Linux kernel version 2.4.mumble has problems and people ask "how can this be releasable". It's not stable!
gcc 3.0 is unstable for the first few releases and folks start whining about how this isn't a production-quality compiler
AbiWord says that they're not providing commercial-grade support services and everyone gets honked off and claims that open source software can't work
Can you all just take a step back, breathe deeply and remind yourselves that in any software organization with more than 10 developers there are two versions of the software (at least):
The development snapshot (or mainline, depending one your local terminology). This is a stable release from the developers to inernal customers such as Q/A, release engineering and perhaps alpha testers for integration testing and embeded product testing.
The release. This is the ready-for-prime-time code that will be supported and maintained by the company.
Are you seeing the parallel here? When Linus releases kernel 2.4.57, he's releaseing a snapshot that lets Q/A (made up of Q/A groups in numerous companies that sell Linux-based products) release engineering (the distribution vendors) and alpha integration testers (embedded systems customers) begin their test and release cycle. Same for AbiWord. Ximian, Red Hat and many others release AbiWord, but I doubt that they ever release it absolutely as shipped. Their Q/A process only begins when AbiSource creates a new version.
So, here's the question of the day: why are people shocked when the developers start acting like developers and say "we're not going to hand-hold you"? Well, there's a few reasons. Obviously there are the folks who just wait for an opportunity to slam OSS. Then there are the people who have become confused and don't realize that the Mozilla developers or the AbiWord developers are just that: developers. Then there are the folks who get their priorities confused. They say that they don't want to deal with "big business software", so they go it alone. This is all well and good, but when you do this, you have to expect the other shoe to drop.
If you're downloading gcc 3.0 the day it comes out because you want the new features fast, great! But, don't be shocked when your code fails to work correctly because you have a hardware combination that was not well tested. If you'd waited for Red Hat 7.2, you would have found the optional gcc 3.0.x binaries with a big old wad of patches. Why? Because they tested it, patched it, and released it.
Get over it. Software support is hard, and there are people in the OSS world that do it well. But, to expect every project to come out the gate with good Q/A and support is just silly.
Larry Wall once said something of the Perl community to the effect of, "My beta test community is either too large or too small, depending on how you look at it." I think this is kind of what Verant ran into with EQ. Personally, I think they should have had a wide-open beta period with anyone who wanted to create a character playing. Yes, this takes some of the suprise out of it, but when you're talking software, suprise is often bad. It would have created more buzz and pushed people who had enjoyed the beta to buy the game and feel that they were involved in it.
The earth is warmer now than it was 100 years ago. And significantly cooler than it was 900 years ago which was warmer than 100 years before that. etc. etc. ad nauseum.
Yep... In fact the earth lost quite a few species and a number of human cultures died out between 300 and 800AD (not sure of exact dates, I'd have to look that up) because of a nasty period of global warming. Of course, that was because of all the CFCs and greenhouse gasses the early Christians were emitting from their land rovers;-)
My terminology was fuzzy. This is a passable "realese" in the sense that they released on time with most features and are fixing the bugs fast. All this, not to mention the fact that there's some pretty cool stuff in this release!
This was a disaster of near epic proportions for the players because of the nature of the game. This is a great example of the dangers of subscriber-ware, and an even better example of the way our paradigm is changing in the gaming world. Once, the phrase "I have version 2.3" meant something. Soon though, I just don't see that it'll be meaningful at all, at least in the gaming world.
Since I submitted this, there have been several developments:
It's becoming obvious that a fair number of Win95 users are pissed that the game they bought and ran under Win95 has suddenly stopped working. Note: this has nothing to do with the fact that the expansion does not run under Win95, but that old, installed versions of EQ were updated (via the patcher) with a game that refuses to run under 95!
Verant has been BUSY! There is now a patch for some of the larger problems that people were seeing. I was personally bit by the auto-follow, crash-to-desktop bug several times:-(
This is not a bad release, really. It's just pointing out a lot of the problems that games now face when everyone gets updated/patched at once. The artificial lines between people running "Luclin" and people running "the old EverQuest" are very thin....
I find it interesting that this would be modded up.
