This is complicity. The editors often comment when they feel compelled to *and* they also edit the comments (often to stir up controversy). That timothy did not just shows that he does not disagree with the gnutella sentiment strongly enough to edit it or comment on it.
One of the huge issues with alternative non-Windows OSes is "does it run Office?"
This is a small step in the major strength of Open Source products like StarOffice: cross platform dominance *because* of portability. It ran on the right platforms so it got chosen.
This bodes well for projects like KDE and Gnome that feel the need to become a common desktop for *ix. This may also do interesting things for MS if they want to keep control of the entire supply chain. Porting apps to *ix (eg: IE) is not new to MS but this may hasten it and it's further involvement with *ix and open source code.
I don't know why you think it's shameful. Mostly the lead developer/founder is stuck with this role be default. I end up doing this as maintainer of my own packages. Usually this gets bundled in with release engineer, developer, tester, document writer, and designer but it's there.
The problem with OSS is that the leaders tend to be very technical and want to keep their hands dirty. For a large project there really isn't time to do everything so you just have to find the right person for the job(s) or sputter along as a project grows out of control.
You just have to go to BSD to see examples of how organizatoins work. While I like FreeBSD and OpenBSD (no opinion on NetBSD, haven't tried it), there is as much division between the groups as their is cooperation. You have an elected selective democracy in the FreeBSD group and a fairly vertical 'benevolent' dictatorship in OpenBSD.
I'd say groupthink is pretty strong in OpenBSD, you'd have to be to focus on their goals and work with Theo. FreeBSD, otoh, would seem to have a much more collaborative environemnt, albeit with a strong central steering committee. I suspect strong centralized leadership doesn't exist in NetBSD, it would seem to be the nature of the beast when you're trying to port to everything, but that's pure speculation on my part..
The real uses I've found tend to be storing addresses and phone numbers, writing quick to-do lists, carrying around a list of good places to eat, and making appointments. Storing personal information (usernames, passwords, accounts for your phone, complex urls, etc) is pretty useful as well. And because you sync with your PC you have a backup of all your data in case something should happen to your handheld.
The occaisional game of chess or minesweeper will help pass the time. Whoop. I know some people who like to read 'eBooks' as well, never tried that.
I also call places around the world and I've got a neat app that tells me what time it is in any part of the world. Definitely useful when. Waking someone up at 3am is a good thing to avoid.
Btw: personally I'd go for a Handspring if you absolutely want one. Yeah, you don't get to run Linux. If you really wanted to run linux as a student then you're just wanting a luxury toy.
Despite the recent departure of Tim Perdue you would do well to ask on phpbuilder.net's forums. Lots of users there, many clueless, but there are about a dozen people who really know what they're talking about.
Well, you just answered why it's an advanced programming topic. Figure out a way to let the languge/OS take care of the nightmare situations you encounter if you would like to make it simpler. I don't see any other way out of it except for using better trained programmers (which won't happen anytime in my lifetime).
Hi, I write GPL code and I dislike the GPL. I also write proprietary code and dislike proprietary restrictions. I also share my proprietary code when convenient.
So now I have the right to complain, joy. Now I'll get back to coding. Thanks for listening.
Reading slashdot to keep familiar with tech is useful to stay on top of the skills for an IT job, not a programming job. I had no problems finding entry level programming opportunities altough some did. There is a common thread inthsoe that get lots of offers: they really like to program. They program in their free time. They make time to program. They don't need someone to tell them that they need to know X language to get a job.
I wish you luck getting a job, the current economy is the reason you're having such trouble. Most employers realize that new hires are essentially untrained monkeys for the first 6 months or so of work. They're just reluctant to do so with the bottom line being so tight.
And I disagree entirely. I used to be obsessed with optimizing code. The type of variable used, whether to use pointers or arrays, etc. At some point I relaized, as do most people, that *it's not worth it*. It's more important to get a program that works correctly than a program that runs efficiently (personal pride and asthetics aside). You can't ship an inccorect program. I've never directly used any knowledge from my assembler course in any of my programming, save one instance where I got to code some simulations on an embedded proc.
I run into performance bottlenecks and I find out where and then optimize those as necessary. I leave you to continue to code in the slowest method possible. I'd like to get stuff done.
Touche! This reminds me that C/C++ also suffers from this problem. Not having ths standard libraries in your include path (on an unfamiliar shared unix machine) is quite fun for the beginner.
