Most truly innovative software was never written for the market in the first place. Most really neat stuff is written by on staff programmers or consultants and never leaves the company. There are so many scientific apps that are used that were never on the shelf and probably never will be.
It's always bothered me that so many OSS zealots are so limited in their vision. They're just as bad as the "Corporate Sheeple" they deride. If you want to make something Innovative, don't just make a themeable Windows. That's quite possibly the stupidest example of creative software. Who gives a fuck if you can make a scroll bar look like it's a leaf if you need software to model water flow through pipes? There's a lot of areas where Open Source could be useful but everyone's jerking off over Windowing Systems and Word Processors. McVoy is saddly right in this regaurd at the moment.
The open source oses are there, it's time to actualy make something worth running on them and not just focusing on the same 5% over and over.
Re:So, you programmers ready to give up your code?
on
McVoy Strikes Back
·
· Score: 1
Why should they? The OSS never paid these people's rent. What's worse is that instead of trying to help developers make a living they give moral lectures why it's wrong for them to expect a paycheck for the work they do.
I find it hard to want to write software when I'm homeless and hungry. To date no one has ever offered to give me food or shelter to write software who's code was open. I have, how ever, payed my bills writing code you'll never see.
The grandparent makes a damn good point where as your just insulting.
In my experience things like autocompletion are nice if your not exactly sure about the syntax or how an api works but otherwise it's just eye candy at best and breaks your flow at worst. The integrated help is nice, but it's nothing different from what's online or in the standard help kit that comes with the downloadable framework. I usually find what's online is much more helpful that Intellisense. I don't just get the basic syntax but how to actually use it and other related functions.
A lot of the tools included in VS.Net are top notch, but they're not the only way to work. Debugging with the debugger is like trying to find a huge billboard with a microscope. There are faster ways than the debugger. Next time you want to see what a variable is doing try putting MessageBoxes or Response.Write(""); in your code to see what's happening. Half the time you'll get it figured out in half the time than if you had used the debugger.
As far as other tools go, you can still; browse databases with other tools. In fact it's usually better to browse them with other tools. I know for MS SQL Server, Enterprise Manager is awesome and Query Analyzer is great for tuning my SQL statements. There are tools for Oracle and PostGreSQL as well that are very handy. Also keeping a copy of your basic unix tools on Windows is also handy. Grep has saved my ass a few times and bob knows I couldn't live without make;)
I too use a lot of XSLT and I couldn't live without it, but in all honesty 98% of the problems I run into aren't syntaticial. That means things like Intellisense are pretty much moot. I have quite a few macros for Gvim for testing XSL documents as well as some other basic tools.
That's why I and many others don't use IDEs but that doesn't mean my way is right for everyone. The best tool for the job is the one that you feel most comfortable with and allows you to solve issues most effectively.
First of all, your idea that the best art is free art flies in the face of hundreds of years of art-making.
That doesn't grok to me. Sure a lot of people made great stuff while being paid for it, but just as many people made stuff to not get paid. Consider Walden by Henry David Thoreau. He wrote that while trying to live out in the woods avoiding making money as much as possible. Consider folk music. Not the kind of popular commercial folk music but the old school traditional kind. People didn't make folk music to make money, but to have a good time.
Secondly, you are incredibly short-sighted if you think all these great movies will be made if it starts to become unprofitable to make movies.
It's equally short sighted to think that great movies are made because it is profitable. Quite a few ground breaking movies were made not because they were going to be profitable, but because everyone involved loved the art form.
The hard belief that everything is done to make money somehow is a rather sad and unstable view of life. Everyday I see people do wonderful things without any regard to weather they're going to make money from this action or not. That's not just art but life itself. Open your eyes,man,and have a little faith in people.
Vi and FireFox is all I need for ASP.Net development. Oh and a.Net compatable web server.
Gvim for Windows is my tool of choice, thou some days, when I'm feeling sassy, I use TextPad. People who can only code in an IDE, tend to produce bad code. Don't get me wrong, IDEs have their place, but you don't actually NEED one.
If you don't understand your platform and tools well enough to be able to code with only a text editor then you may not actually know your platform. Doesn't matter if it's.Net, PHP or Prolog. Know the dark secrets of your tools and you could code everything with nothing more than a shell account and an apple II.