"This self-important bullshit ought to stop"
You are making an assumption about the purpose of Debian. You assume that it's about providing you with a new release. It is not. Debian as a project is about producing a free software operating system. If 1/2 of that definition is not met by 2/3 of the documentation, then it should be of major concern (to at least 1/3 of the team;)
It's not a big deal, get back to hacking code
And what code do you hack? I'm getting rather tired of self-important Slashdot posters who feel that these slackers should go back in the kitchen and bake some pie. We, the coders of various open source and/or free software applications write the code for our own reasons. If you don't like the code or don't feel that it's up to your standards/schedules, then don't use it. We'll be just as happy either way.
I would have a lot more sympathy for your comments if you spent any time acknowledging that these folks have provided you with an awful lot of benefit because they're fanatics who will waste hours/days/years of their lives for the good of the free software cause.
Or powerful text editors? In Windows, you have UltraEdit32, among many others, while in UNIX you need to use either vi or emacs to reach the same editing power, both of which are far harder to use than UltraEdit32 is (albeit even more powerful).
I tried to find UltraEdit32 on my Windows box... it seems to be missing, so I can't comment. On my virgin Linux box (Red Hat 7.2), I have:
vi (vim by default and nvi if you opt for it)
emacs (GNU emacs by default and Xemacs if you opt for it)
gedit
ed (the standard editor;-)
xedit
Those are the ones I know about.
Emacs is hard to use if you're used to using Windows editors (which, suprise!, are hard to use if you're used to using Emacs).
Vim (and vi in general) can be a bit of a pain to learn for anyone (it's the modal editing thing), but there's a lot of flexibility in its paradigm once you come to understand it (I've yet to see an easy way to do the same as "ct;return" in another editor without a lot more convolution.
If you want simplicity, you're looking for Gedit (and I presume KDE has a "kedit"). This is your quick-and-dirty notepad-alike.
Now, I'm sure there are others (for example, joe, jove, pico, and all of the free editors on freshmeat (67 a last count)), but you first have to define what it is you're trying to do.
For example, when many people speak of text editing, they really mean code editing. There are a number of good IDEs out there (assuming Emacs is too difficult for you, which is saying a lot for starters). Check out the list on freshmeat, including Komodo, which is a very slick IDE built on top of the Mozilla platform.
In the realm of editors, about the only thing you can accuse UNIX and Linux of these days is requiring you to do a fair amount of research, since there are so many options. Personally, I see this as a Good Thing(tm).
Um.. you're missing the point. The goal of creating a usable desktop under Linux (and by the extension of clean porting, other UNIX and UNIX-like systems) is to begin taking over the desktop market the same way Linux has been doing in the server market. There's no reason to run Windows if Linux can do everything that Windows can. Right now, I see the following impediments to that:
1. Office file formats. The DoJ needs to make this call. They need to force MS to release specifications for their file formats as part of the settlement. Though, I will note that conversion by programs like AbiWord and Gnumeric are impressive.
2. Games. Linux needs to be more game friendly, including some standards on what a distribution needs to have and how they need to configure to support them. If running EverQuest: Shadow of Luclin were as easy as putting the CD in the drive, I know about 20 people that would never need to run Windows again;-) 3. PR. Red Hat has been doing a good job, but IBM has only been preaching to the choir. We need good Linux PR.
Here's why you don't care about being able to intellegently identify spam:
1. I get a fair amount of real mail from friends of the form "check this out: http://x.y.z/"... I also get a LOT of spam that looks exactly the same.
2. You can catch spam the moment it comes out, by having honeypot-like mailboxes which are for no users at all, but you submit thier addresses to various places spammers look for such addresses.
Given these two, spam filters that don't look at real spam constantly are just hobbling themselves. Perhaps you intelligent filter should start off by running razor-check, and then thinking real hard if Razor says it's not spam...?
No, there are a great many other documents, objects and other "items" which are owned by the FSF and are not covered by the GPL.
The GPL is a tool which was created with one goal: to allow modification and distribution of software. The goal was not (even given the FSF's fondness for recursion) to allow modification of the GPL.
You're reading GNU's Free Documentation License (as you note). This license is intended to apply to user manuals, technical references and such. If you stop to think about the goal of Free Software (I'm not going into OS vs FS here), it's to make sure that Stallman can get the source to his printer drivers, modify them and then give them out to others (imagine a world where that printer had come with source under a BSD-like license!)
