Re:Let's stop worrying about 486's
on
KDE Looks Ahead
·
· Score: 1
Why are people only worried about 486s? What about all those 386s out there? The Linux kernel supports it, therefore every app I could wish to run should work on it just as well as any Athlon. :)
Mow lawns, save up $300 and buy yourself a K6/300 or so. Or that Oracle NC thing a few days ago, they said something like $150 for it...
Re:Some comments... (not supposed to be flamebait)
on
KDE Looks Ahead
·
· Score: 1
Kanossa is basically just a short-cut to embedding KDE stuff inside other KDE apps. CORBA for this has proven to be slow, unstable, etc. CORBA still exists for communicating with outside objects.
What you suggest with kpackage is feasable already with Konqueror. Konq essentially just displays a view when you click on a file. If a view is available (ie plain text, html, pics with kview's view etc) it displays it. So it just takes someone writing the view, I don't think it has one currently.
Re:No, it doesn't make sense...
on
KDE Looks Ahead
·
· Score: 1
And not just a file manager...
You can enter URLs for several network protocols like http, ftp, gopher, pop, (eventually) imap, smb, etc You could read email, check news, read slashdot all in one app.:)
About your video reference, I have a little anecdote. One time I picked up a WinTV card to watch a little TV and convert some tapes to Video CDs. I get home, install all the necessary things on Windows (I've never seen any VideoCD burning utils for Linux, if there are any, please hollar). Now this is a cheap card, only records to AVI, not MPEG.
I hook up a TV/VCR, and start recording. I let it go for about 10 minutes and decide to check out the resulting AVI. Win9x comes back with some file corrupted sort of error. I look at the file size, and it's over 2gigs. So I start searching around the 'net and come to find out AVIs can't be any larger than 2gig. It's been a while, but I believe that also applied to NT. So, sure Windows can create them over 2g, but their useless.:)
Then well, write a real GNOME book. Or offer a draft to be included in the next edition of this book or any of the billions of Linux books. If you say you don't know enough about it to write a book/section, then I'd say you're perfect for the job. Learn and write as you go. It'll give a better view for other unknowledgeable users. It'll give a better view for other unknowledgeable users. There's that big gnome manual around. If it's not enough to fill a reasonably sized book, add to that and then talk with O'Reilly. There is much one could do to improve this situation. If you're just too lazy too do it, then well, there's not much that can be done.
I must say, that's one area that I wouldn't mind distributions leaving as is. We don't want to end up where Windows folks are where any yocal can click around, and voila they're a network administrator.
It's one thing for some guy out there to make a GTK/KDE frontend to ipchains or something. But for a distribution to add these things to their list of features is hazardous. It is far better to go through a few extra steps, read some HOWTOs, learn a good bit about Linux, before setting up a firewall. Once you've become knowledgeable about such things, you will know a lot about TCP/IP in general, security, etc. If there's a simple 1-2-3 approach to firewalling, these "administrators" will not know enough to watch CERT advisories, Bugtraq, or their own distribution's security announcements and they end up getting cracked. And of course, in their inexperience, they blame Linux. What good does that do?
I hereby decree that from the moment Slashdot contains an article detailing a new program, new version, etc, be it free software or commercial, software or hardware, every single person viewing said article MUST, under penalty of going to Windows, download, install, purchase, and use for no less than fifteen minutes per week said software. There will be no exceptions no matter what the reason.
There.. now at least there is a reason for people to bitch and moan about every article on here. "Boo hoo, I use Debian so I'm not going to load Redhat." "Waaa, my mommy said C rules and Delphi is for losers." "Oracle is made by Satan and is demonstrative (notice that word!) of the fall of Linux and mankind." "Rob/Hemos/Roblimo/Katz/etc are lamers because I don't find this topic interesting." "My ftp program beats anything else, so you all can go to hell! You go to hell and you die!" Is there any point to any of these people posting other than to whine? If it doesn't apply to you, just move on with life already!
As the others have said the CDs have the upgrade option right there. But if you're like me and don't want to download 650MB before getting started, you can just download a few packages and install them.
I've done this a few times now. I have many RPMs I've created myself so I like to double check that everything is working before moving on to the next package. The last thing I want is to have all daemons on my system to not be working right because RPM moved their config files to.rpmsave. In fact, I'm right now in the midst of upgrading to Mandrake 6.1 (from what was originally Mandrake 5.3). I'll just download those packages that apply to me, install them, make sure they work, lather, rinse, and repeat. It takes a few days with my 56k modem, but it'll be done eventually...
