Well, Moody's analysis is seriously flawed. You should read both the original article as well as the Bugtraq statistics. The statistics are accompanied by a lot of disclaimers which Moody just casts aside with half a sentence despite their importance.
What is compared is also interesting. Moody's data about Linux vulnerabilities is not about the kernel or the core system. It is not even about a single Linux distribution. It is not even computed correctly. The numbers given by Moody are for the union of all vulnerabilities in all Linux distributions covered by the statistics, and to make it look worse he adds vulnerabilities of Red Hat in once more for good measure.
Moreover, if you consider that Linux covers allpackages in all distributions and hence contains a lot more software than the standard Windows (NT or 98) distribution, the number become even more meaningless.
Basically, the article is a bad example of how to lie with statistics.
I do care. Your exhaust is what I have to breathe (well, metaphorically, assuming that you are in the US), your car's footprint is in my way, and in a collision your 4 tons of scrap metal will do an inane amount of damage to the colision partner.
Using a reasonable tool for a given task should be as self-evident in traffic as it is in programming.
In fact, LISP compilers have come a long way. For many tasks, compiled LISP or Scheme is as fast or faster than C or Fortran. We are currently reimplemting a high-performance prover in Scheme, and we have noted a 30 % speed increase compared to the old C version (which became a nightmare to maintain). Functional programs are not slow.
I recently read an interesting paper doing a comparison between LISP, Java and C++. That paper came to the conclusion that a good LISP implementation will usually be only very slightly slower than a good C++ implementation. However, an average LISP implementation usually is faster than an average C++ implementation. As for Java, forget about it;-)
But there's a reason they haven't caught on for non-trivial real world applications.
Emacs
AutoCAD
I write highly performance-oriented code in C. However, in Scheme I am a lot more productive for most other issues (amd I have a lot more experience with C than with Scheme). Map, lambda, and tail recursion are very powerful tools.
This claim "not a democracy, but a republic" is getting stupid. Yes, there is a web site that quotes a definition of "republic" with rather positive connotations from Webster's and a definition of "democracy" with rather negative connotations from some army handbook. Did you ever guess why they did not quote Webster's on democracy?
From Merriam-Webster Online (my printed copy is away from the computer), on the vile democracy:
Main Entry: de-moc-ra-cy Function: noun Inflected Form(s): plural -cies Etymology: Middle French democratie, from Late Latin democratia, from Greek dEmokratia, from dEmos + -kratia -cracy Date: 1576 1 a : government by the people; especially : rule of the majority b : a government in which the supreme power is vested in the people and exercised by them directly or indirectly through a system of representation usually involving periodically held free elections 2 : a political unit that has a democratic government 3 capitalized : the principles and policies of the Democratic party in the U.S. 4 : the common people especially when constituting the source of political authority 5 : the absence of hereditary or arbitrary class distinctions or privileges
And for your precious republic:
Main Entry: re-pub-lic Function: noun Etymology: French republique, from Middle French republique, from Latin respublica, from res thing, wealth + publica, feminine of publicus public -- more at REAL, PUBLIC Date: 1604 1 a (1) : a government having a chief of state who is not a monarch and who in modern times is usually a president (2) : a political unit (as a nation) having such a form of government b (1) : a government in which supreme power resides in a body of citizens entitled to vote and is exercised by elected officers and representatives responsible to them and governing according to law (2) : a political unit (as a nation) having such a form of government c : a usually specified republican government of a political unit "the French Fourth Republic" 2 : a body of persons freely engaged in a specified activity "the republic of letters" 3 : a constituent political and territorial unit of the former nations of Czechoslovakia, the U.S.S.R., or Yugoslavia
And if you look at the roots of the words, Republic means "Public thing", Democracy means "Rule by the people". The meanings of the two words as used to describe a political system overlap so strongly that in nearly any context they can be used interchangably.
Borland has no claim to your source code, hence you need no license from them to distribute it in the first place. The license above is ambiguous, but what probably is intended (and legally enforcable) is that you can distribute Borlands libraries et al only as a part of your compiled executable.
It's still fairly stupid, but such is the world of commercial software.
Let's face it, we don't get treated to descriptions of Stallman and Torvaaldes' dress sense and grooming.
Perhaps you don't. I have seen quite some of these. In addition to Stallman's ever-popular T-Shirts and hair style (kind of like Samson's, I suppose), I have seen more than one article commenting on Linus' fine sense of style, in particular his excellent combination of white socks with sandals.
