I live in a house my family has lived in for over 60 years, with the same old phone line and it's NEVER GONE DOWN IN SIXTY YEARS!
It's not as simple as you describe. For example, in the United States at least, a large number of landlines were unable to initiate any phone calls on September 11, 2001, whereas internet based services such as e-mail had no problems on that day.
Even for people who need a landline for 911, VoIP is still a useful complement for a landline. You can use VoIP for calling overseas, and the landline for local calls. In fact, you don't even need to subscribe to a VoIP service -- any calls that you place overseas through your "old phone line" are probably already being relayed by your long distance carrier over VoIP, without you knowing it.
The Israeli airline has been profiling passengers all sorts of ways for decades. This sounds a lot like one of the methods they employ.
The difference is that the Israeli airline agents interview you one-on-one for an hour or two during the process, which is a lot more reliable than judging someone based on one or two face glimpses. Also, there are no artificial setups such as having an agent pretend to be another passenger, which was one of the things described in the article. In the Israeli airlines you know what's going on and they are up front about it.
If you think Apple can do no wrong (lest their programmers get headaches), then why are Mac windows only resizeable from the lower right corner? Do you have some twisted excuse why that's a better design than how Windows lets you resize a window from any corner or edge?
It's even better on Linux: you can resize a window without even navigating to the edge. All modern Linux desktops allow you to resize any window by holding the Alt key while dragging the mouse in the window. The window will grow/shrink in the direction of the corner closest to your mouse cursor. From a usability standpoint (Fitt's law) it's a hell of a lot easier to find the Alt key than to find a window edge.
I know a local architectural firm [turned filters off] after a purchase order was false-positived.
The calculation is not as simple as you imply. If you get a million spams per day (like this guy), then you're probably better off with the spam filter, since without it your chances of catching the one purchase order hiding in 1000000 spams is pretty slim.
Spam filtering becomes worth it when the error rate of the filter is lower than the error rate of a human sorting through the same mail. That level of performance is pretty easy to achieve.
I'll begin by saying that I totally agree with you in that Windows and Linux both have their own purposes and that it is insane to suggest that either represents a good fit for everyone. My objection is to your reasoning, not your conclusion.
Now, please understand that although I mean this in the kindest possible way, your response indicates a complete lack of understanding of the basic principles of software administration. To me, reading your response is like listening to a typewriter user explain why he doesn't need computers because a typewriter does everything that he needs it to do.
I cannot blame you for this failure, because the truth is that Linux's model for managing software is so dramatically opposed to Windows that you cannot possibly appreciate the difference unless you have worked in Linux for some time.
Let's start with the following claim of yours:
I install stuff if I use it in Linux. Sometimes by the package fetcher, sometimes by a downloaded package + manager, sometimes by source. Oh, looks there's lots of different ways there too.
In Windows it usually just involves wisards with extremely similar interfaces, where if you want you can put in the CD and keep clicking "next" until done...
It is true that Linux allows several different methods for installing software. Heck, Linux even allows wizards too (for example nvidia drivers use them). But using this flexibility as a basis for your comparison is a fallacy. In reality, to a 99% approximation, Linux users use the package fetcher, and that's it.
A package fetcher really does make installation easier, because if nothing else it puts everything in one place -- you don't have to put in a different CD or visit different web sites to obtain different programs. However, what you may fail to understand is that a package manager represents a big win, not for installation, but for maintenance of software. To give just one simple example, if I have a shared library (a.k.a. a DLL) and I want to know, right now, what programs on my system use this library, I can type, literally:
rpm -q --whatrequires libz.so.1
and it will show me in less than five seconds all 181 programs on my system that use the library. In other words, bye-bye DLL hell. This kind of information gathering is very very difficult on Windows, and indeed it is easy to argue that any reasonable solution to this problem on Windows would amount to implementing a package system.
Let's look at another of your claims, regarding updating Linux programs:
Yes, because if you have Linux installed, with Xorg 6.9, you will *never* have to upgrade to 7.x to use version 7.x! It's *magical*
Again, if you really think that package management is worthless here, it can only be because you have never learned what package management can do. Imagine for a moment if Windows Update worked for all software that you have installed on your system, including not just Windows and Office but also Firefox and Photoshop and your wireless drivers and the random shareware program you found on the street. While you're at it, you also need to pretend that the updates never break anything, which is palpably false on Windows but is close to the truth on Linux, especially the expensive versions such as Red Hat Enterprise Linux where you are basically paying them money in exchange for them not breaking things.
The point is that in Linux it is actually true that you never have to upgrade to 7.x to use version 7.x. The system by default will automatically upgrade to 7.x for you, and the next time you log in you'll be on 7.x, and it will work.
Unfortunately, even this kind of thought exercise does not truly capture the added capabilities of package management, because it is easy to get the false impression that such a magical system represents the end in itself. The truth is that y
Really, there aren't any worries if you don't have a firewall and connect the default Ubuntu to the 'net. Pretty much all services that would be exposed to network are disabled or not installed by default anyway.
What you say is true, and it's also worth mentioning that many other Linux distros are even better than Ubuntu in this regard. For example, a default Fedora install has no external services and enables the firewall and activates SELinux in enforcing mode. You could set up one of these machines unattended on the internet for years without having the system compromised, even with no updates.
Firefox is probably the biggest security weakness for a typical Linux desktop today, not because Firefox is particularly exploitable in an absolute sense but just because all other avenues of attack are so difficult that Firefox vulnerabilities have become the weakest link.