Granted I'm probably preaching to the choir here;)
At least I hope I am......
I should probably go to bed, post-24 hour codeing sessions make for crappy posts on slashdot.
Granted the grand parent is trolling but holy crap man, way to prove a troll right. Almost everything you just said was a bunch of buzzwords and miss informed incoherent babble. First off, POSIX has nothing to do with weather or not an os is a microkernel or a monolithic kernel. It's a standardized api for unix system calls.
Next up Windows has progressively grown larger and larger since it's release. By your logic it bloated up in 95, shrank in 98 and then it bloated up in both the Me version and 2000 version (which was a completely different os). Internet Explorer wasn't tied to the kernel ever. It was, however, added to the gui api in 98 and then ported over to NT. The kernel can live without IE and it can be removed. It's just not a good idea since a lot of apps use the IE widgets (it is kinda handy at times).
As for Outlook and MSN being tied to the OS to the level that IE is kinda says you know fuck all about Windows. It is possable to remove both Outlook and MSN in Windows without it comeing to a grinding halt.
You may have had an OS class in collage but I seriously hope your professor failed you because you didn't learn a single thing.
Because there's 2 models for ASP.Net. The standard way (which if your using VS.Net your probably doing) is to use code behind pages to separate the code from mark up. The code is in a separate file and everything is inherited from the page object and compile-able. Nice but when you build, your only building the code behind pages and components.
I've been porting about 8 years of old asp stuff recently which doesn't really compile. Lots of inline code, well written stuff (most of it) so it's not too hard to port or maintain, it just doesn't compile. Not exactly Microsoft's recommended method for ASP.Net, but it does have it's good points. Minor increase in speed and easier to understand and manage code. You do kinda have to deal with managing your own forms and such, but that's not to hard and your pretty much screwed if you try to use VS.Net, but I use gvim anyway so it's a non issue. Seems inline code was scrapped from asp 3.0 to ASP.Net 1.0 and 1.1 but it was requested enough that they put together a compiler for the.Net 2.0 framework that bulk pre-compiles all pages for a app, inline and code behind.
actualy "porting" 32-bit apps to amd64 is pretty much a non-issue. The 64-bit Long Mode is just 32-bit Protected Mode with the ability to optionaly have longer word size and more registers. It's absolutely nothing like moveing from 16-bit Real Mode to 32-bit Protected Mode.
We've been updating our old asp 3.0 code to asp.net. The big problem we've run into is we want to precompile our asp.net but can't with the 1.1 framework. We can with the beta asp.net 2.0 compiler. It requires the asp.net 2.0 runtime as well and that's not so common. I don't know what Mono's 2.0 support is like, if anything yet, but if I could this tool could help compile the inlined pages as well as "port" it to other platforms that would rule.
Also the title of this story is incredibly stupid. What the hell is a VS.Net application? I know what a.Net app is. I know I have Visual Studio.Net on my work machine, but I've rarely use VS.Net when writing.Net code....
I'd be really distraught. I'd probably spend a week not eating, drinking heavily and listen to really depressing music. I'd also give up computers, move to a mountain and become a monk of some esoteric religion. I'd have to after havening finished my life long quest so abruptly.
Blah, their just copying OpenBSD. You can harden up Windows. Shit, something like 80% of Window's security issues go away by throwing up a simple firewall. Not using IE helps , thou now more Firefox issues are coming up so even that's not so true anymore. The biggest security flaw out there that's even cross platform is thinking that anything default will protect you. Any time you think "oh, well I'm secure now" you've probably just been rooted.
Basicly Operating systems are flawed at this point but linux is proactive adding features to stop the flaws even if the happen.
Linux isn't proactive, they're reactive. You must remember, Linux wasn't designed for security, it was designed so that Linus could run Unix on his PC at home for free. Fixes happen because flaws were FOUND not to keep them from being found.
Probably because that would set a legal precedence for suing the creators of ever OSS Operating System every security flaw ever exploited in their software as well. Spam has to be relayed somehow.....
If they're guru's... why are there still issues after hundreds of "fixes" over the years?
The same reason Linus and hundreds of other people still do patches to Linux. No software is truly finished and secure. Not even Hello World. There's a really nasty buffer overflow in that one. I don't even know why people still use it.
