I use a PowerBook G3 400 because it meets my day-to-day needs; runs Mac OS 9, Windows 98 (in emulation fast enough for most of the games I play), and UNIX (Mac OS X); allows me to do computational modeling and analysis (neural modeling); has region-free DVD; and is pretty well invulnerable to most security problems. Even when I upgrade next week to a G4 500, I'll keep it around to watch DVDs on.
iapetus wrote: "Progress in computer hardware has followed this curve and continues to do so. Progress in computer intelligence however, hasn't. Computers are still stupid. They can now be stupid more quickly. This isn't going to produce super-human intelligence any time soon."
The problem from the perspective of a working neuroscientist is that we don't yet understand how the brain is intelligent. On the other hand, things are starting to fall into place. For example, we have a hint of why neural synchronization occurs in the brain, because we're beginning to realize that time synchrony is something many neurons are very good at detecting. We're also beginning to understand memory formation in the cortex. It seems to involve the creation of clusters of synapses, and those clusters get activated by time-synched signals. There's some evidence for analog computation, and there's some evidence for almost quantum computation. So we're beginning to understand how to build a brain. That seems to be the hump, so I'm fairly confident I'll live to see computers at least as intelligent as I am. And I'm 54.
There was a discussion of this on MWJ some time ago, and the editor indicated the reason is MPAA paranoia about DVDs being played on any machine where a debugger (e.g., gdb) can be used to examine the decoding code when it's executing. Hence no DVD playback on UNIX, Linux, or Mac OS X, and MacsBug is disabled when you run the Apple DVD Player under MacOS 9. So it's not a rumor.
I teach Java and C++ at the university level. C++ is more powerful, since it supports non-OOP design paradigms, but for the introductory courses, I actually prefer Java. Java, once you get beyond the froth, is a decent language that is a bit safer than C++ and avoids pointers. I'd start with two semesters of Java or C++ to teach basic programming skills, and then I'd move on to C++ for data structures, algorithms, and generic programming.
I'm inclined to encourage MS to take this step. The reason, you ask? I think it's probably a bad idea for them in the long run. It reduces the barriers to entry and is likely to irritate corporate customers. Corporate customers are the ones they can't afford to irritate.
In the past, the relative security of Macintosh systems was due to the lack of remote access and the generally unattractive nature of a user community that was willing to give you a barbed-wire enema if you crossed it. (Well, not literally, but there were enough people in the community who were willing to go after a virus-maker that it was a bit risky.) I wonder if either factor has changed much. Mac OS X 10.0.1 comes with standard telnet, rlogin, rsh, etc., disabled and SSH installed instead, and the user community remains small and intolerant.
The severities listed are essentially similar to those I've seen used by the FAA. Software is not deliverable with any severity 1 or 2 faults and the severity 3 fault density is limited to no more than one per KSLOC.
I'm a computational scientist, so I'm one of those people that actually do science using computers. The problem is that there are a large number of interesting systems that can only be modeled in crude approximation. (Your computational costs go to infinity much faster than you can improve your model.) These are called 'complex systems', and there is theoretical reason to believe they can't be modeled. Computer scientists rarely understand real science, and I suppose that's the reason for this belief in particular.
That's how brains seem to work. When synchronization is seen in the brain, it is suspected to reflect multiple regions attending to a single object. Note that synchronization in the brain is self-organized, rather than driven by a clock.
The idea in negotiating a business deal is to first find the kernel that maximizes total value. This idea comes from non-zero-sum game theory. If you try to work a deal that doesn't maximize total value, someone disgruntled with the current proposal can sell an alternative to the other players that increases their cut and is likely to reduce your cut. Maximizing the total value maximizes the value to you.
See a lawyer. Then negotiate, with the help of the lawyer. The goals are (1) to maximize the total benefit to all parties, and (2) get a reasonable share of that benefit. Don't be a pig, but also don't let yourself be run over.
I seriously can't credit this report unless I see a scientific paper with the data. There _is_ evidence that a rich environment improves cognition and memory, so I suspect some other cause--perhaps increased exposure to neuroactive substances or carbon monoxide, or increased stress-related anxiety. Alternatively, there may be some environmental pollutant entering via the olfactory system and damaging the memory systems.
If you use a Mac, download icab. It supports them all and has versions for MacOS 9 and X. It's currently in beta, but it will cost you about $30 once it's released.