It's not even clear that PGP would help with spam. A malware program can just as easily be programmed to steal a PGP key in addition to bank passwords and whatever else they are presently stealing. If PGP usage were widespread, you can bet that spammers would be focusing their efforts on compromising the keys. Judging from the current dismal state of computer security, I have no doubt that they would succeed.
It's a thin client OS running virtually everything off a server through RDP via a technology and software MS "licensed" from someone else. Ooooh, but you can run Telnet local!!!
Have you ever actually used Windows Fundamentals for Legacy PCs, or are you just spouting your mouth off? You can run firefox locally on WinFLP. See screenshot here.
The 9.x branch of Windows should be left as dead. The NT/2000/XP branch has more stability than the the 9.x branch. MS should actually create a slim downed version of NT/2000/XP for low end systems.
Believe it or not, Microsoft has actually done this. It's called Windows Fundamentals for Legacy PCs. The reason you haven't heard about it is because Microsoft does not advertise it at all, and also because it's only available to corporate customers.
It is so strange that you bring up coal. I did not.
Even though you did not bring up coal, the fact of the matter is that for large scale energy production the only options are coal and nuclear. Solar? Don't make me laugh. Even the simplest back of the envelope calculation shows that there does not exist the theoretical production capacity to make enough solar panels for everyone at current rates of consumption.
Should we wait for nuclear power to kill even more people in its inevitable way, and at a level that dwarfs any single coal related incident before doing what needs to be done?
As many others here have already pointed out, you can't use Chernobyl as a representative example of nuclear safety. In the 21 years since Chernobyl, the total number of accidental deaths caused by nuclear energy plants, worldwide, is two.
Read that again. Two deaths worldwide in the last 21 years, for an industry that provides one sixth of the world's electricity. That is an astoundingly good safety record. I wouldn't be surprised if solar panel manufacturers and suppliers were found to be covering up two deaths worldwide in the past 21 years. This is without even considering that solar power provides much less than one sixth of the world's present electricity.
Even if solar eventually proves to be the superior technology, fossil fuels are so mind-bogglingly bad that it's worth it to switch to nuclear power immediately just for the sake of the short term benefit of getting away from fossil fuels. Unlike renewable energy, nuclear power has been proven to work on a large scale. We can switch to it now. We should.
Conservatively 4000 to 9000 deaths. [from all nuclear accidents, in all of history]
I hope you're aware that the coal mining industry suffered over 6000 accidental deaths in 2004 alone. Given the relative safety records of coal vs. nuclear energy, it's not even close. Nuclear energy is safer by a long shot.
So how much money did it cost to pull staff off other projects and put em on this Jupiter diversion? Is it really economical to pop by just to pick up 3 years? It's not like there's a time to market here.
There is a deadline here, and the deadline is a natural one. Right now Pluto is near its perihelion, which means it is (just barely) warm enough to have an atmosphere. There are many many things you can learn scientifically from an atmosphere. However, if the space probe takes too long to arrive at Pluto, the atmosphere will be gone by the time it gets there. In that case, we'll have to wait a cool 200 years before Pluto comes around to perihelion again.
Scientists believe that as Pluto continues its 248-yearlong orbit around the sun, its tenuous atmosphere eventually will freeze and collapse to the surface. Pluto has been racing away from the sun since its closest approach in 1989 and scientists do not know how much time remains before Pluto's atmosphere collapses. Once that happens its atmosphere is not expected to re-emerge for about 200 years.
"Some people think its 20 years off and some people think its five years off," said Stern. "No one really knows when Pluto's atmosphere will snow out and collapse."
The major premise of wikipedia functionality is that it can be edited by anyone, yes?
By far my biggest concern about this scandal is that your premise is actually false, and the falsity of your premise is directly related to the negative consequences of this affair in a very intimate way.
I understand that in an ideal world, anything on Wikipedia can be edited by anyone with no censorship whatsoever, and in an ideal world, two conflicting edits are resolved on the basis of the actual contributions with no regard to credentials or background or the identities of the contributors involved. Unfortunately, Wikipedia falls far short of this ideal in many important instances, and (ironically) the most serious shortcomings emerge during the most serious cases.
For example, let's look at Essjay's talk page as of today, 12:40am eastern time. This is an important article for anyone wishing to voice their opinion on the very matter that we are discussing now. Yet, despite the presence of multiple commentors on that page claiming that content is king and credentials don't matter, the simple fact is you cannot edit that page at all unless you already have an account which has been active for some amount of time, because the page is protected.
This blows a big hole in your assertion that Wikipedia can be edited by anyone at any time. I cannot edit this page at the present time, because I don't have an account, and even if I were to create an account, I would have to wait some amount of time before the account would be considered active long enough to edit that page.
Although you may like to think that an obscure user's talk page is not important enough to be considered representative of Wikipedia as a whole, the fact is that the large majority of so-called controversial pages are kept in protected status, with the result that outsiders cannot edit the page.
The sheer hypocrisy of Wikipedia's stance in this matter is astounding. It is far worse than anything I have seen in other notoriously hypocritical arenas such as presidential politics. Wikipedia is saying that, on the one hand, your (academic) credentials are actively immaterial, but on the other hand it considers your (Wikipedia account owning) credentials so essential that it won't even let you post on important matters unless you have a sufficient amount of the latter. If there is a more insidious and adversarial display of censorship to be found anywhere else in the world, I have not seen it.
