The problem here is not just one of marketing. They have spread the idea that the BSA and Microsoft have been cramming down everyone's throats, that software should be evaluated on the merits of the product, ignoring the merits of the system involved in crating the product. This is like a construction company insisting that when a building project goes out to bid the only thing evaluated are price and the size of the building created with no constraints on the construction methods used, safety, environmental impact, or maintainability. They want eliminate concideration of evaluation criteria where they inevitably lose. The software companies are doing the equivalent of building the building in such a way that only they can repair it when it fails and ensuring that it will fail over time and when it does, they insist that it can't be repaired, it needs to be torn down and rebuilt. This is not the right model for infrastructure.
Software companies are afraid that governments are going to start to realize they have done the equivalent of paying a company to build their roads while giving them the rights to put tollbooths wherever they want. It's ridiculous. There is a lot of infrastructure software out there (operating systems, relational databases, office suites, etc.) that have been paid for 100 times over by governments and they still don't own them.
Yea. Except T1's do go down. Also, every ISP sells more downstream bandwith than they have upstream. A difference that makes no difference is no difference at all. When cable bandwidth is shared among neighbors it's never 2.5 mbit shared among a bunch of people, it's more like 100 mbit shared among a bunch of people who are each capped at 2.5 mbit. That's the same thing T1 providers do.
The real issue with cable is crap upstream bandwith. Comcast. Let me pay an extra $10/month for 1.5 mbit upstream and I'll do it!
All we need now is for the government to put up a DirecTV like satellite and use it like it uses the regular public airwaves. Anyone can broadcast but they can't charge for people to receive it. Then everyone in the country could have 300 digital channels and no monthly bill.
Then we get a little data over power lines technology and you can have your power company be your ISP and your phone company. Or start using a cellular Internet connection. Then the cable company would have to really work to get your internet business because it would be the only way they could make money!
Competition does make for a good market as long as there is lots of it.
There is no one answer. If security is your only concern you should have as many layers of security as possible with firewalls between each layer locked down as tight as possible. That said, security is never your only concern. Cost, ease of maintenance, performance, and flexibility are all important in a network design. After all, the purpose of your company is probably to get something accomplished, not to avoid getting hacked. There are times when every different network configuration is appropriate from super secure to a cable modem router to a windows box right on the internet. There is no one answer.
Application layer firewalls are another layer above port filtering. They can increase security and could, in theory, make it worthwhile to share a service hosted on a machine that is inside your network. I would only do that if you trusted the security of your internal network. Most network designs assume that once you get in to the "internal network" there is no more security and all your deepest company secrets are available to anyone browsing around. If this is true, you've probably made some bad decisions somewhere along the way and you should address those before you open any holes. If you are willing to maintain strict security on your internal network then the added simplicity of allowing Internet access to machines on it can be worth the risk. This can be a lot easer than setting up a dmz.
Usually layers do make sense though, even if one of the layers is just a Linux box doing firewalling, routing and serving some services. One thing I like to do is to mix operating systems at different layers. That way if you get a worm of some kind that gets into one layer it won't penetrate to the layer behind. For example, internet facing servers are Linux based, desktops are Windows based.
Another thing I have done when I absolutely needed a Windows based web server is to setup Apache as a reverse-proxy only forwarding requests to a particular subdirectory to the Windows server. This filtered out all the standard buffer overload attacks since none of them referred to that subdirectory name. It also made sure the requests were relatively well behaved and buffered outgoing data for the Windows box, reducing connection counts when it was under high load. This is an easy way to do an application layer firewall and if you are firewalling with a Linux box you can do it right on the firewall.
I take it back. Perl 5.8.1 doesn't bring a fix for the memory leaks. This program will still make perl blow up:
use threads; use threads::shared; my @shared : shared; @shared = (1) while (1);
Re:Can we use languages not by lunatics?
on
Perl 5.8.1 Released
·
· Score: 1
Great post. I epecially liked the Aristotle quote. I'll use that when trying to explain why it's not a bad idea to use Perl to a PHB some time.
Furthermore, If I were going to teach programming to anyone who was new at it I would absolutely, positively, teach them Perl. It's an awesome starter language and, unlike most good starter languages, it's a awesome full power language that you can build real usable software with.
There are too many languages now a days. Please eliminate 3. I am not a crackpot!
With Perl 5.8.1 we finally get usable threads that efficently communicate and cooperate. There is a fix for the memory leak that existed when using shared queues, arrays and hashes in Perl 5.8.0. This starts a whole new era for Perl programs. My little language is finnally growing up (sniffle.)