So, with that goal in mind, how would you construct a license that is both modifiable under the terms of the GFDL (which you quote) and still accomplishes the stated goal? The GPL can be used as a guide in creating your own license. This has certainly been done often enough. But, to modify the license itself would hurt the aim of Free Software.
I'm also not certain what the legal implications are if a license agreement affords me the right to modify it. The GPL has teeth that come directly from copyright law. Under copyright law, you are not allowed to modify or distribute the code except in accordance with fair use doctrin. The GPL acknowledges this fact, and then offers you an "out" in the form of a license (this is in direct contrast to EULAs and other "shrink-wrap licenses", which require you to accept the license before USING the software) which you can take or leave as you see fit.
Now, if you were allowed to modify the license, your software would have to refer to some "license archetype", perhaps backing that usage up using trademark (e.g. you can modify the GPL, but only if you give it your own name). This is sticky, and keep in mind that the GPL was a daring bit of legal hackery that has still yet to be tested in court. To add yet another complication to the core oddity of defending right-to-modify with copyright law would risk the basic goal by making the GPL harder to defend than it already was.
All that asside, I think it's of questionable value to refer to the restrictions on the GPL as hypocritical. The GPL is a software license, not a work of art or engineering. I'm not quite certain why you feel that it would be hypocritical to say that software is an area of human endevor where freedom to modify is important but contracts and licenses are not. You may disagree, and you are most welcome to. But even if I accept that the two should be treated the same (and I do not, obviously), you make a challenge of hypocricy here which I do not believe you have explained.
I'm getting sick of this constant stream of freshmeat-like announcements of Linux-specific junk. You know there's more in the world than just Li... oh, you said OpenBSD!;-)
-Aaron, who has seen too many serious posts that began with similar statements
You've never seen the support costs for an IBM mainframe, have you? Suffice to say that our customers are quite happy with the cost differential, but they do it more so because of the access to modern technology in terms of software, networking and support.
AIX will still run on clients high-end mission critical servers. Linux will run on small to midrange servers.
Absolutely. I can assure you that you are correct (for the most part) because I've been working in a company whose job it is to take behemoth companies running ancient software on IBM mainframes and convert their production operations over to our software on farms of small Intel/Linux boxen.
I don't see anyone running Linux on huge hardware. They're all converting to farms of tiny, redundant, replacable, damn-near-disposable servers. Of course, in our case tiny means dual-processor boxes with 4GB of RAM, but for our customers that is tiny.
So, as with the mainframe vs. mini and mini vs. midrange arguments of yesteryear, you will begin to see a new trend in corporate computing: the micro-farm. It will not simply replace larger systems. It will become the standard by which the next stage of hardware downsizing (remember when that didn't mean layoffs?) will be judged.
In the long haul, however, I'd feel more comfortable if there were something open, free and distributed that accomplished the same thing. Just in case any of those good testers at RH, SuSE, Mandrake, Caldera, Debian [...]
Umm... since when is Debian not open, free and distributed? Did they get bought by IBM or something?!
Sure, what you propose would be a great adjuct to what Debian does now (perhaps they already do, I'm not much of a Debian guy 'cause I've never really had the time).
What's the best way to tell which kernel is best? Run it for about 2 months on a wide variety of hardware, with a wide variety of software loads. Record incidents and map those against known problems, apply available patches for those that will impact you the most. Re-test.
Then again, your distribution vendor already does this, so why would you be grabbing the latest development release (don't let the term "stable" fool you, that refers to interfaces, not field performance)? Red Hat is now up to 2.4.9. I know that there's a lot of work going on in the VM world, and it seems to have been sorted out, but as you are noticing, there are other things in the kernel besides VM. If you want a kernel whose performance charactaristics are known, and whose primary bugs have been addressed, you have to sacrifice bleeding-edge fixes.
Not an easy pill for the "I want my tarball now!" world of Open Source, is it? Look on the bright side, 2.4.9 updates from Red Hat on 11/2 beats the heck out of the too-little-too-late geological updates from any closed-source proprietary OS vendor. Q/A is hard work and cannot happen in zero-time.