Moreover, even if I were to by some stroke of fortune create an account and wait the minimum amount of waiting time necessary to post on that page, I would still be attacked on the grounds of having an account that is too new for my comments to merit consideration. See for instance the comment where Netscott dismisses the opinion of Snackycakes on this very basis. Again, it is hard for me to reconcile this blatantly hostile stance with Wikipedia's official (and largely ficticious) policy of honoring contributions based solely on content.
However, on top of this (already long) rant, the absolute worst part is that Essjay is an administrator and a member of the oversight committee, and as such, he has more power on Wikipedia than all but five other people in terms of deciding which pages to protect, which users to ban, and which comments to delete. In other words, Essjay, the very user whose integrity I feel is justifiably subject to question, is in a strong position to disproportionately influence this debate about himself, not because of the merit of his contributions to the debate in question, but because of his...
credentials.
I should close by saying that I am not by any means the anti-Wikipedia zealot that this post makes me out to be. As a matter of fact, I am a founding member of PlanetMath and a strong supp
I think we'd be more amazed to discover a planet with an ozone layer, period.
This is more true than probably most slashdotters realize. Ozone is the only chemical indicator of life that we can reliably detect across long distances.
Ozone, unlike oxygen itself, has a strong absorption spectrum in the infrared wavelengths. A space-based infrared telescope (like Spitzer, but better) is exactly the right tool for detecting the presence of ozone. (A ground based telescope will not do, since infrared is absorbed by the atmosphere.) Finding ozone on a planet is just like finding oxygen -- the two compounds are so closely related that you can't have one without the other. And oxygen is a very volatile compound that reacts with almost anything else if you leave it alone. The only way for a planet to have free floating oxygen is if something on the planet is producing it.
As far as we know, the only way to sustain an oxygen atmosphere on a planetary scale is with life. So, yes, finding ozone on a distant planet would be a very exciting discovery indeed.
If you hold the MPAA to the law on this, then it is a tacit acknowledgement that I.P. is a valid thing. You can't have it both ways, Slashdot.
Stop being ridiculous. Slashdot is not one person. Slashdot is a diverse group of users and is entitled to hold a diversity of opinions.
The MPAA, on the other hand, is a single legal entity. It is the MPAA who cannot have it both ways. Yet somehow you seem to think that what's okay for the MPAA is not okay for Slashdot. For you to hold Slashdot to a stricter standard than the MPAA is, quite simply, mind boggling.
Well, "low" spam volume is a relative term here. A few years ago, 1000 spams per day would have been considered high. Back then, everyone who used filters was seeing perfect performance, and a few optimistic folks thought that spammers could not raise their spam volumes to compensate. However, that theory isn't holding true anymore.
Nowadays the main effect of filters is that they improve the spam situation from totally unmanageable to borderline tolerable. The result is that we have two tiers of email users: those who use filters can actually for the most part continue to use email as they have been using it all along, with relatively minor differences (but still with enough inconvenience that I don't consider the spam problem "solved", even for this class of user). OTOH, those who do not use filters have been forced to drastically curtail the types of things they do with email, sometimes to the point of periodically abandoning email addresses that have become overrun with spam.
I don't think there is a way to solve the spam problem at this point. Conceptually, the set of solutions can be divided into two types: elimination approaches, where you try to reduce the amount of spam sent, and blocking approaches, where the spam has already been sent and you try to distinguish the spam from the non-spam. The experience of the past 20 years tells me that the elimination approach is impossible unless we change the underlying protocol, which is not going to happen. Since reducing or eliminating spam is impossible, a consequence is that any sort of blocking technique would have to be perfect in order to solve the problem, because for any nonzero amount of error rate that you have, the spammers will gladly raise their spam volumes high enough to compensate.
This analysis may be flawed, in the sense that there may be some physical upper limit to the amount of possible spam in the universe, but from where I'm sitting I don't see it (okay, I'm joking here, but only slightly).
A well managed bayesian filter is a lot better than nothing, but I do not agree that it is good enough.
I use spamassassin, like you, and I've been using it for years. I am also fairly open with my email address (for example, my email is displayed publicly on my slashdot posts, of all places). The problem with a bayesian filter is that it is not perfect. Given enough spam volume, such as the amounts we're seeing today, any filtering approach is bound to make errors, and beyond a certain point any attempts to improve the performance (for example by tuning parameters) end up destroying the usefulness of the filter.
The reason I use the filters is because they are the least bad option out of a sea of sucky alternatives. For example, if I were to sort email by hand, my own rate of human error would already be higher than the error rate of the filter. But to say that filters are the least sucky option is a far cry from saying that filters solve the problem. At best, it might be true that at "low" spam volumes such as ~1000 spams per day, a filter can perform perfectly. However, at higher volumes even the filters start to break, and since the amount of spam is increasing at a phenomenal rate, this is not a battle that can be won.
Simply put, filters do not solve the spam problem because they still inevitably make errors, and every such error represents a failure of the filter to eliminate the major harmful effect of spam, which is losing legitimate email (I'm ignoring secondary harmful effects such as bandwidth consumption, which others have already addressed).
Filters do help to solve the spam problem, because with good filters in place it is no longer necessary to suppress all sources of spam. It is only necessary to suppress spam sources to such an extent that the resulting volume of spam lies within the range that can be effectively handled by filters. Unfortunately, at the present time the internet as a whole is failing to achieve even this modest goal.