The voice communications system used by the FAA enroute centers is currently transitioning from Tandem to Windows NT for the control subsystem. Think about it.
Some systems have to be on the internet despite the fact that they work with sensitive data or provide critical functions. The right approach is to start with a strong firewall with a man in the loop. There are a number of other design rules, including the use of internal compartmentalization and redundancy, that can help. Just remember that no automated system can deal reliably with the threat posed by hacker creativity. There has to be a guard on duty somewhere.
The issue of the article is that a hacker outside MS has seen the source code, which makes it easier for him to design attacks. The real problem, however, is not this loss of security, but the general vulnerability of software crufted by the MS development process. Most software developed for the the Gummint is required to have a remaining software fault rate (read 'bug rate') around 1.5 per KSLOC and is supposed to take security seriously. I doubt MS code meets those standards.
American software companies look at two things in choosing new graduates: your grades and any internships you've done. Whether business or research internships are better depends on the company.
Re:I work in a place where CS = CIS = MIS
on
CS vs CIS
·
· Score: 1
Try this on:
I'm a senior software engineer and next month will be moving to a senior lectureship in computing. I've taught core CS for seven years. I don't have a CS or CIS degree.
What I do have are a BS, MA, and ABD in math, and a natural science PhD (in computational neuroscience that I earned in a psychology laboratory). I also have 32 years of experience as a software systems engineer. I know how to solve problems and how to teach others how to solve problems.
The message--it doesn't matter what your degree is; someday you'll be out there in the real world solving problems. So study the things you will need to know to solve problems (lots of applied math) and plan to go back to school from time to time for a refresher.
In the US, most employees are employed 'at will'. So it doesn't have to make sense if it's what management wants to do.
That said, I don't blame you for not wanting to be a manager. It looks like it's time to move on.
By the way, I'm taking a 50% pay cut in my new job, as it short-circuits 11 years of academic career (2 post-docs + 1 assistant professorship). Sometimes the other factors are very important.
Slashdot Mirror
I use a PowerBook G3 400 because it meets my day-to-day needs; runs Mac OS 9, Windows 98 (in emulation fast enough for most of the games I play), and UNIX (Mac OS X); allows me to do computational modeling and analysis (neural modeling); has region-free DVD; and is pretty well invulnerable to most security problems. Even when I upgrade next week to a G4 500, I'll keep it around to watch DVDs on.
The problem from the perspective of a working neuroscientist is that we don't yet understand how the brain is intelligent. On the other hand, things are starting to fall into place. For example, we have a hint of why neural synchronization occurs in the brain, because we're beginning to realize that time synchrony is something many neurons are very good at detecting. We're also beginning to understand memory formation in the cortex. It seems to involve the creation of clusters of synapses, and those clusters get activated by time-synched signals. There's some evidence for analog computation, and there's some evidence for almost quantum computation. So we're beginning to understand how to build a brain. That seems to be the hump, so I'm fairly confident I'll live to see computers at least as intelligent as I am. And I'm 54.
There was a discussion of this on MWJ some time ago, and the editor indicated the reason is MPAA paranoia about DVDs being played on any machine where a debugger (e.g., gdb) can be used to examine the decoding code when it's executing. Hence no DVD playback on UNIX, Linux, or Mac OS X, and MacsBug is disabled when you run the Apple DVD Player under MacOS 9. So it's not a rumor.
I teach Java and C++ at the university level. C++ is more powerful, since it supports non-OOP design paradigms, but for the introductory courses, I actually prefer Java. Java, once you get beyond the froth, is a decent language that is a bit safer than C++ and avoids pointers. I'd start with two semesters of Java or C++ to teach basic programming skills, and then I'd move on to C++ for data structures, algorithms, and generic programming.
Remember the water-powered computer (see the Hacker's Dictionary)?
I'm inclined to encourage MS to take this step. The reason, you ask? I think it's probably a bad idea for them in the long run. It reduces the barriers to entry and is likely to irritate corporate customers. Corporate customers are the ones they can't afford to irritate.
In the past, the relative security of Macintosh systems was due to the lack of remote access and the generally unattractive nature of a user community that was willing to give you a barbed-wire enema if you crossed it. (Well, not literally, but there were enough people in the community who were willing to go after a virus-maker that it was a bit risky.) I wonder if either factor has changed much. Mac OS X 10.0.1 comes with standard telnet, rlogin, rsh, etc., disabled and SSH installed instead, and the user community remains small and intolerant.