Perl has been a great language for people who just want to "get it done." It get's the job done and usually get's it done fast. It's great at making what would be a complicated thing easy. If you want to reference your song data by artist album and track number you just say $song_data{$artist}{$album}[$track_number]. What is happeneing behind the scenes is complicated, that's actually a hash of hashes of arrays possibly pointing at another data structure underneath but conceptually it all optimizes out to simply storing by artist album and track number. Most programmers aren't used to having that option. They don't understand it and when they go to figure it all out they get mired in the way the references are being stored and what is pointing at what and lose sight of the simplicity, efficiency, and power of the whole thing.
One of the arguments I've heard here a lot is that Perl is "messy", that it's too hard to understand and that any large projects become a pile of unintelligable stuff. I think that's mostly because the people who usually program in perl are young and inexperienced and not a symtom of the language itself. I've seen many truely crappy C implementation in my day. I've also seen very well done Perl code. If you use strict; and put things in objects when approprate (it's not hard, really!) you end up with very clean, easy, reusable code. Since the data structures in perl tend to get complicated (since it's often useful and usually easy) you really need to document your data structures. This is one thing that people tend to forget when programming. The code tells you what it's doing. The data structures don't tell you what the data in them means. Document it!
Perl is growing. Perl 5 has been great at integrating into different environments and with different libraries and you can now tackle a huge variety of tasks using just Perl. It's about to get better too. Perl 6 will be to.net what.net was to Java. It embraces the model and extends it just a little more. Soon you will be able to write in one language and compile to perl byte code (actually parrot byte code) C code,.net programs, Java programs, etc. Perl is beginning to be a meta language. A way to define a way to get the job done that can then be converted and used any place it's needed.
Oh, and the 3 languages I'd dispatch with? 1. C# 2. PHP 3. LISP Most projects written in them should really be done in Perl.
Encryption just means that you can't listen in on the communications. It doesn't mean that those communications aren't running over TCP/IP using a stack with security holes in it. I can have all my comm's going over superSSH3 if I want but if I'm running sendmail with a root exploit it doesn't save my machine from getting broken into. If Blaster or CodeRed or some other worm breaks it's way in and creates a root exploit you're ATM machine could be at the mercy of whoever else manages to wander on to the network regardless of what encryption the ATM transactions use. Though the risk of this causing any major harm is low it is there. I could see some giddy hackor figuring out that they can walk around through Citizen's ATM network and inserting a program that makes every ATM machine periodically dump a wad of cash out the front. That would be a "bad thing" especially if you are the bank. Worse, they could install software that would log card numbers and pin numbers at the source before they are encrypted. If they cleaned up behind themselves they chould be stealing money using that information for years. It could be disasterous.
Sometimes the efficiency of using well esablished standard software and protocols is not worth the risk. I wouldn't use Windows. I wouldn't use RedHat. I would, however, build something simple on top of a Linux kernel making sure I was careful every step of the way.
I'd run Linux just not a full fledged OS. No rpm or grep or vi or any of that. Just one program that replaces init that handles all the comms and interface. A kernel module for the encrypted networking and maybe a second program that can upgrade the first one and you are all done. You could fit it on a flash chip.
What are the odds that there is a subcontractor for a subcontractor who is working on finding and "documenting" vulnerabilites in linux right now? It's unethical but it would make good business sence. Those two things seem to be required for anything to happen within Microsoft these days so it must be true. (grin)
As a Linux supporter I would certainly like to see this happen but I sincerely doubt this will be the case. I do, however, believe Microsoft lock-in is doomed to failure and the faster and harder they try for it the more obvious the intentions, the swifter people will move to replace their software with open alternatives.
I think it's foolish to think that a few major corporations switching to linux will rapidly change the speed of advancement. The effects of such a move will take years to trickel down to the average Linux end user, but they will be there and they will be there long term. Once they go down this route it will cost Ford a lot of money to switch out of Linux, which means they will be willing to spend a lot of money to develop software that makes staing in Linux possible.
It is great to see a big corporation decide to go this route. I have long believed that a very large corporation could easily pay for the development of the applications they use in a Linux environment for less than what they pay to MS each year to upgrade their software. As the Linux versions of software that they use on a day to day basis gets closer to being what they need this switch will continue to look better. The question is not *if* the big companies switch to Linux, the question is when will they? How long will MS convince them to keep giving them truckloads of money for esentially doing nothing. Ford is really an early adopter of this which is great to see since they will probably help speed other companies adoption. Let's hope they weren't too early though since a study suggesting that switching to Linux cost Ford money would be bad for Linux adoption as a whole.
