ironically, at the same time it leads to proliferation of many magazines, supposingly with general audience, using sexually explicit titles or at times photos in their cover page to attract attention.
> why would he have to quit? Just don't do it. What is the worst that can happen? they fire you?
I'd agree, and would add: "and do as good as reasonably possible for every (other) legal and reasonable thing they ask you to do". There is a cost associated to your company firing anybody, including you. Add it to the value that you give to the company, it won't do so just because you refuse to do one of the many things they want you to do. It will instead resolve their problem in another way, whether it find somebody else to do it, or it follows your suggestion to do something else. Things do not always go to the legal department.
On the other hand, if most of the things you're asked to do involve illegal work, you should quit. Not just for moral reasons, but also because there is real danger that your company collapses in front of you, and you want to exit the company before that happens.
Hm... you seem to think assembly is more important than it really is.
> You can't write a compiler
More exactly, you can't write a complete compiler that compiles to assembly language (although you can write the parser and tokenizer). Nowadays few does that, more write a compiler to something else (e.g., to byte-code, to XSLT, etc), and lack of knowledge have little effect for those.
> You can't debug C/C++ programs
C/C++ compilers are good enough these days that I no longer think it is relevant. As long as you can use a good source debugger (gdb is enough), you should know everything in your virtual memory that is relevant to solving your problem. (And they don't normally show you assembly, either!)
> You don't really know why buffer overflows are bad > You don't really understand memory management and what the heap and stack really are > You won't ever understand garbage collection
I think this has to do with memory model, rather than assembly language. It is not difficult to abstract the memory model without touching assembly language at all, especially in language like C/C++ that you can cast a pointer to an integer and print that out in hex. On the other hand, yes, most likely you cannot "exploit" buffer overflow.
> You don't really know why threads are different than processes
I believe that even this is actually about memory model, although the "memory" here must now include cache memory and registers.
> You can't write a device driver
My belief is that if somebody tells you that he cannot write a device driver because he doesn't know assembly, he is just lying because he is lazy. Most of the device drivers in the Linux kernel has no line of code that is assembly.
> You don't know any computer architecture at any depth that matters > You don't know how your CPU works > You won't think the movies with "hacking" in them are as funny as the rest of us do.
These are mostly true, though. My belief is that you need to use assembly if you are doing one of the following activities:
- Understanding, modifying or writing the code generator of a compiler that targets to machine code.
- Understanding, modifying or writing a virus or other similar code that must be attached to such machine code.
- Understanding, modifying or writing operating system routines or library routines that deals directly with registers, e.g., to do process creation, context switching or mutual exclusion.
- Understanding, modifying or writing performance critical library routines that need to use special machine instructions (e.g., MMX).
- Designing or implementing the instruction set of a CPU. (No, I don't think you really need to know assembly to implement a hardware device to be attached to a bus.)
In other cases, you learn assembly mainly for two reasons: curiosity ("why the hell can human beings design machines from such silly transistors that are so reliable and efficient!"), and unwillingness to accept that something is out of your capability because you don't know assembly. Of course, programmers with these two traits are more likely to be good programmers, so if you are hiring you will choose one who know assembly language. =)
My first feeling is that if children are not "exercising", they are "playing" something else, usually spending just as much energy, perhaps in other parts of the body. Very few parents will allow their children to just "sleep all day" instead, which will actually save energy.
> Maybe it's just me, but I think that putting all your eggs in one basket is not smart.
I don't know whether it is just you, but I'm sure I'm not with you.
First, the eggs are "quantum encumbered", breaking one in a basket is equivalent to breaking the corresponding one in another basket. Most of the time, if you configure multiple ways store keys to access something sensitive (say, a server), the same access required in multiple ways, so you end up having the key (or keys of equivalent functionalities) in multiple places, breaking any of them allows attacker to access what you considered sensitive.
Second, the baskets are quantum encumbered as well, in such a way that if one breaks, the probability that the other will break as well is higher. This is because in an open-source world, code get shared.
Third, you have about the same amount of time to care the baskets no matter how many baskets you want to manage. No matter whether you have a single key ring application or 3, you only want to update your computer once per day, you only have those few hundred (say) of security experts capable of looking for security holes, etc.