Wow, and I thought the no compiler thing was just for the free version of Solaris I have. With that I can see they want to limit what you can do. They'd prefer you buy the full thing. But no compiler and all with the full version too? That's plain crazy.
And if anyone out there is contimplating the free Solaris, just don't. You can download some packages for gcc/egcs or whatever. But I never could find a proper binutils sort of package, so one major hinderance was compiling libs (no ar commnad). Maybe there is one out there somewhere, but after a couple days of playing with it and searching, my opinion (as many others have said) it's just too slow on common x86 hardware that you use Linux with currently.
What do you expect for $10, a perfect image equal to or better than a $1500 21" monitor? If it were that simple, would monitors that big actually cost a few grand?
As I found from someone else in this forum, here is a link to Netscape's site that explains many things you can put in your ~/.netscape/preferences.js. The problem is it's geared to Windows users, so many things do not work despite it saying it works with all versions.
I have gotten Netscape to replace things in the Help menu with my own links (menu.help.item*), put items in the Guide (toolbar.places.*, and turn on the Guide with setting browser.chrome.useGuideButton to true, it will then replace My Netscape). Then I broke open the netscape bin with the nifty khexedit hex editor, and started searching for the text strings for the buttons and tooltips. You just have to use the same number of letters/spaces for them. Seeing the guy above talking about Xresources, some of these may be settable with that rather than hex editing, I'll have to try it. Then the Search and Shop buttons can be turned off or have their URLs replaced with more useful sites like Slashdot. Now, if only the button images could be replaced as well...
I believe these things are what the customization kit does for Windows folk. Seeing as they have trouble editing text files and all.:)
Well, it doesn't do away with the button altogether, but still nice to put linuxtoday.com in.:)
Chmod -w the preferenses.js and you won't have to worry about Netscape deleting anything. I got tired of Netscape doing that to the little settings I like. And I don't have a need to change settings permanently too much, just on a per-session basis maybe.
But, where do you find such tips? Does Netscape publish every possible configuration item somewhere?
Announcing security holes publicly (via bugtraq or whatever) motivates people to fix the software. Now, if you are stuck on a closed-source system, you have to wait for the company to put together a patch/update. Who knows how long that can take with some companies. Or you can use Open Source software and be assured a fix in hours, or minutes if you fix it yourself.
This is perhaps one of the strongest arguments to use Open Source software on sensitive equipment. You are not stuck disabling a service or just hoping you aren't attacked if the service is important to your business, while waiting for the company to get together a proper fix.
Well la-dee-da. Jacoplane doesn't need XYZ program. What on Earth is Inprise/Borland doing making these programs then?? Aren't they aware that Jacoplane won't be purchasing them?? Time to sell off all the Inprise stock, as obviously they are complete morons doing something without asking Jacoplane first. Sheesh...
Like the subject, I was asking for VCD creating programs for Linux. Sure you can do it in windows, but I don't really want to unless there are none for Linux. So, I asked, are there?
Sure many of us would prefer the free tools that already exist. But there will always be suit-types that do not want to either hire new competent staff to use the free software or retrain their employees to break their dependance on Visual Basic. Some people just will never grasp the idea of C/C++ or whatever. Also many of them don't understand free software, and so are too scared to think about releasing their coveted products with any sort of Open Source license.
With products like Borland's programming tools, Oracle's database, etc all those windows people that can use these products suddenly can be fairly useful in Linux as well. Then as they get their feet wet, some will want to learn more about Linux and we've got ourselves another recruit. World domination is an incremental process, but we have the better technology. We just have to convice people to make that first leap, and making it as easy as possible for those who are scared.
Is there any (commercial or free) programs that can do what's necessary to burn VCDs? I've found some free MPEG1 encoders, but they do only video (not sound+video). And then comes the burning process, which I haven't found a thing...
Or are there VCD players for Linux? Ones I have burned in Windows only work for a track or two. After that they give read errors, so my DVD is the only thing I can use to watch them.
