The short answer is the modern java garbage collector is tuned to deal with lots of small, short lived objects so odds are you're all right. An interesting thing to think about is when you do invokeLater(), a common style is to create a new Runnable().. I don't know if you're doing this but if you are then you're creating a new Runnable object everytime anyway.. Also know that there is some overhead to switching threads.
If you want to now where you're being in-efficient, you've got to profile, baby... Don't guess!!!! find out for sure.
( You might have a forgotten sleep statement somewhere.. I've seen this quite a few times actually !:) )
If it's not functional it will have very good support for many of aspects of a functional language. One of the first things one realizes after dealing with concurrency for a while is immutable objects are a good thing. Mutable states are bad..
I can't see automatic parallelization happening for a long time. Too hard.. What I expect to see is a trend toward functional style programming.. Also maybe things like composable memory transactions.. Maybe thread based synchronization (which is a term I've been using to describe a technique I've been using. It's hard to explain but works quite nicely...)...
Javascript is netscape's (and now mozilla's) implementation of ECMAScript. If you want to become a master of the language then reading the ECMAScript's spec is an excellent way of doing it since both Javascript and JScript (explorer's implementation of ECMAScript) will follow the EMCAScript spec more than less...
The really difficult thing with javascript (I really mean ECMAScript but meh. It's common usage to say javascript when really talking about EMCAScript since ECMAScript is a pain in the ass to say) is the DOM. The DOM spec is maintained by these guys: http://www.w3.org/DOM/.. The two major browsers actually implement most of the DOM and implement it the same way. So you can write to the spec on the w3's web site and not worry about (too much) about compatibility.. Well, some things aren't very well supported, like events, but most of the official DOM spec is supported by both major web browsers anyways..
No one should be modding up the parent. It is offtopic!
There are only theories in science. The best you can get is to show that X theory models the real world (or "serves as a model for the real world") to some extent or other.
The main difference between the heliocenter solar system model and evolutionairy model, is that the solar system is a linear / closed system in the time scale we're observing and the level of detail the model describes. This means that any errors will add up over time in a nice linear way. This makes it quite powerfull and impressive because it can predict things with a great deal of accuracy. Evolution, on the other hand, is trying to model a none-linear / open system, which mean you can't use the theory to make perticularly accurate predictions because of the enherent nature of the phenomenon it's describing (evolution in the natural world is a chaotic system in a mathematical sense).
Evolution has been proven in the sense that there have yet to be any observations that can be shown to contradict it. Which is generally how most if not all scientific theories are proven.
Absolutely. RSI is a very bad thing. Possibly career ending injury.
I have, for the first time in my life, experienced RSI. It's nasty. About 8 weeks ago I had some tingling in my arms. I started to ease off the computer and by the weekend my arms felt fine again so I did some coding on Sunday. On Monday I was in pain. I could barely move my right hand. I couldn't believe it. Out of nowhere, suddenly my right hand was nearly useless. It had never done this before. It had always been a nice hand and did what it was told. But now it didn't. I stopped using the computer for a week (this is quite hard to do when you're a professional programmer). When Friday came around and my right hand was still not better (in fact my left hand was starting to hurt) I decided to see a doctor.
At the end of every week I kept thinking: "This can't possibly go on for another day! It's just freaking muscle pain! I've had worse than this...".. I mean, I am pretty athletic too. I am doing some sport everyday: x-country skiing, squash, hockey... I do 160kms bike rides in the summer.. but the pain did go on and on and on and so I started taking it seriously and going to psychical therapy. It has taken until now (2 months later!!) to get back enough endurance to go back to a 40 hour work week (well, we'll see). I used to do around 60 hours of computing a day (my day used to consist of computing, doing some sport and sleeping:-). )
What seemed to happen was my hands/lowers arms just became so weak that any type of movement damaged the muscle and tendons. The solution consisted of building up the muscle and tendons using various techniques. It worked and, as a bonus, I am now much stronger... I can hit the ball very hard now in squash and have more control to boot.
Anyway: the moral is RSI is a very serious problem. Go get help.