So I'll instead take the mantra "keep all your eggs in the same basket, and look after it carefully" instead of "keep your eggs in many different baskets".
> you have a nice start, but it has some weird consequences.
> 1. Sending email gets infuriating as your machine slows to a crawl anytime > someone hasnt whitelisted you.
This really does not need to be slow. There are many "trapdoor functions" for you to build your challenge response. Given the answer (that you pick randomly) you can easily generate the question (and therefore can easily check whether it really answer the question), but given the question it takes an arbitrarily long time to compute the answer. The server only need to run the "fast" routes, but the client need to run the "slow" route.
> 2. Maintaining a Taint Free Whitelist gets to be a bit tricky.
Most people only have so many people contacting him, so building a white-list is not as horrible as it sounds. Not to say that this is particularly interesting for support staff of companies, though, who constantly need to respond to emails from complete strangers. And you can build software that if a challenge-response is answered correctly, and, after reading the mail, you do not click on the delete button before you do other actions (e.g., to archive it or to place it into some folder), the sender is added to the white-list automatically.
> 3. How is this going to work for services like Gmail and Yahoo? A minute of > chug time on a machine is expensive if your offering it for free. If you > whitelist them it doesnt do much good because then spammers just use those > accounts
That's for Gmail and Yahoo to think about. If the market is full of solutions that have negligible spam problem, they will be creative to think about solutions that they can adopt as well. The current problem is that the whole market is filled with solutions that has "features" (like unauthenticated and deniability) that some people somehow like while at the same time encourage spam.
> 4. How does this work for people in poor areas of the world using some antique > machine (like a Pentium 200 mmx) where email would take 30 minutes a peice to > send.
Hm... if you mean that those computers need that amount of time to just send e-mails, then those systems are not usable today already, no need to cater for them. If you mean that those computers need that amount of time to answer a challenge-response, then the answer is simple: just have that person to talk to the one he want to contact in some other means, like a phone call, so that the address is put into a white-list.
Now my take for adding one more question:
5. This doesn't solve the problem when your friend's mail account is broken (by virus or trojan horse), and subverted into a spam sending center.
When is your 99% number coming from? By the way, TSP doesn't restrict you to "travel in straight lines", the distance measure is arbitrary. Indeed, if you restrict to problems in the euclidean space (i.e., distances are always measured in straight line distance), you can get an approximation scheme, i.e., you can approximate as close as you want. But not for the general TSP problem.
If the smokers do not have the right to smoke there anyway, what is the use of that silly right of the property owner? To tell the smokers that "I'm not the one forbidding you, the government is."? By the way, here the government is supposed to be the one who actually stop a violator, not the property owner, so the property owner do not lose the right to "allow" their guests to smoke, only that other guests can complain and have the government prosecuting the violator. (However, I know places where this is not true, the property owner will lose their license to be a dining place by not stopping smokers, Australia being one example.)
> Private establishments now, believe it or not, even in New Orleans (last bastion of sin and > freewill), will not allow you to smoke in private establishments if serving food is their > primary form of business....places that are primarily restaurants.
Things gone the same way in HK now, the ban happened a bit more than one month ago. Every non-smoker (and some smokers!) applauded.
> I say this is WRONG. If a proprietor wants to allow smoking in his place, then those that > prefer not to smoke...can either deal with it, or take their hard earned dollars elsewhere.
Unluckily, before that ban, most restaurants allowed smoking. We could choose to dine elsewhere, but then we would have to give up our choice for good food. And how many people we entertained by allowing people to smoke? In HK, around 12% of the population are smokers. Are you meaning that because 12% of the population want to be able to choose whether to smoke, the other 88% of the population must not be able to choose the food they consider to be good if they don't accept getting harmed by second hand smoke? I don't quite buy that argument.
> For reference, I'm a recently reformed smoker myself...and not having any smokers around IS > easier for me...but, I'd rather have others have the choice than my having less temptation.