There is another problem with filters, one that does not affect you and me, but does affect the vast majority of internet users, and that is the issue of user expertise. It takes quite a bit of sophistication and computer savvy on the part of the end user in order to properly run and manage a bayesian filter. I'm not saying the task is very demanding on an absolute scale, but remember, we're talking about users who can't even keep their own computers free of spyware and trojans, which is administratively a far easier task than maintaining a bayesian spam filter. For this reason I think that, even if filters do end up solving the spam problem, it will take quite a bit of work to implement that solution in a manner that benefits the average internet user.
Call me a total thicky, but can't we strengthen any application that uses a hash by using several different hashes?
This exact proposal shows up, like clockwork, literally dozens and dozens of times for each slashdot story about hash functions. Since the number of people who know why this proposal fails is miniscule compared to the number of people who think of the idea, it is literally impossible to respond to all the people who keep suggesting this idea. I mean, even if all of us spent literally every minute of every day responding to people who suggest this idea, we would still not have time to reply to every single post.
Here is an old post on slashdot explaining exactly why this idea doesn't work. The post has some details wrong... for example, the correct security strength of the combined md5+sha1 hash is in reality 2^80 + 160*2^64, which is much weaker than even the already weakened security level cited in the post. However, the general idea is correct, and if you google for the title of the paper cited in that post, you can find much more information.
I hope that this reply helps to educate at least one poster, but judging by the regularity with which this idea keeps reoccurring, it's a little bit like rearranging chairs on the Titanic.
The length of your reply, in and of itself, is sufficient to debunk your original claim that copyright law is "simple and easy."
The United States has a strong right to contract - you can sell just about anything if you make it a contract. If I have an iPod, and it's mine, I can sell it to you. That's a contract.
The vast majority of rights cannot be sold by contract. For example, the right to life and freedom cannot be sold. Property rights can be sold, which is why proponents of strong copyright try to conflate the notions of copyright and property, but nothing in the Constitution indicates that copyright and property are connected.
All copyright says is that whatever you right is yours, from the moment of inception. Simple and easy. What about that are you against?
Unfortunately, you're wrong. Copyright law in the US does not say that what you write is yours. More often than not, what you write belongs to the publisher. That's why you see musicians complaining bitterly that "the band owns none of its work" (quoted directly from that article).
You have to go to Europe in order to find countries where the copyright laws stipulate that copyrights always belong to the artist. I urge you to take a long hard look at the implications of a society such as the US where copyrights in practice do not belong to the artists.
One interesting side note: the Constitution of the United States says that
The Congress shall have power... To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries... (emphasis added) The contrast between the US Constitution (which favors authors and inventors) and US law (which favors publishers and employers) could not be more striking.
Let me make sure I have this straight... commercial licensing is bad, and we should do all we can to defeat it, because licensing is bad, I own my content, the revolution will not be televised!!111!!!, etc. But breaking a Creative Commons license is bad and wrong?
Let me get this straight... a company proposes (or seems to propose, it's not really clear from the story) to violate CC licensing, while insisting on license protections for its own works, and yet your objection is to the people complaining about the company, rather than the company itself?
Keep in mind that slashdot is not one person -- there are hundreds of thousands of users on slashdot, each with their own opinions. It boggles me that you would expect such a diverse population of users to be of one mind with respect to this issue while allowing a corporation to get off scot free for the same behavior.
64bit benchmarks???!? how many 64bit applications are you running/ are there in usful production?
everyone's got the 64bit crazy Ive had 64bit technology a long time can you say risc?
This is clearly a troll post, since you denigrate the availability of 64-bit computing in your first paragraph and then contradict yourself by claiming you jumped on the 64-bit bandwagon before everyone else, but I'll squash your post anyway.
Only a windows user (or possibly a Mac user) would treat 64-bit computing as useless or unavailable. Linux has been available in 64-bit versions since the days of the DEC Alpha, or since 2003 if you count only AMD64. Almost every Linux application benefits from recompiling for AMD64 as opposed to x86, because of the increased register space, and the nature of open source guarantees availablity of such versions. Compute-intensive applications such as cryptography (ssh/scp over gigabit ethernet) and media encoding (ogg, mp3, mpeg) exhibit performance gains of over 100% with 64-bit operations owing to the quadratic nature of block multiplication.
Scientific applications such as Mathematica and Maple, which I require for my job, have been available for AMD64 almost from the beginning days of the hardware platform, and gain rather a lot from AMD64 not only in terms of CPU performance but also from the larger virtual address space.
Even if all of these things weren't true and Linux didn't exist, the fact is that Windows Vista (vaporware jokes aside) really is coming out in five months and really does spell the end of 32-bit computing for mainstream performance applications. Windows Vista isn't some half-unfinished 64/32 bit mixture like Windows 95 was a half-unfinished 32/16 bit mixture -- Vista is 64-bit through and through.
The fact that your elitist risc platforms had 64-bit addressing some 30 years ago is not relevant to this discussion. Like it or not, x86 has "won" the platform battles, and x86-64 (unlike Alpha, IA64, Sparc) is the first and only 64-bit computing platform that will be relevant for general purpose computing.
I'm disappointed to see that as usual the review contains no mention of 64-bit performance. Does anyone know any place that provides 64-bit benchmarks for core 2 duo?
As much as it's done for us in the last 20 years, 32-bit x86 is not the future. Linux was AMD64-ready three years ago and Windows Vista which is just around the corner already puts more emphasis on the x86-64 platform than x86. Reviewing the 32-bit performance of core 2 duo is like reviewing Pentium processers based on 16-bit performance. Let's get some forward looking reviews instead of backward looking reviews, please!