I have my doubts that people who say they just changed mousing hands and the problem went away have fixed much. If the problem comes back if you switch the mouse back then you probably still need to increase the strength of your hands. While you're at it increase the strength of your arms and back too. My local gym (the one at which I play squash) also has a fitness center with treadmills and rowing machines and weights etc.. I asked them about a training regiment that could prevent common office injuries and they set me up. I have made it a part of my routine. I consider it the price I have to pay for being a programmer. Some jobs kill your back, or your feet.. Some jobs can get you killed. Programming can ruin your arms and turn you into a pear - it's an occupational hazard.
From what I gather from your post, you're taking the position that Linux, as an OS, is more focused on technical achievements or software listener idealism than pragmatic end-user concerns, for that matter, broad apeal. While I have to say that this is more true than it is false, I don't believe that Linux would have been better off it did try to be an end-user or broad apeal OS. The strength in Linux vs microsoft windows is the difference focus between the two OSes. Consider this:
Developement resources are not infinite and building a mass market OS is extremely difficult.
One of the most important features of a mass market OS is how well it's supported. By that I mean, how easy is it to get help? Will this or that device work on it? Is it likely my problem has been solved before using this system? How safe a choice is it?.. etc... In fact, I would argue that this is probably the most important aspect of an OS for 95% of the total potential market (see "crossing the chasm" http://www.amazon.com/exec/obidos/tg/detail/-/0060 517123/103-2472006-2694248?v=glance) . If Linux was positioned purely as a mass-market OS it would have spent much of its development time trying to get, and keep things running and the things that draw new people to the platform today would probably not have been developed.
Linux has grabbed something very valuable: a niche. It got this niche because of who built it. The people who built it did so because of how it was developed. It will expand in this niche because it is the right tool for the job.
That said, what other niches will it expand to? The server, developer, uber-geek market is substantial, but not the whole market... It's something to watch. Linux and its nebulous cloud of distributions has my moral support. I gotta say though, even though I quite squarely fit into the developer and uber-geek categories, I still prefer to use either of MacOS X and Windows 2000 when I'm at home... I just hate the command line.
It's rather a your post is marked as flamebait. I don't agree with that.
the Java API version numbering scheme, from what I can tell, works like this
The first digit denotes platform changes that break backwards compatibility
The second digit denote platform changes that break forward compatibility.
The last digit details changes in implementation (in the java code)
the _0X digit denotes changes in the VM native / compatibility code.
Unfortunately, the public does not understand this versioning scheme since it's not how things usually work. Things usually work
First number - get REALLY excited Second number - we made a bug fix you should get exited about Last number - woop, we made a small goof, we fixed it here bug don't get too excited.
Since Java's none-backwards compatible changes line up nicely with the releases people should get exited about they are calling the java 1.5.X branch java 5.
Syntactic sugar is a language feature that only makes the code look nicer and / or to saves some typing. To say that some new language feature is NOT syntax sugar because it makes the code shorter or nice is silly. That's the whole point of syntac surgar.
Here are two examples of language features ->
The for loop interractor construct is syntactic sugar because you could do an iterator loops before. The only difference is now it's shorter to write and makes the loop look nicer.
The generics feature is not syntatic sugar because it adds to java the possibility for the compiler to track type information through generic containers. You couldn't do that before. Now you can. So it's not syntax sugar.
This has to stop. Guys, you don't have to set all references to null. Setting references to null is only needed in one special case (see links below) and even then the reasone for it should be obvious (you're retaining a reference). Garbage collectors can determin which objects are reachable. You don't need to tell them.
That does neither. Java has a max heap size. Deaful is 64 megs. It will grow 'till whatever the max heap size is then just throw out and out of memory exception and kill the thread.
Having used both mac and PCs I'm quite surprised anyone actually cares about dialog box button order. I find the close/minimize/resize (maximize)) button order differences between macOS 9, MacOS X and windows much more jarring.
Then there are things like ctrl-q not quitting all apps and ctrl-w not closing all windows while alt-F4 does.. alt-tab working slightly differently between all platforms.. the tendency of auto-selecting everything in a field when activated.. copy-paste action having to do with the active selection... apps not bringing all relevent windows forward when one is activated... empty space in a List causing or not causing all items to deselect.. etc.. yuck... and people are complaining about button order.. woop teee dooo..
There is no remote exploit to a machine that is not connected to any network, likewise the only security holes you have are the ones you make.