For years, the government in HK are in the same line of reasoning as you suggested. People should choose not to smoke, smoking should be reduced by education rather than by regulation. But the actual facts are that the government is not the only one trying to convince people. The cigarette manufacturers are also trying to do the same (well, the opposite). The net effect was that whether the number of smokers increase or decrease in a year depends on which of the two sides are doing better.
Does the society want such a game? At times, "freedom" and "collective benefits" need to have some sort of balance. If "freedom" for 12% of the population (or some manufacturers, for sake of argument) has to make the whole society pay big for the medical expenses that the "freedom" implies, and if that 12% of the population are actually harming themselves when they exercise their "freedom", and if many of that 12% of people are hoping and waiting for a day when that "freedom" is "deprived" from them so that they have a better reason to stop being addicted to that "freedom", and if the remaining 88% of the people will have to pay for the "freedom" of that 12% by not just their money but their health as well, the law makers should know which side of the balance the law should put us at.
> You can still smoke in bars and casinos, and the street down here....but, how much longer > till they try to regulate that? Smoking is still a legal activity...if they want to ban > smoking, then try to make cigarettes illegal? Why not protect everyone? See? That just > doesn't make sense either....at least not to me.
I still believe that banning cigarettes altogether is the simplest and most effective solution, and at the same time it is ethical. If drug usage is illegal in HK, why smoking should be legal? Answer: some big manufacturers are behind the scene, they will lose big if the law go that way, so they tried all possible ways to stop that, including to put some of their key person into the law making machinery. And some country will lose big as well, so they put heavy political pressure in that machinery at the same time. That is the real meaning of "freedom"?
On TV, many people can enjoy viewing ads. On the web, few does. I think it is more than just the "it's annoying" reason.
The first thing that come to mind is the relevancy. In TV, you won't be bombarded by a stream of ads for services that is available "in US and Canadian only" if you are viewing from London. In the web, you definitely will, most of the time.
Then comes the trustworthiness issue. If you see a TV ad that gives misleading information, or spread its news in an inappropriate way, you complain to your local authority, and that ad get removed. If you see a web ad that does the same, there is nothing that you can do. After a while you get to assume all web ads have the same standard.
No less important is the lack of innovations in web ads. Most are plain boring. When watching TV, even if you are not interested in a product, you will continue to view an ad selling it, just because it is amusing, because you see your movie star there, or something else. When viewing the web and find an ad, you are simply not interested most of the time.
Then comes the intrusion issue. In TV, the designer of the programmes know when will be the slot for ads, and tries to fit their programme so that the time of an ad is exactly the time for the next excitement, or a time to rest, etc. People will not be too much annoyed by such an ad. In the web, ads are placed at a place that cause you trouble reading the actual contents you want, either by reducing the space available for displaying your information, or sit at the front of the page so that it is shown first rather than your desired content, or even keep moving "in front of" the content you need, obscuring them. No wonder people won't want watching such ads.
Let's hope that the web advertisers continue to have no way to deal with all these. If they learn that to maximize their revenue it is best to make viewers like the ads rather than trying (unsuccessfully) to force users to view them, our web peace will suffer severely.
Slashdot Mirror
> The effect in Java is like every pointer (or object reference or whatever you want to call
> them) were shared_ptr (except better).
Except that the GC implies that there is no destructors, and you never know whether finalizers are called, and if they do, when. That alone barred Java from the neat RAII technique that C++ programmers are so accustomed to.
The problem is that people in FSF believe (and I think the belief is correct) that most people writing GPL code does not think that this is their intention.
Why did you think that there had been nobody crazy enough to try challenging GPL? See
l
http://www.linuxdevices.com/news/NS8564956607.htm
Easy: have a ROM with hard-coded public key that it will use. Boot from the ROM. Immediately read the entire flash RAM with its signature. Verify it using the public key in the ROM, and continue if and only if it checks correctly. With a public key system, nothing in your setup reveals the private key. Unless the user can install another ROM, he is stuck with whatever signed by the vendor's private key, which only the vendor possess. There are quite a few things in the world doing similar things (see FON).