Microsoft Research hires math majors. I know this because they hired me, and I am a math major. The only catch is that you have to have a Ph.D to work in research. Still, it is definitely something worth considering, if you have interests in both math and CS and want to major in math.
Of course, it takes quite a few years and a lot of work to get a Ph.D, so take that into consideration -- make sure you like it.
Another option that a few people have mentioned is financial services. A lot of brokerage firms on Wall Street love to hire math majors with CS knowledge. The math is needed for financial modeling and the CS is needed in order to implement those models in actual trading programs. Out of all of my classmates who work at companies, about half have chosen this route. Contrary to what a lot of the comments here have stated, it is in fact quite possible to get a job in a trading firm with only a bachelors degree, but of course your salary will be lower than if you had a Ph.D.
It is definitely not true that math majors have no jobs. If your parents need convincing, look up the recent Business Week cover story from a few months ago about the exploding number of job opportunities for math majors in the current information driven economy. The myth that math majors only have teaching jobs is something that was maybe possibly true 20-30 years ago when computers were not a big deal and manufacturing was dominant, but it's not true anymore. Nowadays math majors are in higher demand than ever because tech skills are so complex that employers increasingly look for more foundational training such as a mathematics background as opposed to someone who has already specialized in some narrow subject area before even leaving school.
One thing the article doesn't really make clear is that the remote control cable for the landing gear is not connected by default. The cable by default is in an unattached state. In order for the landing gear to be lowered by remote control, an astronaut on board has to first attach the cable. Presumably this would only be done if the shuttle was found to be damaged while in space.
What this all means is that manual intervention, in a way, is still required in order to deploy the landing gear. There is no chance that the remote control mechanism will inadvertently trigger deployment.
Slashdot Mirror
It's not as simple as you describe. For example, in the United States at least, a large number of landlines were unable to initiate any phone calls on September 11, 2001, whereas internet based services such as e-mail had no problems on that day.
Even for people who need a landline for 911, VoIP is still a useful complement for a landline. You can use VoIP for calling overseas, and the landline for local calls. In fact, you don't even need to subscribe to a VoIP service -- any calls that you place overseas through your "old phone line" are probably already being relayed by your long distance carrier over VoIP, without you knowing it.
The difference is that the Israeli airline agents interview you one-on-one for an hour or two during the process, which is a lot more reliable than judging someone based on one or two face glimpses. Also, there are no artificial setups such as having an agent pretend to be another passenger, which was one of the things described in the article. In the Israeli airlines you know what's going on and they are up front about it.
It's even better on Linux: you can resize a window without even navigating to the edge. All modern Linux desktops allow you to resize any window by holding the Alt key while dragging the mouse in the window. The window will grow/shrink in the direction of the corner closest to your mouse cursor. From a usability standpoint (Fitt's law) it's a hell of a lot easier to find the Alt key than to find a window edge.
The calculation is not as simple as you imply. If you get a million spams per day (like this guy), then you're probably better off with the spam filter, since without it your chances of catching the one purchase order hiding in 1000000 spams is pretty slim.
Spam filtering becomes worth it when the error rate of the filter is lower than the error rate of a human sorting through the same mail. That level of performance is pretty easy to achieve.
Now, please understand that although I mean this in the kindest possible way, your response indicates a complete lack of understanding of the basic principles of software administration. To me, reading your response is like listening to a typewriter user explain why he doesn't need computers because a typewriter does everything that he needs it to do.
I cannot blame you for this failure, because the truth is that Linux's model for managing software is so dramatically opposed to Windows that you cannot possibly appreciate the difference unless you have worked in Linux for some time.
Let's start with the following claim of yours:
I install stuff if I use it in Linux. Sometimes by the package fetcher, sometimes by a downloaded package + manager, sometimes by source. Oh, looks there's lots of different ways there too.
In Windows it usually just involves wisards with extremely similar interfaces, where if you want you can put in the CD and keep clicking "next" until done...
It is true that Linux allows several different methods for installing software. Heck, Linux even allows wizards too (for example nvidia drivers use them). But using this flexibility as a basis for your comparison is a fallacy. In reality, to a 99% approximation, Linux users use the package fetcher, and that's it.
A package fetcher really does make installation easier, because if nothing else it puts everything in one place -- you don't have to put in a different CD or visit different web sites to obtain different programs. However, what you may fail to understand is that a package manager represents a big win, not for installation, but for maintenance of software. To give just one simple example, if I have a shared library (a.k.a. a DLL) and I want to know, right now, what programs on my system use this library, I can type, literally:
and it will show me in less than five seconds all 181 programs on my system that use the library. In other words, bye-bye DLL hell. This kind of information gathering is very very difficult on Windows, and indeed it is easy to argue that any reasonable solution to this problem on Windows would amount to implementing a package system.
Let's look at another of your claims, regarding updating Linux programs:
Yes, because if you have Linux installed, with Xorg 6.9, you will *never* have to upgrade to 7.x to use version 7.x! It's *magical*
Again, if you really think that package management is worthless here, it can only be because you have never learned what package management can do. Imagine for a moment if Windows Update worked for all software that you have installed on your system, including not just Windows and Office but also Firefox and Photoshop and your wireless drivers and the random shareware program you found on the street. While you're at it, you also need to pretend that the updates never break anything, which is palpably false on Windows but is close to the truth on Linux, especially the expensive versions such as Red Hat Enterprise Linux where you are basically paying them money in exchange for them not breaking things.
