its missing stability. its missing polish. granted there are parts that are very nice and very polished but the whole is not.
Id would say that gnome lacks the potential of kde4 but it has much more of that completeness, that cohesive environment as of right now.
I really prefered kde3.x over gnome. The switch the kde4 was poorly timed I think. Too many distros switched way before it was ready. I understand that you need to get the code out there to be tested but there was so much that obvoiusly needed/needs cleaned up before it is production ready that it was just a release of some early beta with a tag on it that said it was final.
too me, right now, production linux is one of the 3 (ubuntu, rh/fed/cent, suse) with gnome.
dont get me wrong, I would love to see kde4.x come back with a vengence. I have no hard feelings for the kde team, i just dont see kde4 as anywhere near ready for a production environment.
oh, and a 'sleeper' like a chevy citation is a novel joke, not a prized automobile.
ditto. money drives careers, and $40k at a moderately sized school is just not enough money for someone who is gifted. $40k is average pay. teachers almost by definition should be above average in ability(and theirfore wages)
I am a IT exec/tech (yeah, Exec and Tech. medium business + im awesome) and I will never teach even though I have a lot to offer because there is no money in it.
Obviously people that are interested in making money are lousy at the jobs they apply for. In fact, we should only hire teachers that have no intention to make any money whatsoever. Thats how we know that they will be a quality teacher.
Yes, raising pay could attract worse teachers that just do it for the money but it also will attract quality people. We are talking volume here, not good-to-bad ratio.
How about paying quality teachers an appropriate wage, paying terrible teachers a crap wage, and firing the ones that need fired. A good teacher can teach a student more in 1/2 a year than a bad teacher will teach in their entire run with a student.
and dont bother picking apart my paragraphs. I am a science guy, which predisposes me to bad spelling.
for those who will likely flame me out here, I am a huge KDE3.x fan. Back in my ricer days, it was custom compiled everything with a tricked out KDE desktop all the way up to and including some of the beryl/compiz stuff.
unfortunately, KDE4 came along. Sure its got that makings of a real next-gen desktop with eyecandy and function as far as the mouse can click, but today it sucks.
KDE4 is a gift to Gnome!
I have a number of machines I use regularly and have settled into a nearly stock ubuntu/gnome system for most of my work. I make a few minor usability tweaks to compiz and to gnome but otherwise it does the job and does it well.
Does it have KDE4 level eye candy? maybe not, but it looks pretty good once I change to color scheme away from the stock camel-poop-brown theme.
A social networking desktop? how about a usable desktop first.
This is the equivilent of poping a 600HP engine in a chevy citation. worthless.
I DVR my tv so what I do with the time wasted by commercials is skip it entirely.
I do agree that quicker boot isnt my highest priority for the future of linux but I certainly like my system to come up as fast as possible.
I do suspend my system and have it sitting on a very nice APC 1500VA UPS that lets me run for nearly an hour or will let me system sit in suspend for longer than I care to test.
kernel or driver updates that cannot be dynamically unloaded and reloaded require a reboot. If you want to keep your system patched up to fix bugs and insecurities you will need to reboot at some point. Might sound like a trivial scenario but a long boot process can deter you from doing the necessary update.
Laptops and other mobile devices should boot quickly. Granted, suspend and hybernate are good options but these systems do need to be rebooted on occasion and nothing sucks more than a 5 minute reboot when your battery is low and you are on an airplane. believe it or not, sometimes X crashes and the system cant shake the leftovers because of an nvidia driver quirk.
virtual machine environments where bringing up more computing resources in a hurry is necessary and a fast booting VM image is necessary.
and lastly, the drive to make things work faster is what keeps the gears of progress spinning. If we get content to have slow bootup then we will always have slow bootup.
I came into a job where the previous guy had installed upwards of 300 copies of MS Office 2000 Pro and a number of other programs such as terminal emulators.
I went to the management with this and got pretty much nowhere. I did win on the fact that I would not under any circumstances install software without a license so I have a solution moving forward.
For all those machine without proper licenses I went to the software company and explained the issue and that I would like to bring the company into compliance if they would be willing to give me their discounted upgrade rate. I replaced all of the Office 2000 installs with open office and got the vendor of a terminal emulator to make me a good deal.
We are now 100% compliant and migrating towards more open source software.
