Somewhat of a matter of approach. Most of them keep iterating on the same techniques in a constant arms race. Just so hard to make much more than an incremental progress and differentiate from everyone else.
I like the approach being used by Atriceps (http://www.atriceps.com)-- they're flipping the problem on it's head with a consent-based approach that looks for valid content and reputation, rather than exclusion.
Disclaimer: My company was in the Fall 2015 cohort of the Mach37 Cyber Accelerator with them.
It's interesting that in EvE, the most vulnerable alliances are the ones that are most democratic.
The article is a really horrible description of what happened. The goons launched a combination of raw propoganda as well as propoganda targetted to specific events. Defeats for the goons were absorbed and made part of the culture. Wins were beaten into the ground as a failing on the part of the enemy. An impressive spy and saboture network was fully exploited in terms of economic and military assistance, as well as in the propoganda.
"Failure Cascade" was a term coined by The Mittani, the leader of the Goon Intelligence Agency (GIA). It seems to apply mostly to the application of public opinion and propoganda to widen rifts and blame games within an enemy organization. If the logistics people make a mistake, any military victories are met with comments on the pointlessness of fighting when the territory-holding infrastructure won't hold. Same thing applies to military losses in the face of stout infrastructure. Pretty soon the fighters and logisticians are distrustful and burnt out, not trusting the others to do their jobs.
Enemy command structures are infiltrated and often the goon populace knows as much or more about internal workings as the rank and file members. After the fighters begin got get more confirmed leaks from their leadership on public boards than they do on private boards, a rift is formed and further exploited. At some point, these rifts become self sustaining -- a "failure cascade". It's not unrecoverable.
Democratic alliances are more vulnerable because there are more rifts. People bring up and participate and lose in the democratic process, causing a LOT of "I told you so"s and "If you'd only gone my way..." to exploit. An alliance with ideals presents a target for showing hypocracy within the leadership. The best defense seems to be playing in a largely amoral, berserker-don't-give-a-damn-about-loses style.
I found that the C++ development was difficult if the project size was too large. The "build automatically" feature was hopeless if it was turned on. It fires every time you save a file and usually didn't finish before the next save if you were doing tweaks. Autocompletion got to be extremely slow as the symbol tables increased, though that might have been partially to the templates I was using. Without those features, a lot of the appeal for Eclipse went away for me.
The caveats are that I was using the gcj compiled version that came with FC4, and was using this when FC4 was current. It may have gotten better, but C++ is such a complicated language there may still be issues.
Frankly, I think the best answer to the question is not a straight answer, but a "I would research it to find the cannonical answer with the best complexity for our expected data, but for a general first-pass solution I would try X". The sophomore study group problem is about asking for programming trivia rather than programming approaches. Programming is easy. Managing the complexity of a software project is hard. A real question would be bring in real situations and ask for an approach:
We have a C++ cross compiler to a custom platform. Template support is not sufficient to allow usage of the STL. Our model will require a lot of insertions, removals, and reversing of an ordered but unsorted list. At a later date, the list may need to be sorted quickly, but that is not a requirement for revision 1.0. How would you meet the requirements for this basic data structure and manage future needs?
A good interviewer asks open-ended questions that prompt the interviewee to explain his accomplishments or show off his ability to produce in a tight situation. If their answer is "google it," then a follow on to that answer must contain an explaination of validation, testing, and maintenance. Simply asking how to reverse a list tells you nothing about their abilities except they remembered the right paragraph in the right volume of Knuth.
The software isn't tricky, it's the politics. The FBI is a feifdom of petty departmental tyrants and ancient practices. Several friends of mine have worked on the case file system in various incarnations, the problem is that the COTRs come in and define the requirements to be "exactly the same thing as this 25 year old main frame, but on a web page". One guy was having problems because his COTR was telling him that it had to be green text on a black background. That may have been an exageration, but based upon my experience it's quite belivable.
I don't care how good of an engineer you are, you can't build a product for a customer can not or will not help you determine what their needs are.
A friend of mine is a reviewer for a gaming site that does their reviews on a [1,10] scale. Apparently the editors tell the writers to rate them on a [1,5] scale, then they just tack on a bonus 5 points to whatever the writers say.
How does a copyright banner affect any technical decision of the software? It doesn't affect my data model, algorithms, security model, or any other aspect of implementation in the least. In addition, it doesn't affect the original programmer in the slightest, it merely says that when you modify someone's code, you can't remove it if it exists.
How does the user's right to modify the software in any way affect technical decisions about the software? It doesn't affect my data model, algorithms, security model, or any other aspect of implementation in the least.
And when the copyright owner does this, he's making a political decision, not a technical one. Linus's whole point is that the GPL 3 dictates technical details of projects that use it, where V2 didn't.
