Finally, someone who has a clue! Thanks, Dr. Bent, for the insightful command. They are few and far between in on this topic.
I can't believe how many times I hear people that are supposed to be independent, free-thinking professionals use the term "PHB", referring perjoratively to their managers, and then complain about getting let go!
Do you need to be beaten with a Clue Stick?
I ran my own consulting business for 6 years (not in the computer industry), and one thing I learned -- the vast majority of people don't take responsibility for their actions, don't invest the time and effort to learn the business they are in, and blame others when things go wrong -- especially when you have to let them go. I finally folded my business, because I figured out that I would rather program than lead people. But, I learned a few things along the way by wearing the "leadership" mantle.
If you are in the technical industry, and are competent, and are working for a manager who is incompetent, it is YOUR FAULT! Do you know why? Because it is always your fault. Period. No matter what happens. It doesn't mean you are a bad person. It's just the way it is. Life isn't fair. After I shut down my business, me and my wife were 1 week away from living in a ditch. Was it my fault? Yes. Could I have blamed my people, my mentors, my wife, someone 3000 miles away that could convince someone that they could do the same think I do, but cheaper? Sure! What the heck good would that do? Oh right, it would make me feel better. But, we'd still end up living in a ditch.
However, would you get into a car with a drunk behind the wheel, then complain when the car crashes, and you get injured? Hmmm....
First, start of respect your leadership. Even if they are "PHB"s. Because, like it or not, they know how to do some things better than you -- maybe only the ability to bullshit and shmooze their way to the top -- but they are there, and not you. If you are so smart, go and take their jobs. If you are even smarter, arrange for someone better than you at leadership to take their jobs. If you don't want to do that, shut your pie hole.
Second, get a clue about your business. If you can't present yourself in such a way that you are not perceived as more valuable than some faceless name 3000 miles away, then you need to take a serious look at your skill set -- and not just your technical skill set. Do you know what your manager finds most important about the area your are involved in? If not, Why Not? If your boss doesn't realise how important your position/department, are you embarking on an extended, intense education campaign to make certain he learns this? If not, once again: Why Not?
Or, are you like most people, and believe that the "Education Fairy" is going to sprinkle magic "Clue Dust" on your boss?
So, there is/are some large investor(s) that suddenly excersised huge financial instruments, exposing the RBOC to USD$50M in potential losses, should SCOX increase in value?
This is the only rationalle that I can think of, for a big bank to buy SCOX; banks are typically not big risk takers.
I wonder if there is any way to discover who bought "Call" options on SCOX from the Royal Bank? Perhaps some investor whose name starts with M? Maybe it didn't take a rocket scientist corporate finance guy to figure out how to force an "independent" third party to buy a big chunk of SCOX, if you don't want to do it personally...
And in all those rewrites, still no simple way to control the various timeouts that TCP/IP imposes on the user, making TCP/IP virtually unusable for mission critical, deterministic applications, such as SCADA.
Don't get me wrong; TCP/IP is great -- for browsing the web. Don't get your web page in a reasonable time, or the connection dies? Fine. Try that approach when you are trying to control a remote piece of equipment that, say, prevents a guyser of oil from appearing in someones back yard...
Hopefully, Sun will provide support for strong determinism in their TCP/IP stack. Until then, we're forced to implement custom protocols using UDP/IP, and provide our own determinism, error detection/correction, encryption, etc...
For those of you that might be interested, SCADA is an acronym for
Supervisory Control And Data Acquisition
It defines nothing about whether or not COM, DCOM, OPC (Ole for Process Control), or any other proprietary communication framework is used (contrary to some other highly moderated statements you are likely to see in this discussion)
The bulk of serious SCADA systems in place are probably legacy systems of some kind, including many variants of UNIX systems (we have old micro VAX systems still in use). Many of the newer systems are Windows based, and are obviously subject the standard Windows viri, worms, etc.
Worse yet, these systems are very difficult to upgrade or patch, due to the critical nature of their duties. It is not unlikely that a large portion of the Windows based SCADA systems in the world remain unpatched, and are "safely" firewalled off from the internet.
Of course, the problem is that much of the monitoring gear used to diagnose network issues is also Windows based, and carries with it the standard retinue of Windows viruses and worms, right into the heart of the control center.