Yes. He abused someone's username and password to help get more information. Nobody died, nobody lost $300,000. There is no damage here. Is there a crime, almost definitely, but the claim of damages is rediculous.
The difference is if you steal gas, the gas station doesn't have it anymore. If you steal my money, I don't have the money anymore. If you "steal" LexusNexus service everyone still has LexusNexus service. It's not stealing. In my book, in order for something to qualify as stealing somone must not have something that they had before and that thing must have value. Copying music isn't stealing, hooking up cable isn't stealing. It's illegal and arguably immoral but it's not stealing, it's something else.
From one of RMS's better rants:
When it comes to copying, this analogy disregards the crucial difference between material objects and information: information can be copied and shared almost effortlessly, while material objects can't be. Basing your thinking on this analogy is tantamount to ignoring that difference. (Even the US legal system does not entirely accept the analogy, since it does not treat copyrights or patents like physical object property rights.)
What this guy did was most definitely illegal and wrong. It's wrong in the way that peeping in people's windows while they are undressing is wrong though, not in the way that robbing a bank or stealing a car is wrong. It's mildly annoying and should be stopped but it's not a major offense, nobody got hurt, nobody died, nobody is scarred for life, and he definitely did not steal $300,000. He should be arrested, given community service, and released and we should take this as a healthy reminder to close the security holes that we know are in our networks.
Also, don't fool yourself into thinking that if you just through the book at everyone who get's cautht hacking, your security problems will go away. That's not the way it works and it's foolish to think that.
I keep wondering how "news" ended up simply being a mouthpiece for whomever wants to create a press release or make a legal claim in court. The old unbiased reporting where balance was applied to issues has turned into todays lazy equivilant where hevily spun news items are repeated verbatum by thousands of news channels. When the RIAA sues someone for billions of dollars the number is never followed by "which is obviously a rediculously inflated value" which in the end gives the appearance that this person stole billions of dollars from music companies. The news has more or less stopped reporting and insead just acts as a mouthpiece for all kinds of powerful organizations.
The $300,000 figure is rediculous. If this guy could break in then someone else could also so any review process that needed to happen based on his intrusion should have been made anyway. The use of the LexusNexus service is a greyer area but in reality he didn't steal anything, he didn't hurt anyone, he didn't make a single person lift a finger. He made some machines work a little bit more instead of doing nothing. The Times won't have to pay the bill sine they didn't actually use the services they are being asked to pay for. $300,000 is a *huge* ammount of money and unless someone died, it is obviously rediculous.
It's sad that we live in a time where practically the only critical analysis of the crud that passes as news is on a fake comedy news program simply because in order to make fun of the news you need to criticise it.
There are always 3 things involved with any good system of rules:
1. A set of rules. 2. Circumstances where the rules shouldn't apply. 3. A person or persons in charge of enforcing the
rules that understand both 1 and 2.
Not every Linux box has a compiler on it. There *are* people who save the massive disk space and time it takes to install a compiler. If I need Time::HiRes on a machine that has no compiler I'm facing a nasty prospect of building it myself on another machine, hand copying it over and installing it. I could also take the time and disk space to install a compiler that will possibly only ever be used for about half a second. Now I can just install an RPM.
More than once I've sworn that there weren't RPM's for all this stuff. I'm happy now, the world got a little better and I didn't have to do it.
How about an internal monitoring system for the human body that could record data in real-time so that rates of change and other dynamic information can be collected and it could transmit the data to the outside world using a wireless link. If adopted, this technology would lead to huge improvements of the understanding of all aspects of human health and medicine.
Along the same lines there could be an auto-activating emergency beacon that would transmit when it detects something went wrong inside you (rapid blood pressure drop, pulse way too fast/slow, temperature dangerously high/low) recharging itself from your blood supply. It could use the celular network to send a distress signal or it could use some other triangulatable signal on it's own frequency. This could even be combined with the medical monitoring system into an all-in-one implant.
How about a brain UPS, a simple device that pumps blood to your brain and lungs when your real heart stops working (heart attack) drawing power from the blood it's pumping keeping your brain alive until someone can get your real heart back working. If made safe it could be put into people at high risk for heart attacks.
Re:Some Interesting New Products...
on
Powered by Blood
·
· Score: 3, Funny
Maybe this is what they had in mind when they created Bender in Futurama.
ZigBee (unlike X-10) has acknowledgement built in like TCP so it will retransmit until the message get's through. ZigBee has lower latency than X-10. Lots of people use X-10 for light switches because it's very cool. ZigBee sounds like it will be cooler.