Networking, DNS/DHCP administration, network security, etc are things that should NOT be left to Windows dialog boxes and wizards. The person in charge of these should study, and learn about them before trying to use them. After that is done, compiling and configuring Bind and dhcpd to do these DYNDNS updates is trivial. My original point was that the technology exists for any mildly competent person in charge of DHCP/DNS on a Unix box, despite the PCWeek author's claim that it just does not exist.
For adequate security models, I'll trust Bugtraq and the dozens of other mailing lists/newsgroups far over MS's little bug page which takes 3-4 weeks to acknowledge security problems, and another 3-4 to come out with a workaround like "don't use this option." If a business wants to protect their networks, they MUST hire a competent person to do the job (I'm available if anyone's looking:)), and not rely on the OS manufacturer to secure their systems.
Running network services like these on Windows just doesn't promote the Unix concepts of RTFM. Explaining to my brother the concept of mapping hostnames to an IP and likewise that IP to the hostname, or what an MX record is, was made terribly difficult because of what Microsoft has done.
There is a patch I pulled from the dhcp-server mail list to do it now. Check their archives. I use ISC's dhcpd 2.0b1pl18 with it. Well, it sometimes forgets to delete hosts that are no longer connected. And I haven't checked for any newer versions in like 8 or more months. But overall it works very nicely.
As for no competition, I've enjoyed HalfLife a heck of a lot more than any quake. Just as I finish the single player game and get bored of plain deathmatch online games, they come out with TFC. After a few months of that, I discovered the CounterStrike mod. The fun just never ends.
With TF2 coming out this fall, you can bet what I'll be buying. I tried the Q3 test when it first came out, and I was just underwhelmed. It was just too much of the same old brown graphics, race for the rocket launcher, etc. Of course, if HalfLife had a Linux client it would be the perfect game, but oh well...
There are plenty of quotes around to answer any question these "news" sites have. Let's face it, they don't ask questions that are very intelligent (they should read a FAQ/HOWTO/etc). I'd personally only be able to take a few times before getting pissed at them.:) Linus doesn't cover a lot of what Linux is now about, desktop environments, Apache losing to IIS, and the like. These sites like ZDNet are more targetted to the using end of Linux, not the development. I'm sure many of their questions would be better directed at the appropriate groups.
If there's some big Linux news, like a new kernel release (ie 2.4, 3.0), there will be appropriate press releases for their quoting needs.
Linus, keep up the good work and never mind these wackos.
My only concern is people jumping on this and seeing 100 Linux distributions by Christmas. The more distributions is not necessarily the better. Now survival of the fittest says the best will win out in the end, no so if someone can copy and re-package with new utilities.
Of course if there were 100+ distributions not all of them will survive for long. Users will pick what they like, gives them the features they need/want, etc.
Mandrake started as a Pentium optimized, more customized GUI, KDE added, more internationalized RedHat. It turned out there was a substantial market for just a thing. And Mandrake has done very well. Now with custom programs like the repartitioner, GUI installer, hardware detecting thing (lothar). All of which can put utilized by other distributions, even RedHat. If another group of people see a hidden market, they can tweak Mandrake or whatever and have a shot at stardom. That is exactly what survival of the fittest means. A pure copy, just changing "Mandrake" with "Foodrake" will die because people will see the fraud. But if they add the Foo office suite, translations of all HOWTOs and whatnot into the Fooian language, specially optimized for Foo-based hardware, etc, etc and the Fooites like such a thing, they have a good chance at surviving.
So how can a distribution be commercially viable in this atmosphere. Granted there are the Debians and Slackwares, but it is the commercial distributions that have really added the simple tools and polish that make Linux a little bit more productive, they have to in order to stand out.
A distribution has to have some good points to make it standout and survive, they've got to get a user base and convince us to convert (or try them first) from other distributions. And if they GPL those programs, all the better for everyone, not just those that use their distribution. No one loses. As for profits, I read somewhere that RedHat's income is more from their support contracts and the like, not so much from the boxes people buy at Best Buy. Maybe there IPO filings have more breakdowns on like their last year's income. The boxes at Best Buy and the features of the distribution get the name recognition, so big businesses will spend moolah on the support contracts.
The most important advantage of Linux over many other OSes, is this free software idea. The GPL doesn't discriminate on who you are. It's just free. That's what we need to concentrate on. It's free for everybody, even Bill Gates.
