People brining in their art projects to do at work. Boss running around like a chicken with its head cut off. No one empowered to do anything and the boss to busy to give them anything to do. Yeah... sure... it was the/music/ that was holding them back.
At least it's possible in linux to hack the config file when the auto-detection fails you. In windows you're pretty much out of luck if it detects your video card as being less capable than it really is. If it's not on the drop down box its becomes/very/ hard to convince windows otherwise. When I went from 95 to 98 there was a conciderable about of time spend tearing my hair out screaming things like "but it did 1024x768 BEFORE!!!".
"So only God has been able to build such a device (the atom with its ever moving cloud of electrons)..." Actually God's following 'his own rules' on this one. Atoms arn't perpetual motion machines. If it wasn't for all those suns doing matter to energy conversions (and the huge energy surplus we started with at the big bang) the whole universe would eventually drop to 0 Kelvin and the motion of those electons would stop.
This is not (nessissarily) a kernel problem. 2.4.* does (I beleive) use 32 bit uids but if your userspace applications don't then you've gained naught. It your chown command uses a 16bit uid then it will set it wrong. And if your ls uses a 16bit uid it will read it back wrong.
Density/does/ want to be pretty. If you should happen to be as think as two short planks you're not going to get real far in life unless you're nice to look at.
"I'll do a make and make install and *boom* suddenly it overwrites the rpm's files." You can hardly expect the package manager to keep track of things when you don't use it. If you make an end run around rpm and modify files directly how is it meant to know? That said "rpm --justdb -e binutils" will let the rpm database know that the bin utils package is no longer installed and leave your own files alone. If you later reinstall the RPM your config files will be backedup. "also, rpm -qpl should put a header" rpm --queryformat 'Package: %{NAME}-%{VERSION}\n' -qlp *
Re:What's up with the piss-poor reporting?
on
BSD Quickies
·
· Score: 1
They're not releaseing RPC source. They're releasing the NFS interface/protocol. And really for anyone wanting to write NFS clients and servers the protocol is the hardest bit. We have no trouble at all making the protocol to OS hooks - we've written lots of file systems before. This is not a non-trivial release.
Well I don't mind saying I'm delighted VA Linux is going to be the senior partner in this arangement. No offence to the slashdot folk or the freshmeat guys (guy?) - I'm just pleased to see the 'heavy hitters' of Linux content under the banner of a company thats proved itself capable of more than just buying other people. Congrats to slashdot for making andover buyable:)
"The GPL will prevent him from leveraging and improving upon publicly available code to make money" Stopping people unfairly 'leveraging' our code is precisily what the GPL is/meant/ to do. If you want to make money by standing on our back you have to play by our rules. The GNU exists (and hence the GPL) specifically to counteract commerical software not to give it a helping hand. The 'little guy' has only to join the team. Without teamwork there isn't any code in the first place. Producing standard and compatible code is what RFC's are for. If you don't want to give your improvements back then you don't deserve to take our code in the first place. Its very nice that BSD are willing to let people do this but that's up to them - personally I think they're nuts. You can argue that the goals are not worthwhile but arguing that it achieves what it sets out to do is a bit pointless.
Somehow I think he's missing the point of open source. We're not here to get big, make money and go home with the girl. No the folk who are acutally making a difference. This guy is talking as if open source was a scarse resource. Its only exists/at/all/ becuase the people who made it wanted it. The corporate raiders can come in, strip the land scape bare, and when they leave open source will still be here just like it was before. I guess sucks if you were looking to get rich out of linux. But if you just wanted an OS that got the job done - no ones taking that away. Moneys nice - but it's not really the goal is it? If it's your goal you deserve what you get.:P
General opinions about the quality of microsoft software aside this'll be great. Only - its not going to happen. Empty, vauge promises about the distant future are so very very easy to make and just as easy to cast aside. How often have microsoft 'changed plans' about their flag ship products? Again and again and again.. If they are that flexible with their core products, how likely are they going to be to carry plans for a Linux media player all the way to fruition. The article spoke about 6mths for the Mac version, the linux version an unspecified amount of time after that. I'm not even sure it can fairly be called vapourware. Its just that 6mths is a _eon_ online - especially multimedia technologies. If the last few months with napster and mp3 and decss et all have show us anything its the volitility of the marketplace. I'll be optimistic and say MS are actualling going releases it but I'm not going to hold my breath. I'll be _really_ optimistic and say that in 6mnths I won't care about their product any more.:)
Link to the article itself.
on
On to Mars
·
· Score: 5
Was there meant to be a link to the article and not just the mag? No missing this time either. Anyway... http://www.weeklystandard.com/magazine/ mag_5_19_00/krauthammer_cov_5_18_00.html
The article meantions non-voting members of the house of reps. What are these? Are there two types of reps or are these non-voting members 'special provisions'?