These UNIX systems have run (and will contiue to run) uninterrupted for years at a time. We have calendar alerts in place to tell us to go manually reboot unix machines after months of uptime, just to ensure that their SCSI drives will spin back up (in case of a control center power outage, etc.)
Somehow, I don't think that is an issue for some vendor's SCADA systems based on more popular OS's, but I might be wrong...
I'm using:
$ uname -a
Linux tramp 2.6.0-test2-1-386 #1 Sat Aug 2 10:49:01 EST 2003 i686 GNU/Linux
on my laptop (600MHz OmniBook 6000), and it is awesome! Much, much better than 2.4.21. It literally feels like a machine with twice the CPU it had before. Mouse motion is incredibly smooth, switching focus is much quicker and less jumpy, interactive applications are... interactive.
The only issue I've run into is related to when it powers down the display; sometimes (about 5% of the time) the display won't come back on. I'll have to switch into the console (CTRL-ALT-1), log in, and reboot, blind. Other than that, I've run into no issues at all. I've used Open Office, emacs, mozilla, g++ extensively on the machine, with no glitches.
I installed using
$ apt-get install kernel-image-2.6.0-test2-1-386
and then I had to add the following to my/etc/modules:
mousedev
psmouse
rebooted, and voila! Brand new user experience. I would recommend it highly. I have my filesystem NFS mounted from a nice, Debian "stable" firewall system, though...;^)
"The open source license is not open, because you can't take it and ever use it in a job-creating activity," Gates said.
Translation:
"The open source license is not open, becuase you can't take it and ever use it in a job-creating activity at Microsoft"
One of the best features of open source, is its ability to melt away unnecessary expenditures of money on software not directly related to the business goals or your company. It is inconceivable that any right-headed CIO or CFO would spend penny one on a "Word Processor", for example. The ONLY company that this decision would hurt is Microsoft. A company frees up virtually 100% of their software dollars to hire real, local software developers to develop solutions to their own, personal, business problems.
Wow! That has to be one of the most eloquent things I've read in a long time... and on the 'net, ever! Did you actually just write that, "on-the-fly", or have you been working on that for some time?
Consider this, though. If what you say is true, that "Society's laws grow from its mores", then can't we expect the same results as the Romans? The "mores" of Greed, Power, and Lust surely will overwhelm those of responsibility, honour and bravery. The former are "natural"; the latter -- they must be forged through much discomfort.
I think that you might actually be wrong. I think that people actually still (for this and perhaps one more generation...) have respect for the value of "a work", even if it is easily distributable. I think people still have at least a vague sense "payment for value"; What the masses are actually rebelling against is the extraction of wealth that occurs between those that Invent, Dream, and Create, and those that enjoy the results of that creation.
Generations of Management Schools have produced a stratum of society that cannot Create, but can only think in terms of Arbitrage -- Buying Low and Selling High -- who are prepared to feed the lusts of those in power, as long as they promise to protect these Manager's ability to ride the industries from which they derive their wealth. Adding nothing, but taking a slice of everything.
Unfortunately for this great mass of "Middlemen", the internet is becoming the great solvent of unnecessary power structures. When the mass of Creators and Enjoyers can be brought face-to-face with virtually zero cost and delay, the only thing left for the "Nation of Salesmen" to do is something -- anything -- to prevent it.
Well, those of us that are both technical and creative must fight back. Not with a bit of file sharing "freedom" (remember, the RIAA still holds most artists by the short and curly, and they *will* squeeze), but with a real, practical mechanism to provide direct, instantaneous compensation from the user to the creator.
You want to break the backs of those who provide nothing but Arbitrage? Then create a system that provides a stable, reliable market for the exchange of ideas and value. Create a system that pays the authors more than RIAA, and the artists will abandon them. On the other hand, if we continue on the path we are on, the artists will only hold on more tightly to their only semi-reliable source of income -- the recording industry.
The future of the creative commons is in our hands.
It's pretty tough to beat Debian Linux + hostap + shorewall + a few NICs to build a nice Linux 802.11b Access Point/firewall/webserver/fileserver. Runs for months at a time, serving up two external web sites, ~1/2 dozen household linux and WinXP clients and mobile zauruses and laptops.
