I don't see it ending any time soon. IBM currently holds the largest market share for UNIX servers and Oracle would rather see their DB on those rather than DB2. As long as there is still a good market share in the high end server market, then Oracle will continue to provide it just as they do for HP and did for Sun before buying them.
Why is it people assume you have to run Linux on Power hardware for Oracle? AIX fully supports Oracle RDBMS solutions, including combining them high availability products such as HACMP. You buy Power hardware for the reliability and performance in mid-large scale computing. Someone that can afford to put Oracle on that kind of hardware usually runs AIX vs Linux.
My solution to it given IT often has long and somtimes odd hours with various maintenance windows, events, late night coding sessions and what have you is to use my 24/7 gym at off hours. In general, I see 4-5 people acutally in my gym at night, not including myself and one other buddy I go with. So that leaves plenty of space, equipment and stuff for you to do at nights without the mass of folks using the gym like say at 6AM or 5PM time frames. With this huge 60'000 sq ft gym, 4 people is nothing at all and its very comfortable. After all, those that go at 2 or 3 AM in the mornings, tend to mind their own business.
'Well, IBM complied with legal orders when they cooperated with Nazi Germany. Those were legal orders under the Nazi German system.'
IBM Germany division simply had no choice. What is not talked about much is that IBM's US division was ashamed at that forceful compliance and used their factories to manufacture weapons for the US military. It was a sort of paying their dues for their German counterparts assisting the Nazis in WWII. There are still many IBM stamped guns out there. Most are M1 Carbine rifles.
If a company is truely global, it also means they have to apply to the laws of their respective countries they do business with. If the US doesn't want US based companies to play ball with China's restrictions, they simply could put a trade restriction against China. It would be like what has happened with Cuba then. That wouldn't be the best of moves since China is a huge manufacturing country that makes a lot of products we import. You simply can't ban Internet companies ability to do trade with China and not ban everyone else as well. If they don't like the sensorship China requires by law, then take it up with China and the UN to pressure them. China, regardless of acceptance, makes the laws that pertain to their country and others either abide by them if they want to deal with China or stop doing business with them.
OS/2 fell because of bad marketing and low sales in comparison to Microsoft Windows. Linux is used on IBM p5 line, but it is hardly replacing AIX. AIX is still IBM's flagship product for UNIX servers. No matter how much they support the Linux community, they don't sell Linux as the solution for UNIX servers, they sell AIX. Also, why would Apple compete with Dell, HP, etc... They have a pretty looking over priced PC now. Dell and the like can take some designers and make their own pretty looking PC cases and cram hardware into it just as well. If they dumped OS X, they would also likely dump the Mac. Which would leave them in one place, iPods.
One other thing to remember is that they have a small subset of hardware to support. Meaning their driver development and overall system support is limited to the small line they produce. He is arguaing like they would attempt to support all PC platforms which doesn't seem like it's in their plans.
They may eventually support MS Windows on it, but the only way to make a difference in their PC versus another is the OS. It reminds me a lot of BeOS. Fast, reliable, well coded OS, but support and lack of drivers for new hardware and costs overall killed it. Apple could still keep up, but who know if they will.
After working for Lowe's Companies for 5.5 years at their corporate offices, I have seem much more in the realm of distribution. Newegg's distribution seems like childs play compared to larger retailers like Lowe's, Walmart, Home Depot, etc... Think of 1.8 Million sq ft of warehousing. Products that are already barcoded by machine, flying down the highest speed conveyors past scanners to reach their appropriate shipping area. They service hundreds of semi-trucks a day and ship quite a bit more. Imagine instead of people doing sorting, labeling, etc, your system solves the distribution nightmares by solving the travelling saleman problem for all your trucks going to thousands of locations that sell to millions of customers. Quantity, routing, and distribution is all computerized and selected based on market sales, in-stock needs, and store orders. The only need for people is to palletize the product and get it on a truck. The sheer size of "big box" retail stores are as large as the larger of Newegg's warehouses.
For some reason after seeing this in action I'm not impressed. If you want to see some amazing things, try and get Walmart, Lowe's or similar companies to do a show and tell of their facilities. It's even more impressive when you get into the data centers they use as well. Granted, much of the marketing and business decision software is secretive and tightly kept, but the systems, remote DR fiber links sychronizing TB's of storage accross thousands of miles, and the raw power of computing is impressive. I think that's why I went to work for Lowe's in the first place. I mean, coding for IBM SP frames right out of college seemed rather appealing. Heh, and at one point I thought they were just a company that sold hammers and stuff. Who knew that retail could be so innovative with their implementations and solutions.