And we would be right - it isn't ready for prime time, developers only - howerver Win2k is a shipping product. Now if 2.4 ships, and a bug is found in khttpd or knfs (I can't think of any other in kernel servers), and they default to on when you build your kernel, then yes that would be something to scream about. But that isn't going to happen. Half a dozen beta realeases can not begin to compare to a hundred plus development releases when it comes to peer review.
Personally I'm -very- excited about this one. Enterprise level mass storage (there's more to veritas than just filesystems) is one of few sticking points for linux making the move from utility to work horse. Having the same bullet proof, flexible FS on my linux box as I have on my Solaris box.... drool....:) http://www.computerworld.com/home/print.nsf/al l/000121E29E
Well bugger me - my script for downloading UF isn't 2000 complient. Oh the shame...:) wget -nc http://...chives/${year#19*}$mon_name/$date.html
Re:"Long uptime is evil" or "gee isnt my system op
on
Server Uptimes Ranked
·
· Score: 1
> 2. Besides, as the Linux kernel works with jiffies, on a 32 bit arch, your uptime is limited to 497 days, not?
Actually your box won't crash when your jiffies wrap - the kernel will mostly keep on trucking. But 'strange and mysterious things' will begin to happen in user land. Most obviously 'ps' and its friends will get stuck at the momment of wrap. Hmm... sometime ago, when the idea of setting your HZ=1000 was trashing about the kernel dev list some posted a patch for all those user land things. That was diffinitly pre 2.0.0 though so who knows what state its in today.
I've been using Redhat for.. well forever. I tried making a switch to Debian at 2.0 (what was that called?) but never made it past the install. The mind boggling array and internconnectednessness of packages proved to much for me. I guess I didn't have a high enough geek quoient at the time.;) Now 18 months further down the road Redhat is increasingly frustrating with just a few too many bags on the side for my liking. That debian have taken this long to get their next 'blessed' release out bodes well for their quality control. I can't wait to give it another whirl. Aptget sounds like the closest thing the Linux world has to FreeBSD's cvsupdate. Does anyone know how Debian handle things like PAM and sysv vs. bsd style init scripts?
That was very convient - thank you Mr Lawyer. for i in `grep gz$ legal-info.txt`; do wget http://$i done There, lovely! - didn't have to go searching or anything.
Re:Fast. Buggy. Weird UI design. Windows-like widg
on
Opera Beta Released
·
· Score: 1
Ahh... so it has a name. MDI huh? I used it for a few minutes and thought 'ugh - just like star office. I'm never going to be using this one.' Kind of a shame - it sure did load fast.
"How hard is it to find a cheapo modem and set up a quickie ISP account? That explanation just doesn't fly." And when randomuser@some.random.isp emails you claiming to be your webadmin you're going to beleive them? If the answers yes I have a bridge you might be intrested in...
For my money the voyger probes have got to be one the all time great hacks. Not so much the probes themselves (although they've long since proved themselves to be hacks in their own right) but the trajectories. If we launch at -this- time we can go here, here, here, and here, and then fling ourselves out into deep space with a tiny little probe. That's the quintessential elegent hack to me.
Actually I see this more as a union than further fragmentation - a bridge between linux and freebsd if you like. Your server room could have various BSD and Linux boxes that all have the same 'feel' to them. Nice if your an admin. Personally I can't see more than a very marginal gain. Except for perhaps hardware incompatilities or boxes that are brushing the outside of the envelope the respective kernels are much of a muchness. They work. They work as close to _all_ the time as makes no odds. It's all the other things on the box that differenciate the products. Init scripts, user space programs, system utilites. Take linux user space and put it on bsd, take the bsd use space and put it on the linux kernel if you really want - what did you gain that you didn't already have?
Argh! Feel... tension... rising... must... flame... I don't know what gets under my skin more - the folks saying NTFS does 'full logging' when it doesn't or the folk saying ext3 doens't when it does. Ext3 logs both the meta-data and the file data. It's even concidered a 'bug'.:) From the latest alpha's To Do list: * Journaling of metadata only. Currently everything is journaled, incuding data, resulting in a performance drop as all data gets written twice. Journaling of metadata only is supported but is not enabled. It turns out to involve several extra complications in the journaling buffer state, so I'm testing the simpler case first to get that reliable on its own. I've been using the alpha for a while now and there is a big performace hit with full logging. I can't wait until we get a choice of full or meta-only logging. Eventually we should be able to have full logging on those rarely written to partitions like/usr and / (the hit for reading is very slight approaching 0) and meta only for the rest. Way cool.:) It's an alpha product - flaming it or praising it are both a bit premature - but if we must talk about it at least we could talk about what it -acutually- does.
