I think the important part is that the installation scripts and the source they are using is not obtainable. With other distributions, I can at least get most of the source for non-proprietary products.
However, i'm afraid they will opt out of this problem through a cheap method, as all other things relating to this company. They will provide a 56k website where you can download the source, hence, almost unattainable.
While I know that other companies are doing such things as closed, binary-only distributions, I believe that we need some way to stop these morons from tarnishing the good name of Linux. And this might have been an option.
It should be pointed out in the quest for free software, that better software does not mean a more successful company.
Take Company A (lets say RedHat). They make a great product, they put it out there, people can get what they need done, they don't need to call support. Any support they do need, they get in the free first 180 days of support.
Now take Company B (lets say LinuxOne). They have a buggy LinuxOne Lite, which people need a lot of support for. They have a repackage of a repackage of a distribution. They are missing help files. There's an excellent chance that people won't be able to get to their support forums. They will most likely need support.
Now, Company A's software is more likely to be used in the first place, but there's a good chance that someone could be using Company B's software, throw it away, and just go with Microsoft. Either way gives a bad name for Linux, but Company B is more likely to make money as they have users that 'require' support. And if any of the hype of the LinuxOne Marketing Machine ('You too can have a successful company through marketing!') realize that there might be people out there that are almost forced to use this distribution.
Anyway, with free software, all bets are off in the idea of 'what makes a successful company.' Sometimes, technological markets have a stranger basis than technical merits. I mean, look at DirectPlay from Microsoft. It is *not* better than id's networking code, but it sure is used a lot more...
Do you think we can send Linus' lawyers after them for distributing a new distribution with closed source parts to it? They have some parts that are 'binary only' and I believe that RMS would explode and make a huge mess if this is true...
There was a time when the college I was attending took this to an extreme that had very negative side effects.
Now imagine that you wanted to block all game, non-school related ports. 5190 (everquest), 26900 (quake), etc... you'd end up with hundreds, possibly thousands, right? Now say you grab some of the common dcc ports, add in irc ports, etc. You end up with this huge list of ports.
Now, to be sure you arn't getting any non-school related data, you block it on both sides.
Wow, that's strange, all of a sudden only 1 in 5 attempts at going to a web site actually works! You end up with it constantly trying to get a tcp connection, you hit stop a couple times, it tries on some new tcp connections, and you finally get through. ftp's are usually fine, as once you get connected, you don't have to disconnect. but web was just horrible.
The reason? The destination port might be port 80, but the source port changes with whatever OS you were using (AIX didn't have a problem most of the time, but windows sure did!). hence, you were using a source port that some non-school related app might have been using! and this is why they were timing out.
After presenting this arguement to the community network at school, the network admin would not admit this was the problem, but he did agree to lift all but the major ports. and wow, the problem went away.
So, if you hear about these kind of problems at other institutions, remember that for every tcp connection, there's a source and destination port. The destination port is stable, and with some programs (like quake) the source port is stable, but others use the OS-defined TCP sequence port. (Which, nmap uses to do OS fingerprinting). And it's excellent reasoning to do prohibitive routing against specific programs, not everything.
Of course, at home, we have port 5190 routed to 127.0.0.1 so.... `8r) No nEverQuit at our house!
I had also wrote in about this story as well, and I thought it was funny that the Catholic Church were that internet savy.
``This is not what Catholics would understand as confession. Confession cannot be done by telephone, e-mail or proxy,'' a church spokesman told the Daily Telegraph newspaper in Thursday's editions.
So it's obviously not valid if you are behind a firewall, but it should be fine without one, because it's port 80. Email might be ports 25, 109, 110, or 143. And remember that God will know the difference, because he most likely has passed the CCIE.
"In the first day, you will be given nothing, and you will be required to create the heavens and the earth. You have 4 hours to complete this test. Tomarrow, the Devil will break it and you will have to fix it once again, possibly using a virgin birth."
I would never have even considered buying one of those magazines. They strike me as aimed at teenagers, let alone just men.