"You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium..." -- GNU General Public License
Most of the programs in Linux distributions are under the GPL or similar licenses. Where in these licenses does it say, ".. except if someone else has already distributed the program?" I write program ABC, I release it under the GPL. Everyone in the world is free to copy it, mirror it on ftps all around the globe, modify it, even put it on a distribution CD. Are these programs only to be distributed by one vendor?
Yet, there is a ton of people that come out for EVERY Mandrake Linux story here, bitching and moaning about how they just "stole" everything from RedHat, and they do not even deserve the right to exist. With Corel's upcoming distribution based on Debian, will this same situation arise? I sure hope not.
We release programs under the GPL so that they are able to be used/copied/etc for everyone. So where on Earth does this hostility for a small group following the copy clause of the GPL have ANY legitimacy?? If you do not want your program to be copied freely, give it a license that states something like, "This program can only be used on Debian systems, and not any stupid copy-cat distributions either, just the genuine Debian one." As author of a program, that's your right. But you have no right to say, "These people should not copy programs, because well, RedHat already did it."
If you come up with conspiracy theories like stockholders demanding RedHat not allow this sort of behavior, then the stockholders should have read RH's IPO statement, should learn about a company and how it gets the programs on its CDs, etc, before investing any money in it. If this homework is done (primarily reading the GPL a few times), you will see the copy clauses and either believe in them and invest your money, or get scared by them and not invest. If RedHat did anything so close as to make a distribution unable to be copied (they couldn't use any GPLed code, so that would require totally reengineering EVERYTHING), they would get hounded so horribly they would have to go out of business as I doubt anyone in this community would ever trust them again. That is a much larger risk for them than shooting down everyone who wants to copy the distribution.
The official people involved, both RedHat and Mandrake, have said they do not mind either side. Bob Young is always asked this, when I saw him at Comdex this Spring and again at LinuxWorld just a few days ago.
What's this mean for the Linux community at large? Yes, you can copy a Mandrake CD and slap a label on it that says, "Slappy Joe's Spiffy Linux," and sell it for $20, $50, even $199.99. But, to win the hearts of anyone, and make a name for yourself you will have to put in work and make it something that is noticable. Copying alone doesn't win you Product of the Year awards
So, shut your pieholes and do something useful with your time.
Slashdot Mirror
Why are people only worried about 486s? What about all those 386s out there? The Linux kernel supports it, therefore every app I could wish to run should work on it just as well as any Athlon.
:)
Mow lawns, save up $300 and buy yourself a K6/300 or so. Or that Oracle NC thing a few days ago, they said something like $150 for it...
Kanossa is basically just a short-cut to embedding KDE stuff inside other KDE apps. CORBA for this has proven to be slow, unstable, etc. CORBA still exists for communicating with outside objects.
What you suggest with kpackage is feasable already with Konqueror. Konq essentially just displays a view when you click on a file. If a view is available (ie plain text, html, pics with kview's view etc) it displays it. So it just takes someone writing the view, I don't think it has one currently.
And not just a file manager...
:)
You can enter URLs for several network protocols like http, ftp, gopher, pop, (eventually) imap, smb, etc You could read email, check news, read slashdot all in one app.
About your video reference, I have a little anecdote. One time I picked up a WinTV card to watch a little TV and convert some tapes to Video CDs. I get home, install all the necessary things on Windows (I've never seen any VideoCD burning utils for Linux, if there are any, please hollar). Now this is a cheap card, only records to AVI, not MPEG.
:)
I hook up a TV/VCR, and start recording. I let it go for about 10 minutes and decide to check out the resulting AVI. Win9x comes back with some file corrupted sort of error. I look at the file size, and it's over 2gigs. So I start searching around the 'net and come to find out AVIs can't be any larger than 2gig. It's been a while, but I believe that also applied to NT. So, sure Windows can create them over 2g, but their useless.
Then well, write a real GNOME book. Or offer a draft to be included in the next edition of this book or any of the billions of Linux books. If you say you don't know enough about it to write a book/section, then I'd say you're perfect for the job. Learn and write as you go. It'll give a better view for other unknowledgeable users. It'll give a better view for other unknowledgeable users. There's that big gnome manual around. If it's not enough to fill a reasonably sized book, add to that and then talk with O'Reilly. There is much one could do to improve this situation. If you're just too lazy too do it, then well, there's not much that can be done.