ZigBee-compliant products operate in the unlicensed bands worldwide, including 2.4GHz (global), 915Mhz (Americas) and 868Mhz (Europe). Raw data throughput rates of 250Kbs can be achieved at 2.4GHz (10 channels), 40Kbs at 915Mhz (6 channels) and 20Kbs at 868Mhz (1 channel). Transmission distance is expected to range from 10 to 75 meters, depending on power output and environmental characteristics.
I could see lots of nice things you could do with this. Your alarm clock could hop on the network and allow you to synch it's time to other devices (or vice/versa if it's a clock that set's itself from the Colorado time signal like mine)
I could see joysticks using this.
Light switches ala X-10.
Water meters, power meters, gas meters, wireless thermometers and other sensors.
VCR's could use it as an interface to allow configuration from a computer.
TV's could use it as a way to implement a universal RF remote control.
Apparently they already thought of some of these ideas. From the ZigBee FAQ:
* Wireless home security
* Remote thermostats for air conditioner
* Remote lighting, drape controller
* Call button for elderly and disabled
* Universal remote controller to TV and radio
* Wireless keyboard, mouse and game pads
* Wireless smoke, CO detectors
* Industrial and building automation and control (lighting, etc.)
Then you could combine a few of these things to implement something the detects when it's too hot inside and it's colder outside and the humidity outside isn't too bad, turn on a fan. This is otherwise very complicated but hook up a few thermometers, a humidity sensor and a switch that are all accessible from a computer and it gets very easy.
It's not just speed that makes it important to shrink the drive technology. Just like it's cheaper to produce 3.5" drives than 5.25" drives, it's cheaper to produce 2.5" than 3.5" (assuming the same volume). Less raw materials, less machining, cheaper shipping, storage, etc. Eventually the added expense of manufacturing on the smaller scale get's reduced to the point where it makes sense. It is inevitible that we will end up with 2.5" harddrives in desktops and servers. Smaller devices are often easier to make more reliable and durable (drop a matchbox car and a real car from 6 feet up onto it's roof and see which one is cheaper to repair)
Storage system scalability is another reason to reduce the form factor of the drives. You can fit a lot more 2.5" drives in 1U than 3.5".
This diet is popular because people want to lose weight, not because they want to be healthy. Therefore, the burden of proof is on Dr. Atkins to prove it is healthy. If you come out with some diet that you are advocating, it is your responsibility to show it won't kill you in 5 years.
Actually, I believe most people try the Atkins diet in an attempt to improve their overall health and not simply lose weight. It is part of the stated effects of the diet in his book and he talks at length about the overall health implications. Also, from what I can tell, Atkins did as much as can be expected to show that his diet was healthy and much more than almost anyone else who has produced a dietary recommendataion. It seems his patients were monitored carefully and the evidence was quite strong that his diet improved overall health. That is much more than can be said for most diets including the food pyramid. Now it's up to the doctors and nutritionists who are standing up and saying that it shouldn't be followed because it is unhealthy to prove what they are claiming. The embarrasing fact is they base most of what they are doing on little or no experimental evidence. They make the same assumptions they have been making for years about how the body works. They do not know that the diet is unhelathy. The evidence seems to show that it, in fact, is very healthy. Mostly the evidence I have heard is from doctors saying it's unhealthy because they assume it is based on their understanding of how the body works.
If you don't know fast food is bad for you then you are an idiot.
I disagree with this on a lot of levels. First, just because it's quick to prepare and serve doesn't make it bad for you. There are few people who would argue that picking up a grilled chicken bacon ranch salad and a diet coke at McDonalds is eating unhealthily. It's not 'fast' food that's bad it's the types of food we eat. I would argue that much of the blame on fast food is misplaced. Our problems have more to do with Special K and Cheerios than BigMac's and fries and a whole lot more to do with Coke than either.
It comes down to the question of what parts of the meals are bad for you. How do you pick and chose. Is it better to get a double quarter pounder with cheese (760 calories) or a Big Mac (580 calories) or simple hamburger and small fries with only (490 calories)? The people who sit there and say "if you intake more calories than you burn, you gain weight" would probably say go with the hamburger and fries. The answer is more complicated than that though. For me, I'll take the double quarter pounder with cheese. Why? The double quarter pounder with cheese will fill me up and not spike my blood sugar which would make me hungy again soon. A hamburger and fries has 61g of carbs, the Big Mac has 47g of carbohydrates where a DQPwC has only 38g. (from here) Caloric intake vs output is as relavent to weight loss in much the same way that water input vs output is to water pressure in a water system. While the rule is very true you don't get much usefull information as to how to design a good system from it.
