Really? I don't see Microsoft going out of business anytime soon. Of course, they have closed-source applications software and other businesses such as WebTV and MSN that make them a bundle (or have the potential to make them a bundle) as well.
They're just making themselves look bad. They put this piece out precisely because they DO think that Linux is a threat to server markets.
Just about every journalist, pundit, etc. sees Linux as a direct threat to other server operating systems, including Windows NT, Netware, and other Unixes like SCO. In fact, many contend the exact opposite of SCO's position that Linux is better for desktops than for servers.
I tend to agree with the most of the press: Linux is a *great* server OS and is only good on desktops for true power users. John Q. Public doesn't have the first clue what to do if someone tells him he has to modify/etc/XF86Config by hand, or even worse that he has to recompile his kernel.
OTOH, servers are typically installed and run by competent professionals who would have no trouble tinkering under the hood of the OS.
But I digress: my point is that by giving Linux this negative press, they make themselves look far worse than they make Linux. Customers of SCO products are likely to pass brochures like that along to their sysadmins who are likely to look at it and laugh.
Star Office is great for generic office package. Its bloated, its slow, but it has MS Office compatibility which is a must in today's business world.
As for databasing, you should probably write your own, if at all possible. The FSF has some great database routines for C programs. Couple that with a MySQL setup on the server...and you're in business. Thats what most companies, including Burlington Coat Factory, which has a Linux-based solution, have done.
Java or Perl and HTML might not be bad for a Web-based intranet client. However, IMHO, Web- based clients generally aren't robust enough due to security and other limitations...
Lets not forget that FoxPro runs on SCO and (I think) AIX. And you can get a stripped-down version of SCO for free....
GM and Ford don't lay off engineers: they take the higher paid ones near retirement age and offer them a nice retirement bonus and do anything it takes to make retirement seem good.
Then they replace them with contractors, which don't have to be laid off because they are temporary employees. A lot of GM engineers work for contract houses like MSX International, and more and more engineering work is being done by Tier One suppliers.
I work with a lot of such Tier One suppliers in the Body-In-White Assembly Tooling community. A lot of tooling work used by done by GM's Metal Fabricating Division (MFD) but is increasingly being done by Tier One.
While SGI's have been the overwhelming popular choice in Hollywood for doing CGI (as in computer generate imagery, not in the WWW sense), this is changing. More and more of this is being done on workstations running NT with packages like 3D Studio Max or the Crystal Graphics package (I can't think of its name right now, and last I remember it was a DOS app, but someone told me its now NT). Some of this includes some of the work on the Terminator 2 and Terminator 3 movies, The Abyss, Titanic, and others. (Ok, I admit these are all James Cameron movies...but there are others) No, SGIs aren't going away in Hollywood just yet, not by a long shot, but Intel-based hardware is becoming capable enough for production use.
Well, at GM there is a disclaimer on all login prompts that says something to effect that ALL communications are monitored. The company policy is essentially that because GM pays for bandwidth, equipment, etc., that personal communications of ANY kind are strictly forbidden and that all e-mail, WWW traffic, etc. can be scanned.
That being said, I don't think anyone at EDS (who does most sysadmin work at GM) actually scans the network traffic unless they believe there is a security breach of some kind.
How GM deals with the issue is that 1) it assumes that GM employees and contractors are professionals and as such are somewhat trusted to behave professionally and 2) not everyone automatically has Internet access, including access to e-mail: you basically have to have a business case. Most people with PCs have e-mail, but this is not the case for other forms of Internet access.
Finally, when it comes down to it, if you simply cannot morally abide by it, either A) refuse and stand up for your morality and get fired if thats what it takes (at least you will have your integrity) or B) tell them you're scanning but don't. B is a cop-out, IMHO: that option, in and of itself is morally corrupt.
Another option is to simply quit: there are lots of other system administrator positions available. But don't count on the other company to not put you in the same situation: its becoming increasingly commonplace for companies to scan their network activity.
... "Every" phone conversation? Do you realize the kind of resources it would take to police EVERY SINGLE phone conversation on the planet? There isn't enough people or equipment in the entire NSA/CIA/FBI/whatever to listen to every single phone conversation. That being said, the phone system isn't secure enough for truly private conversations.