You know Microsoft has a Linux lab, right? The problem is they probably could answer your questions and possibly seal up a few security issues that could have bitten you in the ass later. Your right about Windows being a flawed model, but they said the same thing about Unix 20 years ago. All security models are flawed that allow people in to do things like "run programs" and "view data".
I've yet to see a secure os and it's not from lack of effort. I've been looking for an os that doesn't suck for years.
The problem is in the real world people don't understand that there are razor blades in the room and get cut the hell up even thou they can see them. If you can't see the razor blades but know they are there you tend to be a bit more careful. Of course in both situations you can avoid being cut by knowing that there are razor blades around.
The biggest issue I've seen with admins who use any product isn't weather the source is open or not, but weather they know what the hell their doing in the first place. In my experience admins usually aren't programmers, so even if they have the code, they don't know what their looking at. Elitist, yes, but sadly true in a lot of cases.
Linux does get hacked and you can lock down Windows. The only logical conclusion is that all operating systems sucks and if your ever stuck using one, be aware that there will always be problems.
Man, I thought I was the only one here to caught that.
I do know this guy who's been stuck for months in some airport in some random Chinese city right now because he was writing software to track genetic defects in rabbits that were bred for medical testing. Seems PETA didn't like that and pulled out their navy and threated to blockade Chinese ships if they didn't release something like 20,000 rabbits. I didn't know PETA had a navy much less one that could scare the Chinese government.... creepy.
How is epolling and microcode updating a killer app? Name one application that uses either of those features that does something that doesn't exist on any of the other major operating system? Deleting all your files easily isn't exactly something no other operating system doesn't do. Most of these features in the kernel don't really matter if there's nothing truly new that only runs on it. Who gives a fuck if you can alter the microcode for jz if all the cool new apps run on something else?
It's a huge one. Someone has to come up with a really cool idea that's never been done before and implement it before some for profit, not open source company comes up with the same idea. Not only that but they have to create and implement this and stay years ahead of any commercial app that does create the same kind of software. All in all it's not an easy task.
I'd imagine so. I don't know if I can help you with that one but there are options. Like all Open Source software, there's no guarantee that any of these will work of provide you with the features you need. If you know c or c++ well, I'm sure whatever something doesn't do, you could add to it. This might be the way to break Linux into your industry. Find and upgrade one to the Killer App (tm by someone I'm sure) for CAD users.
I have a feeling that will get sorted out in a few years. I think it would encourage developers to work on the Linux UI if they were to say "I really love and need this one app but the rest is crap, I think I'll help fix it so I won't have to deal with it anymore". Even without that the awesome developers who work on Linux's UI are doing a good job. They'll get it eventually, and when they do, I'm switch my dad over so he'll stop calling me about spyware;)
I don't know, Autocad for Windows users are pretty hardcore. Even GIS is pretty Windows-a-riffic (at least it was 6 years ago).
I'm thinking we're just going to have to create a whole new type of app. Not just another CAD, music composition, graphic design or accounting package. Something truly innovative and extreamly needed.
I should have been more clear. What I ment was Windows 98 and below on a Dos kernel. And Kde and Gnome have the same amount of functionality (note: not just usablity) of Windows 98.
You can have an open source project that's gpled that is very difficult to port. If you don't write it in a scripting language like perl or python and use kernel calls instead of generic libraries you could have the source be open to all but not very portable. Case in point: Rosegarden.
Rosegarden is a gpled sound sequencer for Linux that's tied to a very low level audio library. To make a port to windows they'd have to rewrite large parts of the system which would destroy the speed of the app for limited portability.
Other ways to make porting a bitch is to write parts of the app in assembly. The system calls for Windows and Linux are worlds apart. The program could be completely open source under the gpl which anyone could look at but 95% of Linux USERS wouldn't be able to understand while looking right at it. Not very nice but even that's not required. Just use a library like ALSA and your pretty much stuck with Linux (unless someone ports ALSA, but I'm not sure that'll ever happen).
It's the nature of the beast and it killed my social life but paid my rent.
Slashdot Mirror
Most truly innovative software was never written for the market in the first place. Most really neat stuff is written by on staff programmers or consultants and never leaves the company. There are so many scientific apps that are used that were never on the shelf and probably never will be.
