Err... Perhaps you should advise your relative to suggest to HR to hire some competent admins. They can effectively seal the computers they use in their work from access from the outside world. All it takes is a protocol to communicate the results to the outside. Put a firewall in front of all those computers (preferably one that does NAT and protocol analysis), close every single port and protocol, then open only the one that passes the data through.
Then they wouldn't even need to patch those computers.
They could justify the expense of keeping these machines for data analysis only (i.e. as extra computers) by simply calculating the costs they incur with every worm/virus/etc cleanup.
I find this hard to believe. Did you read the article ? Didn't you just confuse them with self-checkouts, which have, indeed, been available for a while now ?
There's a difference: the scanner is mounted on the shopping cart, so you scan the products as you put them in the cart.
That contrasts with the self-checkouts I believe you're talking about, which are regular checkout points, where you take your cart and scan the products yourself, while being supervised by some employee (the gain being that one employee can supervise 4-8 scanners).
Technically speaking, self-checkouts don't really help with reducing the length of the lines, because regular people scan slower than experienced employees. However, they do reduce the number of point of sale employees, which saves money to the store.
Also, they are a blessing for the average male who has to pick up tampons for his girlfriend;)
Most admin tasks on windows can be done via Telnet. Hell, as you well know you can run BASH and SSH if you feel like it. A "good" admin can deal with as many Unix systems as Windows systems. </B>
While it is true that it is possible to do some administration of windows via command line, the tools available on windows are rather limited.
<B> We're talking about servers here. In a well designed domain no one has the rights to the server systems required to infect them with anything. </B>
<B>You</B>'re talking about servers here. Desktops run Linux just as well as "servers" you know... Heck, what's the difference between a server and a desktop, anyway ? Think a bit about it, and you'll see it's a rather blurry line. All are computers, all can run the same software if needed. The only truly important thing is that these computers will allow their users to communicate with each other. In my broad definition, to communicate would include the process of active communication (obviously), but also all other processes necessary to determining the content of the communication (e.g. ways of processing the data to be discussed).
Since most standards used in the Linux world are open, Linux offers more flexibility, especially when seen in a long-term perspective.
<B> Right. Because all Linux admins are realy high level coders that can debug kernel conflicts in their spare time </B>
Kernel conflicts ? Are you trying to be somewhat specific and you're not very successful at it ? Or are you just ignorant and like to use big words you don't understand ?
There's more to Linux than just the kernel. Heck, it would be wonderful if we only had to deal with the bugs in the kernel - then proving that Linux is technically superior to anything else would be a breeze.
Just like there's more to Windows than its kernel. Did you know that Windows also has a kernel ?
Since you brought this up: how many applications does the XP SP2 break (as in, applications that change their behaviour or completely stop working after installing SP2) ? Here's a link, in case you were curious. Note several big names, including several anti-virus programs and ZoneAlarm.
<url:http://support.microsoft.com/default.aspx?k bi d=884130&product=windowsxpsp2>
The number is common knowledge in US, although I doubt many know its meaning.
It actually means a concentration of 0.08% of alcohol in the blood stream. In some countries this is scaled to permille, which, of course would be 0.8 0/00 which, equates, in international units with 0.8 ml/l. As you can see, the measurement units cancel each other and the value itself is a number.
The reason this number is so popular in US is that in most states this concentration is the threshold above which people are considered legally drunk.
In general, for a 90 kg male, it would take approximately 5 US beers (5 * 330ml = 1.65l) drank within one hour to achieve this alcohol concentration.
Look it up on the net for more details - they're readily available.
Then again, perhaps it's just that I know how to pick a good web browser and set it up so that it runs all Javascript functions only on sites I explicitely allow it to... All others are confined to not running a bunch of functions (e.g. openWindow) unless I actually click on the link.
While this is true, there are a few things to be said about the Athlon 64.
I have a 3400+ in a HP Pavilion zv5000, running a 64-bit SuSE Linux. So far, I am very pleased with how well everything runs: fast, reliable. I primarily use it for my own applications, which are quite demanding. On the 32-bit front I can do pretty much everything I please, including running 32-bit windows programs (Warcraft III, for instance) under wine.
It's a beautiful machine for those who need both the compact design of a laptop, but quite a bit of processing power.
On the 32-bit front it feels great: it actually seems to perform a little better than a Pentium 4 @ 2.6 GHz w/ HT. Admittedly, this is my personal feeling about how Warcraft III plays using wine, so feel free not to consider this remark.
However, it's drains the battery quite quickly. At full power, a 12-cell battery lasts for about 2.5 hours. With aggressive power saving (processor running at 800Mhz instead of 2.4GHz), hard disk spinning down almost as soon as it finished reading, little wireless card usage, I managed to bring it to about 5 hours, which is not great. It still feels sufficiently fast, the most annoying incident being the delay until the hdd spins up.
All in all, it's a great architecture, but it has its problems, of which the main I would say is the power consumption.
And simulated annealing is just as good as genetic algorithms. Instead of setting the mutation rate and crossover probability, you have to set the probability to accept a worse solution that takes you out of the local minimum, and you have to define a neighbouring function (which can be a bitch!! - more so than defining the crossover).