Bingo. This is the point she missed. As more marketers realize this untapped resource of adult computer gamers, we'll see more appropriate magazines emerge. In fact, I wouldn't be surprised if we had a business magazine take up the slack.
I have to say mutt is the best geek tool out there. It is the most configurable mail tool i've ever used, can do everything i want, and is still in development!
All you geeks unhappy with pine and elm, try mutt! it rocks!
What I was pointing out is that higher numbers get noticed as more developed by people who just care about the name on the product, not what it actually does.
Newbies to Linux can look at all the distributions and go: "Hrm, Mandrake 7.0, Slackware 7.0, RedHat 6.1, SuSE 6.2,... whoa, Debian 2.1 and OpenLinux 2.3 are way behind!" They won't take the time to consider the two low version numbers...
This is how version numbers are designed to be used, and when we have distributions totally running wacky with their version numbers, the other distributions will try to keep up, just so they can appear to be current, even if they are just doing point releases on incorporated products.
I don't particularly care if one particular product decides to accelerate their versioning, but it's a bad trend for the industry. Espcially with lay people read version numbers (higher is better!) compared with techies reading the same numbers (ah, this is a point release, a minor upgrade).
It's just a bad trend that I don't want to see used in the future. I think Linus has it down pretty well, with many releases between whole number changes. The differences between 2.0 and 2.2 are actually pretty minor in the grand scheme of things, but the actual number of differences are staggering. This is more the idea I want to enforce.
And if you don't care what I think, *shrug* life goes on.
ugh. this is more an aspect of the fragmentation of Linux than a version inflation issue.
I don't believe the addition of a few installation scripts warrant a whole version number. And yes, that might be what makes Mandrake unique, but doing things in custom ways is another 'issue' i have with the linux community.
I guess I feel they should abide by the 'release early and release often' philosophy, because a decent linux installation should be upgradable between the different versions. And you won't get 10 billon small changes incorporated into a major release, and 10 billon things that can go wrong.
The numbers should not matter, but they are still taken into account by users. For instance, someone telling me redhat was more developed because they were already up to 6.x.
Linus stated in his speech that the chip will be 'essentially software-powered'. And it is supposed to run Windows and Linux applications, though most likely Linux applications faster because it would not have to be interpreted.
However, what about the other x86 products? I would think they would include those products in the release as well, such as *BSD and Solaris. This would just broaden the market. and of course, these forms of Unix might be able to be ported to the new chipset natively.
Once again, if there was a change from glibc 2.0 to 2.1, and it required a whole number redhat change, why didn't they change glibc from 2.0 to 3.0? Obviously the glibc people didn't feel it required a whole number change.
If the change of kernel's from 2.0 was to 2.2 required a whole number change, why didn't Linus change 2.0 to 3.0? I mean, if it was that major of a change?
And I don't think the people at RedHat know Linux better than Linus.
I am so freakin' tired of people using a poor numbering scheme of their products! Numbers do help keep people notified of which version they are using, but they are also used to signify just how different the product is. For instance, if I'm using version 3.0 and version 5.0 is out, there should be some signficant differences!
Good examples:
Redhat 4.2 to 5.0 - glibc2
Triarch 3.0 to 4.0 - Completely different distribution method
GameSpy 2.1.8 to 3.0 - Back and front end revolution
Linux 1.2.13 to 2.0 - completely reworked
Solaris 2.7 to 2.8 - IPv6 added
NetHack 3.1.1 to 3.2 - Monsters, code, and items reworked.
Half-Life 1.0.0.6 to 1.0.0.10 - bug fixes
But many people in the industry have forgotten this fact. They put out a new release all the time, and just name it whatever. and what's worse is that people use these numbers to indicate how much more advanced it is! I was told that RedHat was better than Debian just because it was on version 6.0 already and Debian was on 2.0!
Bad Examples:
Mandrake 6.1 to 7.0 - a new installer. whoo.
RedHat 5.0 to 6.0 - new X components. whoo.