And the NSA and the CIA can only listen in on foreign phone conversations, they can't spy on ordinary citizens..
also, "hand-coding" promotes cleaner OO than the slap-everything-on-a-form-and-be-done-with-it attitude of Delphi. you simply cannot easily subclass, say, TPanel and stuff some widgets in and use it once, like you can with the gtk+, I think that's one reason why software written with Delphi generally is of lower quality than software written with what people would consider more spartan tools.
Well, most people don't use RAD ("slap-it-on-a-form") tools for general-purpose applications building. Their typically used for two reasons: one is for prototyping applications that will later be rewritten in C/C++, and another is for mission-critical, specific use custom-built applications that need to be developed in a hurry. Tools like VB and Delphi are favorites of contract programmers, for instance.
And glade is nowhere near the robust tool Delphi is. (Flames >/dev/null)
What are their credentials? Where is there evidence? Where is their proof? What methods were used in their study? Why should we believe them at all?
I love these journalists that go spouting off about studies because they happen to fit with their own mode of thinking. They'll have all of us on some syndrome sooner or later...We'll all be victims and their will be no accountability for anyone's actions.
Sheesh. Has anyone heard of responsible journalism?
Be skeptical whenever you read anything in the press... don't let people tell you what you should think. Use your brain.
Object Pascal by far more suited to RAD than C/C++.
Pascal has strong type checking, and the structure of the language basically FORCES you to write good code. C/C++ is a good choice for professional/sem-professional developers that need a good, robust language.
But for a one-man programming project, particularly when RAD methods are to be used, Delphi and Object Pascal puts C++ to waste.
Most Linux programmers, OTOH, are experienced hackers that think that definition of "Languages of Choice" certainly does not include Pascal. In hacker-style coding, C++ is certainly preferable, but for mission-critical code that must be developed in a hurry, one certainly cannot beat Delphi. The only other choice for RAD is Visual Basic, and we certainly won't see Linux programmers flocking to that any time soon, for obvious reasons.:-)
Sun's largest markets are in the high-end server and high-end engineering and animation workstation markets. Sun is also primarily a hardware company, not a software company.
While Linux certainly has the ability to run on high-end hardware, most folks buying Sun hardware are going to be running Solaris. Linux has some scalability and other problems in this market.
For instance, one of the more popular engineering applications is Unigraphics, which is used extensively in the automotive and aerospace industries. UG will probably never run on Linux, especially considering Unigraphics Solutions' recent NT port. Currently HP and Sun have the market cornered on UG workstations.
Also, consider high-end server usage. Linux has no support for a journaling file system. Filesystem redundancy on Linux is lacking in general, in fact (ie Linux has support for software RAID, but only a handful of hardware RAID drivers exist...on a busy server, software RAID is practically useless due to its much slower speed.)
So all of you who are concerned with Sun destroying Linux need not be. Its not going to happen. Sun does not picture itself to be in the low-end of the market (Java was created to sell servers).
They can't speed it up -- principle rule of software development is that software never gets smaller or faster (That software is faster is only the result of faster hardware.)
Your "principle rule" is simply not true. It is in fact possible to speed up software. 2.2.x versions of the Linux kernel are certainly faster than 2.0.x version--and not just because of the multiprocessing capabilities. I'd do a little homework before I made blanket statements like the one you just made.
Well, its only been possible to convict someone of piracy about a year... Prior the new law put on the books last year, piracy was only a civil offense, not a criminal one. (IOW, you could get sued for piracy, but not put in jail over it.)
Yes, Merced *WILL* run x86 code, but its "native" 64-bit code is based on HP's PA-RISC architecture, found in many of HPs Unix workstations and servers.
Besides, Linux is still primarily x86. One nice thing about Linux is that it runs well on cheap hardware. Ultimately it won't be the processor that matters, but the software.
Besides the obvious (there are lots of libraries that implement win32 on native Linux) there is no source code available for most Windows apps.