Want to stay up-to-date with the latest security fixes? apt-get update; apt-get upgrade Done.
Especially when I have to reboot my "professionally administered by a multi-million dollar corporate ID deparment" work Window NT box at least weekly -- and I only use it to read my email! Flaky Microsoft garbage...
Last I heard, there were >30 million millionaires in india (I don't have a reference to back this up, though.) That's more millionaires than there are people in the country of Canada! I think there is enough of a market there to make a cellular network profitable...
Perhaps you misunderstood the logical consequences of this statement. The author was NOT saying that a trusted computer professional cannot have non-computer hobbies. He was saying:
The vast majority of what I have been able to provide my employers as a computer professional is derived from what I absorb about my craft outside of working hours. I am continually astonished at what self-purporting professionals in the computer industry don't know... And then I remember I picked it up in my spare time, just thinking about things while on the road, tinkering away on a side-project, reading a physics/medical/electronics/etc. publication. I can't imagine being able to do what I can do now, without having been obsessed with my field, for a significant portion of my life.
For similar reasons, the examples of professionals from other fields is false, unless I'm misunderstanding you.
Tell me; why would you want a lawyer that doesn't spend a signifianct percentage of his waking hours thinking about law? Any lawyer that doesn't is probably working as a public defender, and just putting in his/her hours...
As far as sales goes, I ran a rather large sales company for some time, and hired and trained at least a hundred full- and part-time people to do technical sales. I could tell within a week (with one exception that I still haven't quite figured out) whether they were going to be worth spit -- did they eat, sleep and breath the business, 24/7, until they began dreaming about it? If not, they were going to be toast. Did they memorize sales masters like Tom Hopkins, until they actually began sounding like them? If not -- done for.
The same goes for computer professionals (except, perhaps, for "Cracker Jack" box "professionals", like MCSEs, etc.) But really, is having the capability to install an OS, or configure a small LAN, really what we're talking about here?
Perhaps you meant to say that you pay 7 bucks/mo more for 2 distinct IP addresses? You probably only have one connection (hole in your wall, ADSL/Cable modem, etc).
The technique describes depends on two very simple mechanisms; A) assuming that a NAT router will decrement each packet's Time-To-Live (TTL), thus exposing its presence, and B) searching for independent, incrementing sequences if IP packet ID's, to estimate the number of hosts behind the NAT router.
The time before there are "fixed" versions of both NAT (which don't decrement TTL), and of IP packet ID's (changing all ID's into a single monotonically increasing order, or randomizing them) will be measured in hours.
Hopefully the authors of this paper aren't doing research for a living...
We've just purchased a load of HP 2025 20.1" LCD 1600x1200 displays for our workplace --- and we've run into an interesting challenge... None of the high-end multi-headed DVI-output video cards we've bought support 1600x1200 in DVI!!! They only support "single-link" DVI output, which maxes out at 1280x1024. They support 1600x1200 in VGA, just fine. Fortunately, the HP 2025's sync up and display the analog 1600x1200 signal quite will, but it's still disappointing.
We think that the HP 2025's will support 1600x1200 if we get video cards that have a true "dual-link" DVI output, AND we obtain some "dual-link" DVI cables (the HP 2025's only came with "single-link" cables; go figure...)
However, it is virtually impossible to get an assurance of whether any particular video card is actually capable of producing "dual-link" DVI output! So far, we have deduced through testing that the Matrox 200 4-head, 400 2-head and new 550 2-head cards do NOT output "dual-link" DVI.
We have not been able to find any absolute assurance that any of the NVidia or ATI offerings have true "dual-link" DVI outputs, capable of driving a 1600x1200 display.
Do any of you have any direct, personal experience driving single- or multiple-headed 1600x1200 LCD displays using DVI video cards? We'll probably buy a bunch, if we can find one that works!
Software is... Soft. Any problem simple enough and static (unchanging) enough to be "Engineered" is put into hardware (see "Bridge", "Dam", etc.). You leave all the stuff that is dynamic, or just plain "unknown", to the programmers -- who you hope can figure out a way to solve the problem, at least semi-reliably. Many people who have "programmed", have only really ever tried to solve very simple, well defined problems. They are the ones who complain about how shoddy and unreliable programmers seem to be. Anyone who has every really tried to tackle a large, poorly understood problem can appreciate the "magic" behind "good programming".