Nothing wrong with running it off line, then using the found parameters. However, there may be major drawbacks: a possibly good solution for a particular load, might turn out to be unusually bad for a slightly different load. Given that there seems to be a need for genetic algorithms to search for the solution, it follows that the search space is sufficiently complex and not all that "well behaved". This means that the solutions will be rather unstable. In this case, perhaps it's better to accept a solution that's moderately bad for all tasks, instead of picking one that's excellent for a rare configuration and quite bad for most.
Moreover, the theory that genetic algorithms are easy is a myth perpetuated by those with shallow understanding of the concept. Sure, the concept and the code are simple. However, setting the parameters for genetic algorithms is a bitch and their meaning is not entirely related to the problem at hand.
So, all that this guy managed (with virtually no improvement) was to get rid of some understandable parameters coded in the original scheduler, and replace them with the mutation rate, crossover probability and population size, which don't have any meaning in terms of process scheduling.
And, no, kernel hacking does not imply any knowledge of statistics or statistical experiment design.
Your point is short sighted. How many times do you actually have to run the optimiser to adapt the parameters to the current load ? If you had to do so one hundred times per minute (not unlikely on a web server, for instance, where processes constantly start and end), then the overhead from the GA would become quite substantial.
A method that takes a microsecond to run once, will take a millennium to finish if ran sufficiently many times.
From his email address, the guy works at IBM, perhaps in some research department. I would suspect this was published because the project needed more money and they had to show they produced something. Along the lines of "our results are inconclusive. Please give us more money to continue".
As the recent trends show it, in both academia and industry, it's better to publish crap than not to publish at all. It is actually commonly said "publish or perish".
So, the work of this guy is marginal at best. It un-necessarily complicates the code in the scheduler while the gains are minimal.
Even your 5-10% would be useless. They'd be easily compensated for by a slightly faster CPU.
All in all, this seems to follow a trend in OS research nowadays: there are preciously few who have good ideas, and the rest believe they are improving things by throwing in something exotic (e.g. genetic algorithms) they don't even understand, hoping that everyone will be impressed.
It's a shame when a story is posted as it was published on cnet. This should be a site for nerds and stuff that matters. Some editing would have been necessary, as the cnet story is neither clear nor terribly informative.
Although perhaps it's silly to with that the moderators spent a little more time doing some very basic research (yes, that could be "googling") to find more accurate information. Otherwise, why do they exist ?
Here are two more interesting links, with some more detail about the ideas of the KDE team (don't expect very many details, as it's just vapourware):
I think there are a couple of interesting things to do out there. Your indecision seems te me to be the problem. I would suggest you to pick something that previously caught your attention and try to add a touch of AI in it. It is not of the uttermost importance to be something that will change the course of the world. Remember - you are only one person, and setting a huge goal may end up by doing nothing. A long journey begins with a simple step (no guys, this is not a chinese proverb - it's just common sense). I can't tell you what to pick - I have no idea what your interests are. But I can tell you that I found a very rewarding "landscape" in the field of computer vision. I had to make such a project myself when I graduated, and I wrote an OCR But as I said - it doesn't really matter to write something huge - it is more important that you learn from it. And, yes, I'd like though to make a suggestion somewhat related to the previous answers. The paperclip in Word has some good AI behind it. Don't flame me for this - I usually don't put the words Microsoft and good in the same sentence. But in this case it's not the AI that's bad - it's the idea of having a paperclip dancing while I am doing some work. I mean, why the hell is that thing stretching on my screen while I am trying to do some work ? Conclusion : do something, don't set a goal higher than you could possibly reach (for now), and don't make it annoying.
In my opinion this is the best story since that scary law for controlling the web pages anonimity passed in France. This only proves the Murphy's law : never put the blame on bad intent when you can blame stupidity.
I guess this could be rephrased a bit like this : Shoot yourself in the leg - perhaps the enemy won't shoot you. Although I can imagine a scene where all the north korean generals are maneouvering playstations:).
A sure path to disaster
on
RMS On eBooks
·
· Score: 2
I suppose noone will argue the point that books are carriers of ideas, items used to share experiences (even when it comes to fantasy stories), sources of enllightment and education. Since Guttenberg invented his primitive press, humans have been attracted by the books, by the information inside. Reading is a privilege. Of the one that knows to read, of the one that has the means to buy books and so on. About a hundred years ago, many educated people were proud for having read about 10 books. Nowadays, because books are affordable we read 10 of them before we go to school. This is progress. Still, there are many among us that are too poor to afford spending their money on books, instead on their supper. And for years, there existed for them a way to have access to this information : libraries. And God only knows how many brilliant scientists were poor beyond belief in their youth. Their only chance was that they had access to a library Nowadays, in France, the Louvre Museum opens its gates for everyone (withouy any cost) on the first Sunday of every month. Are they losing money ? Probably. But in exchange they allow the entrance to poor students that are interested in art but have no means to pay the entrance fee. Why are they doing this, and not allow only students to enter for free ? (something like academic prices). Because having janitors to check for id's would be even more expensive. Granting access to everyone is, then, something desirable not only from the visitor's point of view
Why not do the same here, in US ? Because apparently companies cannot see beyond the immediate gain. Because they don't really care, and because sometimes (too often unfortunately) they are stupid. Because the costs are probably even more impressive (although this is something genuinely american : exagerrated governmental costs). Because companies are taking the bureaucratic model from the government, no matter the cost. It easier to impose the existing regulations (even if they don't really fit) and pay more money, instead of thinking about new ones who may turn to be better.