Linux and AIX you mean. IBM doesn't support running Sun Solaris and I am sure Sun has no plans to port to Power architecture from Sparc/x86 any time soon. Although, it would be rather fun to port OpenSolaris to the Power platform just for the hypocracy in it...
Windows may hold the Enterprise Desktop environment, but they are a far cry from holding the server market in business. IBM is the leader there. They build the top end, most reliable, and highly available systems short of a custom engineered product for complete failure detection. In short, IBM is doing more for large scale business than anyone else these days. If you were running a large company, say a Fortune 50 retailer, would you install a bunch of Windows servers and setup nightly reboot jobs to keep things running smoothly, or would you setup AIX servers with HACMP (IBM's high availability offering) to run even if the hardware dies, sites go down, net connections drop, etc... you get the drift. It's all about what people can afford and the best is IBM.
5. Continue hiring mostly foreign workers on H1B to train.
6. Crush all other ISP's
7. Outsource everything to India to reduce expenses by 400%
8. Laugh all the way to the bank and when other countries are in complete control of our internet services...
Oh and almost forgot
9. Deploy orbital brain lasers
As I have before and he now takes great pride in my accomplishments. Although he is mostly worried now about my current work environment and stress levels... Ah, the life of an engineer...
On the mainframe, a programmer must have the concerns of system resources and getting it right the first time. Jobs for mainframe are scheduled to optimize system usage and consistantly run at a set capicity designated in planning. This said the worst thing that happens to a mainframe programmer is an abend (a job that fails and requires interaction, often causing other jobs to get backed up). Loss of system time means a significant loss of money in the mainframe world. That directly affects the process they must take to submit and run jobs. If they are coding JCL or COBOL, their code must be reviewed and run on a test mainframe if possible. It must be exact which takes a certain kind of programmer to get it right. As a suggestion in handling mainframe programmers, it is important to improve upon the stress that comes from deadlines and being right the first time. Most mainframe programmers I know are over stressed, over worked and need the support and coaching from their manager that will make their lives a little better. Be cautious of resource usage and allocations though and make sure they have a procedure for implemtation that is strictly followed. As a manager over mainframe teams, you should be strict with procedures, but don't hold it over your employees heads. They have a hard enough job to be constantly docking them in their reviews because of not quite getting it 100% every time. After all there are not as many mainframe programmers as they used to be, so keep the ones doing a good job.
UNIX programmers on the other hand depend on what their designation is. System level coders usually have attitude to cope with and you have to take that in stride. On the System Engineers level, you have to also manage resources and allocations. It is important for them to worry about a balanced system performance and ensuring that a poorly written application can be dealt with, not causing pain to the other applications since it is a multi-user system. Proper tuning will handle this. On the application programmers side, the most important thing is to make sure the "big picture" gets taken care of. You need an architectural layout and design of your applications that will fit into the business model over all. The right people need to be consulted to ensure your programmers are not writing a peice of code that doesn't fit well. Code reviews and some level of strictness are required, but not to the level of a mainframe programmer. In UNIX you can recover and risk far less losses with application downtime than on the mainframe. Restarting applications here is not as costly.
On the Windows side, it's about interaction and getting the product moving quickly. Windows programmers are either concerned about the tightness of the application or the features it provides. To properly write and deploy a Windows application, strict QA must be implemented and some minimal level of profiling must be done. The important factor is user perception in this case. You want the end user to see it run fast and have tons of features. It will have bugs and the bugs are not looked at interactively due to it running on a user system. You people need to be geared towards debugging problems with the minimal interaction available. This requires that your coders know their code in and out. You want to develop and culture that ability and make sure you keep track of bugs for improvements.
Overall, people are people, but the stress levels of a mainframe coder are for getting it exact and right to prevent downtime and backing up jobs. For UNIX engineers and applications programmers, it's in stability of the system and proper interaction with the system and business architectures that is stressful. With Windows programmers it's in the end users perspective and timeline to release to market that is the most stressful. Realizing those things, you can appropriately develop a plan to help manage and cope with the managerial needs of your people.
Keep in mind, if you are a resource and career manager, make sure you keep your people
Programming beyond the "Here is how you get this output" method is much different. You can show someone functions, structure of the language, even useful API's to accomplish most things. What you can't show them is how to design, and develop within the art of programming. There is a huge difference between programmers who create and programmers who emulate. Given a good foundation of skills, information, education, and basic experience in programming, one can start the life long development of creating programs. There are always new techniques, new tricks, better ways, things to use, more creative designs, and more ways to write your code than one normally imagines.