I wish that there were direct OSS replacements for everything I run but there are not. I need perfect VT400 emulation and I have not found an OSS that does that. Putty is about 95% but that other 5% doesnt allow me to have the proper keys mapped to the proper location.
Good luck and be on Buddha's side. Stick to your principals.
Considering that you are likely the Cable TV vendor and being an ISP is a method to keep your customers on your wire and not on Satellite you should lean on your boss again to keep your customers from jumping ship.
Im wondering what you have for backbone that you are 70:1 oversubscribed. If you deploy 768/256 connections with 400 customers sounds like a whopping 3 T1 lines (~4.5Mb/s). if you do a more standard 1.5MB thats 6 T1 lines(~9Mb/s).
Maybe you should look at your upstream provider and see if you can get a fractional T3 to replace the T1s if my math is anywhere near correct. You will likely have a longer contract to sign but you may be able to pull in 10Mb/s for less than you currently pay. Then you could try to match the current expense.
There are other ways to trim back your backbone usage. Consider a cluster of transparent proxy servers. You can get pretty aggressive with the cacheing mechanise in squid and you can easily balance the cluster with DNS and not have to worry about session awareness as clients also cache DNS temorarily so each client will use the same proxy for their browsing session.
Certainly some sort of QoS will work for you and lessen the need to directly throttle.
If you just throw some proxying in there and give http and https higher priority and do some packet inspection to sniff out the P2P traffic and drop it down a level you will put off the inevitable need to grow your bandwidth for a while.
if my math is correct on 1.5Mb/s cable, you look like you have a per users upstream cost of just $7.50 each. That is pretty low. Too low.
that would limit the user from even viewing anything else. This would not work. The user cannot see the gnome libs so cannot use gnome, cannot see X so cannot run X, etc etc.
You must not be a admin. The computers are there for users to use. There is no option to not let users use terminals. Their is no 'trust' in the IT vs users relationship. Notice that this is a IT 'VS' users relationship. Take that litterally.
User education is an oxymoron. Users are not educated and are somewhat uneducatable. They are hired for their skills at a certain position and not for their skills at running a computer. You will always find employees that know just enough to be dangerous. This is one of the 'fools can be fiendishly distructive' thinks.
-noexec means no mode 7. It does not mean no 'bash scriptname'
you could change all executables in user readable directories to 750 and do some group management.
the only problem is that this is going to get cumbersome as the number of workstations increase and the variety of programs that must be run by certain groups and not by other increases. now we come to the idea of central workstation management.
agreed. I am quite at home with bash scripts and can do a ton of mischief on a wide open Linux desktop. how about denial of service attacks from within the network? how about creating a local email relay by creating an ssh forward and connecting out? think you can clock it? how about running ssh over 443 so that you can skip the firewall?
Normal business is when a virus spreads. Scanning for viruses is not a bad thing and performance should not trump security. This is called being pro-active which is ideal when dealing with computer security. Only scanning for virus's at night is call reactive, which is bad when dealing with computer security.
Also, the IT department is responsible for the network and security of the network. If they make a policy that no linux machines can be on the network then what is the issue? Tight control over computer resources by IT staff is certainly best practices for a secure network.
Granted, Linux desktops are more likely to be safe than Windows desktops, but administration time is also very important. Centralized policies such as a Windows Domain is much easier to manage than a hodgepodge of various desktops with no way to enforce policy.
You can certainly install an app into your home directory. An app that for instance reads the input on/dev/something and sends that off to some remote person.
Average users can email off the system typically so a hacker could infiltrate a network for the purpose of it becoming a mail relay without ever gaining root access.
Wouldn't it be better to block executable files in user's directories?
I run a few 2k8 servers and must say that there are very few features that distinguish it from 2k3. For me, those are the new remote-apps terminal server feature and hyper-v. not a whole lot has changed other than rearranging a bunch of stuff.
The money WAS in clients. The margins are shrinking and Microsoft has to reduce prices to keep linux from becoming the obvious choice for a desktop os. How will a $100 windows license fly on a $100 computer? it wont. shrinking margins.
The server market on the other hand has big margins and microsoft can argue manageability vs initial price. That argument can work on business types.
darwin is truly a unix as much as *bsd is a unix. lay that to rest.
yes, some unixes were half baked but those that survived really were rock solid if not user friendly, stubborn, and difficult at every opportunity (sco).