And I can't stop wondering how many books have read in the late 10 years the ones that are trying to impose these regulations.
Anyway, imposing such draconian regulations would have as direct effect a drastical reduction of the number of readers in the poor side of the population. By cutting the access to culture to someone, one is denying a fundamental right. The right to be informed, the right become aware and to be able to make good decisions. Then again, isn't a stupid and blind population the ideal for a government ?
I confess : i haven't played with either Napster or Gnutella. Mostly because of lack of time, still I haven't used them So what am I doing here in the first you might ask. I'm not going to try to explain why the companies are so pissed about it. It's pretty much obvious and I think that one way ore another they won't succeed. It also happens that I think they are stupidly unsatiable. And this is what is going to bring them down. I won't even comment what Metallica did : after all they aren't what they used to be (I liked them, I still do but not that much) - no flames please. So, if I ignore all these things, my post will be offtopic, right ? Well, perhaps : but I have something else to say that could be somewhat related. What is Gnutella anyway, but a distributed filesystem ? Ok, with some restrictions (you are not allowed to modify the file on the remote computer). Right now, it is very difficult to implement robust filesystems (and the existing ones have their flaws). What if we could use the code of gnutella, napster etc. to create a robust NFS, that could insure data persistency in the same way gnutella does (quote from their site : f, say, New York was hit by Dr. Evil, GnutellaNet probably wouldn't even notice. So you lose your "Gnutella friends" in New York. Big deal. It's unlikely that New York is the only place where people share the recipe for strawberry-rhubarb pie. ) Think about it. A toy could become something really important.
I guess the post here it way too enthusiastic, since even in the original article they never mentioned some basic facts. Read it well : it's only theory. And there are a few questions to be asked : What about the stability of the memory ? What are the operating conditions ? If it has to be kept below, say 200K it will be quite difficult to use it, wouldn't it ? What would be the needs for producing such a memory ? Because it's quite hard to find pure silicon in nature (and producing itis quite expensive I imagine) And last but not least : how about compatibility with the actual technology ? I guess none of you is actually imagining that this will be joyfully embraced by companies that are strong on the memory/processors market. This may actually end up being buried by marketing, because nowadays an university doesn't really have the possibility (financially) of designing a competitive chip/architecture. So I guess I will be a bit reluctant towards its success until I see it on the commercial ads of AMD or Intel.
I couldn't say I wasn't impressed about this. Especially because I know the hype 20 years ago when the first hdd appeared. It had about 5Mb and everyone was wondering who could ever fill this huge space. Yet, let's not be too enthusiastic because for the moment this technology has some possible problems (at least in what concerns the military - which seem to be the first interested in it). Being a magnetic storage device a magnetic field can destabilize the data on it, thus making the computer unusable (or unreliable). Hypothetically speaking, it would be quite easy to attack a ship with some sort of radiation that will make it vulnerable to another attack that may destroy it On the other hand, civilian users are quite protected against this, since there aren't many important secrets to be destroyed (I'm not speaking about corporate users). Then again, who will ever use 400Gb (or more) at home ? (ok I may sound silly and repeating the sentence I mentioned in the beginning). There are some limits in one's ability to gather information and one of them is the time. Although, if I were to speak frankly, I imagine Windows 2010 taking 70% of this space (with a 25Gb Solitaire). Hopefully there will be no windows any more in 2010
Has anyone noticed a weird thing ? they say that the compressed file will have 5.5 M (compared to 15M the current version). What does this mean ? Are they stripping a lot of functionalities out of it ? have they given up some more or less useless components (e.g the radio part) ? Will it be linked to some libraries that come with the OS ? If so, which OS ? Windows perhaps ? Linux is not mentioned at all... Will they support it ? (most likely, though, since they mention cross-platform compatibility). As far as I am concerned, I hope that smaller size means less bugs (although we have the perfect example of M$-DOS 1.0- some 300 bugs in 5000 lines of code - great job breeding them guys). Anyways, I surely hope it will have less bugs than the current version.
Welcome to the real world. It's quite amusing to say these words when speaking about wirtual reality, but let's start arguing a bit on the consequences and, of course, on the causes. the internet is becoming more and more popular. Everyone knows about it (at least in US and Europe), everyone is excited about it. Do you think the government is going to sit on its comfortable chair and do nothing about it ? When it could make the chair even more comfortable ? Be serious. Of course they will impose regulations. Why ? Because they have the power to do it. Furthermore, the main internet providers are in US. Which gives the US govt the possibility of gaining a real control over an international system. Impressive, isn't it ? Does the word manipulation mean anything to you ? And it's even cheaper than all the more or less visible agencies. Not to mention military actions. Is this moral ? Probably not. Are we going to gain something from this ? Yeah, right !