The point is that in Linux it is actually true that you never have to upgrade to 7.x to use version 7.x. The system by default will automatically upgrade to 7.x for you, and the next time you log in you'll be on 7.x, and it will work.
Unfortunately, even this kind of thought exercise does not truly capture the added capabilities of package management, because it is easy to get the false impression that such a magical system represents the end in itself. The truth is that y
What you say is true, and it's also worth mentioning that many other Linux distros are even better than Ubuntu in this regard. For example, a default Fedora install has no external services and enables the firewall and activates SELinux in enforcing mode. You could set up one of these machines unattended on the internet for years without having the system compromised, even with no updates.
Firefox is probably the biggest security weakness for a typical Linux desktop today, not because Firefox is particularly exploitable in an absolute sense but just because all other avenues of attack are so difficult that Firefox vulnerabilities have become the weakest link.
It's not even clear that PGP would help with spam. A malware program can just as easily be programmed to steal a PGP key in addition to bank passwords and whatever else they are presently stealing. If PGP usage were widespread, you can bet that spammers would be focusing their efforts on compromising the keys. Judging from the current dismal state of computer security, I have no doubt that they would succeed.
Have you ever actually used Windows Fundamentals for Legacy PCs, or are you just spouting your mouth off? You can run firefox locally on WinFLP. See screenshot here.
Believe it or not, Microsoft has actually done this. It's called Windows Fundamentals for Legacy PCs. The reason you haven't heard about it is because Microsoft does not advertise it at all, and also because it's only available to corporate customers.
See the Wikipedia article for more information.
Even though you did not bring up coal, the fact of the matter is that for large scale energy production the only options are coal and nuclear. Solar? Don't make me laugh. Even the simplest back of the envelope calculation shows that there does not exist the theoretical production capacity to make enough solar panels for everyone at current rates of consumption.
Should we wait for nuclear power to kill even more people in its inevitable way, and at a level that dwarfs any single coal related incident before doing what needs to be done?As many others here have already pointed out, you can't use Chernobyl as a representative example of nuclear safety. In the 21 years since Chernobyl, the total number of accidental deaths caused by nuclear energy plants, worldwide, is two.
Read that again. Two deaths worldwide in the last 21 years, for an industry that provides one sixth of the world's electricity. That is an astoundingly good safety record. I wouldn't be surprised if solar panel manufacturers and suppliers were found to be covering up two deaths worldwide in the past 21 years. This is without even considering that solar power provides much less than one sixth of the world's present electricity.
Even if solar eventually proves to be the superior technology, fossil fuels are so mind-bogglingly bad that it's worth it to switch to nuclear power immediately just for the sake of the short term benefit of getting away from fossil fuels. Unlike renewable energy, nuclear power has been proven to work on a large scale. We can switch to it now. We should.
I hope you're aware that the coal mining industry suffered over 6000 accidental deaths in 2004 alone. Given the relative safety records of coal vs. nuclear energy, it's not even close. Nuclear energy is safer by a long shot.
There is a deadline here, and the deadline is a natural one. Right now Pluto is near its perihelion, which means it is (just barely) warm enough to have an atmosphere. There are many many things you can learn scientifically from an atmosphere. However, if the space probe takes too long to arrive at Pluto, the atmosphere will be gone by the time it gets there. In that case, we'll have to wait a cool 200 years before Pluto comes around to perihelion again.
Quoting space.com:
Scientists believe that as Pluto continues its 248-yearlong orbit around the sun, its tenuous atmosphere eventually will freeze and collapse to the surface. Pluto has been racing away from the sun since its closest approach in 1989 and scientists do not know how much time remains before Pluto's atmosphere collapses. Once that happens its atmosphere is not expected to re-emerge for about 200 years."Some people think its 20 years off and some people think its five years off," said Stern. "No one really knows when Pluto's atmosphere will snow out and collapse."
The major premise of wikipedia functionality is that it can be edited by anyone, yes?
By far my biggest concern about this scandal is that your premise is actually false, and the falsity of your premise is directly related to the negative consequences of this affair in a very intimate way.
I understand that in an ideal world, anything on Wikipedia can be edited by anyone with no censorship whatsoever, and in an ideal world, two conflicting edits are resolved on the basis of the actual contributions with no regard to credentials or background or the identities of the contributors involved. Unfortunately, Wikipedia falls far short of this ideal in many important instances, and (ironically) the most serious shortcomings emerge during the most serious cases.
For example, let's look at Essjay's talk page as of today, 12:40am eastern time. This is an important article for anyone wishing to voice their opinion on the very matter that we are discussing now. Yet, despite the presence of multiple commentors on that page claiming that content is king and credentials don't matter, the simple fact is you cannot edit that page at all unless you already have an account which has been active for some amount of time, because the page is protected.
This blows a big hole in your assertion that Wikipedia can be edited by anyone at any time. I cannot edit this page at the present time, because I don't have an account, and even if I were to create an account, I would have to wait some amount of time before the account would be considered active long enough to edit that page.
Although you may like to think that an obscure user's talk page is not important enough to be considered representative of Wikipedia as a whole, the fact is that the large majority of so-called controversial pages are kept in protected status, with the result that outsiders cannot edit the page.
