If you are looking for a minimal GUI IceWM is the way to go.
Ice's project goals are
Feel good and fast to use, be simple and don't get in the way
Default configuration should be fully usable without tweaking
Mouse is optional
Combine the best features of other window managers and GUIs
Themes can customize the look, user can customize the feel.
One of the really nice features I have noticed is the configuration files are not only very easy to read and customize, but the way it parses it's menu file is advanced enough so that when an exacutable is not in the users path that item is left off the menu.
As per the bugtraq discussion this isn't always possible.
The Tx can be cut on AUI lines since it is not the same line as the keep alive, but on RJ45 Tx and keep alive are the same line so if you cut Tx, the device on the other end will stop sending you data.
Another method that would work better (which was also on Bugtraq) was to sniff on one nic that had an unreachable address (eg 0.0.0.0) and have another nic in the sniffer machine to connect remotely with that has a valid IP. This hack practically requires a rewrite of the tcp driver tho so it would really one be feasable on an Open Source OS.
"Also it should be obvious that this interest in lunux will kill Monterey."
Probably not. Monterey is practically an OS wrapper for huge Databases. Yea, some zealot will run a 1Tb Db on Linux once or twice, but most companies will see that Monterey was designed from the start to only serve Db and they will make an intelligent choice.
This isn't to say that Linux with XFS and IBM's DB2 wouldn't be able to handle it, but Monterey has the advantage of being designed from the start to serve this purpose.
I like GPL because I get to see the code and learn from others. Not because it goes well with my {lack of} religion / political views / ethics.
I could care less who also is using GPL'ed programs its not my place as a user, or even as a developer to censor code that is GPL'ed. The fact remains that if you write something that you GPL'ed you are in fact saying "here everyone you can look at this I don't mind." If you wanted to only give it to hippies / soccer moms / students you need to pick a different license.
One does not have to release the source to previously GPL'ed code that they have modified, but not distributed. This is why the NSA has their own Linux distro, they don't have it for download on ftp.nsa.gov so they don't have to give away the source for their super secret modifications.
Think about it, how would you enforce the stipulation "if at all you ever modify this you have to redistribute your modifications" 1) its infeasible to track down each change. 2) most of the new code would be useless to the project as a whole.
Why not Notes? you will get that kind of scalablity and you can let you boss start you out on WinNT but after they see the light you can switch them to Linux. Domino isn't currently out for Linux, but supposively R5 will be out by the end of the year.
"It's an i386 chip; that's supported. It has 3DNow; that's supported (not by AMD, but hey, it's a compiler problem). What else do you want?"
How about they help work on gcc to make k6, k6-2, k6-3, and k7 optimizations?
Intel releases their own compiler which beats the crap out of gcc for optimization of Intel chips, I would like to see AMD produce optimizing code for gcc.
If AMD did produce an optimized version of gcc that and you could compile a Linux kernel or entire system that ran 20-33% faster on AMD that the equvalent Intel chip they would dominate not only the linux home user market, but also the linux server market which is starting to develop.
High end corporations such as Dell and Compaq would still put Xeon's in their DB servers, but if VA and Penguin started using AMD because it was optimized, AMD would start to penetrate a market that is still owned by Intel.
Be supports telnet "Out of the box" so both remote execution of commands and remote login are possible, it is designed as a single user OS, but it does have those two capablites.
As a Beowulf OS it is most likely not the direction that Be would want to go. Other than render farming there are not many multimedia applications that would benifit from clustering. That is the main design goal of Be, to be the media OS.
a geek union could finally bring meaning to the saying "card carrying geek"
It would be interesting though after seeing how many websites went black to protest the original CDA imagine if all of those companies reported unpresidented numbers of people out sick. These strikes are going to be the hardest to organize, if not for a common website or other way to dessimate the information there would be no way to know that you were supposed to not go into work/school the next day.
The problems with the installer can not be fixed with expert mode. That is why Caldera also distributes a second set of disks for "old style" install via Lisa. If you search through their indices, the second most common way to solve your install problem is to drop the GUI, create new install/modules disks use the text based installer.
This shows that it is not a matter of using the lilo args er=XXX those will not be able to solve all your problems, and if you cannot get past the boot sequence with a correct list of your hardware then you cannot install.
The GUI install is great for one thing, installing via a local atapi cdrom on an IDE x86 box, anything other than that and you are going to have a lot of tweaking ahead of you. That is why I feel the GUI is not only lame, but will end up causing more experienced users to turn away from COL.
Caldera will still gain new home users, and new business users since those are the customers that buy the cd and pretty manuals, but experienced linux users who want to do an ftp install are not going to use Caldera if they keep distributing such a poor installer.
Caldera 2.2 used this installer, and it sucks. I am currently running a basterdized version of 1.2, but when 2.2 came out I tried to install it. I have since tried to install it on 5 other machines, from homebuilt x86 boxen to Dells and Microns, it has yet to install anywhere.
This lame happy window bootup interface does not allow the installer to select "i know more than dirt" mode. as a result selecting modules to use, or where you are going to get the install files from are nearly impossible.