Keep in mind that the BSA's definition of pirated means that you don't have an original receipt or purchase order containing your name, the seller's name, the purchase price, the date, and a couple other odds and ends. The license key doesn't seem to concern them.
Take all the people who work in the computer industry. Take the subset of people who wave their certifications around like they mean something. I'm willing to bet there's a far higher degree of competence in what's left. Putting a cert on your resume is almost an admission that you don't have any experience worth talking about.
So you're criteria would probably work better if you ask them if they have a cert, and went with them if they laughed and said that certs didn't mean all that much, regardless of whether they had one or not.
[...]evolution is an excellent theory to explain the origins of biological diversity, but it has little or no religious significance - it can be placed equally well within an atheistic or theistic context.
Doesn't seem to me like this guy supports Intelligent Design, he's just giving his spin on Gould's Non-Overlapping Magisteria of Science and Religion.
Yeah, that was kinda the point. You pay 1/x^2 in tax instead of 1/x. There's no reason at all for a commuter to be getting less than 20mpg these days. Linear tax just taxes consumption, but a progressive tax would tax efficiency.
I've always thought some stepped price based upon your fuel milage would be interesting. For example, set the current goal at 25 mpg, so a gallon of gas for the a 25mpg car would cost the same as it does now. If you half that milage, double the tax. If you double the milage, halve the tax. Of course, the mechanics of making such a system work are a practical impossibility.
The fact that this company basically patented a software design pattern isn't evidence? Modular achitectures are one of the most basic ideas of software, and this company claims that it owns that idea in the area of web browsing.
What Role are you Playing in FF7 (just an example)?
You're just a passenger watching things go by in an interactive movie with inventory management. There is no Playing of Roles. I am not Cloud, I am some guy who feeds Cloud a healing potion and throws materia into his sword.
There's nothing wrong with that, I rather enjoyed FF7, but calling them Role Playing is a strong misnomer. They're just Kings Quest with newer graphics rather than Ultima IV.
A successfully completed contract would be a "very good reason" for leaving. Reaction will depend upon what you are applying for, too. It's not hard to get typecast as a "short contracts person" and have someone who is looking for a long term duration employee to view that type of history as an ill fit to his needs. The opposite can be true as well. Then again, a lot of places are so hard up for anyone that can correctly spell "IT" that they don't care about your history.
There are obviously exceptions to every rule, but someone who feels comfortable taking a series of short contracts or hopping more frequently probably doesn't need to read advice in a slashdot comment. Like most things in programming, advice and rules are the things you follow until you know why they should be ignored. A path of working at places for at least years is a good rule of thumb for getting that experience if you don't already have it.
Generally it's about what you can rationally explain in an interview. Certainly avoid working somewhere less than a year or two unless you have an extremely good reason. I think you can tolerate a faster jump earlier in your career rather than later. I don't think any employer is going to begrudge you for having a couple 2 year stints early in your career while you explore different areas of your field. As your career progresses, you should look at longer and longer times of service.
I think it's one of those self-limitting things. As you get more experience, you know when it's the best time to leave a job.
VB is a high level programming language in the same way that "a/s/l u want 2 cyber" is high level romantic language. Both express the necessary concepts in a quick form that doesn't require a lot of subtlety, but you're only successful by lowered expectations in both cases.
Re:Educaton is not always that important.
on
RadioShack CEO Resigns
·
· Score: 2, Interesting
I never got a degree, and I'm now a contractor who does very quick turnaround solutions for a customer who needs very specialized and unusual tools (software and hardware). It's very interesting and technical work requiring a fair breadth of knowledge, and there's no room for incompetency (4 man team, combined 70 years of experience). I think I've done quite well for myself.
I say that not to brag (well, maybe a little), but more to drive the point home when I say that a college degree is a very important part of your resume. Practical experience in IT has a hard time truly capturing the theory aspects of what you are doing. Too often it's a "make it work" world where you have enough understanding to navigate the dialogues in windows without understanding why the network is the way it is. How often will you have practical experience in compilers, operating systems, or assembly language from working experience? Yet all of those skills are extremely valuable to be a good software person. The most important thing about formal education is that it exposes you to things that you may not need right away, but later down the road may be useful. I spent a days researching just to find that the algorithm I needed for one problem was a "weighted connected component", and most of that was just in finding the right term to search for. If I had taken a junior level graph theory class, I would have been exposed to it.
You mock the engineers who flash their degrees at you when you fix their network. Does their job entail fixing their network connections? If there's a mess on the floor, I could clean it myself but I'm more likely to call the janitors. Don't get me wrong, there are plenty of really compentent people without degrees. There exists a few of them who are wildly successful. The other 90% of people won't be competent or successful, though, and a most of the remaining would be more competent and successful with a formal education.
Slashdot Mirror
Somewhat of a matter of approach. Most of them keep iterating on the same techniques in a constant arms race. Just so hard to make much more than an incremental progress and differentiate from everyone else.