However, this could also bring some benefits. If the rules are fair we won't have to deal with situations like M$. Remember that they became what they are especially because of the lack of regulations (and the silliness of IBM). I'm not very sure how this will end but it will probably be something like the telephony networks today. Perhaps a bit more complex, but not too much. And I guess this isn't that bad after all...
I've got my share of bad teachers during my school days (which aren't finished yet). Let's face it : there will always be good teachers, bad teachers and "bad and arrogant" teachers. This last category is the worst. I've seen teachers that didn't want to give the maximum grade because they considered that was destined only to them, thus lowering your GPA. I saw teachers that gave you an incomplete even if you did well on the exam just for the fun of having you retaking that exam and spend the summer learning. Some of them made sarcastic comments regarding a wrong answer from their students (like "if you can prove this to me I'll give you a PhD").
At one point or another we all experienced this. It doesn't make it right, but this is the way it is. Does this mean we should simply ignore the problem ? Definitely no. Yet, I read some of the comments of the students, and I think they definitely deserve their teacher. Ok, he's a lousy teacher, but this doesn't give you any right to curse him like that. Besides, if one doesn't understand that cursing someone hardly make him one's friend, then he should avoid expressing himself in public. I suppose the reaction of the teacher is even understandable in this case. Although I can't see why he is sueing the web site. After all, they are a simple messenger. Destroying the site won't change the opinion of his students. On the contrary.
Then again, from my experience I could also tell that although the goal of the site is meritory, it probably won't do too much good. The best professors I've had until now were very special human beings and during the years they earned my respect without being too concerned about this. And this includes my respect both as teachers and as persons. On the other hand, I found that the main characteristic of the bad teachers was that they never tried to meet the requirements of the students, they never tried to "teach". They were exposing the material and that was it. The students always were a "hurd" with whom they had to spend their precious time.
And since the bad teachers don't care about the opinions of their students anyway, why bother giving those grades ?
Well, it's been known for years the legend that princess Anastasia of Russia survived the slaughter of her family in 1916. I am not very sure, but I don't think it's been proven that there aren't any survivors of the Russian royal family Imagine how milions of fans of Elvis would react at the news that their idol is not dead... Or even Jim Morrison... Or...
Or perhaps it's better to leave things as they are... If they're dead, let them rest in peace - it would only be a proof of respect.
OK, suppose it would fork. I don't think this is such a big of a problem. After all we do have several distributions among which some are not that "standard". We have three different flavours of packages, yet nothing stops me from installing any of them on my machine. And if I really don't want to install packages I can always get the source. The issue is not if the Linux breaks into several incompatible distributions. Unless they are incompatible in the Microsoft sense of the word. Meaning that those distributions will not be able to share data between them. It doesn't really matter if that distribution is running a binary of StarOffice and the other is running another type of binary of the same package. As long as StarOffice for Linux1 and StarOffice for Linux2 will share the document formats then there is no problem. After all, how many of you have several phones at home ? Are they all the same ? Probably not, yet you can speak from one to any other.
In a way you're right. However, in life it's not always necessary to know everything, but to know where to efficiently search for your information. I agree, there should be some tests where the students are not allowed to have access to the internet (yes, general history is a good example), but there also should be just as many where they are allowed to do it.
If this is true (I don't see any reason why it shouldn't be) then it's very cute. I can't say it is new (after all, there are take-home exams), it's just that the scenario is slightly different. What it is even more interesting is that in this way foreign or disabled students are allowed access to information in their native language or something that would compensate their disabilities. Doesn't this make much more sense than the LEGO based test presented on this site a couple of months ago ? (sorry I don't have a link to that article).
Besisdes, if we can afford to give every student a laptop imagine how much less paper will be consumed. Respecting the environment is one of our priorities right now (or at least should be) and this may be a first step. The next one is to convince enterprises to give up un printing that much.
I hated the hum! of the article poster. On one hand we want Linux to reach the desktop, on the other we are reluctant to easy installation procedures. I am sorry, but this is bigotry. Besides, why complain since Yast2 comes along with yast1 (yes, text mode, full of features installation software) ? Yast2 was written for those that don't know exactly what Linux is, who don't know what hardware is inside their machines and so on. Or for those that like inserting the cd, writing a few lines and going to smoke a cigarette while the system is installing. Of course, an experienced user won't use this tool (although I can tell it does look cute) because it doesn't offer you the complete control over the installation.
As for myself, I would have liked to see the KDE 2 with the new release of Suse. I know, it won't probably be included until 6.6 - but still, one can hope, can't he ? Right now it's only a simple update of 6.3 (great distribution!). So, if you have enough bandwidth I think it's worth downloading (or even better take advantage of the update from ftp site feature). Wish (read it especially if you work for VA) : please mirror the distribution on your site sooner. ftp.suse.com works only at 50k/s which is a real pain when downloading 6 cd's .
Slashdot Mirror
Err... Perhaps you should advise your relative to suggest to HR to hire some competent admins. They can effectively seal the computers they use in their work from access from the outside world. All it takes is a protocol to communicate the results to the outside. Put a firewall in front of all those computers (preferably one that does NAT and protocol analysis), close every single port and protocol, then open only the one that passes the data through.