Windows 1995 to 1998 - was this really revolutionary?
Windows 2000 - ugh. I hate year based distributions of all sorts.
Worst example:
Solaris 2.6 to Solaris 7 - Sun is forced to keep up by redoing their numbers. ugh. This is the absolute worst example I can think of in this 'Version inflation'.
Now, I know that all you Zealots out there have your arguements of 'major changes' in the way your favorite product handled it's numbers, and quite a few relgious pamphlets to give me about it, but the fact of the matter is there are very few people who are following the correct versions for products any more, and we should encourage correct usage of the version schemes we have been using for years.
It's actually relatively old, just was popularized in cyberpunk novels, such that Gibson, Rucker, and Stephenson write. More people who are readers of cyberpunk have been using them around normals. (Though I don't think my parents understood what 'dermal implant' meant...) I used the phrase because 'IRL' sounds kind of dumb sometimes. (In Real Life)
It should also be pointed out that new terms and ideas are more likely to be noticed by the conscious brain when you first learn them. Hence, if a friend points out that he naturally has redhair, you'll notice more redheads the rest of the week. kind of a subliminal mark on you.
I was at IIT in 92, and you taught my freshman physics course in mechanics. Are you still teaching at IIT or are you back managing a large particle physics lab?
Actually, I'm pretty sure I just assume everyone as asexual. It's not like I do anything special for my male friends on the net either. Though I am more likely to call them 'bitch' when they kick my butt at Quake.
The same might be said for race issues, though I am more likely to say 'whaddup my niggah' to my white friends than my black friends. heh
Plus, I honestly am color blind so... heh... all green and red people look alike to me!
I think the real question for the future will be concerning who should be selected.
We have two aspects of human evolution, we have healthy, strong bodies, and we have the thing that differentiates us from animals, our brains. Now, should modern natural selection be based upon traditional standards, healthy bodies, strength, agility... Or will a factor of intelligence be more of a key, such as making more money, the ability to use power tools, the ability to use computers. And what mix is the best combination.
If you ask any 10 women who their perfect husband is, you'll get 10 different responses, with varying shades of these aspects. Not all will choose Mel Gibson. You ask 10 men however, you'll get answers only a few answers based on the first aspect of health and vigor. Interesting thought, eh?
Modern day humans don't have natural selection, why should their media? We have no method to weed out the weak, we routinely save the sick, and the moronic are glorified in our society. Take a look at Forest Gump for instance...
I'm not saying this is a bad thing, I'm sure some predator would have taken me down in high school. But isn't Katz kind of missing the point by trying to apply biological anologies to an abstract entitity when it doesn't even apply to the biologicals that created the abstract concept?
Slashdot Mirror
However, i'm afraid they will opt out of this problem through a cheap method, as all other things relating to this company. They will provide a 56k website where you can download the source, hence, almost unattainable.
While I know that other companies are doing such things as closed, binary-only distributions, I believe that we need some way to stop these morons from tarnishing the good name of Linux. And this might have been an option.
--
Gonzo Granzeau
my mom is dead.
--
Gonzo Granzeau
Take Company A (lets say RedHat). They make a great product, they put it out there, people can get what they need done, they don't need to call support. Any support they do need, they get in the free first 180 days of support.
Now take Company B (lets say LinuxOne). They have a buggy LinuxOne Lite, which people need a lot of support for. They have a repackage of a repackage of a distribution. They are missing help files. There's an excellent chance that people won't be able to get to their support forums. They will most likely need support.
Now, Company A's software is more likely to be used in the first place, but there's a good chance that someone could be using Company B's software, throw it away, and just go with Microsoft. Either way gives a bad name for Linux, but Company B is more likely to make money as they have users that 'require' support. And if any of the hype of the LinuxOne Marketing Machine ('You too can have a successful company through marketing!') realize that there might be people out there that are almost forced to use this distribution.