Why weren't you actively developing your skills on the side in your free time? I suppose it's all fine to get a degree just to get a job but I can't see any merit in your complaints if that's what you did.
I quickly figured out that the job colleges have is to make you jump through hoops to get a degree which will make you a "computer scientist". I've asked several professors about this when I was in college and basically colleges expect you to do programming on your own time. They just need to teach the curriculum.
Too bad that you feel screwed but you're the only one who can help yourself. College isn't high school anymore. I learned all those diverse languages (and more) and so far (1 year out) I've used the following in real world applications: Java, Assembly, C, C++, php, perl, pascal/Delphi. No tears from me. You learn a lot when you get a degree but if you don't have any plans you might as well not have gone.
If that's your goal then a better choice of a language would probably be lisp or scheme. Some people *cannot* wrap their heads around functional programming. It's much worse than when you first learned how pointers worked. With assembly yuou can usually hack at it until it works. It's just a painful process of following basic steps. Recursive thinking is almost unnatural to most people.
... And several colleges are requiring languages like scheme or haskell as one of their introductory courses. They're experimenting with seeing how introducing funcitonal languages affect students early on in their CS degree. Mostly it's another good weed out gimmick.
Of course, here on slashdot we have to be elite and exclusive so we get this kind of thinking.
Absolutely not. When's the last time you had to think about how many instructions your loop would take or which registers were likely to be swapped out? Certainly there are areas of programming where such optimizations and considerations are important but it's not particularly applicable to anything but low-level systems programming and inner loop optimizations. Frankly, optimizations are an advanced programming topic.
To be a skilled programmer you must know the principles of what good programming is. You learn all the useful information from Computer Architecture and Operating Systems courses which are more advanced courses.
Assembly exposes you to too many specifics to be of any great value. The one general thing it should be useful for is optimization. A few good rules of thumb and most higher level languages will optimize themselves well enough.
I do believe assembly should be taught. It's a (often painful) experience that programmers need to have. But I don't believe it's fundamentally important. Much more important would be for programmers to have the contents of books like Code Complete or taught to them.
Frankly I think users learning languages like perl and php on their own are doing much more harm to themselves thna if they learned Java.
Another key advantage is you don't spend as much time fighting the language. With C/C++ you have wonderful things like linker errors. It might seem like a small thing to the initiated but I know many people who had given up learning C/C++ (at least initially) because they couldn't compile AND link their programs. I've spent many a wasted hour trying combinations of includes to figure out what exactly the linker didn't like. STL linker errors are even more fun! Most people who know C/C++ don't pay much attention to this drawback.
Pretty much all programming languages are fads. Some just last longer than others.
What *doesn't* change are the fundamental concepts of programming and computer science. What you are striving to do is to correctly implement the solution to a problem with limited resources. Languages are the tools that try to increase your productivity with these limited resources. They are the practical illustrations and implementations of the concepts (data models, algorithms, proof techniques).
As time changes the limitations change. Memory and CPU speed used to be a huge limitation. Ditto for storage. The age of small-low level languages. Then we hit 400Mhz CPUs and 128MBs RAM along with CD/DVD and 30GB HDs and 'most' apps are more than happy with that. Thus, languages like Java become more practical. Of course low-level languages continue to excel, except that the limitation now is time. OOP languages and RAD tools take advantage of this and leverage inefficiencies for more efficient use of the programmers time.
What you as a student should take form this is that the fundamental ways of approaching a problem is most important. Personally I feel that all students should learn (or at least be exposed to) a OOP language (Java/C++), a functional language (scheme/lisp, etc), and a low level language (C/Assembly). Probably throw in some interesting languages like Prolog and you're pretty well rounded. Scripting languages like perl/python/php are good to know but won't teach you much about CS/programming that the others won't as well.
IMO, the only thing you get out of learning C/C++ early on as far as 'inner working' of computers is memory mamangement. You also get exposed to a lot of painful (and questoinable but necessary) experiences like memory leaks and linker errors. You don't learn much that is truly useful until you take a Computer Architecture and Operating Systems course which are usually mid or upper level courses. This is where you learn more about how caching works(level1/2/3, TLB?), understand bottlenecks in performance and resources (bus speeds, IO), how memory functions (what is CAS2), and how the CPU functions (instruction ordering, pipelining, etc.).
If the goal is to learn OOP then I would use Java to teach. If at all possible I woud also use C++. My OOP programming class in college did both.