Trying to work from calories has lead to stupid things like people not tosting their bread. Toast, you see, has more calories per gram than bread does and thus is more fattening. Of course that is rediculous since they are the same thing, but toast has less water thus is lighter and contains more calories per gram. This is an extreme example of why the oversimplified models that are commonly used by nutritionists don't work. Calories per gram is obviously not an apropriate way to analyse food'd impact on health or weight loss. Measuring a food's weight or volume is not suficcient to determine how long it will keep you full.
I believe the variables that determine how long something will fill you up will even change from day to day as the populations of bacteria in your d
Slashdot Mirror
The problem here is not just one of marketing. They have spread the idea that the BSA and Microsoft have been cramming down everyone's throats, that software should be evaluated on the merits of the product, ignoring the merits of the system involved in crating the product. This is like a construction company insisting that when a building project goes out to bid the only thing evaluated are price and the size of the building created with no constraints on the construction methods used, safety, environmental impact, or maintainability. They want eliminate concideration of evaluation criteria where they inevitably lose. The software companies are doing the equivalent of building the building in such a way that only they can repair it when it fails and ensuring that it will fail over time and when it does, they insist that it can't be repaired, it needs to be torn down and rebuilt. This is not the right model for infrastructure.
Software companies are afraid that governments are going to start to realize they have done the equivalent of paying a company to build their roads while giving them the rights to put tollbooths wherever they want. It's ridiculous. There is a lot of infrastructure software out there (operating systems, relational databases, office suites, etc.) that have been paid for 100 times over by governments and they still don't own them.
Yea. Except T1's do go down. Also, every ISP sells more downstream bandwith than they have upstream. A difference that makes no difference is no difference at all. When cable bandwidth is shared among neighbors it's never 2.5 mbit shared among a bunch of people, it's more like 100 mbit shared among a bunch of people who are each capped at 2.5 mbit. That's the same thing T1 providers do.
The real issue with cable is crap upstream bandwith. Comcast. Let me pay an extra $10/month for 1.5 mbit upstream and I'll do it!
All we need now is for the government to put up a DirecTV like satellite and use it like it uses the regular public airwaves. Anyone can broadcast but they can't charge for people to receive it. Then everyone in the country could have 300 digital channels and no monthly bill.
Then we get a little data over power lines technology and you can have your power company be your ISP and your phone company. Or start using a cellular Internet connection. Then the cable company would have to really work to get your internet business because it would be the only way they could make money!
Competition does make for a good market as long as there is lots of
it.
There is no one answer. If security is your only concern you should have as many layers of security as possible with firewalls between each layer locked down as tight as possible. That said, security is never your only concern. Cost, ease of maintenance, performance, and flexibility are all important in a network design. After all, the purpose of your company is probably to get something accomplished, not to avoid getting hacked. There are times when every different network configuration is appropriate from super secure to a cable modem router to a windows box right on the internet. There is no one answer.
Application layer firewalls are another layer above port filtering. They can increase security and could, in theory, make it worthwhile to share a service hosted on a machine that is inside your network. I would only do that if you trusted the security of your internal network. Most network designs assume that once you get in to the "internal network" there is no more security and all your deepest company secrets are available to anyone browsing around. If this is true, you've probably made some bad decisions somewhere along the way and you should address those before you open any holes. If you are willing to maintain strict security on your internal network then the added simplicity of allowing Internet access to machines on it can be worth the risk. This can be a lot easer than setting up a dmz.
Usually layers do make sense though, even if one of the layers is just a Linux box doing firewalling, routing and serving some services. One thing I like to do is to mix operating systems at different layers. That way if you get a worm of some kind that gets into one layer it won't penetrate to the layer behind. For example, internet facing servers are Linux based, desktops are Windows based.
Another thing I have done when I absolutely needed a Windows based web server is to setup Apache as a reverse-proxy only forwarding requests to a particular subdirectory to the Windows server. This filtered out all the standard buffer overload attacks since none of them referred to that subdirectory name. It also made sure the requests were relatively well behaved and buffered outgoing data for the Windows box, reducing connection counts when it was under high load. This is an easy way to do an application layer firewall and if you are firewalling with a Linux box you can do it right on the firewall.
I take it back. Perl 5.8.1 doesn't bring a fix for the memory leaks. This program will still make perl blow up:
use threads;
use threads::shared;
my @shared : shared;
@shared = (1) while (1);
Great post. I epecially liked the Aristotle quote. I'll use that when trying to explain why it's not a bad idea to use Perl to a PHB some time.
Furthermore, If I were going to teach programming to anyone who was new at it I would absolutely, positively, teach them Perl. It's an awesome starter language and, unlike most good starter languages, it's a awesome full power language that you can build real usable software with.