This poorly designed install program is leading the company I work for to support only redhat and not caldera simply because they cannot test their product on any calera machines, since they cannot get Linux to install "out of the box" and if they do not want to get into the buisness of distributing lists of how to tweak the Caldera installer to get it to install on your machine.
Yes easy (read gui) installs are needed if you want more and less experienced users to run linux, but if the installer is so simple as to make such advanced options as ftp install impossible it does not bring new users to linux, it only makes them more afraid of it.
I am a Caldera user, I like the distro, but I will probably never install a version of Caldera OpenLinux again simply because 1.x is to outdated, and 2.x's installer sucks, its a shame. I feel bad when people ask me what distro I use and I have to tell them "I use Caldera, but dont try to use it since now the installer sucks, you should use XXX"
Its not if they use it, its if they improve it
on
Feature:GPL vs BSD
·
· Score: 1
If it is true that Microsoft used the FreeBSD tcp stack in WinXX, the worst problem is not that the person who wrote it does not get credit, the problem is that whoever uses the BSDed code can make it better and not have to share their improvements.
The WinXX tcp stack *could* be 5 times more efficent than FreeBSD's, but Microsoft (or anyone else) not only does not have to give credit/royalties to the developer, but they don't have to share their improvements.
There are several examples of where code forks in BSD projects have hurt. Emacs vs XEmacs is one example, and the split that created OpenBSD from NetBSD is another. There most certainly are more, but those were two big projects that had rather nasty code forks. Depending on how you look at either of these code splits you may think they helped the project, but nevertheless, these code splits left both the new project and the existing one to have fewer developers, which is not good for the community.
At the end Graydon suggests another model for contracting software that still has the limitation of the split the money effect.
I can see two possible work around for this.
pay per line of code, this has the disadvantage of code bloat, but if multiple people work on a project each gets compensated for their work in a semi fair manner.
The other method that could be best illistrated by continuing on the "this library is not threadsafe" example would be give a bounty for each api, you get the same effect of people not sharing the work, but in this case the code chunk they would be producing would be so small that sharing would have little advantage. The company would still be able to post a bounty for the whole lib, and bonuses for multiple api's, but those would only endcourage the interested developers to code more.
both of these options should have the effect of allowing the developers to use peer review, without the penalty of loss of compensation.
I am also in a university setting and a similar thing recently happened. Our network had over 100 machines which allowed spam relaying. The owners of these machines were given instructions twice and if they still had not compiled their network ports were deactivated untill the could show that they had taken the steps required.
Also a monthly meetings are schedualed where the University admins show others how to fix common security holes. If a machine is shown to have these problems they are informed via email about the meetings, and if they do not attend, or at least do not give reason for not attending their network ports are deactivated.
It seems that when ever a port is turned off the user will generally fix the problem within a few hours, even if it means using another computer to dl the latest version or patch and using zipdisks to transfer to the machine in question.
Slashdot Mirror
Ice's project goals are
- Feel good and fast to use, be simple and don't get in the way
- Default configuration should be fully usable without tweaking
- Mouse is optional
- Combine the best features of other window managers and GUIs
- Themes can customize the look, user can customize the feel.
One of the really nice features I have noticed is the configuration files are not only very easy to read and customize, but the way it parses it's menu file is advanced enough so that when an exacutable is not in the users path that item is left off the menu.As per the bugtraq discussion this isn't always possible.
The Tx can be cut on AUI lines since it is not the same line as the keep alive, but on RJ45 Tx and keep alive are the same line so if you cut Tx, the device on the other end will stop sending you data.
Another method that would work better (which was also on Bugtraq) was to sniff on one nic that had an unreachable address (eg 0.0.0.0) and have another nic in the sniffer machine to connect remotely with that has a valid IP. This hack practically requires a rewrite of the tcp driver tho so it would really one be feasable on an Open Source OS.
"Also it should be obvious that this interest in lunux will kill Monterey."
Probably not. Monterey is practically an OS wrapper for huge Databases. Yea, some zealot will run a 1Tb Db on Linux once or twice, but most companies will see that Monterey was designed from the start to only serve Db and they will make an intelligent choice.
This isn't to say that Linux with XFS and IBM's DB2 wouldn't be able to handle it, but Monterey has the advantage of being designed from the start to serve this purpose.
This is very similar to how Plan 9 works by offloading computer intensive tasks to computers that can handle such tasks easily.
I like GPL because I get to see the code and learn from others. Not because it goes well with my {lack of} religion / political views / ethics.
I could care less who also is using GPL'ed programs its not my place as a user, or even as a developer to censor code that is GPL'ed. The fact remains that if you write something that you GPL'ed you are in fact saying "here everyone you can look at this I don't mind." If you wanted to only give it to hippies / soccer moms / students you need to pick a different license.
One does not have to release the source to previously GPL'ed code that they have modified, but not distributed. This is why the NSA has their own Linux distro, they don't have it for download on ftp.nsa.gov so they don't have to give away the source for their super secret modifications.
Think about it, how would you enforce the stipulation "if at all you ever modify this you have to redistribute your modifications"
1) its infeasible to track down each change.
2) most of the new code would be useless to the project as a whole.