You can't engineer most software, for the same reason you can't engineer a book; it satisfies emotion, not logic. Most of the time, "success" or "failure" of software cannot even be defined! (see staggeringly successful, stupid and unreliable systems such as Micros~1 Windoze(tm)). Failure in engineering projects is usually, tragically, easy to define (see "Columbia").
It simply takes a different kind of duck to do software than it takes to do engineering. Have you ever met someone who can remember what they had for breakfast on July 23 1993, can remember the specs on a specific type of transistor, but can't come up with a efficient method for solving some seemingly novel problem? Thats an engineer. Ever met someone who, in a "eureka" moment, manages to re-frame a sticky problem so that it suddenly becomes solvable? And then turns out the code to do it in an afternoon, and it works first try? That might be a Programmer. Sometimes, you might even get both in the same package (you'd call him a Genius). But usually, people that are good at engineering, are very frustrated by programming.
Slashdot Mirror
Finally, someone who has a clue! Thanks, Dr. Bent, for the insightful command. They are few and far between in on this topic.
I can't believe how many times I hear people that are supposed to be independent, free-thinking professionals use the term "PHB", referring perjoratively to their managers, and then complain about getting let go!
Do you need to be beaten with a Clue Stick?
I ran my own consulting business for 6 years (not in the computer industry), and one thing I learned -- the vast majority of people don't take responsibility for their actions, don't invest the time and effort to learn the business they are in, and blame others when things go wrong -- especially when you have to let them go. I finally folded my business, because I figured out that I would rather program than lead people. But, I learned a few things along the way by wearing the "leadership" mantle.
If you are in the technical industry, and are competent, and are working for a manager who is incompetent, it is YOUR FAULT! Do you know why? Because it is always your fault. Period. No matter what happens. It doesn't mean you are a bad person. It's just the way it is. Life isn't fair. After I shut down my business, me and my wife were 1 week away from living in a ditch. Was it my fault? Yes. Could I have blamed my people, my mentors, my wife, someone 3000 miles away that could convince someone that they could do the same think I do, but cheaper? Sure! What the heck good would that do? Oh right, it would make me feel better. But, we'd still end up living in a ditch.
However, would you get into a car with a drunk behind the wheel, then complain when the car crashes, and you get injured? Hmmm....
First, start of respect your leadership. Even if they are "PHB"s. Because, like it or not, they know how to do some things better than you -- maybe only the ability to bullshit and shmooze their way to the top -- but they are there, and not you. If you are so smart, go and take their jobs. If you are even smarter, arrange for someone better than you at leadership to take their jobs. If you don't want to do that, shut your pie hole.
Second, get a clue about your business. If you can't present yourself in such a way that you are not perceived as more valuable than some faceless name 3000 miles away, then you need to take a serious look at your skill set -- and not just your technical skill set. Do you know what your manager finds most important about the area your are involved in? If not, Why Not? If your boss doesn't realise how important your position/department, are you embarking on an extended, intense education campaign to make certain he learns this? If not, once again: Why Not?
Or, are you like most people, and believe that the "Education Fairy" is going to sprinkle magic "Clue Dust" on your boss?
So, there is/are some large investor(s) that suddenly excersised huge financial instruments, exposing the RBOC to USD$50M in potential losses, should SCOX increase in value?
This is the only rationalle that I can think of, for a big bank to buy SCOX; banks are typically not big risk takers.
I wonder if there is any way to discover who bought "Call" options on SCOX from the Royal Bank? Perhaps some investor whose name starts with M? Maybe it didn't take a rocket scientist corporate finance guy to figure out how to force an "independent" third party to buy a big chunk of SCOX, if you don't want to do it personally...
And in all those rewrites, still no simple way to control the various timeouts that TCP/IP imposes on the user, making TCP/IP virtually unusable for mission critical, deterministic applications, such as SCADA.
Don't get me wrong; TCP/IP is great -- for browsing the web. Don't get your web page in a reasonable time, or the connection dies? Fine. Try that approach when you are trying to control a remote piece of equipment that, say, prevents a guyser of oil from appearing in someones back yard...