This is in contrast to anti-copy technology where some third party is giving you information but doesn't want you to copy it. If you can read the information, you must be able to copy it. Anti-copy techology usually relies on trying to break the reader in certain controlled ways by taking advantage of common limitations of the perticular reader or device.. There's also the use of encryption, which tends not to work since you need to load the decryptor program to be able to read the data and you have access to programs that have been loaded into RAM by using debuggers..
Really copy-techology is trying to make using the information you have difficult whereas security technology is trying to preserve the-lack-of-ability you had to begin with. Basically trying to root out un-intened concequences... aka bugs that give the persone on the other end of the connection more capabilities then ou thought..
The everyday version gets 60mpg does it? What car might this be? Some sort of miracle car? or is it an insight? You know on Mars we get 100mpg daily. Then again the gravity is only a fourth of what it is on earth so we just pick up our cars and throw them.
Oh yeah, I was rereading your post, a firewall wouldn't be necessary if you shut off both the server and workstation services since Windows won't accept remote connections but it is reasonable to have a firewall on every home system. XP already does this, the SP2 firewall is actually something that works too.
__
I don't expect the end users to install anything. I will push all the patches out with my SUS server after I test them. Its good policy for any OS since I've seen updates disrupt Oracle on my linux boxes.
==
I was referring to people without an admin. People on home LANs. Admin centred solutions need not apply.
__
net stop server does not close all ports. It's also a command line trick no end user is going to know. Similar to turning off port 445 by using regedit or netBios by doing the netBios dance.
Disabling all network activity is silly I might as well unplug the cable.
==
I'll add that all these worm are irrelevant, no workstation on any of my networks has ever gotten any of them because I run something so simple as a firewall. Might add I also make normal users for everyone. No one runs admin unless they are installing something. This has prevented most of the spyware from getting on their systems and the only final step is getting them all to use Firefox.
__
Irrelevant to what? They cause outages to all sort of none-you services both public and private. They cause performance problems over the whole internet. They cause neighbours and friends of mine grief. They sound very relevant to me.
No admin makes them the admin, so that's not and option.
Spyware is a separate plague. And off topic for now.
==
Oh yeah, I was rereading your post, a firewall wouldn't be necessary if you shut off both the server and workstation services since Windows won't accept remote connections but it is reasonable to have a firewall on every home system. XP already does this, the SP2 firewall is actually something that works too.
__
Firewalls or NATs screw up or make impossible a great deal of usefull things. Running an FTP or sFTP server, running an Appleshare server, running an SSHd, running most p2p software, both file sharing and other (without some of the spooky work around those guys use) running most network games, running a web server etc... There is.. or maybe the correct temr now is "was" a great deal of potential to be exploited in the p2p nature of the internet. NATs are destroying this potential.
There comes a point where the concept of a "firewall" and the concept of "sufficient control over what has ports open to the outside" merge. Fine. That is beside the point. ->
- Releasing an effectively ubiquitous operating system written in C/C++ with a bunch of ports open is stupid. - Expecting end users (that's adminless end-users) to do many things that admins consider routine maintenance is unrealistic.
Therefore it's microsoft's fault even if a patch is available.
- Firewalls and NATs in a home environment instead of closing all ports is the wrong solution. - There are many apps that work badly or not at all behind firewalls and NATs. There are many applications that require the p2p nature of the internet. - Doing the wrong thing (firewalls/nats) is actually easier than doing the right thing (turning off software and ports that you don't need).
Therefore I don't like the idea that these problems are the user's fault because they aren't using NATs or firewalls.
No, it's Windows and the just-install-the-patch mentality that causes all the problems. Ugh, I'm so tired of hearing all this admin / end user bashing shit. Creating a program is one thing, having the ability to auto-install it on every computer in the world is another. Windows became a problem when it took an attitude towards network security that was more appropriate to a marginally popular third party software product running on a LAN. All these worms would be irrelevant if windows shipped with 1) no network ports open, 2) Actually allowed the user to close all ports by closing all the services (servers) rather than running the services but adding another layer of software that blocks off the ports. Using a firewall or NAT to block of ports is not sensible-security-precaution for a home LAN. It's a work around to a combination of bugs in the OS / third party software, stupid default settings and a lack of control over what ports are open or for that matter which applications/services are allowed network access.