Then they wouldn't even need to patch those computers.
They could justify the expense of keeping these machines for data analysis only (i.e. as extra computers) by simply calculating the costs they incur with every worm/virus/etc cleanup.
I find this hard to believe. Did you read the article ? Didn't you just confuse them with self-checkouts, which have, indeed, been available for a while now ?
;)
There's a difference: the scanner is mounted on the shopping cart, so you scan the products as you put them in the cart.
That contrasts with the self-checkouts I believe you're talking about, which are regular checkout points, where you take your cart and scan the products yourself, while being supervised by some employee (the gain being that one employee can supervise 4-8 scanners).
Technically speaking, self-checkouts don't really help with reducing the length of the lines, because regular people scan slower than experienced employees. However, they do reduce the number of point of sale employees, which saves money to the store.
Also, they are a blessing for the average male who has to pick up tampons for his girlfriend
Most admin tasks on windows can be done via Telnet. Hell, as you well know you can run BASH and SSH if you feel like it. A "good" admin can deal with as many Unix systems as Windows systems.
k bi d=884130&product=windowsxpsp2>
</B>
While it is true that it is possible to do some administration of windows via command line, the tools available on windows are rather limited.
<B>
We're talking about servers here. In a well designed domain no one has the rights to the server systems required to infect them with anything.
</B>
<B>You</B>'re talking about servers here. Desktops run Linux just as well as "servers" you know... Heck, what's the difference between a server and a desktop, anyway ? Think a bit about it, and you'll see it's a rather blurry line. All are computers, all can run the same software if needed. The only truly important thing is that these computers will allow their users to communicate with each other. In my broad definition, to communicate would include the process of active communication (obviously), but also all other processes necessary to determining the content of the communication (e.g. ways of processing the data to be discussed).
Since most standards used in the Linux world are open, Linux offers more flexibility, especially when seen in a long-term perspective.
<B>
Right. Because all Linux admins are realy high level coders that can debug kernel conflicts in their spare time
</B>
Kernel conflicts ? Are you trying to be somewhat specific and you're not very successful at it ? Or are you just ignorant and like to use big words you don't understand ?
There's more to Linux than just the kernel. Heck, it would be wonderful if we only had to deal with the bugs in the kernel - then proving that Linux is technically superior to anything else would be a breeze.
Just like there's more to Windows than its kernel. Did you know that Windows also has a kernel ?
Since you brought this up: how many applications does the XP SP2 break (as in, applications that change their behaviour or completely stop working after installing SP2) ?
Here's a link, in case you were curious. Note several big names, including several anti-virus programs and ZoneAlarm.
<url:http://support.microsoft.com/default.aspx?
The number is common knowledge in US, although I doubt many know its meaning.
It actually means a concentration of 0.08% of alcohol in the blood stream. In some countries this is scaled to permille, which, of course would be 0.8 0/00 which, equates, in international units with 0.8 ml/l. As you can see, the measurement units cancel each other and the value itself is a number.
The reason this number is so popular in US is that in most states this concentration is the threshold above which people are considered legally drunk.
In general, for a 90 kg male, it would take approximately 5 US beers (5 * 330ml = 1.65l) drank within one hour to achieve this alcohol concentration.
Look it up on the net for more details - they're readily available.
Er... Java script ? Which Java script ?
Then again, perhaps it's just that I know how to pick a good web browser and set it up so that it runs all Javascript functions only on sites I explicitely allow it to... All others are confined to not running a bunch of functions (e.g. openWindow) unless I actually click on the link.
While this is true, there are a few things to be said about the Athlon 64.
I have a 3400+ in a HP Pavilion zv5000, running a 64-bit SuSE Linux. So far, I am very pleased with how well everything runs: fast, reliable. I primarily use it for my own applications, which are quite demanding. On the 32-bit front I can do pretty much everything I please, including running 32-bit windows programs (Warcraft III, for instance) under wine.
It's a beautiful machine for those who need both the compact design of a laptop, but quite a bit of processing power.
On the 32-bit front it feels great: it actually seems to perform a little better than a Pentium 4 @ 2.6 GHz w/ HT. Admittedly, this is my personal feeling about how Warcraft III plays using wine, so feel free not to consider this remark.
However, it's drains the battery quite quickly. At full power, a 12-cell battery lasts for about 2.5 hours. With aggressive power saving (processor running at 800Mhz instead of 2.4GHz), hard disk spinning down almost as soon as it finished reading, little wireless card usage, I managed to bring it to about 5 hours, which is not great. It still feels sufficiently fast, the most annoying incident being the delay until the hdd spins up.
All in all, it's a great architecture, but it has its problems, of which the main I would say is the power consumption.
And simulated annealing is just as good as genetic algorithms. Instead of setting the mutation rate and crossover probability, you have to set the probability to accept a worse solution that takes you out of the local minimum, and you have to define a neighbouring function (which can be a bitch!! - more so than defining the crossover).