This means that you personally cannot recompile your Windows application unless you have written it. And lets not forget that most software companies aren't going to invest the money and energy into "band-aiding" their Windows code to make i9t run on Linux (with the obvious exception of Corel) as it is probably more cost effective in the long run to implement the code the "right" way (written as a native Linux app) since maintaining Windows code compiled for Linux will likely be a bear. Remember that most Windows APIs rely on Windows' poor OS design (ie, too many modules try to do too much...vs. simple one-function programs on Unix/Linux that are really really good at one thing instead of trying to do everything...)
I saw that movie. It was pretty decent for the type of movie it was.
But like most movies about hackers its way off the mark. Even the basic premise is at fault: remember Phil Zimmerman and the whole PGP thing? The NSA is still unable to crack anything but the smallest (56-bit) RSA keys, but RSA encryption is now everywhere...
The Jargon File, aka "The New Hacker's Dictionary" currently written/edited by Eric S. Raymond, paints just the picture you mention. Hardly the "liberal media" that you mention.
You know, this story is greatly distorted. The very large software company's OS did not catch on in large numbers at all until 1990 (Version 3.0 of that operating system). The same OS really didn't gain its near-monopoly status until sometime after 1993. (Before about 1992/1993, most computers ran on the very large software companies earlier efforts, which weren't even graphical.)
So the larger companies graphical OS really didn't catch on until *well* after the earlier companies computer became color-capable.
And the realm of computing is split into far more than two camps. You have the Windows camp. The smaller, but still existant OS/2 camp. The Linux/xBSD camp. The commercial Unix camp. The still-very-much-active Amiga contingent. The BeOS camp. Need I go on?
Don't be so short-sighted. There are many, many more viable platforms than the one you happen to find attractive.
Slashdot Mirror
Really? I don't see Microsoft going out of business anytime soon. Of course, they have closed-source applications software and other businesses such as WebTV and MSN that make them a bundle (or have the potential to make them a bundle) as well.
Yes, but the only restrictions in the GPL are to make it unrestrictive.
They're just making themselves look bad. They put this piece out precisely because they DO think that Linux is a threat to server markets.
/etc/XF86Config by hand, or even worse that he has to recompile his kernel.
Just about every journalist, pundit, etc. sees Linux as a direct threat to other server operating systems, including Windows NT, Netware, and other Unixes like SCO. In fact, many contend the exact opposite of SCO's position that Linux is better for desktops than for servers.
I tend to agree with the most of the press: Linux is a *great* server OS and is only good on desktops for true power users. John Q. Public doesn't have the first clue what to do if someone tells him he has to modify
OTOH, servers are typically installed and run by competent professionals who would have no trouble tinkering under the hood of the OS.
But I digress: my point is that by giving Linux this negative press, they make themselves look far worse than they make Linux. Customers of SCO products are likely to pass brochures like that along to their sysadmins who are likely to look at it and laugh.
Star Office is great for generic office package. Its bloated, its slow, but it has MS Office compatibility which is a must in today's business world.
...and you're in business. Thats what most companies, including Burlington Coat Factory, which has a Linux-based solution, have done.
As for databasing, you should probably write your own, if at all possible. The FSF has some great database routines for C programs. Couple that with a MySQL setup on the server
Java or Perl and HTML might not be bad for a Web-based intranet client. However, IMHO, Web- based clients generally aren't robust enough due to security and other limitations...
Lets not forget that FoxPro runs on SCO and (I think) AIX. And you can get a stripped-down version of SCO for free....
GM and Ford don't lay off engineers: they take the higher paid ones near retirement age and offer them a nice retirement bonus and do anything it takes to make retirement seem good.
Then they replace them with contractors, which don't have to be laid off because they are temporary employees. A lot of GM engineers work for contract houses like MSX International, and more and more engineering work is being done by Tier One suppliers.
I work with a lot of such Tier One suppliers in the Body-In-White Assembly Tooling community. A lot of tooling work used by done by GM's Metal Fabricating Division (MFD) but is increasingly being done by Tier One.
While SGI's have been the overwhelming popular choice in Hollywood for doing CGI (as in computer generate imagery, not in the WWW sense), this is changing. More and more of this is being done on workstations running NT with packages like 3D Studio Max or the Crystal Graphics package (I can't think of its name right now, and last I remember it was a DOS app, but someone told me its now NT). Some of this includes some of the work on the Terminator 2 and Terminator 3 movies, The Abyss, Titanic, and others. (Ok, I admit these are all James Cameron movies...but there are others) No, SGIs aren't going away in Hollywood just yet, not by a long shot, but Intel-based hardware is becoming capable enough for production use.