The sheer hypocrisy of Wikipedia's stance in this matter is astounding. It is far worse than anything I have seen in other notoriously hypocritical arenas such as presidential politics. Wikipedia is saying that, on the one hand, your (academic) credentials are actively immaterial, but on the other hand it considers your (Wikipedia account owning) credentials so essential that it won't even let you post on important matters unless you have a sufficient amount of the latter. If there is a more insidious and adversarial display of censorship to be found anywhere else in the world, I have not seen it.
Moreover, even if I were to by some stroke of fortune create an account and wait the minimum amount of waiting time necessary to post on that page, I would still be attacked on the grounds of having an account that is too new for my comments to merit consideration. See for instance the comment where Netscott dismisses the opinion of Snackycakes on this very basis. Again, it is hard for me to reconcile this blatantly hostile stance with Wikipedia's official (and largely ficticious) policy of honoring contributions based solely on content.
However, on top of this (already long) rant, the absolute worst part is that Essjay is an administrator and a member of the oversight committee, and as such, he has more power on Wikipedia than all but five other people in terms of deciding which pages to protect, which users to ban, and which comments to delete. In other words, Essjay, the very user whose integrity I feel is justifiably subject to question, is in a strong position to disproportionately influence this debate about himself, not because of the merit of his contributions to the debate in question, but because of his...
credentials.
I should close by saying that I am not by any means the anti-Wikipedia zealot that this post makes me out to be. As a matter of fact, I am a founding member of PlanetMath and a strong supp
This is more true than probably most slashdotters realize. Ozone is the only chemical indicator of life that we can reliably detect across long distances.
Ozone, unlike oxygen itself, has a strong absorption spectrum in the infrared wavelengths. A space-based infrared telescope (like Spitzer, but better) is exactly the right tool for detecting the presence of ozone. (A ground based telescope will not do, since infrared is absorbed by the atmosphere.) Finding ozone on a planet is just like finding oxygen -- the two compounds are so closely related that you can't have one without the other. And oxygen is a very volatile compound that reacts with almost anything else if you leave it alone. The only way for a planet to have free floating oxygen is if something on the planet is producing it.
As far as we know, the only way to sustain an oxygen atmosphere on a planetary scale is with life. So, yes, finding ozone on a distant planet would be a very exciting discovery indeed.
Stop being ridiculous. Slashdot is not one person. Slashdot is a diverse group of users and is entitled to hold a diversity of opinions.
The MPAA, on the other hand, is a single legal entity. It is the MPAA who cannot have it both ways. Yet somehow you seem to think that what's okay for the MPAA is not okay for Slashdot. For you to hold Slashdot to a stricter standard than the MPAA is, quite simply, mind boggling.
Nowadays the main effect of filters is that they improve the spam situation from totally unmanageable to borderline tolerable. The result is that we have two tiers of email users: those who use filters can actually for the most part continue to use email as they have been using it all along, with relatively minor differences (but still with enough inconvenience that I don't consider the spam problem "solved", even for this class of user). OTOH, those who do not use filters have been forced to drastically curtail the types of things they do with email, sometimes to the point of periodically abandoning email addresses that have become overrun with spam.
I don't think there is a way to solve the spam problem at this point. Conceptually, the set of solutions can be divided into two types: elimination approaches, where you try to reduce the amount of spam sent, and blocking approaches, where the spam has already been sent and you try to distinguish the spam from the non-spam. The experience of the past 20 years tells me that the elimination approach is impossible unless we change the underlying protocol, which is not going to happen. Since reducing or eliminating spam is impossible, a consequence is that any sort of blocking technique would have to be perfect in order to solve the problem, because for any nonzero amount of error rate that you have, the spammers will gladly raise their spam volumes high enough to compensate.
This analysis may be flawed, in the sense that there may be some physical upper limit to the amount of possible spam in the universe, but from where I'm sitting I don't see it (okay, I'm joking here, but only slightly).
I use spamassassin, like you, and I've been using it for years. I am also fairly open with my email address (for example, my email is displayed publicly on my slashdot posts, of all places). The problem with a bayesian filter is that it is not perfect. Given enough spam volume, such as the amounts we're seeing today, any filtering approach is bound to make errors, and beyond a certain point any attempts to improve the performance (for example by tuning parameters) end up destroying the usefulness of the filter.
The reason I use the filters is because they are the least bad option out of a sea of sucky alternatives. For example, if I were to sort email by hand, my own rate of human error would already be higher than the error rate of the filter. But to say that filters are the least sucky option is a far cry from saying that filters solve the problem. At best, it might be true that at "low" spam volumes such as ~1000 spams per day, a filter can perform perfectly. However, at higher volumes even the filters start to break, and since the amount of spam is increasing at a phenomenal rate, this is not a battle that can be won.
Simply put, filters do not solve the spam problem because they still inevitably make errors, and every such error represents a failure of the filter to eliminate the major harmful effect of spam, which is losing legitimate email (I'm ignoring secondary harmful effects such as bandwidth consumption, which others have already addressed).
Filters do help to solve the spam problem, because with good filters in place it is no longer necessary to suppress all sources of spam. It is only necessary to suppress spam sources to such an extent that the resulting volume of spam lies within the range that can be effectively handled by filters. Unfortunately, at the present time the internet as a whole is failing to achieve even this modest goal.
There is another problem with filters, one that does not affect you and me, but does affect the vast majority of internet users, and that is the issue of user expertise. It takes quite a bit of sophistication and computer savvy on the part of the end user in order to properly run and manage a bayesian filter. I'm not saying the task is very demanding on an absolute scale, but remember, we're talking about users who can't even keep their own computers free of spyware and trojans, which is administratively a far easier task than maintaining a bayesian spam filter. For this reason I think that, even if filters do end up solving the spam problem, it will take quite a bit of work to implement that solution in a manner that benefits the average internet user.