I suggest they provide updates for the pirate users merely because the OS is capable of harming others if unchecked. I don't want zombie computers disrupting my networking functionality.... again....
In the end, expecting an end user to install a patch is not a reasonable expectation. Expecting a user to install and configure a piece of software that blocks ports that shouldn't be open in the first place is not a reasonable expectation. I don't have an admin in my house. I have no interest in applying patches. Security bugs are not my fault or -responsibility- and I resent the implication they are.
Yes, you're completely right. When I wrote my comment I was thinking primarily about flaws in software products or dumb system default on a typical home user's PC.
For large software sites there can be a disconnect between security fantasy and security reality. On the wall of my cubical is a Dilbert cartoon and a memo for IT. The subject of both is password policy. The reason they are on my wall is the new security policy of IT and the fictional one proposed by the IT person in the Dilbert cartoon are identical. I keep both on my wall to remind myself of the chasm that exists between the average user and someone trying to implement a standard security policy. It's similar to the situation between 128-bit encryption and a typical pin number. You have 128 bit string encryption protecting a 30 odd digit card number protected by a 4 digit pin number. It's completely stupid. Reportedly, the pin number is only 4 digits long because anywhere above that size and people started to do things like write it down or forget it consistently or generally make a big fuss. oddly enough, I'm guessing that people made a big fuss about low encryption too.
It's an annoying reality that any security policy has to be compatible with not only all digital hardware on the network but also with the people and their irrational personalities. The sad truth is, legislation is unlikely to correct the problem. I propose massive, systemic, amateur, brain transplants! or perhaps the "store all data in write only memory" trick... Yes, it's a hard problem. Write if you find a workable solution. We'd all like to know:-)...
Well, I might believe that if there were fewer security issues and warnings.
Shipping an OS with ports open is not a prudent security decision.
Shipping an OS with ports open with no way to close them save installing an extra piece of software called a "firewall" is infuriating.
An attitude of security through obscurity a software firm whose software products run on 90% of all desktop computers is naive.
Using an environment that allows the programmer to make an error that allows a hostile data packet to corrupt memory without even so much as a warning is foolish.
Continuing to use said environment after repeated (read hundreds if not thousands) vulnerabilities are discovered in all manner of software is totally irresponsible.
In my mind, the best thing that would come out of making businesses liable for their security failures would be that these businesses would start to demand systems that were designed with security in mind.
You see, the problem isn't simply that people aren't applying patches. The problem is that software is being released without security in mind. Leaving ports open unnecessarily, not letting a user lock down their own machine, creating an operating environment so prone to virus exploits, using C/C++ inappropriately when dealing with potentially hostile IO data etc... represent the root causes of the current batch of problems. For leaders in software industry to be critical of a user for not installing a patch is, in my mind, hypocrisy of the highest order. This is why I say, this idea of making users responsible for failures in a vendor's software is backwards.
So the people that use the software should assume liability for not patching holes but the manufacture assumes no responsibility for leaving security holes in their product to begin with? This sound very backwards to me.
Given that the EU now how a GDP larger than the US, I would say that the EU is now a more influential economic entity than the US. Given the might-makes-right attitude of your post I would expect the US will eventually have to cave into the trade demands of the EU from now on. That would include changing any cultural eccentricities that conflict with EU trade interests. It would greatly expedite things if the US would just stop arguing and become a team player... this is all by your logic, of course.. I, for one, do not hold this view as I find it morally irresponsible, but I could certainly see how it would appeal to citizens of the EU. I don't see why it would appeal to a US citizen, however, unless they were ill-informed about the current state of world economic affairs or possibly masochistic.
(The above comments should be viewed as satirical)
Slashdot Mirror
The short answer is the modern java garbage collector is tuned to deal with lots of small, short lived objects so odds are you're all right. An interesting thing to think about is when you do invokeLater(), a common style is to create a new Runnable().. I don't know if you're doing this but if you are then you're creating a new Runnable object everytime anyway.. Also know that there is some overhead to switching threads.
:) )
If you want to now where you're being in-efficient, you've got to profile, baby... Don't guess!!!! find out for sure.