Anyway, with free software, all bets are off in the idea of 'what makes a successful company.' Sometimes, technological markets have a stranger basis than technical merits. I mean, look at DirectPlay from Microsoft. It is *not* better than id's networking code, but it sure is used a lot more...
--
Gonzo Granzeau
I mean, he does own the name Linux so....
--
Gonzo Granzeau
Now imagine that you wanted to block all game, non-school related ports. 5190 (everquest), 26900 (quake), etc... you'd end up with hundreds, possibly thousands, right? Now say you grab some of the common dcc ports, add in irc ports, etc. You end up with this huge list of ports.
Now, to be sure you arn't getting any non-school related data, you block it on both sides.
Wow, that's strange, all of a sudden only 1 in 5 attempts at going to a web site actually works! You end up with it constantly trying to get a tcp connection, you hit stop a couple times, it tries on some new tcp connections, and you finally get through. ftp's are usually fine, as once you get connected, you don't have to disconnect. but web was just horrible.
The reason? The destination port might be port 80, but the source port changes with whatever OS you were using (AIX didn't have a problem most of the time, but windows sure did!). hence, you were using a source port that some non-school related app might have been using! and this is why they were timing out.
After presenting this arguement to the community network at school, the network admin would not admit this was the problem, but he did agree to lift all but the major ports. and wow, the problem went away.
So, if you hear about these kind of problems at other institutions, remember that for every tcp connection, there's a source and destination port. The destination port is stable, and with some programs (like quake) the source port is stable, but others use the OS-defined TCP sequence port. (Which, nmap uses to do OS fingerprinting). And it's excellent reasoning to do prohibitive routing against specific programs, not everything.
Of course, at home, we have port 5190 routed to 127.0.0.1 so.... `8r) No nEverQuit at our house!
--
Gonzo Granzeau
Hence, high speed prayers only. And who says that God loves the poor too....*sniff*
--
Gonzo Granzeau
``This is not what Catholics would understand as confession. Confession cannot be done by telephone, e-mail or proxy,'' a church spokesman told the Daily Telegraph newspaper in Thursday's editions.
So it's obviously not valid if you are behind a firewall, but it should be fine without one, because it's port 80. Email might be ports 25, 109, 110, or 143. And remember that God will know the difference, because he most likely has passed the CCIE.
"In the first day, you will be given nothing, and you will be required to create the heavens and the earth. You have 4 hours to complete this test. Tomarrow, the Devil will break it and you will have to fix it once again, possibly using a virgin birth."
--
Gonzo Granzeau
Be glad you're unique.
I would never have even considered buying one of those magazines. They strike me as aimed at teenagers, let alone just men.
Bingo. This is the point she missed. As more marketers realize this untapped resource of adult computer gamers, we'll see more appropriate magazines emerge. In fact, I wouldn't be surprised if we had a business magazine take up the slack.
--
Gonzo Granzeau
All you geeks unhappy with pine and elm, try mutt! it rocks!
--
Gonzo Granzeau
Or that standards that have been in place for a long time that are getting ignored were not put in place for a reason?
--
Gonzo Granzeau
Newbies to Linux can look at all the distributions and go: "Hrm, Mandrake 7.0, Slackware 7.0, RedHat 6.1, SuSE 6.2, ... whoa, Debian 2.1 and OpenLinux 2.3 are way behind!" They won't take the time to consider the two low version numbers...
This is how version numbers are designed to be used, and when we have distributions totally running wacky with their version numbers, the other distributions will try to keep up, just so they can appear to be current, even if they are just doing point releases on incorporated products.
--
Gonzo Granzeau
It's just a bad trend that I don't want to see used in the future. I think Linus has it down pretty well, with many releases between whole number changes. The differences between 2.0 and 2.2 are actually pretty minor in the grand scheme of things, but the actual number of differences are staggering. This is more the idea I want to enforce.
And if you don't care what I think, *shrug* life goes on.
--
Gonzo Granzeau
I don't believe the addition of a few installation scripts warrant a whole version number. And yes, that might be what makes Mandrake unique, but doing things in custom ways is another 'issue' i have with the linux community.