The severities listed are essentially similar to those I've seen used by the FAA. Software is not deliverable with any severity 1 or 2 faults and the severity 3 fault density is limited to no more than one per KSLOC.
I'm a computational scientist, so I'm one of those people that actually do science using computers. The problem is that there are a large number of interesting systems that can only be modeled in crude approximation. (Your computational costs go to infinity much faster than you can improve your model.) These are called 'complex systems', and there is theoretical reason to believe they can't be modeled. Computer scientists rarely understand real science, and I suppose that's the reason for this belief in particular.
That's how brains seem to work. When synchronization is seen in the brain, it is suspected to reflect multiple regions attending to a single object. Note that synchronization in the brain is self-organized, rather than driven by a clock.
The idea in negotiating a business deal is to first find the kernel that maximizes total value. This idea comes from non-zero-sum game theory. If you try to work a deal that doesn't maximize total value, someone disgruntled with the current proposal can sell an alternative to the other players that increases their cut and is likely to reduce your cut. Maximizing the total value maximizes the value to you.
See a lawyer. Then negotiate, with the help of the lawyer. The goals are (1) to maximize the total benefit to all parties, and (2) get a reasonable share of that benefit. Don't be a pig, but also don't let yourself be run over.
I seriously can't credit this report unless I see a scientific paper with the data. There _is_ evidence that a rich environment improves cognition and memory, so I suspect some other cause--perhaps increased exposure to neuroactive substances or carbon monoxide, or increased stress-related anxiety. Alternatively, there may be some environmental pollutant entering via the olfactory system and damaging the memory systems.
CSS is currently at the top of the queue.
If you use a Mac, download icab. It supports them all and has versions for MacOS 9 and X. It's currently in beta, but it will cost you about $30 once it's released.
The voice communications system used by the FAA enroute centers is currently transitioning from Tandem to Windows NT for the control subsystem. Think about it.
Some systems have to be on the internet despite the fact that they work with sensitive data or provide critical functions. The right approach is to start with a strong firewall with a man in the loop. There are a number of other design rules, including the use of internal compartmentalization and redundancy, that can help. Just remember that no automated system can deal reliably with the threat posed by hacker creativity. There has to be a guard on duty somewhere.
The nice thing about open source is that there are many eyes verifying that the code is solid.
The issue of the article is that a hacker outside MS has seen the source code, which makes it easier for him to design attacks. The real problem, however, is not this loss of security, but the general vulnerability of software crufted by the MS development process. Most software developed for the the Gummint is required to have a remaining software fault rate (read 'bug rate') around 1.5 per KSLOC and is supposed to take security seriously. I doubt MS code meets those standards.
American software companies look at two things in choosing new graduates: your grades and any internships you've done. Whether business or research internships are better depends on the company.
I'm a senior software engineer and next month will be moving to a senior lectureship in computing. I've taught core CS for seven years. I don't have a CS or CIS degree.
What I do have are a BS, MA, and ABD in math, and a natural science PhD (in computational neuroscience that I earned in a psychology laboratory). I also have 32 years of experience as a software systems engineer. I know how to solve problems and how to teach others how to solve problems.
The message--it doesn't matter what your degree is; someday you'll be out there in the real world solving problems. So study the things you will need to know to solve problems (lots of applied math) and plan to go back to school from time to time for a refresher.
In the US, most employees are employed 'at will'. So it doesn't have to make sense if it's what management wants to do. That said, I don't blame you for not wanting to be a manager. It looks like it's time to move on.
By the way, I'm taking a 50% pay cut in my new job, as it short-circuits 11 years of academic career (2 post-docs + 1 assistant professorship). Sometimes the other factors are very important.
1. Heinlein was not libertarian. He had a high regard for character--that is, initiative and a sense of responsibility to society.
2. He was into nudism and nudist photography.
3. He was anti-religion.
4. He was an individualist.
5. He had a pessimistic attitude towards the future, and
6. He lacked a high regard for editors...
His later works reflected his real attitudes even when he was writing his earlier stuff. Read the end of Beast and ask what is a "John Thomas"?
If MS goes down the .NET path, will we ever see a standard-compliant VC++?