Spam actually, really, from AOL accounts, or spam with "@aol.com" forged into the headers?
How much spam do you get per user? How does this compare to other ISP's?
I have observed three different phases. First, a lot of spam came from AOL throwaway-accounts. This never quite stopped, but came down to a dribble. After that, very many spammers faked AOL from-addresses. However, in the last couple of days I got a lot of spam from users with what looks like dynamically allocated AOL IP-addresses. I suspect that this new wave of spam triggered ORB.
In the last two weeks, I got some spam via an open relay in Spain, some via some obscure mail servers in China, and about 50% via AOL. I have no idea about the number of users in China (1.2 billion? *grin*) or using the Spanish ISP (actually, it was a bank), but AOL certainly is one of the major spam sources at the moment.
I expect AOL to improve fast - while their user base still sucks, they did get a quite good support team.
You can call Stallman a lot of lot of things, some of which I even might agree with (bone-headed comes to mind). However, he is about the least hypocritical person on this or any other known planet. While the FSF is selling the emacs manual (as well as emacs itself), it is freely available from this ftp site or any number of other GNU mirrors, and comes with the permission to copy and modify.
Yes, you can I print verbatim copies of RMS's Emacs manual and sell it for less than the FSF charges for the original. However, RMS is of course free to sign or not sign any copy, depending on his fancy. In fact, what he did is quite common: Using your popularity to raise money for your favourite charity.
Did somebody else visit their website with Javascript enabled? First thing that happens is a new window wich contains nothing but a single form field, a submit button, and the request to "sign our online petition". And in case you forego this great opportunity, you find a similar field and button on the home page, still without any explanation about what exactly you are supporting.
Criticizing some aspects of US society is not equivalent to "US bashing". There is plenty of stupidity in most gouvernments, and I reserve the right to point this out regardless of which country is affected.
That said, nearly all western European countries have large numbers of potential immigrants despite the fact that, contrary to the US, they usually do not consider themselves as immigration countries and hence have much stricter requirements for permanent resident or citizen status.
More likely they've got a custom built supercomputer runnning a custom OS, to squeeze every last cycle into effiency when running huuuuuge decrypts and other spookish stuff.
Only if they are stupid. Today is usually is much cheaper to throw additional hardware at a problem than to try to optimize the operating system. For compute-intensive tasks like data analysis and code cracking, I seriously doubt that the operating system even accounts for 5% overhead. Squeezing those 5% out of the code is prohibitively expensive, particularly for one-shot installations.
Uh, if we can solve problems in NP as if they were all in P, that's bad news for users of discrete one-way functions like modular exponentiation. . ..
Not necessarily. If the lower complexity bound for a given NP problem turns out to be O(P(n)^2012) instead of O(2^Q(n)) (for some poynomials P and Q), most problems will still be "hard" to crack in the practical sense.
Please note that factoring into primes is not an NP-complete problem. It is in NP and there does not exist any known deterministic polynomial algorithm for it (so it is not known to be in P).
It is one of the very few candidates for problems neither in P nor in NPC, but proving it to be in P will not collapse NP to P. Proving that there is no problem in (NP\NPC)\P, however, will show that P==NP.
I do not now what everyone thinks. However, it is a fairly simple conclusion of special relativity (no need to go to general relativity) that any faster-than-light communication is incompatible with causality ("implies time travel", if you prefer it that way).
For the most simple case consider a space-time diagram with two different inertal frames of reference and assume instant communication. You can immediately see that a message with infinite speed in one frame of reference travels back in time in the other one.
"The problem with Solaris" is indeed on low end machines. However, "low end" for SUN machines means a 40 MHz SparcStaiton Classic with 16 MB or something like that - machines that usually are 5 or more years old (and have usually been running more or less continously), but still work perfectly.
Solaris works perfectly fine on any UltraSPARC, including the 143 MHz Ultra 1 with 64 MB. Linux may beat it out in single processor benchmarks, but I doubt that you will notice significant differences in real usage.
Just a quick note: The verdict against the Compuserve manager was struck down after a double appeal (both prosecution and defense argued for the case to be dismissed) about two weeks ago.
Well, "Solaris includes SunOS" only if you buy into SUN's historical revisionism. I have started with SunOS 4.x on SUN 3/50 and SUN 3/60, and experienced the transition to SunOS 5.x. "Solaris 1" is a pure retroactive marketing idea - I have never heard the word Solaris before the SYSV based Solaris 2.0.