I guess I feel they should abide by the 'release early and release often' philosophy, because a decent linux installation should be upgradable between the different versions. And you won't get 10 billon small changes incorporated into a major release, and 10 billon things that can go wrong.
--
Gonzo Granzeau
The numbers should not matter, but they are still taken into account by users. For instance, someone telling me redhat was more developed because they were already up to 6.x.
--
Gonzo Granzeau
However, what about the other x86 products? I would think they would include those products in the release as well, such as *BSD and Solaris. This would just broaden the market. and of course, these forms of Unix might be able to be ported to the new chipset natively.
Thoughts?
--
Gonzo Granzeau
Obviously a minor change. only a
And they added GNOME.
A new X component. Which could have been added by an RPM.
I stand by this example.
--
Gonzo Granzeau
Once again, if there was a change from glibc 2.0 to 2.1, and it required a whole number redhat change, why didn't they change glibc from 2.0 to 3.0? Obviously the glibc people didn't feel it required a whole number change.
--
Gonzo Granzeau
And I don't think the people at RedHat know Linux better than Linus.
--
Gonzo Granzeau
I am so freakin' tired of people using a poor numbering scheme of their products! Numbers do help keep people notified of which version they are using, but they are also used to signify just how different the product is. For instance, if I'm using version 3.0 and version 5.0 is out, there should be some signficant differences!
Good examples:
But many people in the industry have forgotten this fact. They put out a new release all the time, and just name it whatever. and what's worse is that people use these numbers to indicate how much more advanced it is! I was told that RedHat was better than Debian just because it was on version 6.0 already and Debian was on 2.0!
Bad Examples:
- Mandrake 6.1 to 7.0 - a new installer. whoo.
- RedHat 5.0 to 6.0 - new X components. whoo.
- Windows 1995 to 1998 - was this really revolutionary?
- Windows 2000 - ugh. I hate year based distributions of all sorts.
Worst example:Now, I know that all you Zealots out there have your arguements of 'major changes' in the way your favorite product handled it's numbers, and quite a few relgious pamphlets to give me about it, but the fact of the matter is there are very few people who are following the correct versions for products any more, and we should encourage correct usage of the version schemes we have been using for years.
--
Gonzo Granzeau
It should also be pointed out that new terms and ideas are more likely to be noticed by the conscious brain when you first learn them. Hence, if a friend points out that he naturally has redhair, you'll notice more redheads the rest of the week. kind of a subliminal mark on you.
--
Gonzo Granzeau
I was at IIT in 92, and you taught my freshman physics course in mechanics. Are you still teaching at IIT or are you back managing a large particle physics lab?
--
Gonzo Granzeau
The same might be said for race issues, though I am more likely to say 'whaddup my niggah' to my white friends than my black friends. heh
Plus, I honestly am color blind so... heh... all green and red people look alike to me!
--
Gonzo Granzeau
Sorry, I was addressing the 'bigger picture'. Which Katz seems to routinely miss. Also, bad anologies bother me. a lot.
--
Gonzo Granzeau
We have two aspects of human evolution, we have healthy, strong bodies, and we have the thing that differentiates us from animals, our brains. Now, should modern natural selection be based upon traditional standards, healthy bodies, strength, agility... Or will a factor of intelligence be more of a key, such as making more money, the ability to use power tools, the ability to use computers. And what mix is the best combination.
If you ask any 10 women who their perfect husband is, you'll get 10 different responses, with varying shades of these aspects. Not all will choose Mel Gibson. You ask 10 men however, you'll get answers only a few answers based on the first aspect of health and vigor. Interesting thought, eh?
--
Gonzo Granzeau
I'm not saying this is a bad thing, I'm sure some predator would have taken me down in high school. But isn't Katz kind of missing the point by trying to apply biological anologies to an abstract entitity when it doesn't even apply to the biologicals that created the abstract concept?
--
Gonzo Granzeau