What she did was start her journey to that ability. She has learned how to do the base set, let's see if kids her age that have that ability can show their creativity as well. So far I am only impressed at the fact she has taken the tests, some started programming (myself included) even earlier than that. At the time I was 6 years old, reading lines of basic from books and seeing what they did was easy to do on my TSR-80, not to mention learning commands. It wasn't until later that I started with useful languages (well if you call pascal useful...) and then progressed to my current professional level. So it is possible for kids to program, and I am happy to see that some parents encourage that. Speaking from personal experience, it's better than having your dad tell you at age 16, "You will never go any where with this linux thing. You spend too much time on the computer..." I can only imagine what would have happened if I had decent programming classes taught back then....
Most people are not noting the primary reasons for the cost increase over building a cheap cluster. Their system can run off one power outlet. It is designed to pull a smaller amount of power than a typical cluster. Imagine the power bill from 96 350 watt power supplies. That's 33.6 kwh if my calculations are correct. They claim on the spec sheet a maximum consumption of 1.5 kwh total (1500 watt). That is considerably less in the grand scheme of things.
The other thing to consider is the cost of engineering in the nodes themselves.
If they had the extra 4 digits (3 for AMEX cards)on the back stored in the mag stripe that is a big no no to banks. CVV (numbers after your account number on the back) data was originally used so that the old swipe impact 3 part forms would not pick up the numbers. This meant that you would not have all the info that the banks had on the card from the imprint of the card numbers from the front. They often called in the card when making a purchase and read off the CVV data to the bank when making a charge.
What they typically use that data for now is online orders or "card not present" orders. It's a way to validate when do a "card not present" transaction. Most banks require this now for retailers doing online transactions. There are only two bits of info that should not be stored on the mag stripe. One is the CVV data and the other is your PIN number.
For those that have access to mag stripe readers, especially ones that use keyboard input, try running your card through and dumping out the data sometime. You will see exactly what is on them (if your reader supports more than track 1 and track 2 reads that is). Last time I did, my AMEX, Discover, and VISA all had my name on there, card number, expiration and a few other numbers noone typically uses for transactions.
That's incorrect. The mag stripe contains only the 16 digits of the card. (I used to work for MasterCard).
*bzzzt* wrong answer. Actually if you read up on mag stripes you will find that they contain more than that. The data (tracks) of the cards often contain your name and other bits of information the bank finds handy. How else do you explain the "Hello Mr/Ms...." on the reciepts some places have.
Back to the article at hand, it as much of a problem for merchants as it is for consumers. For the merchants, they risk charge backs with no "valid" signature to dispute them with. If it says the persons real name, they generally can win on that argument. However, saying that it is a valid charge and reading back "Shamu" as the name doesn't work so well.
We need to move beyond retailers using un-encrypted data to validate accounts and also having them enforce policies that strictly require the checking of cards and receipts. Will this happen? It's unlikely to occur soon.
In the mean time the ultra-paranoid can skip using a credit card/check card and use DEBIT/ATM only. If you have a true DEBIT/ATM only card, the card number does not do a lot of good for anyone. You have to use a pin to use it and from knowing how that works, trust me, it's better;) Unless someone is watching you punch the keypad, has a camera mounted to watch and are capturing your mag stripe data, you are fairly safe.
Change happens. There are a few ways of dealing with it, you chose to move on and quit. Other chose to stick it out, while some stop everything and complain about the fact change happened and why it should happen to them. Sound familiar? There is a book called "Who moved my cheese?" which is more of a social outlook on acceptance of change. People who choose to move on typically have something more favorable either in pay, benefits, or general happiness lined up.
I recently have seen people leave the company I work for on various complaints and reasons. They were unhappy with management and the way they were treated. Others were unhappy with the loss of the extra benefits that used to make my company stand out. While others were unhappy with the work load that detracted from their family and personal life. To this date, I have not seen many of these problems corrected. Leaving because the type of work you are doing changes is not what I would consider a valid reason, but that is me. Every job will have it's moments, but as a friend of mine who left his job here told me once, "You have to come up with a list of why you work where you do." Some people place time with family, friendship with co-workers, money, advancement and career goals, location, or even what language and environment they code in on that list. It's your list, order it how you would like and then decide based on what you value the most from your work. If your top points are not met, then should you leave? Will you be content with staying? That is for each individual to decide.
fork either returns 0 or -1, so a comparison to value 1, then returning 0 should not happen. I believe it would end up returning a NULL value since the if fails... interestingly enough, you would fork four child processes, which would fork 4 child processes, etc, etc... nasty little way to eat or process space. Sort of like the shell script of script a calls script b which calls a.... etc Even worse when you background those and put a wait in there, blamo.