> i wonder, could you sue someone for writing bad code for a project based on
> a GPLv3 because they were unable to explain how it actually works? i mean,
> if you wrote the code bad enough (but still had it somehow work) you'd have
> a DRM scheme more powerful than AACS...
Then the program writers are depending of the assumption that they have better ability than the collective ability of everybody who tries to read the source code. In my understanding, this might succeed only if you are writing a niche system that there are few who are in any way interested in it.
Did you actually read GPLv3? It quite explicitly requires that everybody distribute the software will not just also give out the source code when asked, but also tells how a modification of the software can be installed to the hardware that the software was distributed in (which means it is not legal to use a digital signature to prevent running modified binaries). It does not just prevent one to use "legal means", but one of the "technical means" to achieve DRM. Well... you can do DRM with GPLv3 code, but expect people to modify it and edit it out, install it to his own hardware to make sure it works, and post a patch to the world to tell how to do it in their hardware as well. And according to the license, you must provide a way to do all these. Of course, as long as the software distributed is not the one doing DRM, it doesn't matter, so it is still possible to do DRM (say, in hardware), but not in GPLv3 protected code.
> So I was saying that for small to medium scale projects pretending that there is no centralized server, just
> people's repository is stupid.
I don't think everybody has to use every single feature of every tool or that is "stupid". If you are in a "small" project that you are the only user of it, you can of course have a single repository in your own hard disk. If you are in a project that you are the only developer of it, you can of course have a single repository in the server of your company and have every user checkout from there. It doesn't make use of the distributed feature, but that give you the flexibility to grow to a point where there are multiple developers and there is a QA team who want to have their own repositories.
> For large scale projects it can work, like it does for Linux, but then you have a
> dedicated core team that is necessary in judging what goes in and what stays out.
If you want to "simulate" a CVS or Subversion repository, you can always have a "merge robot" in the repository that everybody uses, which tries to pull the change anybody send to it, and if successful, it is committed. It doesn't work in practice. Not because you cannot write the code, but because there are a lot of crap that people are sending in. The problem you mentioned is not a problem of the repository. It is the problem that every big enough project cannot "simply trust" the code that any programmer sends in.
> It doesn't matter if you call it
> people x's tree or commit right to the central repository. That is the same thing. The terminology Linus uses is
> annoying because it lies. Not Linus, but the terminology.
If it is not for that ability of the individual, perhaps junior, programmers to have write access to their own repository without affect the others and thus can save their partial changes, and if it is not for that ability of the individual programmers to safely merge code from multiple branches at will without worrying that some code in these branches are actually themselves merged from the same branch and thus must not be merged twice again, etc, then you're right. But because of those abilities, I do think that they are perfectly entitled to say "it is different" and "it is distributed, others are centralized".
When every of us have 1, 2 or 4 core/CPUs doing all the tasks that need to be done in the computer from the kernel to the desktop environment to the virus scanner to the servers to the BT clients to the video player to the real application, who need parallel programming? Give every of us 1024 core/CPUs, and every CPU-bound program will be parallel. But that's not a day that everybody see as about to come.
> China and India will both be producing 5-10 times more emissions than they are today,
> and today China and India are producing almost as much as the US. They aren't covered
> by this agreement at all.
I don't know where you get the figures making you think they are emitting as much as US, which doesn't seem very plausible for me. But even if this is true, it doesn't mean that they will be "producing 5-10 times more emissions than they are today". Instead, their economy is much smaller than those in US at this point of time, and if they still make the same amount of emissions, that means they have far more capacity than US to *reduce* their emission. All it takes is to export them the low emission technology. The target is not to make money over it, instead it is to make sure emission is low. Then it doesn't matter for the export. If no company in the US or Europe will trade with any company having a large amount of emission, then the companies in China or India will be forced to use much lower emission equipment even if they don't sign any treaty, since US and Europe are the biggest trade partners with China and India. But the prerequisite is that US and Europe must themselves be energy efficient, must be able to pass laws to make such trading impossible, and have low emission equipments to export!