Fancy != useful
Nothing wrong with running it off line, then using the found parameters. However, there may be major drawbacks: a possibly good solution for a particular load, might turn out to be unusually bad for a slightly different load. Given that there seems to be a need for genetic algorithms to search for the solution, it follows that the search space is sufficiently complex and not all that "well behaved". This means that the solutions will be rather unstable. In this case, perhaps it's better to accept a solution that's moderately bad for all tasks, instead of picking one that's excellent for a rare configuration and quite bad for most.
Moreover, the theory that genetic algorithms are easy is a myth perpetuated by those with shallow understanding of the concept. Sure, the concept and the code are simple. However, setting the parameters for genetic algorithms is a bitch and their meaning is not entirely related to the problem at hand.
So, all that this guy managed (with virtually no improvement) was to get rid of some understandable parameters coded in the original scheduler, and replace them with the mutation rate, crossover probability and population size, which don't have any meaning in terms of process scheduling.
And, no, kernel hacking does not imply any knowledge of statistics or statistical experiment design.
Your point is short sighted. How many times do you actually have to run the optimiser to adapt the parameters to the current load ? If you had to do so one hundred times per minute (not unlikely on a web server, for instance, where processes constantly start and end), then the overhead from the GA would become quite substantial.
A method that takes a microsecond to run once, will take a millennium to finish if ran sufficiently many times.
From his email address, the guy works at IBM, perhaps in some research department. I would suspect this was published because the project needed more money and they had to show they produced something. Along the lines of "our results are inconclusive. Please give us more money to continue".
As the recent trends show it, in both academia and industry, it's better to publish crap than not to publish at all. It is actually commonly said "publish or perish".
So, the work of this guy is marginal at best. It un-necessarily complicates the code in the scheduler while the gains are minimal.
Even your 5-10% would be useless. They'd be easily compensated for by a slightly faster CPU.
All in all, this seems to follow a trend in OS research nowadays: there are preciously few who have good ideas, and the rest believe they are improving things by throwing in something exotic (e.g. genetic algorithms) they don't even understand, hoping that everyone will be impressed.
It's a shame when a story is posted as it was published on cnet. This should be a site for nerds and stuff that matters. Some editing would have been necessary, as the cnet story is neither clear nor terribly informative.
h eeler-search.metadata.interface.elements.php
h eeler-search.metadata.interface.elements.php
Although perhaps it's silly to with that the moderators spent a little more time doing some very basic research (yes, that could be "googling") to find more accurate information. Otherwise, why do they exist ?
Here are two more interesting links, with some more detail about the ideas of the KDE team (don't expect very many details, as it's just vapourware):
http://conference2004.kde.org/cfp-devconf/scott.w
http://conference2004.kde.org/transcripts/scott.w
I think there are a couple of interesting things to do out there. Your indecision seems te me to be the problem.
I would suggest you to pick something that previously caught your attention and try to add a touch of AI in it. It is not of the uttermost importance to be something that will change the course of the world. Remember - you are only one person, and setting a huge goal may end up by doing nothing.
A long journey begins with a simple step (no guys, this is not a chinese proverb - it's just common sense).
I can't tell you what to pick - I have no idea what your interests are. But I can tell you that I found a very rewarding "landscape" in the field of computer vision. I had to make such a project myself when I graduated, and I wrote an OCR
But as I said - it doesn't really matter to write something huge - it is more important that you learn from it.
And, yes, I'd like though to make a suggestion somewhat related to the previous answers. The paperclip in Word has some good AI behind it. Don't flame me for this - I usually don't put the words Microsoft and good in the same sentence. But in this case it's not the AI that's bad - it's the idea of having a paperclip dancing while I am doing some work. I mean, why the hell is that thing stretching on my screen while I am trying to do some work ?
Conclusion : do something, don't set a goal higher than you could possibly reach (for now), and don't make it annoying.
In my opinion this is the best story since that scary law for controlling the web pages anonimity passed in France.
:).
This only proves the Murphy's law : never put the blame on bad intent when you can blame stupidity.
I guess this could be rephrased a bit like this :
Shoot yourself in the leg - perhaps the enemy won't shoot you. Although I can imagine a scene where all the north korean generals are maneouvering playstations
I suppose noone will argue the point that books are carriers of ideas, items used to share experiences (even when it comes to fantasy stories), sources of enllightment and education.
Since Guttenberg invented his primitive press, humans have been attracted by the books, by the information inside.
Reading is a privilege. Of the one that knows to read, of the one that has the means to buy books and so on.
About a hundred years ago, many educated people were proud for having read about 10 books. Nowadays, because books are affordable we read 10 of them before we go to school. This is progress.