A good programmer who codes in C should be able to understand what is going on. Still it is no substitute for real documentation... Someone should let the folks that enter the IOCCC that they have some of the worst self documenting code I have seen though. It's like trying to read the Xlib programming manuals after a few to many drinks (not recommended and may result in a nasty hangover in the morning).
Most larger companies value support of any software product as the critical key element. They want a support contract, they want an end to finger point at when something breaks. Buying a distro like RedHat EL or SuSe Server, etc will allow them to buy a support contract. Having companies like Novell or Redhat backing a product is comforting to large corporations. If you are looking for something for a company that doesn't spend millions in software and support each year, then check out debian, gentoo, slackware, etc... Keep in mind, you want to setup things to make support it easy, rather than a nightmare. Most companies I know of that don't opt for well backed distros like Redhat/Novell Suse will often customize their own distro based off something else and support it internally. The best thing is to look around, try a few, be un-biased, and keep installation, configuration, upgradability and support aspects in mind. People will try to convince you one is better than the other, IE my distro is bigger than yours, etc, etc... In the end, the one that best fits the support and comfort levels of the company is what is best.
It's not water based, but rather an acrylic latex house paint. Check the website for details. They also have additives to mix with other paints instead of using their base coat paint. Say if you wanted to have a sheilded room painted in a nice mauve with white trim or something...
This a very good strategy for IBM. They have a lot of good researchers coming up with ideas that will cost money to develop. Many of those can benefit the computing industry in general and help them in further development of core technologies. This is the easiest way for IBM to, in a sense, get free development. By opening up the patents, they don't have to spend money on implementation which will allow them free use of the technologies to futher their products with no real development cost. The open source community can implement and futher the technology covered by the patents allowing IBM to integrate that back into their core business to further the commercial products without the development cost.
It makes perfect sense if you think about it. Open Source won't earn them money, but will in turn give them the implementations without the expense of their own development team doing it. A community of free development that can implement technology that their researchers create, who could ask for more?
You have some good points. Also it is good to ask what is in it for you. You need to make it clear that you want to start a career, not just a job with them. Having techincal questions, and other things to ask them is helpful.
My first major interview was for the US Naval Academy. I should say it went well since I received a nomination to the Academy from my Congressman. Interviews are largely about proving to the future employer why they should hire you. Sitting back will never accomplish anything. You have to make a point to them that you are the one and humbly show your abilities. Attitude plays a large part of it as well.
Both a combination of my social skills and my techincal abilities helped me prove to the Fortune 50 company I work for that I was the guy they needed. Since then, I have become a valuable asset to the company with a rather impressive resume and 5 years of experience with them alone. Even without a degree, if you have the right skills, the right drive and attitude, you can start a career at a company that has strict requirements. Education is very important and I would definately stress finishing your degree. If you are a CS major and going that route, immediately after you start coding, try to find internships or coops. If you don't start building your job experience up and working for other places while in school, it will be much harder to find a job. Some companies have also been known to do repeat internships if you prove to be a good employee and even offers once you get out of college.
Re:IBM isn't entirely stupid
on
Defining Google
·
· Score: 2, Insightful
AT&T Bell Labs used to hire nothing but the best... Where are they now? I hope the folks at Google make sure they know the mistakes that came from the business there and use their brilliant minds to create real actual products they can pay for those minds with.
There are plenty of articles about the case. The DOJ and FBI have most of it as public information for the search engine savy. The key is, there is a lot of potential damage to end consumers and the company with crimes like this. Considering it was his second offense and he hadn't even gotten off probation for the first, 9 years seems reasonable to me.
I have been following this case online for a while. They were involved with the 2600 scene, attented "Hacker" conventions, and were already known for things that some might view on a grey line (IE telephone companies - See Telcodata.us). The thing is that occasionally there are bad apples that learn from others and commit crimes. Showing a 9 year prison sentence might help keep some of the apples from rotting.