A lot of folks here have been hammering on RHAT for proposing what they did. I'm frankly stunned. No school is forced to install Red Hat Linux, but if they're going to get a free copy of the software (saves them the download) and free RHN access (saves them the constant checking for downloads), this would be pretty sweet for the schools. They could certainly opt out, and buy a competing OS.
The key thing is that in most schools, no one would go out and install Red Hat on all of their systems for fear of the backlash from "concerned parents" who see this as teaching the students with second-best (e.g. cheaper) tools in order to save. No one is going to question the installation of Red Hat, if it's being provided to the schools as part of a major settlement. It would, after all, be a substantial waste of resources to turn it down;-)
I really hope that something like this happens, but I don't know what, besides issuing a press release, RHAT is doing to move this idea forward.
Re:Mozilla (slightly OT too)
on
GTK-- vs. QT
·
· Score: 2
I've played with Komodo, and I have to say that I'm impressed. XUL+Moizilla can certainly be turned to one's advantage. I imagine that the crucial point is where you put the dividing line between your app and the UI. I would think that you'd want to build only the highest level UI pieces in XUL, and then the rest of your app in a low level language.
Re:Mozilla (slightly OT too)
on
GTK-- vs. QT
·
· Score: 3, Interesting
1. Have you used anything built on top of Mozilla (e.g. Komodo) or are you just flaming because it feels good?
2. If you actually needed a UI that was as rich and powerful as XUL can provide, why woulld you want to go off an write it al yourself?
3. Are you thinking in terms of writing web pages and CGI that acts as an application, or are you thinking in terms of building your UI into the XUL interface of Mozilla? These are very different things (and both actually have a place in different applications).
Re:i'm going to suffer for this but...
on
KDE Wins 3 awards
·
· Score: 2
You're correct. And iptables is the packet filtering configuration (aka firewalling rules).
But the effect of the command itself was not the most interesting part. It's the fact that no two of those programs (well, except for the iptables-save and -restore) were designed to work with eachother. They simply do, because the system is designed to work that way. The power in that command -- the part that your average Windows user cannot grasp -- is the power to fully administer the state of a remote system without ever establishing an interactive session on it.
Windows takes the approch that many graphics oriented systems take (e.g. MacOS, even version 10, which is UNIX-based): it provides you a way to do the things that the people who designed the system felt would be most useful. UNIX takes the approach that the designers know what sorts of individual actions may be required, but often not the end-goals of any combination of them.
Is KDE (or Gnome for that matter) "ready for prime time"? Depends on what you need. It's not as polished as Windows or MacOS. However, building on top of what UNIX and/or Linux has to offer leads to a system that is overall more powerful and more flexible. That lack of polish seems a lot less important when you find that, at 6AM you can log into your desktop from work, print a report on your boss' desk and still have time for breakfast because you didn't have to wait for a graphical program to come up over a remote connection....
Re:i'm going to suffer for this but...
on
KDE Wins 3 awards
·
· Score: 2
Why would you suffer for making coherent, rational points? I think your focus is skewed (as most Windows users' foci have been), but you make some good points.
Linux (and all of its ancestors great and mini) has a set of core features that Windows needs, but will almost certainly never get. These features (a command line that's actually useful; functional concurrent multi-user support; etc) lend power to the snappy higher levels that are put on top of them. Yes, KDE is still klunkier than Windows (as is Gnome, which is what I use). But, KDE provides access to all of the underlying power of Linux.
Try to get a Windows user to understand the concept of the init run-levels, and they'll be lost. This is pure, unadulterated power over your environment in ways that a GUI doesn't address. Now try to explain what "iptables-save | ssh -l jonesy host2 sudo iptables-restore" would do! Good luck;-)
You have to look at what she says in context with what the RIAA does. She says she wants to improve access to smaller artists. The RIAA has worked hard to make sure this never happens. But, let me be clear: they're all for access to smaller artists as long as they continue to be able to manufacture "#1" artists at will. They want to keep the cake they have while eating it.
She also says that the people writing such things as Gnutella don't understand that they have the choice to make money or not on software, but music is just "stolen" (infringed to the rest of us). Of course this ignores the decades of warez precident and the BCA's role. This is a totally hollow argument. We write software. We sell it. We get paid. Some poeple will never be willing to pay. We know. None of that means a damn when Microsoft starts alienating their own customers with tactics like the licensing of XP. Even good, faithful customers look for an out in another product. The RIAA has the same problem.