Unless you intend to specialize in a lnaguage you really should arm yourself with as many languages as possible. You see problems from different points of view and can better understand the strengths and weaknesses of the tools you are using.
From reading the (finally posted) update on SF it seems that a staff member's user/password was sniffed through a compromised upstream connection.
Read about it here:
http://sourceforge.net/forum/forum.php?forum_id= 89 285
It's not clear if the ISP was socially hacked or exploited by a technical means.
This seems to be creating a cascading effect as a direct result of having a centralized repository and network of popular open source site as well as most everyone using similar passwords across multiple accounts.
Actually hobbyist programmers will usually duplicate work no matter waht kind of software model you use. Programmers like doing things 'their way' and as long as the complexity isn't gigantic (an OS) they'll happily duplicate work. They can also get more glory and ego gratification this way.
Visit freshmeat and see how many mp3 players, icq clones, text editors, etc. that you can find. Visit Sourceforge and you'll see how many more products were never released (started).
I say this as a programmer that has written closed and open apps and has projects hosted on fresheat and sourceforge.
Mozilla will probably go down in history as an example of how to produce software right as well as (perhaps irrevocably) lose marketshare and mindshare. Frankly, Mozilla *must* "succeed" or most non-windows OSes are screwed (and I expect it will do moderately well).
I see that this is upposed to be funny but it's also wrong.
Software has always been created to serve people, not the other way around. Someof us may work in conditions that are the other way around but it's wrong headed thinking.
Re:Could this be used as a "legitimate use of p2p"
on
Swarmcast GPLed
·
· Score: 1
Uh, since when was peer to peer technology not legitimate? We've been using it for years in local networking and games and so forth.
As an author who has released GPL'ed software I'd rather see this go to court and finally have some rulings on the GPL. It's one thing to be a company with money to spend of legal fees, it's quite another to be an author making no money off of a product. I'd like to finally see the GPL get some legal rulings to give it more weight.
Slashdot Mirror
This is complicity. The editors often comment when they feel compelled to *and* they also edit the comments (often to stir up controversy). That timothy did not just shows that he does not disagree with the gnutella sentiment strongly enough to edit it or comment on it.
One of the huge issues with alternative non-Windows OSes is "does it run Office?"
This is a small step in the major strength of Open Source products like StarOffice: cross platform dominance *because* of portability. It ran on the right platforms so it got chosen.
This bodes well for projects like KDE and Gnome that feel the need to become a common desktop for *ix. This may also do interesting things for MS if they want to keep control of the entire supply chain. Porting apps to *ix (eg: IE) is not new to MS but this may hasten it and it's further involvement with *ix and open source code.
MS's Mac products have usually been developed with Metrowerks Codewarrior. This continues with OSX.
I don't know why you think it's shameful. Mostly the lead developer/founder is stuck with this role be default. I end up doing this as maintainer of my own packages. Usually this gets bundled in with release engineer, developer, tester, document writer, and designer but it's there.
The problem with OSS is that the leaders tend to be very technical and want to keep their hands dirty. For a large project there really isn't time to do everything so you just have to find the right person for the job(s) or sputter along as a project grows out of control.
You just have to go to BSD to see examples of how organizatoins work. While I like FreeBSD and OpenBSD (no opinion on NetBSD, haven't tried it), there is as much division between the groups as their is cooperation. You have an elected selective democracy in the FreeBSD group and a fairly vertical 'benevolent' dictatorship in OpenBSD.
I'd say groupthink is pretty strong in OpenBSD, you'd have to be to focus on their goals and work with Theo. FreeBSD, otoh, would seem to have a much more collaborative environemnt, albeit with a strong central steering committee. I suspect strong centralized leadership doesn't exist in NetBSD, it would seem to be the nature of the beast when you're trying to port to everything, but that's pure speculation on my part..
The real uses I've found tend to be storing addresses and phone numbers, writing quick to-do lists, carrying around a list of good places to eat, and making appointments. Storing personal information (usernames, passwords, accounts for your phone, complex urls, etc) is pretty useful as well. And because you sync with your PC you have a backup of all your data in case something should happen to your handheld.
The occaisional game of chess or minesweeper will help pass the time. Whoop. I know some people who like to read 'eBooks' as well, never tried that.
I also call places around the world and I've got a neat app that tells me what time it is in any part of the world. Definitely useful when. Waking someone up at 3am is a good thing to avoid.