Slashdot Mirror
I don't see it ending any time soon. IBM currently holds the largest market share for UNIX servers and Oracle would rather see their DB on those rather than DB2. As long as there is still a good market share in the high end server market, then Oracle will continue to provide it just as they do for HP and did for Sun before buying them.
Why is it people assume you have to run Linux on Power hardware for Oracle? AIX fully supports Oracle RDBMS solutions, including combining them high availability products such as HACMP. You buy Power hardware for the reliability and performance in mid-large scale computing. Someone that can afford to put Oracle on that kind of hardware usually runs AIX vs Linux.
My solution to it given IT often has long and somtimes odd hours with various maintenance windows, events, late night coding sessions and what have you is to use my 24/7 gym at off hours. In general, I see 4-5 people acutally in my gym at night, not including myself and one other buddy I go with. So that leaves plenty of space, equipment and stuff for you to do at nights without the mass of folks using the gym like say at 6AM or 5PM time frames. With this huge 60'000 sq ft gym, 4 people is nothing at all and its very comfortable. After all, those that go at 2 or 3 AM in the mornings, tend to mind their own business.
'Well, IBM complied with legal orders when they cooperated with Nazi Germany. Those were legal orders under the Nazi German system.'
IBM Germany division simply had no choice. What is not talked about much is that IBM's US division was ashamed at that forceful compliance and used their factories to manufacture weapons for the US military. It was a sort of paying their dues for their German counterparts assisting the Nazis in WWII. There are still many IBM stamped guns out there. Most are M1 Carbine rifles.
If a company is truely global, it also means they have to apply to the laws of their respective countries they do business with. If the US doesn't want US based companies to play ball with China's restrictions, they simply could put a trade restriction against China. It would be like what has happened with Cuba then. That wouldn't be the best of moves since China is a huge manufacturing country that makes a lot of products we import. You simply can't ban Internet companies ability to do trade with China and not ban everyone else as well. If they don't like the sensorship China requires by law, then take it up with China and the UN to pressure them. China, regardless of acceptance, makes the laws that pertain to their country and others either abide by them if they want to deal with China or stop doing business with them.
OS/2 fell because of bad marketing and low sales in comparison to Microsoft Windows. Linux is used on IBM p5 line, but it is hardly replacing AIX. AIX is still IBM's flagship product for UNIX servers. No matter how much they support the Linux community, they don't sell Linux as the solution for UNIX servers, they sell AIX. Also, why would Apple compete with Dell, HP, etc... They have a pretty looking over priced PC now. Dell and the like can take some designers and make their own pretty looking PC cases and cram hardware into it just as well. If they dumped OS X, they would also likely dump the Mac. Which would leave them in one place, iPods.
One other thing to remember is that they have a small subset of hardware to support. Meaning their driver development and overall system support is limited to the small line they produce. He is arguaing like they would attempt to support all PC platforms which doesn't seem like it's in their plans.
They may eventually support MS Windows on it, but the only way to make a difference in their PC versus another is the OS. It reminds me a lot of BeOS. Fast, reliable, well coded OS, but support and lack of drivers for new hardware and costs overall killed it. Apple could still keep up, but who know if they will.
After working for Lowe's Companies for 5.5 years at their corporate offices, I have seem much more in the realm of distribution. Newegg's distribution seems like childs play compared to larger retailers like Lowe's, Walmart, Home Depot, etc... Think of 1.8 Million sq ft of warehousing. Products that are already barcoded by machine, flying down the highest speed conveyors past scanners to reach their appropriate shipping area. They service hundreds of semi-trucks a day and ship quite a bit more. Imagine instead of people doing sorting, labeling, etc, your system solves the distribution nightmares by solving the travelling saleman problem for all your trucks going to thousands of locations that sell to millions of customers. Quantity, routing, and distribution is all computerized and selected based on market sales, in-stock needs, and store orders. The only need for people is to palletize the product and get it on a truck. The sheer size of "big box" retail stores are as large as the larger of Newegg's warehouses.
For some reason after seeing this in action I'm not impressed. If you want to see some amazing things, try and get Walmart, Lowe's or similar companies to do a show and tell of their facilities. It's even more impressive when you get into the data centers they use as well. Granted, much of the marketing and business decision software is secretive and tightly kept, but the systems, remote DR fiber links sychronizing TB's of storage accross thousands of miles, and the raw power of computing is impressive. I think that's why I went to work for Lowe's in the first place. I mean, coding for IBM SP frames right out of college seemed rather appealing. Heh, and at one point I thought they were just a company that sold hammers and stuff. Who knew that retail could be so innovative with their implementations and solutions.