( You might have a forgotten sleep statement somewhere.. I've seen this quite a few times actually !
If it's not functional it will have very good support for many of aspects of a functional language. One of the first things one realizes after dealing with concurrency for a while is immutable objects are a good thing. Mutable states are bad..
I can't see automatic parallelization happening for a long time. Too hard.. What I expect to see is a trend toward functional style programming.. Also maybe things like composable memory transactions.. Maybe thread based synchronization (which is a term I've been using to describe a technique I've been using. It's hard to explain but works quite nicely...)...
Javascript is netscape's (and now mozilla's) implementation of ECMAScript. If you want to become a master of the language then reading the ECMAScript's spec is an excellent way of doing it since both Javascript and JScript (explorer's implementation of ECMAScript) will follow the EMCAScript spec more than less...
http://en.wikipedia.org/wiki/JavaScript (first line)
The really difficult thing with javascript (I really mean ECMAScript but meh. It's common usage to say javascript when really talking about EMCAScript since ECMAScript is a pain in the ass to say) is the DOM. The DOM spec is maintained by these guys: http://www.w3.org/DOM/.. The two major browsers actually implement most of the DOM and implement it the same way. So you can write to the spec on the w3's web site and not worry about (too much) about compatibility.. Well, some things aren't very well supported, like events, but most of the official DOM spec is supported by both major web browsers anyways..
No one should be modding up the parent. It is offtopic!
Ah yes, Amish OS 1.0.
Alternatively you can unplug the three pronged virus enabler device that runs from every computer to the electrical socket.
Are they?
There are only theories in science. The best you can get is to show that X theory models the real world (or "serves as a model for the real world") to some extent or other.
The main difference between the heliocenter solar system model and evolutionairy model, is that the solar system is a linear / closed system in the time scale we're observing and the level of detail the model describes. This means that any errors will add up over time in a nice linear way. This makes it quite powerfull and impressive because it can predict things with a great deal of accuracy. Evolution, on the other hand, is trying to model a none-linear / open system, which mean you can't use the theory to make perticularly accurate predictions because of the enherent nature of the phenomenon it's describing (evolution in the natural world is a chaotic system in a mathematical sense).
Evolution has been proven in the sense that there have yet to be any observations that can be shown to contradict it. Which is generally how most if not all scientific theories are proven.
Absolutely. RSI is a very bad thing. Possibly career ending injury.
:-). )
I have, for the first time in my life, experienced RSI. It's nasty. About 8 weeks ago I had some tingling in my arms. I started to ease off the computer and by the weekend my arms felt fine again so I did some coding on Sunday. On Monday I was in pain. I could barely move my right hand. I couldn't believe it. Out of nowhere, suddenly my right hand was nearly useless. It had never done this before. It had always been a nice hand and did what it was told. But now it didn't. I stopped using the computer for a week (this is quite hard to do when you're a professional programmer). When Friday came around and my right hand was still not better (in fact my left hand was starting to hurt) I decided to see a doctor.
At the end of every week I kept thinking: "This can't possibly go on for another day! It's just freaking muscle pain! I've had worse than this...".. I mean, I am pretty athletic too. I am doing some sport everyday: x-country skiing, squash, hockey... I do 160kms bike rides in the summer.. but the pain did go on and on and on and so I started taking it seriously and going to psychical therapy. It has taken until now (2 months later!!) to get back enough endurance to go back to a 40 hour work week (well, we'll see). I used to do around 60 hours of computing a day (my day used to consist of computing, doing some sport and sleeping
What seemed to happen was my hands/lowers arms just became so weak that any type of movement damaged the muscle and tendons. The solution consisted of building up the muscle and tendons using various techniques. It worked and, as a bonus, I am now much stronger... I can hit the ball very hard now in squash and have more control to boot.
Anyway: the moral is RSI is a very serious problem. Go get help.
I have my doubts that people who say they just changed mousing hands and the problem went away have fixed much. If the problem comes back if you switch the mouse back then you probably still need to increase the strength of your hands. While you're at it increase the strength of your arms and back too. My local gym (the one at which I play squash) also has a fitness center with treadmills and rowing machines and weights etc.. I asked them about a training regiment that could prevent common office injuries and they set me up. I have made it a part of my routine. I consider it the price I have to pay for being a programmer. Some jobs kill your back, or your feet.. Some jobs can get you killed. Programming can ruin your arms and turn you into a pear - it's an occupational hazard.