Mental Note: Remember to spell check when I type after 3 AM.
There are too many languages now a days. Please eliminate 3. I am not a crackpot!
.net what .net was to Java. It embraces the model and extends it just a little more. Soon you will be able to write in one language and compile to perl byte code (actually parrot byte code) C code, .net programs, Java programs, etc. Perl is beginning to be a meta language. A way to define a way to get the job done that can then be converted and used any place it's needed.
With Perl 5.8.1 we finally get usable threads that efficently communicate and cooperate. There is a fix for the memory leak that existed when using shared queues, arrays and hashes in Perl 5.8.0. This starts a whole new era for Perl programs. My little language is finnally growing up (sniffle.)
Perl has been a great language for people who just want to "get it done." It get's the job done and usually get's it done fast. It's
great at making what would be a complicated thing easy. If you want to reference your song data by artist album and track number you just say $song_data{$artist}{$album}[$track_number]. What is happeneing behind the scenes is complicated, that's actually a hash of hashes of arrays possibly pointing at another data structure underneath but conceptually it all optimizes out to simply storing by artist album and track number. Most programmers aren't used to having that option. They don't understand it and when they go to figure it all out they get mired in the way the references are being stored and what is pointing at what and lose sight of the simplicity, efficiency, and power of the whole thing.
One of the arguments I've heard here a lot is that Perl is "messy", that it's too hard to understand and that any large projects become a pile of unintelligable stuff. I think that's mostly because the people who usually program in perl are young and inexperienced and not a symtom of the language itself. I've seen many truely crappy C implementation in my day. I've also seen very well done Perl code.
If you use strict; and put things in objects when approprate (it's not hard, really!) you end up with very clean, easy, reusable code. Since the data structures in perl tend to get complicated (since it's often useful and usually easy) you really need to document your data structures. This is one thing that people tend to forget when programming. The code tells you what it's doing. The data structures don't tell you what the data in them means. Document it!
Perl is growing. Perl 5 has been great at integrating into different environments and with different libraries and you can now tackle a huge variety of tasks using just Perl. It's about to get better too. Perl 6 will be to
Oh, and the 3 languages I'd dispatch with?
1. C#
2. PHP
3. LISP
Most projects written in them should really be done in Perl.
Encryption just means that you can't listen in on the communications. It doesn't mean that those communications aren't running over TCP/IP using a stack with security holes in it. I can have all my comm's going over superSSH3 if I want but if I'm running sendmail with a root exploit it doesn't save my machine from getting broken into. If Blaster or CodeRed or some other worm breaks it's way in and creates a root exploit you're ATM machine could be at the mercy of whoever else manages to wander on to the network regardless of what encryption the ATM transactions use. Though the risk of this causing any major harm is low it is there. I could see some giddy hackor figuring out that they can walk around through Citizen's ATM network and inserting a program that makes every ATM machine periodically dump a wad of cash out the front. That would be a "bad thing" especially if you are the bank. Worse, they could install software that would log card numbers and pin numbers at the source before they are encrypted. If they cleaned up behind themselves they chould be stealing money using that information for years. It could be disasterous.
Sometimes the efficiency of using well esablished standard software and protocols is not worth the risk. I wouldn't use Windows. I wouldn't use RedHat. I would, however, build something simple on top of a Linux kernel making sure I was careful every step of the way.
I'd run Linux just not a full fledged OS. No rpm or grep or vi or
any of that. Just one program that replaces init that handles all
the comms and interface. A kernel module for the encrypted networking and maybe a second program that can upgrade the first one and you are all done. You could fit it on a flash chip.
What are the odds that there is a subcontractor for a subcontractor who is working on finding and "documenting" vulnerabilites in linux right now? It's unethical but it would make good business sence. Those two things seem to be required for anything to happen within Microsoft these days so it must be true. (grin)
As a Linux supporter I would certainly like to see this happen but I sincerely doubt this will be the case. I do, however, believe Microsoft lock-in is doomed to failure and the faster and harder they try for it the more obvious the intentions, the swifter people will move to replace their software with open alternatives.
I think it's foolish to think that a few major corporations switching to linux will rapidly change the speed of advancement. The effects of such a move will take years to trickel down to the average Linux end user, but they will be there and they will be there long term. Once they go down this route it will cost Ford a lot of money to switch out of Linux, which means they will be willing to spend a lot of money to develop software that makes staing in Linux possible.