Linux and AIX you mean. IBM doesn't support running Sun Solaris and I am sure Sun has no plans to port to Power architecture from Sparc/x86 any time soon. Although, it would be rather fun to port OpenSolaris to the Power platform just for the hypocracy in it...
Windows may hold the Enterprise Desktop environment, but they are a far cry from holding the server market in business. IBM is the leader there. They build the top end, most reliable, and highly available systems short of a custom engineered product for complete failure detection. In short, IBM is doing more for large scale business than anyone else these days. If you were running a large company, say a Fortune 50 retailer, would you install a bunch of Windows servers and setup nightly reboot jobs to keep things running smoothly, or would you setup AIX servers with HACMP (IBM's high availability offering) to run even if the hardware dies, sites go down, net connections drop, etc... you get the drift. It's all about what people can afford and the best is IBM.
You forgot these...
5. Continue hiring mostly foreign workers on H1B to train.
6. Crush all other ISP's
7. Outsource everything to India to reduce expenses by 400%
8. Laugh all the way to the bank and when other countries are in complete control of our internet services...
Oh and almost forgot
9. Deploy orbital brain lasers
"Ironically, the most readily available sources of accurate online information on bomb-making are the websites of the radical American militia."
Should this be a surprise?
As I have before and he now takes great pride in my accomplishments. Although he is mostly worried now about my current work environment and stress levels... Ah, the life of an engineer...
On the mainframe, a programmer must have the concerns of system resources and getting it right the first time. Jobs for mainframe are scheduled to optimize system usage and consistantly run at a set capicity designated in planning. This said the worst thing that happens to a mainframe programmer is an abend (a job that fails and requires interaction, often causing other jobs to get backed up). Loss of system time means a significant loss of money in the mainframe world. That directly affects the process they must take to submit and run jobs. If they are coding JCL or COBOL, their code must be reviewed and run on a test mainframe if possible. It must be exact which takes a certain kind of programmer to get it right. As a suggestion in handling mainframe programmers, it is important to improve upon the stress that comes from deadlines and being right the first time. Most mainframe programmers I know are over stressed, over worked and need the support and coaching from their manager that will make their lives a little better. Be cautious of resource usage and allocations though and make sure they have a procedure for implemtation that is strictly followed. As a manager over mainframe teams, you should be strict with procedures, but don't hold it over your employees heads. They have a hard enough job to be constantly docking them in their reviews because of not quite getting it 100% every time. After all there are not as many mainframe programmers as they used to be, so keep the ones doing a good job. UNIX programmers on the other hand depend on what their designation is. System level coders usually have attitude to cope with and you have to take that in stride. On the System Engineers level, you have to also manage resources and allocations. It is important for them to worry about a balanced system performance and ensuring that a poorly written application can be dealt with, not causing pain to the other applications since it is a multi-user system. Proper tuning will handle this. On the application programmers side, the most important thing is to make sure the "big picture" gets taken care of. You need an architectural layout and design of your applications that will fit into the business model over all. The right people need to be consulted to ensure your programmers are not writing a peice of code that doesn't fit well. Code reviews and some level of strictness are required, but not to the level of a mainframe programmer. In UNIX you can recover and risk far less losses with application downtime than on the mainframe. Restarting applications here is not as costly. On the Windows side, it's about interaction and getting the product moving quickly. Windows programmers are either concerned about the tightness of the application or the features it provides. To properly write and deploy a Windows application, strict QA must be implemented and some minimal level of profiling must be done. The important factor is user perception in this case. You want the end user to see it run fast and have tons of features. It will have bugs and the bugs are not looked at interactively due to it running on a user system. You people need to be geared towards debugging problems with the minimal interaction available. This requires that your coders know their code in and out. You want to develop and culture that ability and make sure you keep track of bugs for improvements. Overall, people are people, but the stress levels of a mainframe coder are for getting it exact and right to prevent downtime and backing up jobs. For UNIX engineers and applications programmers, it's in stability of the system and proper interaction with the system and business architectures that is stressful. With Windows programmers it's in the end users perspective and timeline to release to market that is the most stressful. Realizing those things, you can appropriately develop a plan to help manage and cope with the managerial needs of your people. Keep in mind, if you are a resource and career manager, make sure you keep your people
Programming beyond the "Here is how you get this output" method is much different. You can show someone functions, structure of the language, even useful API's to accomplish most things. What you can't show them is how to design, and develop within the art of programming. There is a huge difference between programmers who create and programmers who emulate. Given a good foundation of skills, information, education, and basic experience in programming, one can start the life long development of creating programs. There are always new techniques, new tricks, better ways, things to use, more creative designs, and more ways to write your code than one normally imagines. What she did was start her journey to that ability. She has learned how to do the base set, let's see if kids her age that have that ability can show their creativity as well. So far I am only impressed at the fact she has taken the tests, some started programming (myself included) even earlier than that. At the time I was 6 years old, reading lines of basic from books and seeing what they did was easy to do on my TSR-80, not to mention learning commands. It wasn't until later that I started with useful languages (well if you call pascal useful...) and then progressed to my current professional level. So it is possible for kids to program, and I am happy to see that some parents encourage that. Speaking from personal experience, it's better than having your dad tell you at age 16, "You will never go any where with this linux thing. You spend too much time on the computer..." I can only imagine what would have happened if I had decent programming classes taught back then....