I like the approach being used by Atriceps (http://www.atriceps.com)-- they're flipping the problem on it's head with a consent-based approach that looks for valid content and reputation, rather than exclusion.
Disclaimer: My company was in the Fall 2015 cohort of the Mach37 Cyber Accelerator with them.
Modated +1 SUPERSLAM
It's interesting that in EvE, the most vulnerable alliances are the ones that are most democratic.
The article is a really horrible description of what happened. The goons launched a combination of raw propoganda as well as propoganda targetted to specific events. Defeats for the goons were absorbed and made part of the culture. Wins were beaten into the ground as a failing on the part of the enemy. An impressive spy and saboture network was fully exploited in terms of economic and military assistance, as well as in the propoganda.
"Failure Cascade" was a term coined by The Mittani, the leader of the Goon Intelligence Agency (GIA). It seems to apply mostly to the application of public opinion and propoganda to widen rifts and blame games within an enemy organization. If the logistics people make a mistake, any military victories are met with comments on the pointlessness of fighting when the territory-holding infrastructure won't hold. Same thing applies to military losses in the face of stout infrastructure. Pretty soon the fighters and logisticians are distrustful and burnt out, not trusting the others to do their jobs.
Enemy command structures are infiltrated and often the goon populace knows as much or more about internal workings as the rank and file members. After the fighters begin got get more confirmed leaks from their leadership on public boards than they do on private boards, a rift is formed and further exploited. At some point, these rifts become self sustaining -- a "failure cascade". It's not unrecoverable.
Democratic alliances are more vulnerable because there are more rifts. People bring up and participate and lose in the democratic process, causing a LOT of "I told you so"s and "If you'd only gone my way..." to exploit. An alliance with ideals presents a target for showing hypocracy within the leadership. The best defense seems to be playing in a largely amoral, berserker-don't-give-a-damn-about-loses style.
Very cool. I'll have to give it a try this weekend. That was about the only thing that would make me give up Emacs.
I found that the C++ development was difficult if the project size was too large. The "build automatically" feature was hopeless if it was turned on. It fires every time you save a file and usually didn't finish before the next save if you were doing tweaks. Autocompletion got to be extremely slow as the symbol tables increased, though that might have been partially to the templates I was using. Without those features, a lot of the appeal for Eclipse went away for me.
The caveats are that I was using the gcj compiled version that came with FC4, and was using this when FC4 was current. It may have gotten better, but C++ is such a complicated language there may still be issues.
No, IP contains no MAC information.
The MAC would vanish at the first routing point.
Frankly, I think the best answer to the question is not a straight answer, but a "I would research it to find the cannonical answer with the best complexity for our expected data, but for a general first-pass solution I would try X". The sophomore study group problem is about asking for programming trivia rather than programming approaches. Programming is easy. Managing the complexity of a software project is hard. A real question would be bring in real situations and ask for an approach:
We have a C++ cross compiler to a custom platform. Template support is not sufficient to allow usage of the STL. Our model will require a lot of insertions, removals, and reversing of an ordered but unsorted list. At a later date, the list may need to be sorted quickly, but that is not a requirement for revision 1.0. How would you meet the requirements for this basic data structure and manage future needs?
A good interviewer asks open-ended questions that prompt the interviewee to explain his accomplishments or show off his ability to produce in a tight situation. If their answer is "google it," then a follow on to that answer must contain an explaination of validation, testing, and maintenance. Simply asking how to reverse a list tells you nothing about their abilities except they remembered the right paragraph in the right volume of Knuth.
No, you won't.
The software isn't tricky, it's the politics. The FBI is a feifdom of petty departmental tyrants and ancient practices. Several friends of mine have worked on the case file system in various incarnations, the problem is that the COTRs come in and define the requirements to be "exactly the same thing as this 25 year old main frame, but on a web page". One guy was having problems because his COTR was telling him that it had to be green text on a black background. That may have been an exageration, but based upon my experience it's quite belivable.
I don't care how good of an engineer you are, you can't build a product for a customer can not or will not help you determine what their needs are.
A friend of mine is a reviewer for a gaming site that does their reviews on a [1,10] scale. Apparently the editors tell the writers to rate them on a [1,5] scale, then they just tack on a bonus 5 points to whatever the writers say.
Ahem,
How does a copyright banner affect any technical decision of the software? It doesn't affect my data model, algorithms, security model, or any other aspect of implementation in the least. In addition, it doesn't affect the original programmer in the slightest, it merely says that when you modify someone's code, you can't remove it if it exists.
How does the user's right to modify the software in any way affect technical decisions about the software? It doesn't affect my data model, algorithms, security model, or any other aspect of implementation in the least.
And when the copyright owner does this, he's making a political decision, not a technical one. Linus's whole point is that the GPL 3 dictates technical details of projects that use it, where V2 didn't.