Btw: personally I'd go for a Handspring if you absolutely want one. Yeah, you don't get to run Linux. If you really wanted to run linux as a student then you're just wanting a luxury toy.
Well, it's what happens when it's written by a person whose favorite languiage is python.
l j- issues/issue73/3882.html
http://www2.linuxjournal.com/cgi-bin/frames.pl/
Something I came across as I was messing with a Wiki today.
Despite the recent departure of Tim Perdue you would do well to ask on phpbuilder.net's forums. Lots of users there, many clueless, but there are about a dozen people who really know what they're talking about.
You've hit the nail on the head.
This is not a zero-sum game.
MS would like it to be, GPL fanatics would also like it to be.
The BSD (and like) licenses are true licenses of 'trust'. The GPL doesn't trust you; that's why it forces it to remain open.
(disclaimer: I've release GPL code. It suited the situation better)
Well, you just answered why it's an advanced programming topic. Figure out a way to let the languge/OS take care of the nightmare situations you encounter if you would like to make it simpler. I don't see any other way out of it except for using better trained programmers (which won't happen anytime in my lifetime).
Hi, I write GPL code and I dislike the GPL. I also write proprietary code and dislike proprietary restrictions. I also share my proprietary code when convenient.
So now I have the right to complain, joy. Now I'll get back to coding. Thanks for listening.
Reading slashdot to keep familiar with tech is useful to stay on top of the skills for an IT job, not a programming job. I had no problems finding entry level programming opportunities altough some did. There is a common thread inthsoe that get lots of offers: they really like to program. They program in their free time. They make time to program. They don't need someone to tell them that they need to know X language to get a job.
I wish you luck getting a job, the current economy is the reason you're having such trouble. Most employers realize that new hires are essentially untrained monkeys for the first 6 months or so of work. They're just reluctant to do so with the bottom line being so tight.
And I disagree entirely. I used to be obsessed with optimizing code. The type of variable used, whether to use pointers or arrays, etc. At some point I relaized, as do most people, that *it's not worth it*. It's more important to get a program that works correctly than a program that runs efficiently (personal pride and asthetics aside). You can't ship an inccorect program. I've never directly used any knowledge from my assembler course in any of my programming, save one instance where I got to code some simulations on an embedded proc.
I run into performance bottlenecks and I find out where and then optimize those as necessary. I leave you to continue to code in the slowest method possible. I'd like to get stuff done.
Touche! This reminds me that C/C++ also suffers from this problem. Not having ths standard libraries in your include path (on an unfamiliar shared unix machine) is quite fun for the beginner.
College is not a job guarrantee!
Why weren't you actively developing your skills on the side in your free time? I suppose it's all fine to get a degree just to get a job but I can't see any merit in your complaints if that's what you did.
I quickly figured out that the job colleges have is to make you jump through hoops to get a degree which will make you a "computer scientist". I've asked several professors about this when I was in college and basically colleges expect you to do programming on your own time. They just need to teach the curriculum.
Too bad that you feel screwed but you're the only one who can help yourself. College isn't high school anymore. I learned all those diverse languages (and more) and so far (1 year out) I've used the following in real world applications: Java, Assembly, C, C++, php, perl, pascal/Delphi. No tears from me. You learn a lot when you get a degree but if you don't have any plans you might as well not have gone.
If that's your goal then a better choice of a language would probably be lisp or scheme. Some people *cannot* wrap their heads around functional programming. It's much worse than when you first learned how pointers worked. With assembly yuou can usually hack at it until it works. It's just a painful process of following basic steps. Recursive thinking is almost unnatural to most people.
... And several colleges are requiring languages like scheme or haskell as one of their introductory courses. They're experimenting with seeing how introducing funcitonal languages affect students early on in their CS degree. Mostly it's another good weed out gimmick.
Of course, here on slashdot we have to be elite and exclusive so we get this kind of thinking.
Absolutely not. When's the last time you had to think about how many instructions your loop would take or which registers were likely to be swapped out? Certainly there are areas of programming where such optimizations and considerations are important but it's not particularly applicable to anything but low-level systems programming and inner loop optimizations. Frankly, optimizations are an advanced programming topic.
To be a skilled programmer you must know the principles of what good programming is. You learn all the useful information from Computer Architecture and Operating Systems courses which are more advanced courses.
Assembly exposes you to too many specifics to be of any great value. The one general thing it should be useful for is optimization. A few good rules of thumb and most higher level languages will optimize themselves well enough.