Hopefully, Sun will provide support for strong determinism in their TCP/IP stack. Until then, we're forced to implement custom protocols using UDP/IP, and provide our own determinism, error detection/correction, encryption, etc...
For those of you that might be interested, SCADA is an acronym for
Supervisory Control And Data Acquisition
It defines nothing about whether or not COM, DCOM, OPC (Ole for Process Control), or any other proprietary communication framework is used (contrary to some other highly moderated statements you are likely to see in this discussion)
The bulk of serious SCADA systems in place are probably legacy systems of some kind, including many variants of UNIX systems (we have old micro VAX systems still in use). Many of the newer systems are Windows based, and are obviously subject the standard Windows viri, worms, etc.
Worse yet, these systems are very difficult to upgrade or patch, due to the critical nature of their duties. It is not unlikely that a large portion of the Windows based SCADA systems in the world remain unpatched, and are "safely" firewalled off from the internet.
Of course, the problem is that much of the monitoring gear used to diagnose network issues is also Windows based, and carries with it the standard retinue of Windows viruses and worms, right into the heart of the control center.
These UNIX systems have run (and will contiue to run) uninterrupted for years at a time. We have calendar alerts in place to tell us to go manually reboot unix machines after months of uptime, just to ensure that their SCSI drives will spin back up (in case of a control center power outage, etc.)
Somehow, I don't think that is an issue for some vendor's SCADA systems based on more popular OS's, but I might be wrong...
$ uname -a
Linux tramp 2.6.0-test2-1-386 #1 Sat Aug 2 10:49:01 EST 2003 i686 GNU/Linux
on my laptop (600MHz OmniBook 6000), and it is awesome! Much, much better than 2.4.21. It literally feels like a machine with twice the CPU it had before. Mouse motion is incredibly smooth, switching focus is much quicker and less jumpy, interactive applications are... interactive.
The only issue I've run into is related to when it powers down the display; sometimes (about 5% of the time) the display won't come back on. I'll have to switch into the console (CTRL-ALT-1), log in, and reboot, blind. Other than that, I've run into no issues at all. I've used Open Office, emacs, mozilla, g++ extensively on the machine, with no glitches.
I installed using /etc/modules:
;^)
$ apt-get install kernel-image-2.6.0-test2-1-386
and then I had to add the following to my
mousedev
psmouse
rebooted, and voila! Brand new user experience. I would recommend it highly. I have my filesystem NFS mounted from a nice, Debian "stable" firewall system, though...
Translation:
"The open source license is not open, becuase you can't take it and ever use it in a job-creating activity at Microsoft "
One of the best features of open source, is its ability to melt away unnecessary expenditures of money on software not directly related to the business goals or your company. It is inconceivable that any right-headed CIO or CFO would spend penny one on a "Word Processor", for example. The ONLY company that this decision would hurt is Microsoft. A company frees up virtually 100% of their software dollars to hire real, local software developers to develop solutions to their own, personal, business problems.
Consider this, though. If what you say is true, that "Society's laws grow from its mores", then can't we expect the same results as the Romans? The "mores" of Greed, Power, and Lust surely will overwhelm those of responsibility, honour and bravery. The former are "natural"; the latter -- they must be forged through much discomfort.
I think that you might actually be wrong. I think that people actually still (for this and perhaps one more generation...) have respect for the value of "a work", even if it is easily distributable. I think people still have at least a vague sense "payment for value"; What the masses are actually rebelling against is the extraction of wealth that occurs between those that Invent, Dream, and Create, and those that enjoy the results of that creation.
Generations of Management Schools have produced a stratum of society that cannot Create, but can only think in terms of Arbitrage -- Buying Low and Selling High -- who are prepared to feed the lusts of those in power, as long as they promise to protect these Manager's ability to ride the industries from which they derive their wealth. Adding nothing, but taking a slice of everything.
Unfortunately for this great mass of "Middlemen", the internet is becoming the great solvent of unnecessary power structures. When the mass of Creators and Enjoyers can be brought face-to-face with virtually zero cost and delay, the only thing left for the "Nation of Salesmen" to do is something -- anything -- to prevent it.