I must say, that's one area that I wouldn't mind distributions leaving as is. We don't want to end up where Windows folks are where any yocal can click around, and voila they're a network administrator.
It's one thing for some guy out there to make a GTK/KDE frontend to ipchains or something. But for a distribution to add these things to their list of features is hazardous. It is far better to go through a few extra steps, read some HOWTOs, learn a good bit about Linux, before setting up a firewall. Once you've become knowledgeable about such things, you will know a lot about TCP/IP in general, security, etc. If there's a simple 1-2-3 approach to firewalling, these "administrators" will not know enough to watch CERT advisories, Bugtraq, or their own distribution's security announcements and they end up getting cracked. And of course, in their inexperience, they blame Linux. What good does that do?
I hereby decree that from the moment Slashdot contains an article detailing a new program, new version, etc, be it free software or commercial, software or hardware, every single person viewing said article MUST, under penalty of going to Windows, download, install, purchase, and use for no less than fifteen minutes per week said software. There will be no exceptions no matter what the reason.
.. now at least there is a reason for people to bitch and moan about every article on here. "Boo hoo, I use Debian so I'm not going to load Redhat." "Waaa, my mommy said C rules and Delphi is for losers." "Oracle is made by Satan and is demonstrative (notice that word!) of the fall of Linux and mankind." "Rob/Hemos/Roblimo/Katz/etc are lamers because I don't find this topic interesting." "My ftp program beats anything else, so you all can go to hell! You go to hell and you die!" Is there any point to any of these people posting other than to whine? If it doesn't apply to you, just move on with life already!
There
In closing, I'd like to say, "Sheesh."
As the others have said the CDs have the upgrade option right there. But if you're like me and don't want to download 650MB before getting started, you can just download a few packages and install them.
.rpmsave. In fact, I'm right now in the midst of upgrading to Mandrake 6.1 (from what was originally Mandrake 5.3). I'll just download those packages that apply to me, install them, make sure they work, lather, rinse, and repeat. It takes a few days with my 56k modem, but it'll be done eventually...
I've done this a few times now. I have many RPMs I've created myself so I like to double check that everything is working before moving on to the next package. The last thing I want is to have all daemons on my system to not be working right because RPM moved their config files to
Wow, and I thought the no compiler thing was just for the free version of Solaris I have. With that I can see they want to limit what you can do. They'd prefer you buy the full thing. But no compiler and all with the full version too? That's plain crazy.
And if anyone out there is contimplating the free Solaris, just don't. You can download some packages for gcc/egcs or whatever. But I never could find a proper binutils sort of package, so one major hinderance was compiling libs (no ar commnad). Maybe there is one out there somewhere, but after a couple days of playing with it and searching, my opinion (as many others have said) it's just too slow on common x86 hardware that you use Linux with currently.
What do you expect for $10, a perfect image equal to or better than a $1500 21" monitor? If it were that simple, would monitors that big actually cost a few grand?
As I found from someone else in this forum, here is a link to Netscape's site that explains many things you can put in your ~/.netscape/preferences.js. The problem is it's geared to Windows users, so many things do not work despite it saying it works with all versions.
:)
I have gotten Netscape to replace things in the Help menu with my own links (menu.help.item*), put items in the Guide (toolbar.places.*, and turn on the Guide with setting browser.chrome.useGuideButton to true, it will then replace My Netscape). Then I broke open the netscape bin with the nifty khexedit hex editor, and started searching for the text strings for the buttons and tooltips. You just have to use the same number of letters/spaces for them. Seeing the guy above talking about Xresources, some of these may be settable with that rather than hex editing, I'll have to try it. Then the Search and Shop buttons can be turned off or have their URLs replaced with more useful sites like Slashdot. Now, if only the button images could be replaced as well...
I believe these things are what the customization kit does for Windows folk. Seeing as they have trouble editing text files and all.
Well, it doesn't do away with the button altogether, but still nice to put linuxtoday.com in. :)
Chmod -w the preferenses.js and you won't have to worry about Netscape deleting anything. I got tired of Netscape doing that to the little settings I like. And I don't have a need to change settings permanently too much, just on a per-session basis maybe.
But, where do you find such tips? Does Netscape publish every possible configuration item somewhere?
On top of what you said...