Most people are not noting the primary reasons for the cost increase over building a cheap cluster. Their system can run off one power outlet. It is designed to pull a smaller amount of power than a typical cluster. Imagine the power bill from 96 350 watt power supplies. That's 33.6 kwh if my calculations are correct. They claim on the spec sheet a maximum consumption of 1.5 kwh total (1500 watt). That is considerably less in the grand scheme of things.
The other thing to consider is the cost of engineering in the nodes themselves.
If they had the extra 4 digits (3 for AMEX cards)on the back stored in the mag stripe that is a big no no to banks. CVV (numbers after your account number on the back) data was originally used so that the old swipe impact 3 part forms would not pick up the numbers. This meant that you would not have all the info that the banks had on the card from the imprint of the card numbers from the front. They often called in the card when making a purchase and read off the CVV data to the bank when making a charge.
What they typically use that data for now is online orders or "card not present" orders. It's a way to validate when do a "card not present" transaction. Most banks require this now for retailers doing online transactions. There are only two bits of info that should not be stored on the mag stripe. One is the CVV data and the other is your PIN number.
For those that have access to mag stripe readers, especially ones that use keyboard input, try running your card through and dumping out the data sometime. You will see exactly what is on them (if your reader supports more than track 1 and track 2 reads that is). Last time I did, my AMEX, Discover, and VISA all had my name on there, card number, expiration and a few other numbers noone typically uses for transactions.
That's incorrect. The mag stripe contains only the 16 digits of the card. (I used to work for MasterCard).
...." on the reciepts some places have.
;) Unless someone is watching you punch the keypad, has a camera mounted to watch and are capturing your mag stripe data, you are fairly safe.
*bzzzt* wrong answer. Actually if you read up on mag stripes you will find that they contain more than that. The data (tracks) of the cards often contain your name and other bits of information the bank finds handy. How else do you explain the "Hello Mr/Ms
Back to the article at hand, it as much of a problem for merchants as it is for consumers. For the merchants, they risk charge backs with no "valid" signature to dispute them with. If it says the persons real name, they generally can win on that argument. However, saying that it is a valid charge and reading back "Shamu" as the name doesn't work so well.
We need to move beyond retailers using un-encrypted data to validate accounts and also having them enforce policies that strictly require the checking of cards and receipts. Will this happen? It's unlikely to occur soon.
In the mean time the ultra-paranoid can skip using a credit card/check card and use DEBIT/ATM only. If you have a true DEBIT/ATM only card, the card number does not do a lot of good for anyone. You have to use a pin to use it and from knowing how that works, trust me, it's better
Change happens. There are a few ways of dealing with it, you chose to move on and quit. Other chose to stick it out, while some stop everything and complain about the fact change happened and why it should happen to them. Sound familiar? There is a book called "Who moved my cheese?" which is more of a social outlook on acceptance of change. People who choose to move on typically have something more favorable either in pay, benefits, or general happiness lined up.
I recently have seen people leave the company I work for on various complaints and reasons. They were unhappy with management and the way they were treated. Others were unhappy with the loss of the extra benefits that used to make my company stand out. While others were unhappy with the work load that detracted from their family and personal life. To this date, I have not seen many of these problems corrected. Leaving because the type of work you are doing changes is not what I would consider a valid reason, but that is me. Every job will have it's moments, but as a friend of mine who left his job here told me once, "You have to come up with a list of why you work where you do." Some people place time with family, friendship with co-workers, money, advancement and career goals, location, or even what language and environment they code in on that list. It's your list, order it how you would like and then decide based on what you value the most from your work. If your top points are not met, then should you leave? Will you be content with staying? That is for each individual to decide.
fork either returns 0 or -1, so a comparison to value 1, then returning 0 should not happen. I believe it would end up returning a NULL value since the if fails... interestingly enough, you would fork four child processes, which would fork 4 child processes, etc, etc... nasty little way to eat or process space. Sort of like the shell script of script a calls script b which calls a.... etc Even worse when you background those and put a wait in there, blamo.