I am a unix (tru64, HPUX, sco, freebsd) admin as well as a linux admin (rhel5/centos5, debian/ubuntu) and prefer linux 10:1 to most unix though freebsd is very nice and a close second to debian for me.
linux != unix, linux > unix
though
freebsd = unix, freebsd > unix
OSX is a unix by heritage but it is a Desktop OS. Apple might try to present it as some sleek server unix but it is a Desktop OS sitting on unix, which is a different creature all together.
now Microsoft knows that big money is in the server market. a single server installation with SQL server is more profit than 15 desktop sales, and there is less rampant piracy in the server market.
Microsoft vs Apple is a battle on a single front. Apple doesnt have a strong flanking maneuver in its OSX server product.
Microsoft vs Linux is a battle on the server front that Microsoft is not winning and Linux is improving on the desktop front with improvements happening far faster than Microsoft could have anticipated or even keep pace with. They have never been able to deliver an updated desktop OS on a schedule anywhere near Apple or Linux.
Since OSX came out there have been 5 full releases and twice as many dot releases, each with some noticable and desirable improvement in function AND performance. Linux is such a multifaceted movement that every 6 months there is a dot release of the main components and hundreds of fixes and tweaks. Microsoft is 2 full releases and 3 service packs in that same timeframe.
I also admin a few Windows Servers (2 2k8 and 1 2k3) and they are reliable systems but the heavy lifting in our datacenter is done by linux and the rock solid legacy systems are unix. I have unix systems that are sitting on decade old hardware and have unlimited uptime only interupted by schedule maintenance.
Microsoft is right to fear the triple threat (remember the mobile market) from linux. Apple is such a niche player and seems satisfied with that.
you could certainly install a 4GB+ USB key and mount any filesystem that was not critical for the OS to be up. If you wanted to get creative you could install the system and then dump it all to cramfs. Then mount the cramfs and use unionfs on top of that. I would suspect that many of your files would be static after the initial setup and only configuration files in/etc would change. cramfs provides a read-only compressed filesystem and unionfs lays a read-write layer on top of that so the system thinks that it has real read-write access to the filesystem.
Slashdot Mirror
its missing stability. its missing polish. granted there are parts that are very nice and very polished but the whole is not.
Id would say that gnome lacks the potential of kde4 but it has much more of that completeness, that cohesive environment as of right now.
I really prefered kde3.x over gnome. The switch the kde4 was poorly timed I think. Too many distros switched way before it was ready. I understand that you need to get the code out there to be tested but there was so much that obvoiusly needed/needs cleaned up before it is production ready that it was just a release of some early beta with a tag on it that said it was final.
too me, right now, production linux is one of the 3 (ubuntu, rh/fed/cent, suse) with gnome.
dont get me wrong, I would love to see kde4.x come back with a vengence. I have no hard feelings for the kde team, i just dont see kde4 as anywhere near ready for a production environment.
oh, and a 'sleeper' like a chevy citation is a novel joke, not a prized automobile.
ditto. money drives careers, and $40k at a moderately sized school is just not enough money for someone who is gifted. $40k is average pay. teachers almost by definition should be above average in ability(and theirfore wages)
I am a IT exec/tech (yeah, Exec and Tech. medium business + im awesome) and I will never teach even though I have a lot to offer because there is no money in it.
Obviously people that are interested in making money are lousy at the jobs they apply for. In fact, we should only hire teachers that have no intention to make any money whatsoever. Thats how we know that they will be a quality teacher.
Yes, raising pay could attract worse teachers that just do it for the money but it also will attract quality people. We are talking volume here, not good-to-bad ratio.
How about paying quality teachers an appropriate wage, paying terrible teachers a crap wage, and firing the ones that need fired. A good teacher can teach a student more in 1/2 a year than a bad teacher will teach in their entire run with a student.
and dont bother picking apart my paragraphs. I am a science guy, which predisposes me to bad spelling.
for those who will likely flame me out here, I am a huge KDE3.x fan. Back in my ricer days, it was custom compiled everything with a tricked out KDE desktop all the way up to and including some of the beryl/compiz stuff.
unfortunately, KDE4 came along. Sure its got that makings of a real next-gen desktop with eyecandy and function as far as the mouse can click, but today it sucks.
KDE4 is a gift to Gnome!
I have a number of machines I use regularly and have settled into a nearly stock ubuntu/gnome system for most of my work. I make a few minor usability tweaks to compiz and to gnome but otherwise it does the job and does it well.