Announcing security holes publicly (via bugtraq or whatever) motivates people to fix the software. Now, if you are stuck on a closed-source system, you have to wait for the company to put together a patch/update. Who knows how long that can take with some companies. Or you can use Open Source software and be assured a fix in hours, or minutes if you fix it yourself.
This is perhaps one of the strongest arguments to use Open Source software on sensitive equipment. You are not stuck disabling a service or just hoping you aren't attacked if the service is important to your business, while waiting for the company to get together a proper fix.
Well la-dee-da. Jacoplane doesn't need XYZ program. What on Earth is Inprise/Borland doing making these programs then?? Aren't they aware that Jacoplane won't be purchasing them?? Time to sell off all the Inprise stock, as obviously they are complete morons doing something without asking Jacoplane first. Sheesh...
Like the subject, I was asking for VCD creating programs for Linux. Sure you can do it in windows, but I don't really want to unless there are none for Linux. So, I asked, are there?
Sure many of us would prefer the free tools that already exist. But there will always be suit-types that do not want to either hire new competent staff to use the free software or retrain their employees to break their dependance on Visual Basic. Some people just will never grasp the idea of C/C++ or whatever. Also many of them don't understand free software, and so are too scared to think about releasing their coveted products with any sort of Open Source license.
With products like Borland's programming tools, Oracle's database, etc all those windows people that can use these products suddenly can be fairly useful in Linux as well. Then as they get their feet wet, some will want to learn more about Linux and we've got ourselves another recruit. World domination is an incremental process, but we have the better technology. We just have to convice people to make that first leap, and making it as easy as possible for those who are scared.
Is there any (commercial or free) programs that can do what's necessary to burn VCDs? I've found some free MPEG1 encoders, but they do only video (not sound+video). And then comes the burning process, which I haven't found a thing...
Or are there VCD players for Linux? Ones I have burned in Windows only work for a track or two. After that they give read errors, so my DVD is the only thing I can use to watch them.
nah, he gets glasses. That makes people look totally different, like the SNL opening monologue when Terry Hatcher was hosting.
A Windows interface does not a LAN Admin make.
:)), and not rely on the OS manufacturer to secure their systems.
Networking, DNS/DHCP administration, network security, etc are things that should NOT be left to Windows dialog boxes and wizards. The person in charge of these should study, and learn about them before trying to use them. After that is done, compiling and configuring Bind and dhcpd to do these DYNDNS updates is trivial. My original point was that the technology exists for any mildly competent person in charge of DHCP/DNS on a Unix box, despite the PCWeek author's claim that it just does not exist.
For adequate security models, I'll trust Bugtraq and the dozens of other mailing lists/newsgroups far over MS's little bug page which takes 3-4 weeks to acknowledge security problems, and another 3-4 to come out with a workaround like "don't use this option." If a business wants to protect their networks, they MUST hire a competent person to do the job (I'm available if anyone's looking
Running network services like these on Windows just doesn't promote the Unix concepts of RTFM. Explaining to my brother the concept of mapping hostnames to an IP and likewise that IP to the hostname, or what an MX record is, was made terribly difficult because of what Microsoft has done.
There is a patch I pulled from the dhcp-server mail list to do it now. Check their archives. I use ISC's dhcpd 2.0b1pl18 with it. Well, it sometimes forgets to delete hosts that are no longer connected. And I haven't checked for any newer versions in like 8 or more months. But overall it works very nicely.
As for no competition, I've enjoyed HalfLife a heck of a lot more than any quake. Just as I finish the single player game and get bored of plain deathmatch online games, they come out with TFC. After a few months of that, I discovered the CounterStrike mod. The fun just never ends.
With TF2 coming out this fall, you can bet what I'll be buying. I tried the Q3 test when it first came out, and I was just underwhelmed. It was just too much of the same old brown graphics, race for the rocket launcher, etc. Of course, if HalfLife had a Linux client it would be the perfect game, but oh well...
There are plenty of quotes around to answer any question these "news" sites have. Let's face it, they don't ask questions that are very intelligent (they should read a FAQ/HOWTO/etc). I'd personally only be able to take a few times before getting pissed at them. :) Linus doesn't cover a lot of what Linux is now about, desktop environments, Apache losing to IIS, and the like. These sites like ZDNet are more targetted to the using end of Linux, not the development. I'm sure many of their questions would be better directed at the appropriate groups.
If there's some big Linux news, like a new kernel release (ie 2.4, 3.0), there will be appropriate press releases for their quoting needs.