She comments that she's excited about the possibilities of P2P. Heh, even in the client-server model of digital music, the RIAA freaked out when artists started putting their own music up for download (members did, that is).
Bottom line: read my lips, music sharing will happen. Movie sharing will happen. People will continue to share what they believe (rightly or wrongly) to be theirs. What the RIAA should be doing is coming up with a better way to take advantage of that momentum. Create a better mousetrap and the world will beat a path to your door. The corollary to that is that if you just stand around yelling at the manufacturers of poor mousetraps, you eventually get ignored.
Slashdot Mirror
- Linux kernel version 2.4.mumble has problems and people ask "how can this be releasable". It's not stable!
- gcc 3.0 is unstable for the first few releases and folks start whining about how this isn't a production-quality compiler
- AbiWord says that they're not providing commercial-grade support services and everyone gets honked off and claims that open source software can't work
Can you all just take a step back, breathe deeply and remind yourselves that in any software organization with more than 10 developers there are two versions of the software (at least):- The development snapshot (or mainline, depending one your local terminology). This is a stable release from the developers to inernal customers such as Q/A, release engineering and perhaps alpha testers for integration testing and embeded product testing.
- The release. This is the ready-for-prime-time code that will be supported and maintained by the company.
Are you seeing the parallel here? When Linus releases kernel 2.4.57, he's releaseing a snapshot that lets Q/A (made up of Q/A groups in numerous companies that sell Linux-based products) release engineering (the distribution vendors) and alpha integration testers (embedded systems customers) begin their test and release cycle. Same for AbiWord. Ximian, Red Hat and many others release AbiWord, but I doubt that they ever release it absolutely as shipped. Their Q/A process only begins when AbiSource creates a new version.So, here's the question of the day: why are people shocked when the developers start acting like developers and say "we're not going to hand-hold you"? Well, there's a few reasons. Obviously there are the folks who just wait for an opportunity to slam OSS. Then there are the people who have become confused and don't realize that the Mozilla developers or the AbiWord developers are just that: developers. Then there are the folks who get their priorities confused. They say that they don't want to deal with "big business software", so they go it alone. This is all well and good, but when you do this, you have to expect the other shoe to drop.
If you're downloading gcc 3.0 the day it comes out because you want the new features fast, great! But, don't be shocked when your code fails to work correctly because you have a hardware combination that was not well tested. If you'd waited for Red Hat 7.2, you would have found the optional gcc 3.0.x binaries with a big old wad of patches. Why? Because they tested it, patched it, and released it.
Get over it. Software support is hard, and there are people in the OSS world that do it well. But, to expect every project to come out the gate with good Q/A and support is just silly.
- Offtopic=1, Flamebait=1, Troll=1, Insightful=8, Interesting=1, Overrated=2, Underrated=1, Total=15
Interesting...Larry Wall once said something of the Perl community to the effect of, "My beta test community is either too large or too small, depending on how you look at it." I think this is kind of what Verant ran into with EQ. Personally, I think they should have had a wide-open beta period with anyone who wanted to create a character playing. Yes, this takes some of the suprise out of it, but when you're talking software, suprise is often bad. It would have created more buzz and pushed people who had enjoyed the beta to buy the game and feel that they were involved in it.
Your points are well taken.
My terminology was fuzzy. This is a passable "realese" in the sense that they released on time with most features and are fixing the bugs fast. All this, not to mention the fact that there's some pretty cool stuff in this release!
This was a disaster of near epic proportions for the players because of the nature of the game. This is a great example of the dangers of subscriber-ware, and an even better example of the way our paradigm is changing in the gaming world. Once, the phrase "I have version 2.3" meant something. Soon though, I just don't see that it'll be meaningful at all, at least in the gaming world.
- It's becoming obvious that a fair number of Win95 users are pissed that the game they bought and ran under Win95 has suddenly stopped working. Note: this has nothing to do with the fact that the expansion does not run under Win95, but that old, installed versions of EQ were updated (via the patcher) with a game that refuses to run under 95!
- Verant has been BUSY! There is now a patch for some of the larger problems that people were seeing. I was personally bit by the auto-follow, crash-to-desktop bug several times
:-(
This is not a bad release, really. It's just pointing out a lot of the problems that games now face when everyone gets updated/patched at once. The artificial lines between people running "Luclin" and people running "the old EverQuest" are very thin....I find it interesting that this would be modded up.