The artice is quite well done, but it does have a couple of historical gotchas. The one thing I noted most is the bungled history of the Emacs/XEmacs split. Lucid Emacs was a split from a very early GNU Emacs 19 version (at a time where everyone was using Emacs 18, the first one with some X-support). Due to this, Lucid Emacs (later XEmacs) always was GPL'ed. Pre-GPL Emacs versions are fairly rare, the canonical example is Gosmacs. I don't know if there still is a semi-current non-free Emacs version around.
Well, it will increase the markedt in terms of the number of potential viewers. But if illegal copying becomes widespread, it reduces the markets value - which is what most companies are interested in. Given that nearly all other media are copyable already, and that e.g. the CD market is alive and well, I doubt that this will happen. But this is due to other factors - primarily copyright enforcement by legal and moral (as opposed to technical) means.
Perhaps it wouls be more appropriate to play up the reverse engineering aspect of this. Is it illegal for a non licenced manufacturer to design and sell replacement door panels for your car? Of course not! What the manufactures of those door panels do is exactly the same as what these people did.
Yes, they had to break some (pretty weak, and bungled) encryption, but is that any different from the door manufacture not releasing the specifications of a special bolt needed to attach the door to the car? Not really - and it was perfectly legal to do.
I'm not going to defend the DVD consortium. However, the situations are not fully analogical. One of the reasons for creating free software is that information is not a natural property, but only an artificial one. This is due to the fact that making additional copies has little cost. The maximum benefit that society can get from a piece of information is obtained if it is free.
The same argument works the other way. If I believe in IP, I have to protect it adequatly. Reverse engineering a panel means I still have to be competitive in producing and marketing it - there are significant real costs. Breaking the DVD protection means that (barring other copying problems) the market for content breaks down.
Slashdot Mirror
What is compared is also interesting. Moody's data about Linux vulnerabilities is not about the kernel or the core system. It is not even about a single Linux distribution. It is not even computed correctly. The numbers given by Moody are for the union of all vulnerabilities in all Linux distributions covered by the statistics, and to make it look worse he adds vulnerabilities of Red Hat in once more for good measure.
Moreover, if you consider that Linux covers allpackages in all distributions and hence contains a lot more software than the standard Windows (NT or 98) distribution, the number become even more meaningless.
Basically, the article is a bad example of how to lie with statistics.
Using a reasonable tool for a given task should be as self-evident in traffic as it is in programming.
I recently read an interesting paper doing a comparison between LISP, Java and C++. That paper came to the conclusion that a good LISP implementation will usually be only very slightly slower than a good C++ implementation. However, an average LISP implementation usually is faster than an average C++ implementation. As for Java, forget about it ;-)
Emacs
AutoCAD
I write highly performance-oriented code in C. However, in Scheme I am a lot more productive for most other issues (amd I have a lot more experience with C than with Scheme). Map, lambda, and tail recursion are very powerful tools.
From Merriam-Webster Online (my printed copy is away from the computer), on the vile democracy:
Main Entry: de-moc-ra-cy
Function: noun
Inflected Form(s): plural -cies
Etymology: Middle French democratie, from Late Latin democratia, from Greek dEmokratia, from dEmos + -kratia -cracy
Date: 1576
1 a : government by the people; especially : rule of the majority b : a government in which the supreme power is vested in the people and exercised by them directly or indirectly through a system of representation usually involving periodically held free elections
2 : a political unit that has a democratic government
3 capitalized : the principles and policies of the Democratic party in the U.S.
4 : the common people especially when constituting the source of political authority
5 : the absence of hereditary or arbitrary class distinctions or privileges
And for your precious republic:
Main Entry: re-pub-lic
Function: noun
Etymology: French republique, from Middle French republique, from Latin respublica, from res thing, wealth + publica, feminine of publicus public -- more at REAL, PUBLIC
Date: 1604
1 a (1) : a government having a chief of state who is not a monarch and who in modern times is usually a president (2) : a political unit (as a nation) having such a form of government b (1) : a government in which supreme power resides in a body of citizens entitled to vote and is exercised by elected officers and representatives responsible to them and governing according to law
(2) : a political unit (as a nation) having such a form of government c : a usually specified republican government of a political unit "the French Fourth Republic"
2 : a body of persons freely engaged in a specified activity "the republic of letters"
3 : a constituent political and territorial unit of the former nations of Czechoslovakia, the U.S.S.R., or Yugoslavia
And if you look at the roots of the words, Republic means "Public thing", Democracy means "Rule by the people". The meanings of the two words as used to describe a political system overlap so strongly that in nearly any context they can be used interchangably.