I think so. All you do is to modify the intensity of the 4 sub-pixels to have carefully chosen relative intensity, so that the first one is of intensity 1 in each of its levels, the second intensity 64, the third intensity 64*64, and the last intensity 64*64*64. Of course, you will need to calibrate each of the pixels very exactly, so that, e.g., the 100th level of the brightest pixel of intensity 100*64*64*64, rather than 100*64*64*64+1. You also have to mix the sub-pixels good enough not to create really bad artifacts. You also have to make sure that you create bright enough pixels so that quantum effects do not run in. I wonder whether anybody in the industry will try such a thing, perhaps you rock every of us by doing so.
> Time to learn binary my friend. A 6 bit binary number can only represent the
..., 62.75, 63; which are 253 different levels of intensity.
> values 0 through 63.
Time to learn counting my friend. Counting from 0 through 63 in steps of 0.25 (because each pixel is created by 4 sub-pixels), you get 0, 0.25, 0.5, 0.75,
I still don't quite understand why the hell somebody would like to use scripts in their office documents, especially if they are so capable. Documents are things people send and read without even thinking about, with no expectation that there can be any harm. Why program must be attached to those, rather than some extensions that is independent of the document, which people can only install when they can be expected to know the possible harm?
March 2007 issue of Scientific American has an article "Diesels come Clean", suggesting that a number of models of SUVs are being manufactured to changed that (with improved Diesel and engine design). So things might change in the (not so distant) future.
Let's look at what TFA points out (disclaimer: my biology studies ends on secondary 5, equivalent to grade 11 in US systems). For epigenetics, it says:
Epigenetics is the study of changes in gene regulation that occur without a change in DNA sequence (genotype). When a cell undergoes such an epigenetic change, it is the phenotype of the cell which is affected.
Epigenetic events during embryo development lead to differentiation of the fetal cells. The process fetal development with the differentiation of cells and organs is called epigenesis.
In biology, while the subject of genetics focuses on how organisms can inherit traits by inheriting genes from their parent(s), which encode information for cell function as sequences of DNA, epigenetics is sometimes used to refer to additional methods of biological inheritance that do not directly relate to the inheritance of collections of genes, or soft inheritance.
Huh? What's the problem? The single first paragraph effectively says everything that his cited alternative tells except the example. I don't know what is phenotype, but now I learn that it is something about "gene characteristics other than DNA". The second sentence is hard for me, but clearly it is about some mechanisms that give rise to the behaviour, so I blame myself for not knowing enough biology rather than the article. The third paragraph supplements the first that sometimes things outside the gene can be considered epigenetics as well. I guess a normal person who has any training of technical reading should not have problem getting this far.
Now let's turn to the next, Fluid mechanics. Wiki has it this way:
Fluid mechanics is the subdiscipline of continuum mechanics that studies fluids, that is, liquids and gases. It can be further subdivided into fluid statics, the study of fluids at rest, and fluid dynamics, the study of fluids in motion. Modern applications use the computational approach to develop solutions to fluid mechanics problems; the discipline concerned with this is the CFD, Computational Fluid Dynamics.
Exactly one term I don't quite understand, "continuum mechanics". If I just ignore this little problem of terms, the remainder doesn't seem to have a problem to understand. That term is a link, so I click on it, getting this at the beginning:
Continuum mechanics is a branch of physics (specifically mechanics) that deals with continuous matter, including both solids and fluids (i.e., liquids and gases).
The fact that matter is made of atoms and that it commonly has some sort of heterogeneous microstructure is ignored in the simplifying approximation that physical quantities, such as energy and momentum, can be handled in the infinitesimal limit.
Okay, intuitive and I need nothing more. Understanding completed. My guess is that if somebody has problem doing just that (click on a link to find out more about something he doesn't completely understand), he should be having problem surfing web at all.
Now the last one, "Mitochondrial DNA". Its entry looks like this:
Mitochondrial DNA (mtDNA) is DNA that is located in mitochondria. This is in contrast to most DNA of eukaryotic organisms, which is found in the nucleus. Nuclear and mtDNA are thought to be of separate evolutionary origin, with the mtDNA being derived from bacteria that were engulfed by early precursors of eukaryotic cells. ...