;)
"This self-important bullshit ought to stop"
You are making an assumption about the purpose of Debian. You assume that it's about providing you with a new release. It is not. Debian as a project is about producing a free software operating system. If 1/2 of that definition is not met by 2/3 of the documentation, then it should be of major concern (to at least 1/3 of the team
It's not a big deal, get back to hacking code
And what code do you hack? I'm getting rather tired of self-important Slashdot posters who feel that these slackers should go back in the kitchen and bake some pie. We, the coders of various open source and/or free software applications write the code for our own reasons. If you don't like the code or don't feel that it's up to your standards/schedules, then don't use it. We'll be just as happy either way.
I would have a lot more sympathy for your comments if you spent any time acknowledging that these folks have provided you with an awful lot of benefit because they're fanatics who will waste hours/days/years of their lives for the good of the free software cause.
In other news, the "This Is Not an Argument" sketch from Monty Python's Flying Circus appears on slashdot.
If you wish to refute a point, please do so without resorting to hypothetical hyperbole.
I tried to find UltraEdit32 on my Windows box... it seems to be missing, so I can't comment. On my virgin Linux box (Red Hat 7.2), I have
- vi (vim by default and nvi if you opt for it)
- emacs (GNU emacs by default and Xemacs if you opt for it)
- gedit
- ed (the standard editor
;-) - xedit
Those are the ones I know about.Emacs is hard to use if you're used to using Windows editors (which, suprise!, are hard to use if you're used to using Emacs).
Vim (and vi in general) can be a bit of a pain to learn for anyone (it's the modal editing thing), but there's a lot of flexibility in its paradigm once you come to understand it (I've yet to see an easy way to do the same as "ct;return" in another editor without a lot more convolution.
If you want simplicity, you're looking for Gedit (and I presume KDE has a "kedit"). This is your quick-and-dirty notepad-alike.
Now, I'm sure there are others (for example, joe, jove, pico, and all of the free editors on freshmeat (67 a last count)), but you first have to define what it is you're trying to do.
For example, when many people speak of text editing, they really mean code editing. There are a number of good IDEs out there (assuming Emacs is too difficult for you, which is saying a lot for starters). Check out the list on freshmeat, including Komodo, which is a very slick IDE built on top of the Mozilla platform.
In the realm of editors, about the only thing you can accuse UNIX and Linux of these days is requiring you to do a fair amount of research, since there are so many options. Personally, I see this as a Good Thing(tm).
So, you were saying?
" Not until it runs on Windows and Mac... "
;-)
Um.. you're missing the point. The goal of creating a usable desktop under Linux (and by the extension of clean porting, other UNIX and UNIX-like systems) is to begin taking over the desktop market the same way Linux has been doing in the server market. There's no reason to run Windows if Linux can do everything that Windows can. Right now, I see the following impediments to that:
1. Office file formats. The DoJ needs to make this call. They need to force MS to release specifications for their file formats as part of the settlement. Though, I will note that conversion by programs like AbiWord and Gnumeric are impressive.
2. Games. Linux needs to be more game friendly, including some standards on what a distribution needs to have and how they need to configure to support them. If running EverQuest: Shadow of Luclin were as easy as putting the CD in the drive, I know about 20 people that would never need to run Windows again
3. PR. Red Hat has been doing a good job, but IBM has only been preaching to the choir. We need good Linux PR.
Here's why you don't care about being able to intellegently identify spam:
1. I get a fair amount of real mail from friends of the form "check this out: http://x.y.z/"... I also get a LOT of spam that looks exactly the same.
2. You can catch spam the moment it comes out, by having honeypot-like mailboxes which are for no users at all, but you submit thier addresses to various places spammers look for such addresses.
Given these two, spam filters that don't look at real spam constantly are just hobbling themselves. Perhaps you intelligent filter should start off by running razor-check, and then thinking real hard if Razor says it's not spam...?
No, there are a great many other documents, objects and other "items" which are owned by the FSF and are not covered by the GPL.
The GPL is a tool which was created with one goal: to allow modification and distribution of software. The goal was not (even given the FSF's fondness for recursion) to allow modification of the GPL.