It's still fairly stupid, but such is the world of commercial software.
Perhaps you don't. I have seen quite some of these. In addition to Stallman's ever-popular T-Shirts and hair style (kind of like Samson's, I suppose), I have seen more than one article commenting on Linus' fine sense of style, in particular his excellent combination of white socks with sandals.
In the last two weeks, I got some spam via an open relay in Spain, some via some obscure mail servers in China, and about 50% via AOL. I have no idea about the number of users in China (1.2 billion? *grin*) or using the Spanish ISP (actually, it was a bank), but AOL certainly is one of the major spam sources at the moment.
I expect AOL to improve fast - while their user base still sucks, they did get a quite good support team.
<sarcasm>Sorry, you can't do that. DeCSS is illegal. You can't even look at the code, because it's banned from all web sites.</sarcasm>
Yes, you can I print verbatim copies of RMS's Emacs manual and sell it for less than the FSF charges for the original. However, RMS is of course free to sign or not sign any copy, depending on his fancy. In fact, what he did is quite common: Using your popularity to raise money for your favourite charity.
Furfu!
That said, nearly all western European countries have large numbers of potential immigrants despite the fact that, contrary to the US, they usually do not consider themselves as immigration countries and hence have much stricter requirements for permanent resident or citizen status.
Only if they are stupid. Today is usually is much cheaper to throw additional hardware at a problem than to try to optimize the operating system. For compute-intensive tasks like data analysis and code cracking, I seriously doubt that the operating system even accounts for 5% overhead. Squeezing those 5% out of the code is prohibitively expensive, particularly for one-shot installations.
Not necessarily. If the lower complexity bound for a given NP problem turns out to be O(P(n)^2012) instead of O(2^Q(n)) (for some poynomials P and Q), most problems will still be "hard" to crack in the practical sense.
It is one of the very few candidates for problems neither in P nor in NPC, but proving it to be in P will not collapse NP to P. Proving that there is no problem in (NP\NPC)\P, however, will show that P==NP.
Unfortunately, this does not seem to be easy.
For the most simple case consider a space-time diagram with two different inertal frames of reference and assume instant communication. You can immediately see that a message with infinite speed in one frame of reference travels back in time in the other one.
No, I'm not going to fraw this in ASCII *grin*.
Solaris works perfectly fine on any UltraSPARC, including the 143 MHz Ultra 1 with 64 MB. Linux may beat it out in single processor benchmarks, but I doubt that you will notice significant differences in real usage.
NFS
YP/NIS
I don't think there is any commercial UNIX vendor that does not profit from these things.
Just a quick note: The verdict against the Compuserve manager was struck down after a double appeal (both prosecution and defense argued for the case to be dismissed) about two weeks ago.
Well, "Solaris includes SunOS" only if you buy into SUN's historical revisionism. I have started with SunOS 4.x on SUN 3/50 and SUN 3/60, and experienced the transition to SunOS 5.x. "Solaris 1" is a pure retroactive marketing idea - I have never heard the word Solaris before the SYSV based Solaris 2.0.
The artice is quite well done, but it does have a couple of historical gotchas. The one thing I noted most is the bungled history of the Emacs/XEmacs split. Lucid Emacs was a split from a very early GNU Emacs 19 version (at a time where everyone was using Emacs 18, the first one with some X-support). Due to this, Lucid Emacs (later XEmacs) always was GPL'ed. Pre-GPL Emacs versions are fairly rare, the canonical example is Gosmacs. I don't know if there still is a semi-current non-free Emacs version around.
Well, it will increase the markedt in terms of the number of potential viewers. But if illegal copying becomes widespread, it reduces the markets value - which is what most companies are interested in. Given that nearly all other media are copyable already, and that e.g. the CD market is alive and well, I doubt that this will happen. But this is due to other factors - primarily copyright enforcement by legal and moral (as opposed to technical) means.
The same argument works the other way. If I believe in IP, I have to protect it adequatly. Reverse engineering a panel means I still have to be competitive in producing and marketing it - there are significant real costs. Breaking the DVD protection means that (barring other copying problems) the market for content breaks down.