Slashdot Mirror
People brining in their art projects to do at work. Boss running around like a chicken with its head cut off. No one empowered to do anything and the boss to busy to give them anything to do. Yeah... sure... it was the /music/ that was holding them back.
At least it's possible in linux to hack the config file when the auto-detection fails you. In windows you're pretty much out of luck if it detects your video card as being less capable than it really is. If it's not on the drop down box its becomes /very/ hard to convince windows otherwise. When I went from 95 to 98 there was a conciderable about of time spend tearing my hair out screaming things like "but it did 1024x768 BEFORE!!!".
The only time this matters is in comparisons and sorting. Perl has explicit numeric and string tests for both of these activities.
:- sort numericaly
:- sort as strings
:- compare as strings
:- compare numericaly
$a <=> $b
$a cmp $b
$i lt $j
$i < $j
... if someone uses the wrong test they've no one but themselves to blame.
"So only God has been able to build such a device (the atom with its ever moving cloud of electrons) ..."
Actually God's following 'his own rules' on this one. Atoms arn't perpetual motion machines. If it wasn't for all those suns doing matter to energy conversions (and the huge energy surplus we started with at the big bang) the whole universe would eventually drop to 0 Kelvin and the motion of those electons would stop.
This is not (nessissarily) a kernel problem. 2.4.* does (I beleive) use 32 bit uids but if your userspace applications don't then you've gained naught.
It your chown command uses a 16bit uid then it will set it wrong. And if your ls uses a 16bit uid it will read it back wrong.
Density /does/ want to be pretty. If you should happen to be as think as two short planks you're not going to get real far in life unless you're nice to look at.
"I'll do a make and make install and *boom* suddenly it overwrites the rpm's files."
You can hardly expect the package manager to keep track of things when you don't use it. If you make an end run around rpm and modify files directly how is it meant to know? That said "rpm --justdb -e binutils" will let the rpm database know that the bin utils package is no longer installed and leave your own files alone. If you later reinstall the RPM your config files will be backedup.
"also, rpm -qpl should put a header"
rpm --queryformat 'Package: %{NAME}-%{VERSION}\n' -qlp *
They're not releaseing RPC source. They're releasing the NFS interface/protocol. And really for anyone wanting to write NFS clients and servers the protocol is the hardest bit. We have no trouble at all making the protocol to OS hooks - we've written lots of file systems before. This is not a non-trivial release.
Well I don't mind saying I'm delighted VA Linux is :)
going to be the senior partner in this arangement.
No offence to the slashdot folk or the freshmeat
guys (guy?) - I'm just pleased to see the 'heavy hitters' of Linux
content under the banner of a company thats proved
itself capable of more than just buying other people.
Congrats to slashdot for making andover buyable
"The GPL will prevent him from leveraging and improving upon publicly available code to make money" /meant/ to do. If you want to make money by standing on our back you have to play by our rules. The GNU exists (and hence the GPL) specifically to counteract commerical software not to give it a helping hand. The 'little guy' has only to join the team. Without teamwork there isn't any code in the first place. Producing standard and compatible code is what RFC's are for. If you don't want to give your improvements back then you don't deserve to take our code in the first place. Its very nice that BSD are willing to let people do this but that's up to them - personally I think they're nuts.
Stopping people unfairly 'leveraging' our code is precisily what the GPL is
You can argue that the goals are not worthwhile but arguing that it achieves what it sets out to do is a bit pointless.
Somehow I think he's missing the point of open source. /at/all/ becuase the people who made it wanted it. The corporate raiders can come in, strip the land scape bare, and when they leave open source will still be here just like it was before. :P
We're not here to get big, make money and go home with the girl. No the folk who are acutally making a difference.
This guy is talking as if open source was a scarse resource.
Its only exists
I guess sucks if you were looking to get rich out of linux. But if you just wanted an OS that got the job done - no ones taking that away.
Moneys nice - but it's not really the goal is it? If it's your goal you deserve what you get.
General opinions about the quality of microsoft software aside this'll be great. Only - its not going to happen. :)
Empty, vauge promises about the distant future are so very very easy to make and just as easy to cast aside.