I find it quite difficult to understand. But then, why I'd like to know anything about "Mitochondrial DNS" if I know nothing about "mitochondria"? It seems to be that clicking on that link will give me some
What a billiant example of Slashdotters who don't know the topics he post on. Please point your browser at the following and read the first two paragraph:
_ Kong
http://en.wikipedia.org/wiki/Demographics_of_Hong
What do you think about it? 0.9? It's less than 1, meaning at least 1 in 10 couples never give birth to any children in their whole life span, the number is likely much more since there are couples giving birth to more than one! According to your theory, now Hong Kong should start criminalizing all porns, so that the people has to "do" it rather than "watch" it, and accidentally give birth to children? (According to recent survey, around 40% of people in Hong Kong who did give birth to children *regret* about it!)
> One problem here that I see, and which I can't see any commenting on, is that this
> is Hong Kong.
But this is the actual problem. It is in HK. Hong Kong is probably the most conservative cities among all modern cities concerning public expression of any content slightly remotely related to sex, with the only exception being academic studies. It has nothing to do with returning to China, its people are very conservative. In fact, I consider it even more conservative than mainland China.
Ironically, at the same time it leads to proliferation of many magazines, supposingly with general audience, using sexually explicit titles or at times photos in their cover page to attract attention.
> why would he have to quit? Just don't do it. What is the worst that can happen? they fire you?
I'd agree, and would add: "and do as good as reasonably possible for every (other) legal and reasonable thing they ask you to do". There is a cost associated to your company firing anybody, including you. Add it to the value that you give to the company, it won't do so just because you refuse to do one of the many things they want you to do. It will instead resolve their problem in another way, whether it find somebody else to do it, or it follows your suggestion to do something else. Things do not always go to the legal department.
On the other hand, if most of the things you're asked to do involve illegal work, you should quit. Not just for moral reasons, but also because there is real danger that your company collapses in front of you, and you want to exit the company before that happens.
Hm... you seem to think assembly is more important than it really is.
> You can't write a compiler
More exactly, you can't write a complete compiler that compiles to assembly language (although you can write the parser and tokenizer). Nowadays few does that, more write a compiler to something else (e.g., to byte-code, to XSLT, etc), and lack of knowledge have little effect for those.
> You can't debug C/C++ programs
C/C++ compilers are good enough these days that I no longer think it is relevant. As long as you can use a good source debugger (gdb is enough), you should know everything in your virtual memory that is relevant to solving your problem. (And they don't normally show you assembly, either!)
> You don't really know why buffer overflows are bad
> You don't really understand memory management and what the heap and stack really are
> You won't ever understand garbage collection
I think this has to do with memory model, rather than assembly language. It is not difficult to abstract the memory model without touching assembly language at all, especially in language like C/C++ that you can cast a pointer to an integer and print that out in hex. On the other hand, yes, most likely you cannot "exploit" buffer overflow.
> You don't really know why threads are different than processes
I believe that even this is actually about memory model, although the "memory" here must now include cache memory and registers.
> You can't write a device driver
My belief is that if somebody tells you that he cannot write a device driver because he doesn't know assembly, he is just lying because he is lazy. Most of the device drivers in the Linux kernel has no line of code that is assembly.
> You don't know any computer architecture at any depth that matters
> You don't know how your CPU works
> You won't think the movies with "hacking" in them are as funny as the rest of us do.
These are mostly true, though. My belief is that you need to use assembly if you are doing one of the following activities:
- Understanding, modifying or writing the code generator of a compiler that targets to machine code.
- Understanding, modifying or writing a virus or other similar code that must be attached to such machine code.
- Understanding, modifying or writing operating system routines or library routines that deals directly with registers, e.g., to do process creation, context switching or mutual exclusion.
- Understanding, modifying or writing performance critical library routines that need to use special machine instructions (e.g., MMX).
- Designing or implementing the instruction set of a CPU. (No, I don't think you really need to know assembly to implement a hardware device to be attached to a bus.)