It's always bothered me that so many OSS zealots are so limited in their vision. They're just as bad as the "Corporate Sheeple" they deride. If you want to make something Innovative, don't just make a themeable Windows. That's quite possibly the stupidest example of creative software. Who gives a fuck if you can make a scroll bar look like it's a leaf if you need software to model water flow through pipes? There's a lot of areas where Open Source could be useful but everyone's jerking off over Windowing Systems and Word Processors. McVoy is saddly right in this regaurd at the moment.
The open source oses are there, it's time to actualy make something worth running on them and not just focusing on the same 5% over and over.
Why should they? The OSS never paid these people's rent. What's worse is that instead of trying to help developers make a living they give moral lectures why it's wrong for them to expect a paycheck for the work they do.
I find it hard to want to write software when I'm homeless and hungry. To date no one has ever offered to give me food or shelter to write software who's code was open. I have, how ever, payed my bills writing code you'll never see.
The grandparent makes a damn good point where as your just insulting.
In my experience things like autocompletion are nice if your not exactly sure about the syntax or how an api works but otherwise it's just eye candy at best and breaks your flow at worst. The integrated help is nice, but it's nothing different from what's online or in the standard help kit that comes with the downloadable framework. I usually find what's online is much more helpful that Intellisense. I don't just get the basic syntax but how to actually use it and other related functions.
;)
A lot of the tools included in VS.Net are top notch, but they're not the only way to work. Debugging with the debugger is like trying to find a huge billboard with a microscope. There are faster ways than the debugger. Next time you want to see what a variable is doing try putting MessageBoxes or Response.Write(""); in your code to see what's happening. Half the time you'll get it figured out in half the time than if you had used the debugger.
As far as other tools go, you can still; browse databases with other tools. In fact it's usually better to browse them with other tools. I know for MS SQL Server, Enterprise Manager is awesome and Query Analyzer is great for tuning my SQL statements. There are tools for Oracle and PostGreSQL as well that are very handy. Also keeping a copy of your basic unix tools on Windows is also handy. Grep has saved my ass a few times and bob knows I couldn't live without make
I too use a lot of XSLT and I couldn't live without it, but in all honesty 98% of the problems I run into aren't syntaticial. That means things like Intellisense are pretty much moot. I have quite a few macros for Gvim for testing XSL documents as well as some other basic tools.
That's why I and many others don't use IDEs but that doesn't mean my way is right for everyone. The best tool for the job is the one that you feel most comfortable with and allows you to solve issues most effectively.
First of all, your idea that the best art is free art flies in the face of hundreds of years of art-making.
,man ,and have a little faith in people.
That doesn't grok to me. Sure a lot of people made great stuff while being paid for it, but just as many people made stuff to not get paid. Consider Walden by Henry David Thoreau. He wrote that while trying to live out in the woods avoiding making money as much as possible. Consider folk music. Not the kind of popular commercial folk music but the old school traditional kind. People didn't make folk music to make money, but to have a good time.
Secondly, you are incredibly short-sighted if you think all these great movies will be made if it starts to become unprofitable to make movies.
It's equally short sighted to think that great movies are made because it is profitable. Quite a few ground breaking movies were made not because they were going to be profitable, but because everyone involved loved the art form.
The hard belief that everything is done to make money somehow is a rather sad and unstable view of life. Everyday I see people do wonderful things without any regard to weather they're going to make money from this action or not. That's not just art but life itself. Open your eyes
Vi and FireFox is all I need for ASP.Net development. Oh and a .Net compatable web server.
.Net, PHP or Prolog. Know the dark secrets of your tools and you could code everything with nothing more than a shell account and an apple II.
;)
Gvim for Windows is my tool of choice, thou some days, when I'm feeling sassy, I use TextPad. People who can only code in an IDE, tend to produce bad code. Don't get me wrong, IDEs have their place, but you don't actually NEED one.
If you don't understand your platform and tools well enough to be able to code with only a text editor then you may not actually know your platform. Doesn't matter if it's
Granted I'm probably preaching to the choir here
At least I hope I am......
I should probably go to bed, post-24 hour codeing sessions make for crappy posts on slashdot.