From what I gather from your post, you're taking the position that Linux, as an OS, is more focused on technical achievements or software listener idealism than pragmatic end-user concerns, for that matter, broad apeal. While I have to say that this is more true than it is false, I don't believe that Linux would have been better off it did try to be an end-user or broad apeal OS. The strength in Linux vs microsoft windows is the difference focus between the two OSes. Consider this:
0 517123/103-2472006-2694248?v=glance) . If Linux was positioned purely as a mass-market OS it would have spent much of its development time trying to get, and keep things running and the things that draw new people to the platform today would probably not have been developed.
Developement resources are not infinite and building a mass market OS is extremely difficult.
One of the most important features of a mass market OS is how well it's supported. By that I mean, how easy is it to get help? Will this or that device work on it? Is it likely my problem has been solved before using this system? How safe a choice is it?.. etc... In fact, I would argue that this is probably the most important aspect of an OS for 95% of the total potential market (see "crossing the chasm" http://www.amazon.com/exec/obidos/tg/detail/-/006
Linux has grabbed something very valuable: a niche. It got this niche because of who built it. The people who built it did so because of how it was developed. It will expand in this niche because it is the right tool for the job.
That said, what other niches will it expand to? The server, developer, uber-geek market is substantial, but not the whole market... It's something to watch. Linux and its nebulous cloud of distributions has my moral support. I gotta say though, even though I quite squarely fit into the developer and uber-geek categories, I still prefer to use either of MacOS X and Windows 2000 when I'm at home... I just hate the command line.
It's rather a your post is marked as flamebait. I don't agree with that.
the Java API version numbering scheme, from what I can tell, works like this
The first digit denotes platform changes that break backwards compatibility
The second digit denote platform changes that break forward compatibility.
The last digit details changes in implementation (in the java code)
the _0X digit denotes changes in the VM native / compatibility code.
Unfortunately, the public does not understand this versioning scheme since it's not how things usually work. Things usually work
First number - get REALLY excited
Second number - we made a bug fix you should get exited about
Last number - woop, we made a small goof, we fixed it here bug don't get too excited.
Since Java's none-backwards compatible changes line up nicely with the releases people should get exited about they are calling the java 1.5.X branch java 5.
Syntactic sugar is a language feature that only makes the code look nicer and / or to saves some typing. To say that some new language feature is NOT syntax sugar because it makes the code shorter or nice is silly. That's the whole point of syntac surgar.
Here are two examples of language features ->
The for loop interractor construct is syntactic sugar because you could do an iterator loops before. The only difference is now it's shorter to write and makes the loop look nicer.
The generics feature is not syntatic sugar because it adds to java the possibility for the compiler to track type information through generic containers. You couldn't do that before. Now you can. So it's not syntax sugar.
This has to stop. Guys, you don't have to set all references to null. Setting references to null is only needed in one special case (see links below) and even then the reasone for it should be obvious (you're retaining a reference). Garbage collectors can determin which objects are reachable. You don't need to tell them.
a ry /j-jtp01274.htmle chTips/1997/tt0903 .html
Explicit Nulling:
http://www-106.ibm.com/developerworks/java/libr
http://java.sun.com/developer/T
That does neither. Java has a max heap size. Deaful is 64 megs. It will grow 'till whatever the max heap size is then just throw out and out of memory exception and kill the thread.
Having used both mac and PCs I'm quite surprised anyone actually cares about dialog box button order. I find the close/minimize/resize (maximize)) button order differences between macOS 9, MacOS X and windows much more jarring.
Then there are things like ctrl-q not quitting all apps and ctrl-w not closing all windows while alt-F4 does.. alt-tab working slightly differently between all platforms.. the tendency of auto-selecting everything in a field when activated.. copy-paste action having to do with the active selection... apps not bringing all relevent windows forward when one is activated... empty space in a List causing or not causing all items to deselect.. etc.. yuck... and people are complaining about button order.. woop teee dooo..
There is no remote exploit to a machine that is not connected to any network, likewise the only security holes you have are the ones you make.