In other cases, you learn assembly mainly for two reasons: curiosity ("why the hell can human beings design machines from such silly transistors that are so reliable and efficient!"), and unwillingness to accept that something is out of your capability because you don't know assembly. Of course, programmers with these two traits are more likely to be good programmers, so if you are hiring you will choose one who know assembly language. =)
My first feeling is that if children are not "exercising", they are "playing" something else, usually spending just as much energy, perhaps in other parts of the body. Very few parents will allow their children to just "sleep all day" instead, which will actually save energy.
> Maybe it's just me, but I think that putting all your eggs in one basket is not smart.
I don't know whether it is just you, but I'm sure I'm not with you.
First, the eggs are "quantum encumbered", breaking one in a basket is equivalent to breaking the corresponding one in another basket. Most of the time, if you configure multiple ways store keys to access something sensitive (say, a server), the same access required in multiple ways, so you end up having the key (or keys of equivalent functionalities) in multiple places, breaking any of them allows attacker to access what you considered sensitive.
Second, the baskets are quantum encumbered as well, in such a way that if one breaks, the probability that the other will break as well is higher. This is because in an open-source world, code get shared.
Third, you have about the same amount of time to care the baskets no matter how many baskets you want to manage. No matter whether you have a single key ring application or 3, you only want to update your computer once per day, you only have those few hundred (say) of security experts capable of looking for security holes, etc.
So I'll instead take the mantra "keep all your eggs in the same basket, and look after it carefully" instead of "keep your eggs in many different baskets".
> you have a nice start, but it has some weird consequences.
> 1. Sending email gets infuriating as your machine slows to a crawl anytime
> someone hasnt whitelisted you.
This really does not need to be slow. There are many "trapdoor functions" for you to build your challenge response. Given the answer (that you pick randomly) you can easily generate the question (and therefore can easily check whether it really answer the question), but given the question it takes an arbitrarily long time to compute the answer. The server only need to run the "fast" routes, but the client need to run the "slow" route.
> 2. Maintaining a Taint Free Whitelist gets to be a bit tricky.
Most people only have so many people contacting him, so building a white-list is not as horrible as it sounds. Not to say that this is particularly interesting for support staff of companies, though, who constantly need to respond to emails from complete strangers. And you can build software that if a challenge-response is answered correctly, and, after reading the mail, you do not click on the delete button before you do other actions (e.g., to archive it or to place it into some folder), the sender is added to the white-list automatically.
> 3. How is this going to work for services like Gmail and Yahoo? A minute of
> chug time on a machine is expensive if your offering it for free. If you
> whitelist them it doesnt do much good because then spammers just use those
> accounts
That's for Gmail and Yahoo to think about. If the market is full of solutions that have negligible spam problem, they will be creative to think about solutions that they can adopt as well. The current problem is that the whole market is filled with solutions that has "features" (like unauthenticated and deniability) that some people somehow like while at the same time encourage spam.
> 4. How does this work for people in poor areas of the world using some antique
> machine (like a Pentium 200 mmx) where email would take 30 minutes a peice to
> send.
Hm... if you mean that those computers need that amount of time to just send e-mails, then those systems are not usable today already, no need to cater for them. If you mean that those computers need that amount of time to answer a challenge-response, then the answer is simple: just have that person to talk to the one he want to contact in some other means, like a phone call, so that the address is put into a white-list.
Now my take for adding one more question:
5. This doesn't solve the problem when your friend's mail account is broken (by virus or trojan horse), and subverted into a spam sending center.
When is your 99% number coming from? By the way, TSP doesn't restrict you to "travel in straight lines", the distance measure is arbitrary. Indeed, if you restrict to problems in the euclidean space (i.e., distances are always measured in straight line distance), you can get an approximation scheme, i.e., you can approximate as close as you want. But not for the general TSP problem.
If the smokers do not have the right to smoke there anyway, what is the use of that silly right of the property owner? To tell the smokers that "I'm not the one forbidding you, the government is."? By the way, here the government is supposed to be the one who actually stop a violator, not the property owner, so the property owner do not lose the right to "allow" their guests to smoke, only that other guests can complain and have the government prosecuting the violator. (However, I know places where this is not true, the property owner will lose their license to be a dining place by not stopping smokers, Australia being one example.)