This exact proposal shows up, like clockwork, literally dozens and dozens of times for each slashdot story about hash functions. Since the number of people who know why this proposal fails is miniscule compared to the number of people who think of the idea, it is literally impossible to respond to all the people who keep suggesting this idea. I mean, even if all of us spent literally every minute of every day responding to people who suggest this idea, we would still not have time to reply to every single post.
Here is an old post on slashdot explaining exactly why this idea doesn't work. The post has some details wrong ... for example, the correct security strength of the combined md5+sha1 hash is in reality 2^80 + 160*2^64, which is much weaker than even the already weakened security level cited in the post. However, the general idea is correct, and if you google for the title of the paper cited in that post, you can find much more information.
I hope that this reply helps to educate at least one poster, but judging by the regularity with which this idea keeps reoccurring, it's a little bit like rearranging chairs on the Titanic.
The vast majority of rights cannot be sold by contract. For example, the right to life and freedom cannot be sold. Property rights can be sold, which is why proponents of strong copyright try to conflate the notions of copyright and property, but nothing in the Constitution indicates that copyright and property are connected.
Unfortunately, you're wrong. Copyright law in the US does not say that what you write is yours. More often than not, what you write belongs to the publisher. That's why you see musicians complaining bitterly that "the band owns none of its work" (quoted directly from that article).
You have to go to Europe in order to find countries where the copyright laws stipulate that copyrights always belong to the artist. I urge you to take a long hard look at the implications of a society such as the US where copyrights in practice do not belong to the artists.
One interesting side note: the Constitution of the United States says that
The Congress shall have powerLet me get this straight... a company proposes (or seems to propose, it's not really clear from the story) to violate CC licensing, while insisting on license protections for its own works, and yet your objection is to the people complaining about the company, rather than the company itself?
Keep in mind that slashdot is not one person -- there are hundreds of thousands of users on slashdot, each with their own opinions. It boggles me that you would expect such a diverse population of users to be of one mind with respect to this issue while allowing a corporation to get off scot free for the same behavior.
This is clearly a troll post, since you denigrate the availability of 64-bit computing in your first paragraph and then contradict yourself by claiming you jumped on the 64-bit bandwagon before everyone else, but I'll squash your post anyway.
Only a windows user (or possibly a Mac user) would treat 64-bit computing as useless or unavailable. Linux has been available in 64-bit versions since the days of the DEC Alpha, or since 2003 if you count only AMD64. Almost every Linux application benefits from recompiling for AMD64 as opposed to x86, because of the increased register space, and the nature of open source guarantees availablity of such versions. Compute-intensive applications such as cryptography (ssh/scp over gigabit ethernet) and media encoding (ogg, mp3, mpeg) exhibit performance gains of over 100% with 64-bit operations owing to the quadratic nature of block multiplication.
Scientific applications such as Mathematica and Maple, which I require for my job, have been available for AMD64 almost from the beginning days of the hardware platform, and gain rather a lot from AMD64 not only in terms of CPU performance but also from the larger virtual address space.
Even if all of these things weren't true and Linux didn't exist, the fact is that Windows Vista (vaporware jokes aside) really is coming out in five months and really does spell the end of 32-bit computing for mainstream performance applications. Windows Vista isn't some half-unfinished 64/32 bit mixture like Windows 95 was a half-unfinished 32/16 bit mixture -- Vista is 64-bit through and through.
The fact that your elitist risc platforms had 64-bit addressing some 30 years ago is not relevant to this discussion. Like it or not, x86 has "won" the platform battles, and x86-64 (unlike Alpha, IA64, Sparc) is the first and only 64-bit computing platform that will be relevant for general purpose computing.
As much as it's done for us in the last 20 years, 32-bit x86 is not the future. Linux was AMD64-ready three years ago and Windows Vista which is just around the corner already puts more emphasis on the x86-64 platform than x86. Reviewing the 32-bit performance of core 2 duo is like reviewing Pentium processers based on 16-bit performance. Let's get some forward looking reviews instead of backward looking reviews, please!
Of course, it takes quite a few years and a lot of work to get a Ph.D, so take that into consideration -- make sure you like it.
Another option that a few people have mentioned is financial services. A lot of brokerage firms on Wall Street love to hire math majors with CS knowledge. The math is needed for financial modeling and the CS is needed in order to implement those models in actual trading programs. Out of all of my classmates who work at companies, about half have chosen this route. Contrary to what a lot of the comments here have stated, it is in fact quite possible to get a job in a trading firm with only a bachelors degree, but of course your salary will be lower than if you had a Ph.D.
It is definitely not true that math majors have no jobs. If your parents need convincing, look up the recent Business Week cover story from a few months ago about the exploding number of job opportunities for math majors in the current information driven economy. The myth that math majors only have teaching jobs is something that was maybe possibly true 20-30 years ago when computers were not a big deal and manufacturing was dominant, but it's not true anymore. Nowadays math majors are in higher demand than ever because tech skills are so complex that employers increasingly look for more foundational training such as a mathematics background as opposed to someone who has already specialized in some narrow subject area before even leaving school.
What this all means is that manual intervention, in a way, is still required in order to deploy the landing gear. There is no chance that the remote control mechanism will inadvertently trigger deployment.