Still, there are many among us that are too poor to afford spending their money on books, instead on their supper. And for years, there existed for them a way to have access to this information : libraries. And God only knows how many brilliant scientists were poor beyond belief in their youth. Their only chance was that they had access to a library
Nowadays, in France, the Louvre Museum opens its gates for everyone (withouy any cost) on the first Sunday of every month. Are they losing money ? Probably. But in exchange they allow the entrance to poor students that are interested in art but have no means to pay the entrance fee. Why are they doing this, and not allow only students to enter for free ? (something like academic prices). Because having janitors to check for id's would be even more expensive. Granting access to everyone is, then, something desirable not only from the visitor's point of view
Why not do the same here, in US ? Because apparently companies cannot see beyond the immediate gain. Because they don't really care, and because sometimes (too often unfortunately) they are stupid. Because the costs are probably even more impressive (although this is something genuinely american : exagerrated governmental costs). Because companies are taking the bureaucratic model from the government, no matter the cost. It easier to impose the existing regulations (even if they don't really fit) and pay more money, instead of thinking about new ones who may turn to be better.
And I can't stop wondering how many books have read in the late 10 years the ones that are trying to impose these regulations.
Anyway, imposing such draconian regulations would have as direct effect a drastical reduction of the number of readers in the poor side of the population. By cutting the access to culture to someone, one is denying a fundamental right. The right to be informed, the right become aware and to be able to make good decisions.
Then again, isn't a stupid and blind population the ideal for a government ?
I confess : i haven't played with either Napster or Gnutella. Mostly because of lack of time, still I haven't used them
So what am I doing here in the first you might ask.
I'm not going to try to explain why the companies are so pissed about it. It's pretty much obvious and I think that one way ore another they won't succeed. It also happens that I think they are stupidly unsatiable. And this is what is going to bring them down.
I won't even comment what Metallica did : after all they aren't what they used to be (I liked them, I still do but not that much) - no flames please.
So, if I ignore all these things, my post will be offtopic, right ?
Well, perhaps : but I have something else to say that could be somewhat related. What is Gnutella anyway, but a distributed filesystem ? Ok, with some restrictions (you are not allowed to modify the file on the remote computer). Right now, it is very difficult to implement robust filesystems (and the existing ones have their flaws).
What if we could use the code of gnutella, napster etc. to create a robust NFS, that could insure data persistency in the same way gnutella does (quote from their site : f, say, New York was hit by Dr. Evil, GnutellaNet probably wouldn't even notice. So you lose your "Gnutella friends" in New York. Big deal. It's unlikely that New York is the only place where people share the recipe for strawberry-rhubarb pie. )
Think about it. A toy could become something really important.
I guess the post here it way too enthusiastic, since even in the original article they never mentioned some basic facts.
Read it well : it's only theory.
And there are a few questions to be asked :
What about the stability of the memory ? What are the operating conditions ? If it has to be kept below, say 200K it will be quite difficult to use it, wouldn't it ?
What would be the needs for producing such a memory ? Because it's quite hard to find pure silicon in nature (and producing itis quite expensive I imagine)
And last but not least : how about compatibility with the actual technology ? I guess none of you is actually imagining that this will be joyfully embraced by companies that are strong on the memory/processors market.
This may actually end up being buried by marketing, because nowadays an university doesn't really have the possibility (financially) of designing a competitive chip/architecture.
So I guess I will be a bit reluctant towards its success until I see it on the commercial ads of AMD or Intel.
I couldn't say I wasn't impressed about this. Especially because I know the hype 20 years ago when the first hdd appeared. It had about 5Mb and everyone was wondering who could ever fill this huge space.
Yet, let's not be too enthusiastic because for the moment this technology has some possible problems (at least in what concerns the military - which seem to be the first interested in it). Being a magnetic storage device a magnetic field can destabilize the data on it, thus making the computer unusable (or unreliable). Hypothetically speaking, it would be quite easy to attack a ship with some sort of radiation that will make it vulnerable to another attack that may destroy it
On the other hand, civilian users are quite protected against this, since there aren't many important secrets to be destroyed (I'm not speaking about corporate users). Then again, who will ever use 400Gb (or more) at home ? (ok I may sound silly and repeating the sentence I mentioned in the beginning). There are some limits in one's ability to gather information and one of them is the time.
Although, if I were to speak frankly, I imagine Windows 2010 taking 70% of this space (with a 25Gb Solitaire). Hopefully there will be no windows any more in 2010
Has anyone noticed a weird thing ? they say that the compressed file will have 5.5 M (compared to 15M the current version). What does this mean ? Are they stripping a lot of functionalities out of it ? have they given up some more or less useless components (e.g the radio part) ? ... Will they support it ? (most likely, though, since they mention cross-platform compatibility).
Will it be linked to some libraries that come with the OS ? If so, which OS ? Windows perhaps ?
Linux is not mentioned at all
As far as I am concerned, I hope that smaller size means less bugs (although we have the perfect example of M$-DOS 1.0- some 300 bugs in 5000 lines of code - great job breeding them guys).
Anyways, I surely hope it will have less bugs than the current version.
Welcome to the real world. It's quite amusing to say these words when speaking about wirtual reality, but let's start arguing a bit on the consequences and, of course, on the causes.
...
the internet is becoming more and more popular. Everyone knows about it (at least in US and Europe), everyone is excited about it. Do you think the government is going to sit on its comfortable chair and do nothing about it ? When it could make the chair even more comfortable ?
Be serious. Of course they will impose regulations. Why ? Because they have the power to do it.