This is in contrast to anti-copy technology where some third party is giving you information but doesn't want you to copy it. If you can read the information, you must be able to copy it. Anti-copy techology usually relies on trying to break the reader in certain controlled ways by taking advantage of common limitations of the perticular reader or device.. There's also the use of encryption, which tends not to work since you need to load the decryptor program to be able to read the data and you have access to programs that have been loaded into RAM by using debuggers..
Really copy-techology is trying to make using the information you have difficult whereas security technology is trying to preserve the-lack-of-ability you had to begin with. Basically trying to root out un-intened concequences... aka bugs that give the persone on the other end of the connection more capabilities then ou thought..
Waffles? What about Bacon and Eggs.
Yes, he means people who don't visit slashdot...
The everyday version gets 60mpg does it? What car might this be? Some sort of miracle car? or is it an insight? You know on Mars we get 100mpg daily. Then again the gravity is only a fourth of what it is on earth so we just pick up our cars and throw them.
r ticle.html
Top ten mpg cars:
http://www.edmunds.com/reviews/list/top10/47000/a
Oh yeah, I was rereading your post, a firewall wouldn't be necessary if you shut off both the server and workstation services since Windows won't accept remote connections but it is reasonable to have a firewall on every home system. XP already does this, the SP2 firewall is actually something that works too.
__
I don't expect the end users to install anything. I will push all the patches out with my SUS server after I test them. Its good policy for any OS since I've seen updates disrupt Oracle on my linux boxes.
==
I was referring to people without an admin. People on home LANs. Admin centred solutions need not apply.
__
net stop server does not close all ports. It's also a command line trick no end user is going to know. Similar to turning off port 445 by using regedit or netBios by doing the netBios dance.
Disabling all network activity is silly I might as well unplug the cable.
==
I'll add that all these worm are irrelevant, no workstation on any of my networks has ever gotten any of them because I run something so simple as a firewall. Might add I also make normal users for everyone. No one runs admin unless they are installing something. This has prevented most of the spyware from getting on their systems and the only final step is getting them all to use Firefox.
__
Irrelevant to what? They cause outages to all sort of none-you services both public and private. They cause performance problems over the whole internet. They cause neighbours and friends of mine grief. They sound very relevant to me.
No admin makes them the admin, so that's not and option.
Spyware is a separate plague. And off topic for now.
==
Oh yeah, I was rereading your post, a firewall wouldn't be necessary if you shut off both the server and workstation services since Windows won't accept remote connections but it is reasonable to have a firewall on every home system. XP already does this, the SP2 firewall is actually something that works too.
__
Firewalls or NATs screw up or make impossible a great deal of usefull things. Running an FTP or sFTP server, running an Appleshare server, running an SSHd, running most p2p software, both file sharing and other (without some of the spooky work around those guys use) running most network games, running a web server etc... There is.. or maybe the correct temr now is "was" a great deal of potential to be exploited in the p2p nature of the internet. NATs are destroying this potential.
There comes a point where the concept of a "firewall" and the concept of "sufficient control over what has ports open to the outside" merge. Fine. That is beside the point. ->
- Releasing an effectively ubiquitous operating system written in C/C++ with a bunch of ports open is stupid.
- Expecting end users (that's adminless end-users) to do many things that admins consider routine maintenance is unrealistic.
Therefore it's microsoft's fault even if a patch is available.
- Firewalls and NATs in a home environment instead of closing all ports is the wrong solution.
- There are many apps that work badly or not at all behind firewalls and NATs. There are many applications that require the p2p nature of the internet.
- Doing the wrong thing (firewalls/nats) is actually easier than doing the right thing (turning off software and ports that you don't need).
Therefore I don't like the idea that these problems are the user's fault because they aren't using NATs or firewalls.
Wasen't one of there "security updates" to enable the firewall by default? That would kind of make bit torrent useless.