> Private establishments now, believe it or not, even in New Orleans (last bastion of sin and
> freewill), will not allow you to smoke in private establishments if serving food is their
> primary form of business....places that are primarily restaurants.
Things gone the same way in HK now, the ban happened a bit more than one month ago. Every non-smoker (and some smokers!) applauded.
> I say this is WRONG. If a proprietor wants to allow smoking in his place, then those that
> prefer not to smoke...can either deal with it, or take their hard earned dollars elsewhere.
Unluckily, before that ban, most restaurants allowed smoking. We could choose to dine elsewhere, but then we would have to give up our choice for good food. And how many people we entertained by allowing people to smoke? In HK, around 12% of the population are smokers. Are you meaning that because 12% of the population want to be able to choose whether to smoke, the other 88% of the population must not be able to choose the food they consider to be good if they don't accept getting harmed by second hand smoke? I don't quite buy that argument.
> For reference, I'm a recently reformed smoker myself...and not having any smokers around IS
> easier for me...but, I'd rather have others have the choice than my having less temptation.
For years, the government in HK are in the same line of reasoning as you suggested. People should choose not to smoke, smoking should be reduced by education rather than by regulation. But the actual facts are that the government is not the only one trying to convince people. The cigarette manufacturers are also trying to do the same (well, the opposite). The net effect was that whether the number of smokers increase or decrease in a year depends on which of the two sides are doing better.
Does the society want such a game? At times, "freedom" and "collective benefits" need to have some sort of balance. If "freedom" for 12% of the population (or some manufacturers, for sake of argument) has to make the whole society pay big for the medical expenses that the "freedom" implies, and if that 12% of the population are actually harming themselves when they exercise their "freedom", and if many of that 12% of people are hoping and waiting for a day when that "freedom" is "deprived" from them so that they have a better reason to stop being addicted to that "freedom", and if the remaining 88% of the people will have to pay for the "freedom" of that 12% by not just their money but their health as well, the law makers should know which side of the balance the law should put us at.
> You can still smoke in bars and casinos, and the street down here....but, how much longer
> till they try to regulate that? Smoking is still a legal activity...if they want to ban
> smoking, then try to make cigarettes illegal? Why not protect everyone? See? That just
> doesn't make sense either....at least not to me.
I still believe that banning cigarettes altogether is the simplest and most effective solution, and at the same time it is ethical. If drug usage is illegal in HK, why smoking should be legal? Answer: some big manufacturers are behind the scene, they will lose big if the law go that way, so they tried all possible ways to stop that, including to put some of their key person into the law making machinery. And some country will lose big as well, so they put heavy political pressure in that machinery at the same time. That is the real meaning of "freedom"?
On TV, many people can enjoy viewing ads. On the web, few does. I think it is more than just the "it's annoying" reason.
The first thing that come to mind is the relevancy. In TV, you won't be bombarded by a stream of ads for services that is available "in US and Canadian only" if you are viewing from London. In the web, you definitely will, most of the time.
Then comes the trustworthiness issue. If you see a TV ad that gives misleading information, or spread its news in an inappropriate way, you complain to your local authority, and that ad get removed. If you see a web ad that does the same, there is nothing that you can do. After a while you get to assume all web ads have the same standard.
No less important is the lack of innovations in web ads. Most are plain boring. When watching TV, even if you are not interested in a product, you will continue to view an ad selling it, just because it is amusing, because you see your movie star there, or something else. When viewing the web and find an ad, you are simply not interested most of the time.
Then comes the intrusion issue. In TV, the designer of the programmes know when will be the slot for ads, and tries to fit their programme so that the time of an ad is exactly the time for the next excitement, or a time to rest, etc. People will not be too much annoyed by such an ad. In the web, ads are placed at a place that cause you trouble reading the actual contents you want, either by reducing the space available for displaying your information, or sit at the front of the page so that it is shown first rather than your desired content, or even keep moving "in front of" the content you need, obscuring them. No wonder people won't want watching such ads.
Let's hope that the web advertisers continue to have no way to deal with all these. If they learn that to maximize their revenue it is best to make viewers like the ads rather than trying (unsuccessfully) to force users to view them, our web peace will suffer severely.