Keep in mind that the BSA's definition of pirated means that you don't have an original receipt or purchase order containing your name, the seller's name, the purchase price, the date, and a couple other odds and ends. The license key doesn't seem to concern them.
0 612networker3.html
http://www.networkworld.com/newsletters/sbt/2006/
Take all the people who work in the computer industry. Take the subset of people who wave their certifications around like they mean something. I'm willing to bet there's a far higher degree of competence in what's left. Putting a cert on your resume is almost an admission that you don't have any experience worth talking about.
So you're criteria would probably work better if you ask them if they have a cert, and went with them if they laughed and said that certs didn't mean all that much, regardless of whether they had one or not.
Doesn't seem to me like this guy supports Intelligent Design, he's just giving his spin on Gould's Non-Overlapping Magisteria of Science and Religion.
Yeah, that was kinda the point. You pay 1/x^2 in tax instead of 1/x. There's no reason at all for a commuter to be getting less than 20mpg these days. Linear tax just taxes consumption, but a progressive tax would tax efficiency.
I've always thought some stepped price based upon your fuel milage would be interesting. For example, set the current goal at 25 mpg, so a gallon of gas for the a 25mpg car would cost the same as it does now. If you half that milage, double the tax. If you double the milage, halve the tax. Of course, the mechanics of making such a system work are a practical impossibility.
The fact that this company basically patented a software design pattern isn't evidence? Modular achitectures are one of the most basic ideas of software, and this company claims that it owns that idea in the area of web browsing.
What Role are you Playing in FF7 (just an example)?
You're just a passenger watching things go by in an interactive movie with inventory management. There is no Playing of Roles. I am not Cloud, I am some guy who feeds Cloud a healing potion and throws materia into his sword.
There's nothing wrong with that, I rather enjoyed FF7, but calling them Role Playing is a strong misnomer. They're just Kings Quest with newer graphics rather than Ultima IV.
A successfully completed contract would be a "very good reason" for leaving. Reaction will depend upon what you are applying for, too. It's not hard to get typecast as a "short contracts person" and have someone who is looking for a long term duration employee to view that type of history as an ill fit to his needs. The opposite can be true as well. Then again, a lot of places are so hard up for anyone that can correctly spell "IT" that they don't care about your history.
There are obviously exceptions to every rule, but someone who feels comfortable taking a series of short contracts or hopping more frequently probably doesn't need to read advice in a slashdot comment. Like most things in programming, advice and rules are the things you follow until you know why they should be ignored. A path of working at places for at least years is a good rule of thumb for getting that experience if you don't already have it.
Generally it's about what you can rationally explain in an interview. Certainly avoid working somewhere less than a year or two unless you have an extremely good reason. I think you can tolerate a faster jump earlier in your career rather than later. I don't think any employer is going to begrudge you for having a couple 2 year stints early in your career while you explore different areas of your field. As your career progresses, you should look at longer and longer times of service.
I think it's one of those self-limitting things. As you get more experience, you know when it's the best time to leave a job.
What matrix library were you using for C++? Was it hand-rolled, and if so was it faster than C++ matrix libraries like Blitz++?
VB is a high level programming language in the same way that "a/s/l u want 2 cyber" is high level romantic language. Both express the necessary concepts in a quick form that doesn't require a lot of subtlety, but you're only successful by lowered expectations in both cases.
I never got a degree, and I'm now a contractor who does very quick turnaround solutions for a customer who needs very specialized and unusual tools (software and hardware). It's very interesting and technical work requiring a fair breadth of knowledge, and there's no room for incompetency (4 man team, combined 70 years of experience). I think I've done quite well for myself.
I say that not to brag (well, maybe a little), but more to drive the point home when I say that a college degree is a very important part of your resume. Practical experience in IT has a hard time truly capturing the theory aspects of what you are doing. Too often it's a "make it work" world where you have enough understanding to navigate the dialogues in windows without understanding why the network is the way it is. How often will you have practical experience in compilers, operating systems, or assembly language from working experience? Yet all of those skills are extremely valuable to be a good software person. The most important thing about formal education is that it exposes you to things that you may not need right away, but later down the road may be useful. I spent a days researching just to find that the algorithm I needed for one problem was a "weighted connected component", and most of that was just in finding the right term to search for. If I had taken a junior level graph theory class, I would have been exposed to it.
You mock the engineers who flash their degrees at you when you fix their network. Does their job entail fixing their network connections? If there's a mess on the floor, I could clean it myself but I'm more likely to call the janitors. Don't get me wrong, there are plenty of really compentent people without degrees. There exists a few of them who are wildly successful. The other 90% of people won't be competent or successful, though, and a most of the remaining would be more competent and successful with a formal education.
I was going to make a smart-ass comment about it not having lame, but turns out that it does