Linus, keep up the good work and never mind these wackos.
My only concern is people jumping on this and
seeing 100 Linux distributions by Christmas. The more distributions is not necessarily the better. Now survival of the fittest says the best will win out in the end, no so if someone can copy and re-package with new utilities.
Of course if there were 100+ distributions not all of them will survive for long. Users will pick what they like, gives them the features they need/want, etc.
Mandrake started as a Pentium optimized, more customized GUI, KDE added, more internationalized RedHat. It turned out there was a substantial market for just a thing. And Mandrake has done very well. Now with custom programs like the repartitioner, GUI installer, hardware detecting thing (lothar). All of which can put utilized by other distributions, even RedHat. If another group of people see a hidden market, they can tweak Mandrake or whatever and have a shot at stardom. That is exactly what survival of the fittest means. A pure copy, just changing "Mandrake" with "Foodrake" will die because people will see the fraud. But if they add the Foo office suite, translations of all HOWTOs and whatnot into the Fooian language, specially optimized for Foo-based hardware, etc, etc and the Fooites like such a thing, they have a good chance at surviving.
So how can a distribution be commercially viable in this atmosphere. Granted there are the Debians and Slackwares, but it is the commercial distributions that have really added the simple tools and polish that make Linux a little bit more productive, they have to in order to stand out.
A distribution has to have some good points to make it standout and survive, they've got to get a user base and convince us to convert (or try them first) from other distributions. And if they GPL those programs, all the better for everyone, not just those that use their distribution. No one loses. As for profits, I read somewhere that RedHat's income is more from their support contracts and the like, not so much from the boxes people buy at Best Buy. Maybe there IPO filings have more breakdowns on like their last year's income. The boxes at Best Buy and the features of the distribution get the name recognition, so big businesses will spend moolah on the support contracts.
The most important advantage of Linux over many other OSes, is this free software idea. The GPL doesn't discriminate on who you are. It's just free. That's what we need to concentrate on. It's free for everybody, even Bill Gates.
well, some of you :).
"You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium..." -- GNU General Public License
Most of the programs in Linux distributions are under the GPL or similar licenses. Where in these licenses does it say, ".. except if someone else has already distributed the program?" I write program ABC, I release it under the GPL. Everyone in the world is free to copy it, mirror it on ftps all around the globe, modify it, even put it on a distribution CD. Are these programs only to be distributed by one vendor?
Yet, there is a ton of people that come out for EVERY Mandrake Linux story here, bitching and moaning about how they just "stole" everything from RedHat, and they do not even deserve the right to exist. With Corel's upcoming distribution based on Debian, will this same situation arise? I sure hope not.
We release programs under the GPL so that they are able to be used/copied/etc for everyone. So where on Earth does this hostility for a small group following the copy clause of the GPL have ANY legitimacy?? If you do not want your program to be copied freely, give it a license that states something like, "This program can only be used on Debian systems, and not any stupid copy-cat distributions either, just the genuine Debian one." As author of a program, that's your right. But you have no right to say, "These people should not copy programs, because well, RedHat already did it."
If you come up with conspiracy theories like stockholders demanding RedHat not allow this sort of behavior, then the stockholders should have read RH's IPO statement, should learn about a company and how it gets the programs on its CDs, etc, before investing any money in it. If this homework is done (primarily reading the GPL a few times), you will see the copy clauses and either believe in them and invest your money, or get scared by them and not invest. If RedHat did anything so close as to make a distribution unable to be copied (they couldn't use any GPLed code, so that would require totally reengineering EVERYTHING), they would get hounded so horribly they would have to go out of business as I doubt anyone in this community would ever trust them again. That is a much larger risk for them than shooting down everyone who wants to copy the distribution.
The official people involved, both RedHat and Mandrake, have said they do not mind either side. Bob Young is always asked this, when I saw him at Comdex this Spring and again at LinuxWorld just a few days ago.
What's this mean for the Linux community at large? Yes, you can copy a Mandrake CD and slap a label on it that says, "Slappy Joe's Spiffy Linux," and sell it for $20, $50, even $199.99. But, to win the hearts of anyone, and make a name for yourself you will have to put in work and make it something that is noticable. Copying alone doesn't win you Product of the Year awards
So, shut your pieholes and do something useful with your time.