Well, at GM there is a disclaimer on all login prompts that says something to effect that ALL communications are monitored. The company policy is essentially that because GM pays for bandwidth, equipment, etc., that personal communications of ANY kind are strictly forbidden and that all e-mail, WWW traffic, etc. can be scanned.
That being said, I don't think anyone at EDS (who does most sysadmin work at GM) actually scans the network traffic unless they believe there is a security breach of some kind.
How GM deals with the issue is that 1) it assumes that GM employees and contractors are professionals and as such are somewhat trusted to behave professionally and 2) not everyone automatically has Internet access, including access to e-mail: you basically have to have a business case. Most people with PCs have e-mail, but this is not the case for other forms of Internet access.
Finally, when it comes down to it, if you simply cannot morally abide by it, either A) refuse and stand up for your morality and get fired if thats what it takes (at least you will have your integrity) or B) tell them you're scanning but don't. B is a cop-out, IMHO: that option, in and of itself is morally corrupt.
Another option is to simply quit: there are lots of other system administrator positions available. But don't count on the other company to not put you in the same situation: its becoming increasingly commonplace for companies to scan their network activity.
... "Every" phone conversation? Do you realize the kind of resources it would take to police EVERY SINGLE phone conversation on the planet? There isn't enough people or equipment in the entire NSA/CIA/FBI/whatever to listen to every single phone conversation. That being said, the phone system isn't secure enough for truly private conversations.
And the NSA and the CIA can only listen in on foreign phone conversations, they can't spy on ordinary citizens..
I think Transmeta is banking on the fact the Linus Torvalds works there.
:-)
In their mind, if they release a CPU, Amiga or not, a large portion of Linux users will migrate to the new Transmeta technology.
Whether this is true or not, IMHO, depends on exactly what it is they release.
also, "hand-coding" promotes cleaner OO than the slap-everything-on-a-form-and-be-done-with-it attitude of Delphi. you simply cannot easily subclass, say, TPanel and stuff some widgets in and use it once, like you can with the gtk+, I think that's one reason why software written with Delphi generally is of lower quality than software written with what people would consider more spartan tools.
Well, most people don't use RAD ("slap-it-on-a-form") tools for general-purpose applications building. Their typically used for two reasons: one is for prototyping applications that will later be rewritten in C/C++, and another is for mission-critical, specific use custom-built applications that need to be developed in a hurry. Tools like VB and Delphi are favorites of contract programmers, for instance.
And glade is nowhere near the robust tool Delphi is. (Flames >/dev/null)
VB is a RAD tool and Delphi is not entirely unlike it. In fact Delphi was written to compete with VB.
Don't knock VB programming until you've tried it. Besides, there's lots of money for VB coders out there.
What are their credentials? Where is there evidence? Where is their proof? What methods were used in their study? Why should we believe them at all?
... don't let people tell you what you should think. Use your brain.
I love these journalists that go spouting off about studies because they happen to fit with their own mode of thinking. They'll have all of us on some syndrome sooner or later...We'll all be victims and their will be no accountability for anyone's actions.
Sheesh. Has anyone heard of responsible journalism?
Be skeptical whenever you read anything in the press
Object Pascal by far more suited to RAD than C/C++.
:-)
Pascal has strong type checking, and the structure of the language basically FORCES you to write good code. C/C++ is a good choice for professional/sem-professional developers that need a good, robust language.
But for a one-man programming project, particularly when RAD methods are to be used, Delphi and Object Pascal puts C++ to waste.
Most Linux programmers, OTOH, are experienced hackers that think that definition of "Languages of Choice" certainly does not include Pascal. In hacker-style coding, C++ is certainly preferable, but for mission-critical code that must be developed in a hurry, one certainly cannot beat Delphi. The only other choice for RAD is Visual Basic, and we certainly won't see Linux programmers flocking to that any time soon, for obvious reasons.
Sun's largest markets are in the high-end server and high-end engineering and animation workstation markets. Sun is also primarily a hardware company, not a software company.