Furthermore, the main internet providers are in US. Which gives the US govt the possibility of gaining a real control over an international system. Impressive, isn't it ? Does the word manipulation mean anything to you ? And it's even cheaper than all the more or less visible agencies. Not to mention military actions.
Is this moral ? Probably not. Are we going to gain something from this ? Yeah, right !
However, this could also bring some benefits. If the rules are fair we won't have to deal with situations like M$. Remember that they became what they are especially because of the lack of regulations (and the silliness of IBM).
I'm not very sure how this will end but it will probably be something like the telephony networks today. Perhaps a bit more complex, but not too much. And I guess this isn't that bad after all
I've got my share of bad teachers during my school days (which aren't finished yet). Let's face it : there will always be good teachers, bad teachers and "bad and arrogant" teachers.
This last category is the worst. I've seen teachers that didn't want to give the maximum grade because they considered that was destined only to them, thus lowering your GPA. I saw teachers that gave you an incomplete even if you did well on the exam just for the fun of having you retaking that exam and spend the summer learning. Some of them made sarcastic comments regarding a wrong answer from their students (like "if you can prove this to me I'll give you a PhD").
At one point or another we all experienced this. It doesn't make it right, but this is the way it is.
Does this mean we should simply ignore the problem ? Definitely no.
Yet, I read some of the comments of the students, and I think they definitely deserve their teacher. Ok, he's a lousy teacher, but this doesn't give you any right to curse him like that. Besides, if one doesn't understand that cursing someone hardly make him one's friend, then he should avoid expressing himself in public.
I suppose the reaction of the teacher is even understandable in this case. Although I can't see why he is sueing the web site. After all, they are a simple messenger. Destroying the site won't change the opinion of his students. On the contrary.
Then again, from my experience I could also tell that although the goal of the site is meritory, it probably won't do too much good.
The best professors I've had until now were very special human beings and during the years they earned my respect without being too concerned about this. And this includes my respect both as teachers and as persons.
On the other hand, I found that the main characteristic of the bad teachers was that they never tried to meet the requirements of the students, they never tried to "teach". They were exposing the material and that was it. The students always were a "hurd" with whom they had to spend their precious time.
And since the bad teachers don't care about the opinions of their students anyway, why bother giving those grades ?
Well, it's been known for years the legend that princess Anastasia of Russia survived the slaughter of her family in 1916. I am not very sure, but I don't think it's been proven that there aren't any survivors of the Russian royal family ... ... Or...
... If they're dead, let them rest in peace - it would only be a proof of respect.
Imagine how milions of fans of Elvis would react at the news that their idol is not dead
Or even Jim Morrison
Or perhaps it's better to leave things as they are
OK, suppose it would fork. I don't think this is such a big of a problem. After all we do have several distributions among which some are not that "standard". We have three different flavours of packages, yet nothing stops me from installing any of them on my machine. And if I really don't want to install packages I can always get the source.
The issue is not if the Linux breaks into several incompatible distributions. Unless they are incompatible in the Microsoft sense of the word. Meaning that those distributions will not be able to share data between them.
It doesn't really matter if that distribution is running a binary of StarOffice and the other is running another type of binary of the same package. As long as StarOffice for Linux1 and StarOffice for Linux2 will share the document formats then there is no problem.
After all, how many of you have several phones at home ? Are they all the same ? Probably not, yet you can speak from one to any other.
In a way you're right. However, in life it's not always necessary to know everything, but to know where to efficiently search for your information.
I agree, there should be some tests where the students are not allowed to have access to the internet (yes, general history is a good example), but there also should be just as many where they are allowed to do it.
If this is true (I don't see any reason why it shouldn't be) then it's very cute. I can't say it is new (after all, there are take-home exams), it's just that the scenario is slightly different.
What it is even more interesting is that in this way foreign or disabled students are allowed access to information in their native language or something that would compensate their disabilities.
Doesn't this make much more sense than the LEGO based test presented on this site a couple of months ago ? (sorry I don't have a link to that article).
Besisdes, if we can afford to give every student a laptop imagine how much less paper will be consumed. Respecting the environment is one of our priorities right now (or at least should be) and this may be a first step. The next one is to convince enterprises to give up un printing that much.
I hated the hum! of the article poster. On one hand we want Linux to reach the desktop, on the other we are reluctant to easy installation procedures. I am sorry, but this is bigotry.
Besides, why complain since Yast2 comes along with yast1 (yes, text mode, full of features installation software) ? Yast2 was written for those that don't know exactly what Linux is, who don't know what hardware is inside their machines and so on. Or for those that like inserting the cd, writing a few lines and going to smoke a cigarette while the system is installing.
Of course, an experienced user won't use this tool (although I can tell it does look cute) because it doesn't offer you the complete control over the installation.
As for myself, I would have liked to see the KDE 2 with the new release of Suse. I know, it won't probably be included until 6.6 - but still, one can hope, can't he ? Right now it's only a simple update of 6.3 (great distribution!). So, if you have enough bandwidth I think it's worth downloading (or even better take advantage of the update from ftp site feature).
Wish (read it especially if you work for VA) : please mirror the distribution on your site sooner. ftp.suse.com works only at 50k/s which is a real pain when downloading 6 cd's .