- Large anti NAT / Firewall rant deleted -
- Large anti patch-and-forget-it rant deleted -
No, it's Windows and the just-install-the-patch mentality that causes all the problems. Ugh, I'm so tired of hearing all this admin / end user bashing shit. Creating a program is one thing, having the ability to auto-install it on every computer in the world is another. Windows became a problem when it took an attitude towards network security that was more appropriate to a marginally popular third party software product running on a LAN. All these worms would be irrelevant if windows shipped with 1) no network ports open, 2) Actually allowed the user to close all ports by closing all the services (servers) rather than running the services but adding another layer of software that blocks off the ports. Using a firewall or NAT to block of ports is not sensible-security-precaution for a home LAN. It's a work around to a combination of bugs in the OS / third party software, stupid default settings and a lack of control over what ports are open or for that matter which applications/services are allowed network access.
I suggest they provide updates for the pirate users merely because the OS is capable of harming others if unchecked. I don't want zombie computers disrupting my networking functionality.... again....
In the end, expecting an end user to install a patch is not a reasonable expectation. Expecting a user to install and configure a piece of software that blocks ports that shouldn't be open in the first place is not a reasonable expectation. I don't have an admin in my house. I have no interest in applying patches. Security bugs are not my fault or -responsibility- and I resent the implication they are.
Why bother? The mac mouse works on a PC. I can just give you the one that came with my mac. It's not like it's doing anything at the moment.
Yes, you're completely right. When I wrote my comment I was thinking primarily about flaws in software products or dumb system default on a typical home user's PC.
:-)...
For large software sites there can be a disconnect between security fantasy and security reality. On the wall of my cubical is a Dilbert cartoon and a memo for IT. The subject of both is password policy. The reason they are on my wall is the new security policy of IT and the fictional one proposed by the IT person in the Dilbert cartoon are identical. I keep both on my wall to remind myself of the chasm that exists between the average user and someone trying to implement a standard security policy. It's similar to the situation between 128-bit encryption and a typical pin number. You have 128 bit string encryption protecting a 30 odd digit card number protected by a 4 digit pin number. It's completely stupid. Reportedly, the pin number is only 4 digits long because anywhere above that size and people started to do things like write it down or forget it consistently or generally make a big fuss. oddly enough, I'm guessing that people made a big fuss about low encryption too.
It's an annoying reality that any security policy has to be compatible with not only all digital hardware on the network but also with the people and their irrational personalities. The sad truth is, legislation is unlikely to correct the problem. I propose massive, systemic, amateur, brain transplants! or perhaps the "store all data in write only memory" trick... Yes, it's a hard problem. Write if you find a workable solution. We'd all like to know
Well, I might believe that if there were fewer security issues and warnings.
Shipping an OS with ports open is not a prudent security decision.
Shipping an OS with ports open with no way to close them save installing an extra piece of software called a "firewall" is infuriating.
An attitude of security through obscurity a software firm whose software products run on 90% of all desktop computers is naive.
Using an environment that allows the programmer to make an error that allows a hostile data packet to corrupt memory without even so much as a warning is foolish.
Continuing to use said environment after repeated (read hundreds if not thousands) vulnerabilities are discovered in all manner of software is totally irresponsible.
In my mind, the best thing that would come out of making businesses liable for their security failures would be that these businesses would start to demand systems that were designed with security in mind.
You see, the problem isn't simply that people aren't applying patches. The problem is that software is being released without security in mind. Leaving ports open unnecessarily, not letting a user lock down their own machine, creating an operating environment so prone to virus exploits, using C/C++ inappropriately when dealing with potentially hostile IO data etc... represent the root causes of the current batch of problems. For leaders in software industry to be critical of a user for not installing a patch is, in my mind, hypocrisy of the highest order. This is why I say, this idea of making users responsible for failures in a vendor's software is backwards.
So the people that use the software should assume liability for not patching holes but the manufacture assumes no responsibility for leaving security holes in their product to begin with? This sound very backwards to me.
Given that the EU now how a GDP larger than the US, I would say that the EU is now a more influential economic entity than the US. Given the might-makes-right attitude of your post I would expect the US will eventually have to cave into the trade demands of the EU from now on. That would include changing any cultural eccentricities that conflict with EU trade interests. It would greatly expedite things if the US would just stop arguing and become a team player... this is all by your logic, of course.. I, for one, do not hold this view as I find it morally irresponsible, but I could certainly see how it would appeal to citizens of the EU. I don't see why it would appeal to a US citizen, however, unless they were ill-informed about the current state of world economic affairs or possibly masochistic.
(The above comments should be viewed as satirical)