Well, those of us that are both technical and creative must fight back. Not with a bit of file sharing "freedom" (remember, the RIAA still holds most artists by the short and curly, and they *will* squeeze), but with a real, practical mechanism to provide direct, instantaneous compensation from the user to the creator.
You want to break the backs of those who provide nothing but Arbitrage? Then create a system that provides a stable, reliable market for the exchange of ideas and value. Create a system that pays the authors more than RIAA, and the artists will abandon them. On the other hand, if we continue on the path we are on, the artists will only hold on more tightly to their only semi-reliable source of income -- the recording industry.
The future of the creative commons is in our hands.
Want to stay up-to-date with the latest security fixes?
apt-get update; apt-get upgrade
Done.
Especially when I have to reboot my "professionally administered by a multi-million dollar corporate ID deparment" work Window NT box at least weekly -- and I only use it to read my email! Flaky Microsoft garbage...
Last I heard, there were >30 million millionaires in india (I don't have a reference to back this up, though.) That's more millionaires than there are people in the country of Canada! I think there is enough of a market there to make a cellular network profitable...
For similar reasons, the examples of professionals from other fields is false, unless I'm misunderstanding you.
Tell me; why would you want a lawyer that doesn't spend a signifianct percentage of his waking hours thinking about law? Any lawyer that doesn't is probably working as a public defender, and just putting in his/her hours...
As far as sales goes, I ran a rather large sales company for some time, and hired and trained at least a hundred full- and part-time people to do technical sales. I could tell within a week (with one exception that I still haven't quite figured out) whether they were going to be worth spit -- did they eat, sleep and breath the business, 24/7, until they began dreaming about it? If not, they were going to be toast. Did they memorize sales masters like Tom Hopkins, until they actually began sounding like them? If not -- done for.
The same goes for computer professionals (except, perhaps, for "Cracker Jack" box "professionals", like MCSEs, etc.) But really, is having the capability to install an OS, or configure a small LAN, really what we're talking about here?
http://www.qbi.ca
OK, I'm biased; my wife built this site...
Perhaps you meant to say that you pay 7 bucks/mo more for 2 distinct IP addresses? You probably only have one connection (hole in your wall, ADSL/Cable modem, etc).
The time before there are "fixed" versions of both NAT (which don't decrement TTL), and of IP packet ID's (changing all ID's into a single monotonically increasing order, or randomizing them) will be measured in hours.
Hopefully the authors of this paper aren't doing research for a living...
We think that the HP 2025's will support 1600x1200 if we get video cards that have a true "dual-link" DVI output, AND we obtain some "dual-link" DVI cables (the HP 2025's only came with "single-link" cables; go figure...)
However, it is virtually impossible to get an assurance of whether any particular video card is actually capable of producing "dual-link" DVI output! So far, we have deduced through testing that the Matrox 200 4-head, 400 2-head and new 550 2-head cards do NOT output "dual-link" DVI.
We have not been able to find any absolute assurance that any of the NVidia or ATI offerings have true "dual-link" DVI outputs, capable of driving a 1600x1200 display.
Do any of you have any direct, personal experience driving single- or multiple-headed 1600x1200 LCD displays using DVI video cards? We'll probably buy a bunch, if we can find one that works!
You can't engineer most software, for the same reason you can't engineer a book; it satisfies emotion, not logic. Most of the time, "success" or "failure" of software cannot even be defined! (see staggeringly successful, stupid and unreliable systems such as Micros~1 Windoze(tm)). Failure in engineering projects is usually, tragically, easy to define (see "Columbia").
It simply takes a different kind of duck to do software than it takes to do engineering. Have you ever met someone who can remember what they had for breakfast on July 23 1993, can remember the specs on a specific type of transistor, but can't come up with a efficient method for solving some seemingly novel problem? Thats an engineer. Ever met someone who, in a "eureka" moment, manages to re-frame a sticky problem so that it suddenly becomes solvable? And then turns out the code to do it in an afternoon, and it works first try? That might be a Programmer. Sometimes, you might even get both in the same package (you'd call him a Genius). But usually, people that are good at engineering, are very frustrated by programming.
I just hope to sometimes be called a Programmer.