A good programmer who codes in C should be able to understand what is going on. Still it is no substitute for real documentation... Someone should let the folks that enter the IOCCC that they have some of the worst self documenting code I have seen though. It's like trying to read the Xlib programming manuals after a few to many drinks (not recommended and may result in a nasty hangover in the morning).
Most larger companies value support of any software product as the critical key element. They want a support contract, they want an end to finger point at when something breaks. Buying a distro like RedHat EL or SuSe Server, etc will allow them to buy a support contract. Having companies like Novell or Redhat backing a product is comforting to large corporations. If you are looking for something for a company that doesn't spend millions in software and support each year, then check out debian, gentoo, slackware, etc... Keep in mind, you want to setup things to make support it easy, rather than a nightmare. Most companies I know of that don't opt for well backed distros like Redhat/Novell Suse will often customize their own distro based off something else and support it internally. The best thing is to look around, try a few, be un-biased, and keep installation, configuration, upgradability and support aspects in mind. People will try to convince you one is better than the other, IE my distro is bigger than yours, etc, etc... In the end, the one that best fits the support and comfort levels of the company is what is best.
It's not water based, but rather an acrylic latex house paint. Check the website for details. They also have additives to mix with other paints instead of using their base coat paint. Say if you wanted to have a sheilded room painted in a nice mauve with white trim or something...
This a very good strategy for IBM. They have a lot of good researchers coming up with ideas that will cost money to develop. Many of those can benefit the computing industry in general and help them in further development of core technologies. This is the easiest way for IBM to, in a sense, get free development. By opening up the patents, they don't have to spend money on implementation which will allow them free use of the technologies to futher their products with no real development cost. The open source community can implement and futher the technology covered by the patents allowing IBM to integrate that back into their core business to further the commercial products without the development cost.
It makes perfect sense if you think about it. Open Source won't earn them money, but will in turn give them the implementations without the expense of their own development team doing it. A community of free development that can implement technology that their researchers create, who could ask for more?
You have some good points. Also it is good to ask what is in it for you. You need to make it clear that you want to start a career, not just a job with them. Having techincal questions, and other things to ask them is helpful.
My first major interview was for the US Naval Academy. I should say it went well since I received a nomination to the Academy from my Congressman. Interviews are largely about proving to the future employer why they should hire you. Sitting back will never accomplish anything. You have to make a point to them that you are the one and humbly show your abilities. Attitude plays a large part of it as well.
Both a combination of my social skills and my techincal abilities helped me prove to the Fortune 50 company I work for that I was the guy they needed. Since then, I have become a valuable asset to the company with a rather impressive resume and 5 years of experience with them alone. Even without a degree, if you have the right skills, the right drive and attitude, you can start a career at a company that has strict requirements. Education is very important and I would definately stress finishing your degree. If you are a CS major and going that route, immediately after you start coding, try to find internships or coops. If you don't start building your job experience up and working for other places while in school, it will be much harder to find a job. Some companies have also been known to do repeat internships if you prove to be a good employee and even offers once you get out of college.
AT&T Bell Labs used to hire nothing but the best... Where are they now? I hope the folks at Google make sure they know the mistakes that came from the business there and use their brilliant minds to create real actual products they can pay for those minds with.
This has been reported over a year ago. See the DOJ Press release...
DOJ Link for Salcedo here.
Also see the Security Focus article here.
There are plenty of articles about the case. The DOJ and FBI have most of it as public information for the search engine savy. The key is, there is a lot of potential damage to end consumers and the company with crimes like this. Considering it was his second offense and he hadn't even gotten off probation for the first, 9 years seems reasonable to me.
I have been following this case online for a while. They were involved with the 2600 scene, attented "Hacker" conventions, and were already known for things that some might view on a grey line (IE telephone companies - See Telcodata.us). The thing is that occasionally there are bad apples that learn from others and commit crimes. Showing a 9 year prison sentence might help keep some of the apples from rotting.