Granted the grand parent is trolling but holy crap man, way to prove a troll right. Almost everything you just said was a bunch of buzzwords and miss informed incoherent babble. First off, POSIX has nothing to do with weather or not an os is a microkernel or a monolithic kernel. It's a standardized api for unix system calls.
Next up Windows has progressively grown larger and larger since it's release. By your logic it bloated up in 95, shrank in 98 and then it bloated up in both the Me version and 2000 version (which was a completely different os). Internet Explorer wasn't tied to the kernel ever. It was, however, added to the gui api in 98 and then ported over to NT. The kernel can live without IE and it can be removed. It's just not a good idea since a lot of apps use the IE widgets (it is kinda handy at times).
As for Outlook and MSN being tied to the OS to the level that IE is kinda says you know fuck all about Windows. It is possable to remove both Outlook and MSN in Windows without it comeing to a grinding halt.
You may have had an OS class in collage but I seriously hope your professor failed you because you didn't learn a single thing.
Because there's 2 models for ASP.Net. The standard way (which if your using VS.Net your probably doing) is to use code behind pages to separate the code from mark up. The code is in a separate file and everything is inherited from the page object and compile-able. Nice but when you build, your only building the code behind pages and components.
.Net 2.0 framework that bulk pre-compiles all pages for a app, inline and code behind.
I've been porting about 8 years of old asp stuff recently which doesn't really compile. Lots of inline code, well written stuff (most of it) so it's not too hard to port or maintain, it just doesn't compile. Not exactly Microsoft's recommended method for ASP.Net, but it does have it's good points. Minor increase in speed and easier to understand and manage code. You do kinda have to deal with managing your own forms and such, but that's not to hard and your pretty much screwed if you try to use VS.Net, but I use gvim anyway so it's a non issue. Seems inline code was scrapped from asp 3.0 to ASP.Net 1.0 and 1.1 but it was requested enough that they put together a compiler for the
actualy "porting" 32-bit apps to amd64 is pretty much a non-issue. The 64-bit Long Mode is just 32-bit Protected Mode with the ability to optionaly have longer word size and more registers. It's absolutely nothing like moveing from 16-bit Real Mode to 32-bit Protected Mode.
We've been updating our old asp 3.0 code to asp.net. The big problem we've run into is we want to precompile our asp.net but can't with the 1.1 framework. We can with the beta asp.net 2.0 compiler. It requires the asp.net 2.0 runtime as well and that's not so common. I don't know what Mono's 2.0 support is like, if anything yet, but if I could this tool could help compile the inlined pages as well as "port" it to other platforms that would rule.
.Net app is. I know I have Visual Studio.Net on my work machine, but I've rarely use VS.Net when writing .Net code....
Also the title of this story is incredibly stupid. What the hell is a VS.Net application? I know what a
I'd be really distraught. I'd probably spend a week not eating, drinking heavily and listen to really depressing music. I'd also give up computers, move to a mountain and become a monk of some esoteric religion. I'd have to after havening finished my life long quest so abruptly.
Man, that would suck.
Blah, their just copying OpenBSD. You can harden up Windows. Shit, something like 80% of Window's security issues go away by throwing up a simple firewall. Not using IE helps , thou now more Firefox issues are coming up so even that's not so true anymore. The biggest security flaw out there that's even cross platform is thinking that anything default will protect you. Any time you think "oh, well I'm secure now" you've probably just been rooted.
Basicly Operating systems are flawed at this point but linux is proactive adding features to stop the flaws even if the happen.
Linux isn't proactive, they're reactive. You must remember, Linux wasn't designed for security, it was designed so that Linus could run Unix on his PC at home for free. Fixes happen because flaws were FOUND not to keep them from being found.
Probably because that would set a legal precedence for suing the creators of ever OSS Operating System every security flaw ever exploited in their software as well. Spam has to be relayed somehow.....
If they're guru's... why are there still issues after hundreds of "fixes" over the years?
The same reason Linus and hundreds of other people still do patches to Linux. No software is truly finished and secure. Not even Hello World. There's a really nasty buffer overflow in that one. I don't even know why people still use it.