You're reading GNU's Free Documentation License (as you note). This license is intended to apply to user manuals, technical references and such. If you stop to think about the goal of Free Software (I'm not going into OS vs FS here), it's to make sure that Stallman can get the source to his printer drivers, modify them and then give them out to others (imagine a world where that printer had come with source under a BSD-like license!)
So, with that goal in mind, how would you construct a license that is both modifiable under the terms of the GFDL (which you quote) and still accomplishes the stated goal? The GPL can be used as a guide in creating your own license. This has certainly been done often enough. But, to modify the license itself would hurt the aim of Free Software.
I'm also not certain what the legal implications are if a license agreement affords me the right to modify it. The GPL has teeth that come directly from copyright law. Under copyright law, you are not allowed to modify or distribute the code except in accordance with fair use doctrin. The GPL acknowledges this fact, and then offers you an "out" in the form of a license (this is in direct contrast to EULAs and other "shrink-wrap licenses", which require you to accept the license before USING the software) which you can take or leave as you see fit.
Now, if you were allowed to modify the license, your software would have to refer to some "license archetype", perhaps backing that usage up using trademark (e.g. you can modify the GPL, but only if you give it your own name). This is sticky, and keep in mind that the GPL was a daring bit of legal hackery that has still yet to be tested in court. To add yet another complication to the core oddity of defending right-to-modify with copyright law would risk the basic goal by making the GPL harder to defend than it already was.
All that asside, I think it's of questionable value to refer to the restrictions on the GPL as hypocritical. The GPL is a software license, not a work of art or engineering. I'm not quite certain why you feel that it would be hypocritical to say that software is an area of human endevor where freedom to modify is important but contracts and licenses are not. You may disagree, and you are most welcome to. But even if I accept that the two should be treated the same (and I do not, obviously), you make a challenge of hypocricy here which I do not believe you have explained.
I'm getting sick of this constant stream of freshmeat-like announcements of Linux-specific junk. You know there's more in the world than just Li... oh, you said OpenBSD! ;-)
-Aaron, who has seen too many serious posts that began with similar statements
Think Lawyers.
You've never seen the support costs for an IBM mainframe, have you? Suffice to say that our customers are quite happy with the cost differential, but they do it more so because of the access to modern technology in terms of software, networking and support.
I don't see anyone running Linux on huge hardware. They're all converting to farms of tiny, redundant, replacable, damn-near-disposable servers. Of course, in our case tiny means dual-processor boxes with 4GB of RAM, but for our customers that is tiny.
So, as with the mainframe vs. mini and mini vs. midrange arguments of yesteryear, you will begin to see a new trend in corporate computing: the micro-farm. It will not simply replace larger systems. It will become the standard by which the next stage of hardware downsizing (remember when that didn't mean layoffs?) will be judged.
Sure, what you propose would be a great adjuct to what Debian does now (perhaps they already do, I'm not much of a Debian guy 'cause I've never really had the time).
What's the best way to tell which kernel is best? Run it for about 2 months on a wide variety of hardware, with a wide variety of software loads. Record incidents and map those against known problems, apply available patches for those that will impact you the most. Re-test.
Then again, your distribution vendor already does this, so why would you be grabbing the latest development release (don't let the term "stable" fool you, that refers to interfaces, not field performance)? Red Hat is now up to 2.4.9 . I know that there's a lot of work going on in the VM world, and it seems to have been sorted out, but as you are noticing, there are other things in the kernel besides VM. If you want a kernel whose performance charactaristics are known, and whose primary bugs have been addressed, you have to sacrifice bleeding-edge fixes.
Not an easy pill for the "I want my tarball now!" world of Open Source, is it? Look on the bright side, 2.4.9 updates from Red Hat on 11/2 beats the heck out of the too-little-too-late geological updates from any closed-source proprietary OS vendor. Q/A is hard work and cannot happen in zero-time.
A lot of folks here have been hammering on RHAT for proposing what they did. I'm frankly stunned. No school is forced to install Red Hat Linux, but if they're going to get a free copy of the software (saves them the download) and free RHN access (saves them the constant checking for downloads), this would be pretty sweet for the schools. They could certainly opt out, and buy a competing OS.