It is great to see a big corporation decide to go this route. I have long believed that a very large corporation could easily pay for the development of the applications they use in a Linux environment for less than what they pay to MS each year to upgrade their software. As the Linux versions of software that they use on a day to day basis gets closer to being what they need this switch will continue to look better. The question is not *if* the big companies switch to Linux, the question is when will they? How long will MS convince them to keep giving them truckloads of money for esentially doing nothing. Ford is really an early adopter of this which is great to see since they will probably help speed other companies adoption. Let's hope they weren't too early though since a study suggesting that switching to Linux cost Ford money would be bad for Linux adoption as a whole.
Wow. Yea. Now that I look at it that wasn't ridiculous.
Where's the edit previous post button when I need it.
Yes. He abused someone's username and password to help get more information. Nobody died, nobody lost $300,000. There is no damage here. Is there a crime, almost definitely, but the claim of damages is rediculous.
From one of RMS's better rants:
What this guy did was most definitely illegal and wrong. It's wrong in the way that peeping in people's windows while they are undressing is wrong though, not in the way that robbing a bank or stealing a car is wrong. It's mildly annoying and should be stopped but it's not a major offense, nobody got hurt, nobody died, nobody is scarred for life, and he definitely did not steal $300,000. He should be arrested, given community service, and released and we should take this as a healthy reminder to close the security holes that we know are in our networks.
Also, don't fool yourself into thinking that if you just through the book at everyone who get's cautht hacking, your security problems will go away. That's not the way it works and it's foolish to think that.
I keep wondering how "news" ended up simply being a mouthpiece for whomever wants to create a press release or make a legal claim in court. The old unbiased reporting where balance was applied to issues has turned into todays lazy equivilant where hevily spun news items are repeated verbatum by thousands of news channels. When the RIAA sues someone for billions of dollars the number is never followed by "which is obviously a rediculously inflated value" which in the end gives the appearance that this person stole billions of dollars from music companies. The news has more or less stopped reporting and insead just acts as a mouthpiece for all kinds of powerful organizations.
The $300,000 figure is rediculous. If this guy could break in then someone else could also so any review process that needed to happen based on his intrusion should have been made anyway. The use of the LexusNexus service is a greyer area but in reality he didn't steal anything, he didn't hurt anyone, he didn't make a single person lift a finger. He made some machines work a little bit more instead of doing nothing. The Times won't have to pay the bill sine they didn't actually use the services they are being asked to pay for. $300,000 is a *huge* ammount of money and unless someone died, it is obviously rediculous.
It's sad that we live in a time where practically the only critical analysis of the crud that passes as news is on a fake comedy news program simply because in order to make fun of the news you need to criticise it.
There are always 3 things involved with any good system of rules:
1. A set of rules.
2. Circumstances where the rules shouldn't apply.
3. A person or persons in charge of enforcing the
rules that understand both 1 and 2.
Not every Linux box has a compiler on it. There *are* people who save the massive disk space and time it takes to install a compiler. If I need Time::HiRes on a machine that has no compiler I'm facing a nasty prospect of building it myself on another machine, hand copying it over and installing it. I could also take the time and disk space to install a compiler that will possibly only ever be used for about half a second. Now I can just install an RPM.
More than once I've sworn that there weren't RPM's for all this stuff. I'm happy now, the world got a little better and I didn't have to do it.
Support your local wacky perl project. I do!
The subject sais it all except:8 &loc=106
http://www.buy.com/retail/product.asp?sku=3087286
How about an internal monitoring system for the human body that could record data in real-time so that rates of change and other dynamic information can be collected and it could transmit the data to the outside world using a wireless link. If adopted, this technology would lead to huge improvements of the understanding of all aspects of human health and medicine.
Along the same lines there could be an auto-activating emergency beacon that would transmit when it detects something went wrong inside you (rapid blood pressure drop, pulse way too fast/slow, temperature dangerously high/low) recharging itself from your blood supply. It could use the celular network to send a distress signal or it could use some other triangulatable signal on it's own frequency. This could even be combined with the medical monitoring system into an all-in-one implant.
How about a brain UPS, a simple device that pumps blood to your brain and lungs when your real heart stops working (heart attack) drawing power from the blood it's pumping keeping your brain alive until someone can get your real heart back working. If made safe it could be put into people at high risk for heart attacks.
Maybe this is what they had in mind when they created Bender in Futurama.
I love puns.
ZigBee (unlike X-10) has acknowledgement built in like TCP so it will retransmit until the message get's through. ZigBee has lower latency than X-10. Lots of people use X-10 for light switches because it's very cool. ZigBee sounds like it will be cooler.
time to other devices (or vice/versa if it's a clock that set's itself from the Colorado time signal like mine)
I could see joysticks using this.