I do believe assembly should be taught. It's a (often painful) experience that programmers need to have. But I don't believe it's fundamentally important. Much more important would be for programmers to have the contents of books like Code Complete or taught to them.
Frankly I think users learning languages like perl and php on their own are doing much more harm to themselves thna if they learned Java.
Another key advantage is you don't spend as much time fighting the language. With C/C++ you have wonderful things like linker errors. It might seem like a small thing to the initiated but I know many people who had given up learning C/C++ (at least initially) because they couldn't compile AND link their programs. I've spent many a wasted hour trying combinations of includes to figure out what exactly the linker didn't like. STL linker errors are even more fun! Most people who know C/C++ don't pay much attention to this drawback.
Pretty much all programming languages are fads. Some just last longer than others.
What *doesn't* change are the fundamental concepts of programming and computer science. What you are striving to do is to correctly implement the solution to a problem with limited resources. Languages are the tools that try to increase your productivity with these limited resources. They are the practical illustrations and implementations of the concepts (data models, algorithms, proof techniques).
As time changes the limitations change. Memory and CPU speed used to be a huge limitation. Ditto for storage. The age of small-low level languages. Then we hit 400Mhz CPUs and 128MBs RAM along with CD/DVD and 30GB HDs and 'most' apps are more than happy with that. Thus, languages like Java become more practical. Of course low-level languages continue to excel, except that the limitation now is time. OOP languages and RAD tools take advantage of this and leverage inefficiencies for more efficient use of the programmers time.
What you as a student should take form this is that the fundamental ways of approaching a problem is most important. Personally I feel that all students should learn (or at least be exposed to) a OOP language (Java/C++), a functional language (scheme/lisp, etc), and a low level language (C/Assembly). Probably throw in some interesting languages like Prolog and you're pretty well rounded. Scripting languages like perl/python/php are good to know but won't teach you much about CS/programming that the others won't as well.
IMO, the only thing you get out of learning C/C++ early on as far as 'inner working' of computers is memory mamangement. You also get exposed to a lot of painful (and questoinable but necessary) experiences like memory leaks and linker errors. You don't learn much that is truly useful until you take a Computer Architecture and Operating Systems course which are usually mid or upper level courses. This is where you learn more about how caching works(level1/2/3, TLB?), understand bottlenecks in performance and resources (bus speeds, IO), how memory functions (what is CAS2), and how the CPU functions (instruction ordering, pipelining, etc.).
If the goal is to learn OOP then I would use Java to teach. If at all possible I woud also use C++. My OOP programming class in college did both.
Unless you intend to specialize in a lnaguage you really should arm yourself with as many languages as possible. You see problems from different points of view and can better understand the strengths and weaknesses of the tools you are using.
Thanks! So I can take comfort from your statement when my web stats show that 95% of all visitors are using some version of IE? Whoopee
From reading the (finally posted) update on SF it seems that a staff member's user/password was sniffed through a compromised upstream connection.
= 89 285
Read about it here:
http://sourceforge.net/forum/forum.php?forum_id
It's not clear if the ISP was socially hacked or exploited by a technical means.
This seems to be creating a cascading effect as a direct result of having a centralized repository and network of popular open source site as well as most everyone using similar passwords across multiple accounts.
Actually hobbyist programmers will usually duplicate work no matter waht kind of software model you use. Programmers like doing things 'their way' and as long as the complexity isn't gigantic (an OS) they'll happily duplicate work. They can also get more glory and ego gratification this way.
Visit freshmeat and see how many mp3 players, icq clones, text editors, etc. that you can find. Visit Sourceforge and you'll see how many more products were never released (started).
I say this as a programmer that has written closed and open apps and has projects hosted on fresheat and sourceforge.
Mozilla will probably go down in history as an example of how to produce software right as well as (perhaps irrevocably) lose marketshare and mindshare. Frankly, Mozilla *must* "succeed" or most non-windows OSes are screwed (and I expect it will do moderately well).
I see that this is upposed to be funny but it's also wrong.
Software has always been created to serve people, not the other way around. Someof us may work in conditions that are the other way around but it's wrong headed thinking.
Uh, since when was peer to peer technology not legitimate? We've been using it for years in local networking and games and so forth.
As an author who has released GPL'ed software I'd rather see this go to court and finally have some rulings on the GPL. It's one thing to be a company with money to spend of legal fees, it's quite another to be an author making no money off of a product. I'd like to finally see the GPL get some legal rulings to give it more weight.