Does it have KDE4 level eye candy? maybe not, but it looks pretty good once I change to color scheme away from the stock camel-poop-brown theme.
A social networking desktop? how about a usable desktop first.
This is the equivilent of poping a 600HP engine in a chevy citation. worthless.
I DVR my tv so what I do with the time wasted by commercials is skip it entirely.
I do agree that quicker boot isnt my highest priority for the future of linux but I certainly like my system to come up as fast as possible.
I do suspend my system and have it sitting on a very nice APC 1500VA UPS that lets me run for nearly an hour or will let me system sit in suspend for longer than I care to test.
A couple reasons for quicker boot:
kernel or driver updates that cannot be dynamically unloaded and reloaded require a reboot. If you want to keep your system patched up to fix bugs and insecurities you will need to reboot at some point. Might sound like a trivial scenario but a long boot process can deter you from doing the necessary update.
Laptops and other mobile devices should boot quickly. Granted, suspend and hybernate are good options but these systems do need to be rebooted on occasion and nothing sucks more than a 5 minute reboot when your battery is low and you are on an airplane. believe it or not, sometimes X crashes and the system cant shake the leftovers because of an nvidia driver quirk.
virtual machine environments where bringing up more computing resources in a hurry is necessary and a fast booting VM image is necessary.
and lastly, the drive to make things work faster is what keeps the gears of progress spinning. If we get content to have slow bootup then we will always have slow bootup.
I came into a job where the previous guy had installed upwards of 300 copies of MS Office 2000 Pro and a number of other programs such as terminal emulators.
I went to the management with this and got pretty much nowhere. I did win on the fact that I would not under any circumstances install software without a license so I have a solution moving forward.
For all those machine without proper licenses I went to the software company and explained the issue and that I would like to bring the company into compliance if they would be willing to give me their discounted upgrade rate. I replaced all of the Office 2000 installs with open office and got the vendor of a terminal emulator to make me a good deal.
We are now 100% compliant and migrating towards more open source software.
I wish that there were direct OSS replacements for everything I run but there are not. I need perfect VT400 emulation and I have not found an OSS that does that. Putty is about 95% but that other 5% doesnt allow me to have the proper keys mapped to the proper location.
Good luck and be on Buddha's side. Stick to your principals.
capacity wise yes. I have a descent sized vm cluster so tend to run multiple services for redundancy.
Considering that you are likely the Cable TV vendor and being an ISP is a method to keep your customers on your wire and not on Satellite you should lean on your boss again to keep your customers from jumping ship.
Im wondering what you have for backbone that you are 70:1 oversubscribed. If you deploy 768/256 connections with 400 customers sounds like a whopping 3 T1 lines (~4.5Mb/s). if you do a more standard 1.5MB thats 6 T1 lines(~9Mb/s).
Maybe you should look at your upstream provider and see if you can get a fractional T3 to replace the T1s if my math is anywhere near correct. You will likely have a longer contract to sign but you may be able to pull in 10Mb/s for less than you currently pay. Then you could try to match the current expense.
There are other ways to trim back your backbone usage. Consider a cluster of transparent proxy servers. You can get pretty aggressive with the cacheing mechanise in squid and you can easily balance the cluster with DNS and not have to worry about session awareness as clients also cache DNS temorarily so each client will use the same proxy for their browsing session.
Certainly some sort of QoS will work for you and lessen the need to directly throttle.
If you just throw some proxying in there and give http and https higher priority and do some packet inspection to sniff out the P2P traffic and drop it down a level you will put off the inevitable need to grow your bandwidth for a while.
if my math is correct on 1.5Mb/s cable, you look like you have a per users upstream cost of just $7.50 each. That is pretty low. Too low.
that would limit the user from even viewing anything else. This would not work. The user cannot see the gnome libs so cannot use gnome, cannot see X so cannot run X, etc etc.
You must not be a admin. The computers are there for users to use. There is no option to not let users use terminals. Their is no 'trust' in the IT vs users relationship. Notice that this is a IT 'VS' users relationship. Take that litterally.