Light switches ala X-10.
Water meters, power meters, gas meters, wireless thermometers and other sensors.
VCR's could use it as an interface to allow configuration from a computer.
TV's could use it as a way to implement a universal RF remote control.
Apparently they already thought of some of these ideas.
From the ZigBee FAQ:
* Wireless home security
* Remote thermostats for air conditioner
* Remote lighting, drape controller
* Call button for elderly and disabled
* Universal remote controller to TV and radio
* Wireless keyboard, mouse and game pads
* Wireless smoke, CO detectors
* Industrial and building automation and control (lighting, etc.)
Then you could combine a few of these things to implement something the detects when it's too hot inside and it's colder outside and the humidity outside isn't too bad, turn on a fan. This is otherwise very complicated but hook up a few thermometers, a humidity sensor and a switch that are all accessible from a computer and it gets very easy.
It's not just speed that makes it important to shrink the drive technology. Just like it's cheaper to produce 3.5" drives than 5.25" drives, it's cheaper to produce 2.5" than 3.5" (assuming the same volume). Less raw materials, less machining, cheaper shipping, storage, etc. Eventually the added expense of manufacturing on the smaller scale get's reduced to the point where it makes sense. It is inevitible that we will end up with 2.5" harddrives in desktops and servers. Smaller devices are often easier to make more reliable and durable (drop a matchbox car and a real car from 6 feet up onto it's roof and see which one is cheaper to repair)
Storage system scalability is another reason to reduce the form factor of the drives. You can fit a lot more 2.5" drives in 1U than 3.5".
This diet is popular because people want to lose weight, not because they want to be healthy. Therefore, the burden of proof is on Dr. Atkins to prove it is healthy. If you come out with some diet that you are advocating, it is your responsibility to show it won't kill you in 5 years.
Actually, I believe most people try the Atkins diet in an attempt to improve their overall health and not simply lose weight. It is part of the stated effects of the diet in his book and he talks at length about the overall health implications. Also, from what I can tell, Atkins did as much as can be expected to show that his diet was healthy and much more than almost anyone else who has produced a dietary recommendataion. It seems his patients were monitored carefully and the evidence was quite strong that his diet improved overall health. That is much more than can be said for most diets including the food pyramid. Now it's up to the doctors and nutritionists who are standing up and saying that it shouldn't be followed because it is unhealthy to prove what they are claiming. The embarrasing fact is they base most of what they are doing on little or no experimental evidence. They make the same assumptions they have been making for years about how the body works. They do not know that the diet is unhelathy. The evidence seems to show that it, in fact, is very healthy. Mostly the evidence I have heard is from doctors saying it's unhealthy because they assume it is based on their understanding of how the body works.
If you don't know fast food is bad for you then you are an idiot.
I disagree with this on a lot of levels. First, just because it's quick to prepare and serve doesn't make it bad for you. There are few people who would argue that picking up a grilled chicken bacon ranch salad and a diet coke at McDonalds is eating unhealthily. It's not 'fast' food that's bad it's the types of food we eat. I would argue that much of the blame on fast food is misplaced. Our problems have more to do with Special K and Cheerios than BigMac's and fries and a whole lot more to do with Coke than either.
It comes down to the question of what parts of the meals are bad for you. How do you pick and chose. Is it better to get a double quarter pounder with cheese (760 calories) or a Big Mac (580 calories) or simple hamburger and small fries with only (490 calories)? The people who sit there and say "if you intake more calories than you burn, you gain weight" would probably say go with the hamburger and fries. The answer is more complicated than that though. For me, I'll take the double quarter pounder with cheese. Why? The double quarter pounder with cheese will fill me up and not spike my blood sugar which would make me hungy again soon. A hamburger and fries has 61g of carbs, the Big Mac has 47g of carbohydrates where a DQPwC has only 38g. (from here) Caloric intake vs output is as relavent to weight loss in much the same way that water input vs output is to water pressure in a water system. While the rule is very true you don't get much usefull information as to how to design a good system from it.
Trying to work from calories has lead to stupid things like people not tosting their bread. Toast, you see, has more calories per gram than bread does and thus is more fattening. Of course that is rediculous since they are the same thing, but toast has less water thus is lighter and contains more calories per gram. This is an extreme example of why the oversimplified models that are commonly used by nutritionists don't work. Calories per gram is obviously not an apropriate way to analyse food'd impact on health or weight loss. Measuring a food's weight or volume is not suficcient to determine how long it will keep you full.
I believe the variables that determine how long something will fill you up will even change from day to day as the populations of bacteria in your d