You know Microsoft has a Linux lab, right? The problem is they probably could answer your questions and possibly seal up a few security issues that could have bitten you in the ass later. Your right about Windows being a flawed model, but they said the same thing about Unix 20 years ago. All security models are flawed that allow people in to do things like "run programs" and "view data".
I've yet to see a secure os and it's not from lack of effort. I've been looking for an os that doesn't suck for years.
The problem is in the real world people don't understand that there are razor blades in the room and get cut the hell up even thou they can see them. If you can't see the razor blades but know they are there you tend to be a bit more careful. Of course in both situations you can avoid being cut by knowing that there are razor blades around.
The biggest issue I've seen with admins who use any product isn't weather the source is open or not, but weather they know what the hell their doing in the first place. In my experience admins usually aren't programmers, so even if they have the code, they don't know what their looking at. Elitist, yes, but sadly true in a lot of cases.
Linux does get hacked and you can lock down Windows. The only logical conclusion is that all operating systems sucks and if your ever stuck using one, be aware that there will always be problems.
Man, I thought I was the only one here to caught that.
I do know this guy who's been stuck for months in some airport in some random Chinese city right now because he was writing software to track genetic defects in rabbits that were bred for medical testing. Seems PETA didn't like that and pulled out their navy and threated to blockade Chinese ships if they didn't release something like 20,000 rabbits. I didn't know PETA had a navy much less one that could scare the Chinese government.... creepy.
How is epolling and microcode updating a killer app? Name one application that uses either of those features that does something that doesn't exist on any of the other major operating system? Deleting all your files easily isn't exactly something no other operating system doesn't do. Most of these features in the kernel don't really matter if there's nothing truly new that only runs on it. Who gives a fuck if you can alter the microcode for jz if all the cool new apps run on something else?
;)
And rm does run on Windows too
Unless they actualy want to write something that uses the features of stuff like a sound card. Then you've written an app that's not so portable.
Done that, mixing the too makes your feet numb. It's neat but not something you want to do ever.
It's a huge one. Someone has to come up with a really cool idea that's never been done before and implement it before some for profit, not open source company comes up with the same idea. Not only that but they have to create and implement this and stay years ahead of any commercial app that does create the same kind of software. All in all it's not an easy task.
Here's hoping thou...
I'd imagine so. I don't know if I can help you with that one but there are options. Like all Open Source software, there's no guarantee that any of these will work of provide you with the features you need. If you know c or c++ well, I'm sure whatever something doesn't do, you could add to it. This might be the way to break Linux into your industry. Find and upgrade one to the Killer App (tm by someone I'm sure) for CAD users.
I agree.
;)
I have a feeling that will get sorted out in a few years. I think it would encourage developers to work on the Linux UI if they were to say "I really love and need this one app but the rest is crap, I think I'll help fix it so I won't have to deal with it anymore". Even without that the awesome developers who work on Linux's UI are doing a good job. They'll get it eventually, and when they do, I'm switch my dad over so he'll stop calling me about spyware
I don't know, Autocad for Windows users are pretty hardcore. Even GIS is pretty Windows-a-riffic (at least it was 6 years ago).
I'm thinking we're just going to have to create a whole new type of app. Not just another CAD, music composition, graphic design or accounting package. Something truly innovative and extreamly needed.
I should have been more clear. What I ment was Windows 98 and below on a Dos kernel. And Kde and Gnome have the same amount of functionality (note: not just usablity) of Windows 98.
And for the record you can skin the hell out of Windows as well as have virtual desktops.
You can have an open source project that's gpled that is very difficult to port. If you don't write it in a scripting language like perl or python and use kernel calls instead of generic libraries you could have the source be open to all but not very portable. Case in point: Rosegarden.
Rosegarden is a gpled sound sequencer for Linux that's tied to a very low level audio library. To make a port to windows they'd have to rewrite large parts of the system which would destroy the speed of the app for limited portability.
Other ways to make porting a bitch is to write parts of the app in assembly. The system calls for Windows and Linux are worlds apart. The program could be completely open source under the gpl which anyone could look at but 95% of Linux USERS wouldn't be able to understand while looking right at it. Not very nice but even that's not required. Just use a library like ALSA and your pretty much stuck with Linux (unless someone ports ALSA, but I'm not sure that'll ever happen).
It's the nature of the beast and it killed my social life but paid my rent.