While Linux certainly has the ability to run on high-end hardware, most folks buying Sun hardware are going to be running Solaris. Linux has some scalability and other problems in this market.
For instance, one of the more popular engineering applications is Unigraphics, which is used extensively in the automotive and aerospace industries. UG will probably never run on Linux, especially considering Unigraphics Solutions' recent NT port. Currently HP and Sun have the market cornered on UG workstations.
Also, consider high-end server usage. Linux has no support for a journaling file system. Filesystem redundancy on Linux is lacking in general, in fact (ie Linux has support for software RAID, but only a handful of hardware RAID drivers exist...on a busy server, software RAID is practically useless due to its much slower speed.)
So all of you who are concerned with Sun destroying Linux need not be. Its not going to happen. Sun does not picture itself to be in the low-end of the market (Java was created to sell servers).
They can't speed it up -- principle rule of software development is that software never gets smaller or faster (That software is faster is only the result of faster hardware.)
Your "principle rule" is simply not true. It is in fact possible to speed up software. 2.2.x versions of the Linux kernel are certainly faster than 2.0.x version--and not just because of the multiprocessing capabilities. I'd do a little homework before I made blanket statements like the one you just made.
I don't know of a computer geek that doesn't have a harddrive full of MP3s and an illegal copy of windows.
:-)
:-)
Well, I don't have an illegal copy of Windows (I run Linux and it isn't possible to have an illegal copy of Linux, at least last time I checked
As for the MP3s, I plead the fifth...
Legislators, what were you smoking to pass this law
I dunno, but I'll bet it was even more illegal than those MP3s and illegal "warez".
Well, its only been possible to convict someone of piracy about a year... Prior the new law put on the books last year, piracy was only a civil offense, not a criminal one. (IOW, you could get sued for piracy, but not put in jail over it.)
Who said Merced was x86?
Yes, Merced *WILL* run x86 code, but its "native" 64-bit code is based on HP's PA-RISC architecture, found in many of HPs Unix workstations and servers.
Besides, Linux is still primarily x86. One nice thing about Linux is that it runs well on cheap hardware. Ultimately it won't be the processor that matters, but the software.
I could probably get 1000 hours by hooking the thing to my car's battery... :-)
Besides the obvious (there are lots of libraries that implement win32 on native Linux) there is no source code available for most Windows apps.
This means that you personally cannot recompile your Windows application unless you have written it. And lets not forget that most software companies aren't going to invest the money and energy into "band-aiding" their Windows code to make i9t run on Linux (with the obvious exception of Corel) as it is probably more cost effective in the long run to implement the code the "right" way (written as a native Linux app) since maintaining Windows code compiled for Linux will likely be a bear. Remember that most Windows APIs rely on Windows' poor OS design (ie, too many modules try to do too much...vs. simple one-function programs on Unix/Linux that are really really good at one thing instead of trying to do everything...)
It doesn't play on RealPlayer 5, the only RP available for Linux. :-(
I've always known black holes suck...now there's proof! :-)
I saw that movie. It was pretty decent for the type of movie it was.
But like most movies about hackers its way off the mark. Even the basic premise is at fault: remember Phil Zimmerman and the whole PGP thing? The NSA is still unable to crack anything but the smallest (56-bit) RSA keys, but RSA encryption is now everywhere...
The Jargon File, aka "The New Hacker's Dictionary" currently written/edited by Eric S. Raymond, paints just the picture you mention. Hardly the "liberal media" that you mention.
You know, this story is greatly distorted. The very large software company's OS did not catch on in large numbers at all until 1990 (Version 3.0 of that operating system). The same OS really didn't gain its near-monopoly status until sometime after 1993. (Before about 1992/1993, most computers ran on the very large software companies earlier efforts, which weren't even graphical.)
So the larger companies graphical OS really didn't catch on until *well* after the earlier companies computer became color-capable.
And the realm of computing is split into far more than two camps. You have the Windows camp. The smaller, but still existant OS/2 camp. The Linux/xBSD camp. The commercial Unix camp. The still-very-much-active Amiga contingent. The BeOS camp. Need I go on?
Don't be so short-sighted. There are many, many more viable platforms than the one you happen to find attractive.