User education is an oxymoron. Users are not educated and are somewhat uneducatable. They are hired for their skills at a certain position and not for their skills at running a computer. You will always find employees that know just enough to be dangerous. This is one of the 'fools can be fiendishly distructive' thinks.
does not stop 'bash /home/user/scriptname', only stops 'chmod +x scriptname; ./scriptname'
-noexec means no mode 7. It does not mean no 'bash scriptname'
you could change all executables in user readable directories to 750 and do some group management.
the only problem is that this is going to get cumbersome as the number of workstations increase and the variety of programs that must be run by certain groups and not by other increases. now we come to the idea of central workstation management.
agreed. I am quite at home with bash scripts and can do a ton of mischief on a wide open Linux desktop. how about denial of service attacks from within the network? how about creating a local email relay by creating an ssh forward and connecting out? think you can clock it? how about running ssh over 443 so that you can skip the firewall?
Normal business is when a virus spreads. Scanning for viruses is not a bad thing and performance should not trump security. This is called being pro-active which is ideal when dealing with computer security. Only scanning for virus's at night is call reactive, which is bad when dealing with computer security.
Also, the IT department is responsible for the network and security of the network. If they make a policy that no linux machines can be on the network then what is the issue? Tight control over computer resources by IT staff is certainly best practices for a secure network.
Granted, Linux desktops are more likely to be safe than Windows desktops, but administration time is also very important. Centralized policies such as a Windows Domain is much easier to manage than a hodgepodge of various desktops with no way to enforce policy.
You can certainly install an app into your home directory. An app that for instance reads the input on /dev/something and sends that off to some remote person.
Average users can email off the system typically so a hacker could infiltrate a network for the purpose of it becoming a mail relay without ever gaining root access.
Wouldn't it be better to block executable files in user's directories?
I run a few 2k8 servers and must say that there are very few features that distinguish it from 2k3. For me, those are the new remote-apps terminal server feature and hyper-v. not a whole lot has changed other than rearranging a bunch of stuff.
A Dell or HP will mark up some but they arent pulling big margins on windows licenses or CALs.
The money WAS in clients. The margins are shrinking and Microsoft has to reduce prices to keep linux from becoming the obvious choice for a desktop os. How will a $100 windows license fly on a $100 computer? it wont. shrinking margins.
The server market on the other hand has big margins and microsoft can argue manageability vs initial price. That argument can work on business types.
though not all university students are scientists, some are.
darwin is truly a unix as much as *bsd is a unix. lay that to rest.
yes, some unixes were half baked but those that survived really were rock solid if not user friendly, stubborn, and difficult at every opportunity (sco).
Linus + Unix = linux
no acronism involved.
I am a unix (tru64, HPUX, sco, freebsd) admin as well as a linux admin (rhel5/centos5, debian/ubuntu) and prefer linux 10:1 to most unix though freebsd is very nice and a close second to debian for me.
linux != unix, linux > unix
though
freebsd = unix, freebsd > unix
OSX is a unix by heritage but it is a Desktop OS. Apple might try to present it as some sleek server unix but it is a Desktop OS sitting on unix, which is a different creature all together.
now Microsoft knows that big money is in the server market. a single server installation with SQL server is more profit than 15 desktop sales, and there is less rampant piracy in the server market.
Microsoft vs Apple is a battle on a single front. Apple doesnt have a strong flanking maneuver in its OSX server product.
Microsoft vs Linux is a battle on the server front that Microsoft is not winning and Linux is improving on the desktop front with improvements happening far faster than Microsoft could have anticipated or even keep pace with. They have never been able to deliver an updated desktop OS on a schedule anywhere near Apple or Linux.
Since OSX came out there have been 5 full releases and twice as many dot releases, each with some noticable and desirable improvement in function AND performance. Linux is such a multifaceted movement that every 6 months there is a dot release of the main components and hundreds of fixes and tweaks. Microsoft is 2 full releases and 3 service packs in that same timeframe.
I also admin a few Windows Servers (2 2k8 and 1 2k3) and they are reliable systems but the heavy lifting in our datacenter is done by linux and the rock solid legacy systems are unix. I have unix systems that are sitting on decade old hardware and have unlimited uptime only interupted by schedule maintenance.
Microsoft is right to fear the triple threat (remember the mobile market) from linux. Apple is such a niche player and seems satisfied with that.
you could certainly install a 4GB+ USB key and mount any filesystem that was not critical for the OS to be up. If you wanted to get creative you could install the system and then dump it all to cramfs. Then mount the cramfs and use unionfs on top of that. I would suspect that many of your files would be static after the initial setup and only configuration files in /etc would change. cramfs provides a read-only compressed filesystem and unionfs lays a read-write layer on top of that so the system thinks that it has real read-write access to the filesystem.