Why not Notes? you will get that kind of scalablity and you can let you boss start you out on WinNT but after they see the light you can switch them to Linux. Domino isn't currently out for Linux, but supposively R5 will be out by the end of the year.
"It's an i386 chip; that's supported. It has 3DNow; that's supported (not by AMD, but hey, it's a compiler problem). What else do you want?"
How about they help work on gcc to make k6, k6-2, k6-3, and k7 optimizations?
Intel releases their own compiler which beats the crap out of gcc for optimization of Intel chips, I would like to see AMD produce optimizing code for gcc.
If AMD did produce an optimized version of gcc that and you could compile a Linux kernel or entire system that ran 20-33% faster on AMD that the equvalent Intel chip they would dominate not only the linux home user market, but also the linux server market which is starting to develop.
High end corporations such as Dell and Compaq would still put Xeon's in their DB servers, but if VA and Penguin started using AMD because it was optimized, AMD would start to penetrate a market that is still owned by Intel.
Be supports telnet "Out of the box" so both remote execution of commands and remote login are possible, it is designed as a single user OS, but it does have those two capablites.
As a Beowulf OS it is most likely not the direction that Be would want to go. Other than render farming there are not many multimedia applications that would benifit from clustering. That is the main design goal of Be, to be the media OS.
a geek union could finally bring meaning to the saying "card carrying geek"
It would be interesting though after seeing how many websites went black to protest the original CDA imagine if all of those companies reported unpresidented numbers of people out sick. These strikes are going to be the hardest to organize, if not for a common website or other way to dessimate the information there would be no way to know that you were supposed to not go into work/school the next day.
The problems with the installer can not be fixed with expert mode. That is why Caldera also distributes a second set of disks for "old style" install via Lisa. If you search through their indices, the second most common way to solve your install problem is to drop the GUI, create new install/modules disks use the text based installer.
This shows that it is not a matter of using the lilo args er=XXX those will not be able to solve all your problems, and if you cannot get past the boot sequence with a correct list of your hardware then you cannot install.
The GUI install is great for one thing, installing via a local atapi cdrom on an IDE x86 box, anything other than that and you are going to have a lot of tweaking ahead of you. That is why I feel the GUI is not only lame, but will end up causing more experienced users to turn away from COL.
Caldera will still gain new home users, and new business users since those are the customers that buy the cd and pretty manuals, but experienced linux users who want to do an ftp install are not going to use Caldera if they keep distributing such a poor installer.
This is neither great nor news.
Caldera 2.2 used this installer, and it sucks. I am currently running a basterdized version of 1.2, but when 2.2 came out I tried to install it. I have since tried to install it on 5 other machines, from homebuilt x86 boxen to Dells and Microns, it has yet to install anywhere.
This lame happy window bootup interface does not allow the installer to select "i know more than dirt" mode. as a result selecting modules to use, or where you are going to get the install files from are nearly impossible.
This poorly designed install program is leading the company I work for to support only redhat and not caldera simply because they cannot test their product on any calera machines, since they cannot get Linux to install "out of the box" and if they do not want to get into the buisness of distributing lists of how to tweak the Caldera installer to get it to install on your machine.
Yes easy (read gui) installs are needed if you want more and less experienced users to run linux, but if the installer is so simple as to make such advanced options as ftp install impossible it does not bring new users to linux, it only makes them more afraid of it.
I am a Caldera user, I like the distro, but I will probably never install a version of Caldera OpenLinux again simply because 1.x is to outdated, and 2.x's installer sucks, its a shame. I feel bad when people ask me what distro I use and I have to tell them "I use Caldera, but dont try to use it since now the installer sucks, you should use XXX"
If it is true that Microsoft used the FreeBSD tcp stack in WinXX, the worst problem is not that the person who wrote it does not get credit, the problem is that whoever uses the BSDed code can make it better and not have to share their improvements.
The WinXX tcp stack *could* be 5 times more efficent than FreeBSD's, but Microsoft (or anyone else) not only does not have to give credit/royalties to the developer, but they don't have to share their improvements.
There are several examples of where code forks in BSD projects have hurt. Emacs vs XEmacs is one example, and the split that created OpenBSD from NetBSD is another. There most certainly are more, but those were two big projects that had rather nasty code forks. Depending on how you look at either of these code splits you may think they helped the project, but nevertheless, these code splits left both the new project and the existing one to have fewer developers, which is not good for the community.
I can see two possible work around for this.
both of these options should have the effect of allowing the developers to use peer review, without the penalty of loss of compensation.
I am also in a university setting and a similar thing recently happened. Our network had over 100 machines which allowed spam relaying. The owners of these machines were given instructions twice and if they still had not compiled their network ports were deactivated untill the could show that they had taken the steps required.
Also a monthly meetings are schedualed where the University admins show others how to fix common security holes. If a machine is shown to have these problems they are informed via email about the meetings, and if they do not attend, or at least do not give reason for not attending their network ports are deactivated.
It seems that when ever a port is turned off the user will generally fix the problem within a few hours, even if it means using another computer to dl the latest version or patch and using zipdisks to transfer to the machine in question.