;-)
The key thing is that in most schools, no one would go out and install Red Hat on all of their systems for fear of the backlash from "concerned parents" who see this as teaching the students with second-best (e.g. cheaper) tools in order to save. No one is going to question the installation of Red Hat, if it's being provided to the schools as part of a major settlement. It would, after all, be a substantial waste of resources to turn it down
I really hope that something like this happens, but I don't know what, besides issuing a press release, RHAT is doing to move this idea forward.
I've played with Komodo, and I have to say that I'm impressed. XUL+Moizilla can certainly be turned to one's advantage. I imagine that the crucial point is where you put the dividing line between your app and the UI. I would think that you'd want to build only the highest level UI pieces in XUL, and then the rest of your app in a low level language.
1. Have you used anything built on top of Mozilla (e.g. Komodo) or are you just flaming because it feels good?
2. If you actually needed a UI that was as rich and powerful as XUL can provide, why woulld you want to go off an write it al yourself?
3. Are you thinking in terms of writing web pages and CGI that acts as an application, or are you thinking in terms of building your UI into the XUL interface of Mozilla? These are very different things (and both actually have a place in different applications).
You're correct. And iptables is the packet filtering configuration (aka firewalling rules).
But the effect of the command itself was not the most interesting part. It's the fact that no two of those programs (well, except for the iptables-save and -restore) were designed to work with eachother. They simply do, because the system is designed to work that way. The power in that command -- the part that your average Windows user cannot grasp -- is the power to fully administer the state of a remote system without ever establishing an interactive session on it.
Windows takes the approch that many graphics oriented systems take (e.g. MacOS, even version 10, which is UNIX-based): it provides you a way to do the things that the people who designed the system felt would be most useful. UNIX takes the approach that the designers know what sorts of individual actions may be required, but often not the end-goals of any combination of them.
Is KDE (or Gnome for that matter) "ready for prime time"? Depends on what you need. It's not as polished as Windows or MacOS. However, building on top of what UNIX and/or Linux has to offer leads to a system that is overall more powerful and more flexible. That lack of polish seems a lot less important when you find that, at 6AM you can log into your desktop from work, print a report on your boss' desk and still have time for breakfast because you didn't have to wait for a graphical program to come up over a remote connection....
Why would you suffer for making coherent, rational points? I think your focus is skewed (as most Windows users' foci have been), but you make some good points.
;-)
Linux (and all of its ancestors great and mini) has a set of core features that Windows needs, but will almost certainly never get. These features (a command line that's actually useful; functional concurrent multi-user support; etc) lend power to the snappy higher levels that are put on top of them. Yes, KDE is still klunkier than Windows (as is Gnome, which is what I use). But, KDE provides access to all of the underlying power of Linux.
Try to get a Windows user to understand the concept of the init run-levels, and they'll be lost. This is pure, unadulterated power over your environment in ways that a GUI doesn't address. Now try to explain what "iptables-save | ssh -l jonesy host2 sudo iptables-restore" would do! Good luck
You have to look at what she says in context with what the RIAA does. She says she wants to improve access to smaller artists. The RIAA has worked hard to make sure this never happens. But, let me be clear: they're all for access to smaller artists as long as they continue to be able to manufacture "#1" artists at will. They want to keep the cake they have while eating it.
She also says that the people writing such things as Gnutella don't understand that they have the choice to make money or not on software, but music is just "stolen" (infringed to the rest of us). Of course this ignores the decades of warez precident and the BCA's role. This is a totally hollow argument. We write software. We sell it. We get paid. Some poeple will never be willing to pay. We know. None of that means a damn when Microsoft starts alienating their own customers with tactics like the licensing of XP. Even good, faithful customers look for an out in another product. The RIAA has the same problem.
She comments that she's excited about the possibilities of P2P. Heh, even in the client-server model of digital music, the RIAA freaked out when artists started putting their own music up for download (members did, that is).
Bottom line: read my lips, music sharing will happen. Movie sharing will happen. People will continue to share what they believe (rightly or wrongly) to be theirs. What the RIAA should be doing is coming up with a better way to take advantage of that momentum. Create a better mousetrap and the world will beat a path to your door. The corollary to that is that if you just stand around yelling at the manufacturers of poor mousetraps, you eventually get ignored.