How often have microsoft 'changed plans' about their flag ship products? Again and again and again..
If they are that flexible with their core products, how likely are they going to be to carry plans for a Linux media player all the way to fruition. The article spoke about 6mths for the Mac version, the linux version an unspecified amount of time after that.
I'm not even sure it can fairly be called vapourware. Its just that 6mths is a _eon_ online - especially multimedia technologies. If the last few months with napster and mp3 and decss et all have show us anything its the volitility of the marketplace.
I'll be optimistic and say MS are actualling going releases it but I'm not going to hold my breath. I'll be _really_ optimistic and say that in 6mnths I won't care about their product any more.
Was there meant to be a link to the article and not just the mag? No missing this time either. Anyway.../ mag_5_19_00/krauthammer_cov_5_18_00.html
http://www.weeklystandard.com/magazine
The article meantions non-voting members of the house of reps. What are these? Are there two types of reps or are these non-voting members 'special provisions'?
And we would be right - it isn't ready for prime
time, developers only - howerver Win2k is a shipping
product. Now if 2.4 ships, and a bug is found in khttpd or knfs (I can't think of any other in kernel servers), and they default to on when you build your kernel, then yes that would be something to scream about. But that isn't going to happen. Half a dozen beta realeases can not begin to compare to a hundred plus development releases when it comes to peer review.
Personally I'm -very- excited about this one. Enterprise level mass storage (there's more to veritas than just filesystems) is one of few sticking points for linux making the move from utility to work horse. Having the same bullet proof, flexible FS on my linux box as I have on my Solaris box.... drool.... :)l l/000121E29E
http://www.computerworld.com/home/print.nsf/a
Well bugger me - my script for downloading UF isn't 2000 complient. Oh the shame... :)
wget -nc http://...chives/${year#19*}$mon_name/$date.html
> 2. Besides, as the Linux kernel works with jiffies, on a 32 bit arch, your uptime is limited to 497 days, not?
Actually your box won't crash when your jiffies wrap - the kernel will mostly keep on trucking. But 'strange and mysterious things' will begin to happen in user land. Most obviously 'ps' and its friends will get stuck at the momment of wrap. Hmm... sometime ago, when the idea of setting your HZ=1000 was trashing about the kernel dev list some posted a patch for all those user land things. That was diffinitly pre 2.0.0 though so who knows what state its in today.
I've been using Redhat for.. well forever. I tried making a switch to Debian at 2.0 (what was that called?) but never made it past the install. The mind boggling array and internconnectednessness of packages proved to much for me. I guess I didn't have a high enough geek quoient at the time. ;) Now 18 months further down the road Redhat is increasingly frustrating with just a few too many bags on the side for my liking.
That debian have taken this long to get their next 'blessed' release out bodes well for their quality control. I can't wait to give it another whirl. Aptget sounds like the closest thing the Linux world has to FreeBSD's cvsupdate. Does anyone know how Debian handle things like PAM and sysv vs. bsd style init scripts?
That was very convient - thank you Mr Lawyer.
for i in `grep gz$ legal-info.txt`; do
wget http://$i
done
There, lovely! - didn't have to go searching or anything.
Ahh... so it has a name. MDI huh? I used it for a few minutes and thought 'ugh - just like star office. I'm never going to be using this one.' Kind of a shame - it sure did load fast.
"How hard is it to find a cheapo modem and set up a quickie ISP account? That explanation just doesn't fly."
And when randomuser@some.random.isp emails you claiming to be your webadmin you're going to beleive them? If the answers yes I have a bridge you might be intrested in...
For my money the voyger probes have got to be one the all time great hacks. Not so much the probes themselves (although they've long since proved themselves to be hacks in their own right) but the trajectories. If we launch at -this- time we can go here, here, here, and here, and then fling ourselves out into deep space with a tiny little probe. That's the quintessential elegent hack to me.
Actually I see this more as a union than further fragmentation - a bridge between linux and freebsd if you like. Your server room could have various BSD and Linux boxes that all have the same 'feel' to them. Nice if your an admin.
Personally I can't see more than a very marginal gain. Except for perhaps hardware incompatilities or boxes that are brushing the outside of the envelope the respective kernels are much of a muchness. They work. They work as close to _all_ the time as makes no odds. It's all the other things on the box that differenciate the products. Init scripts, user space programs, system utilites. Take linux user space and put it on bsd, take the bsd use space and put it on the linux kernel if you really want - what did you gain that you didn't already have?
Argh! Feel... tension... rising... must... flame... :) From the latest alpha's To Do list: /usr and / (the hit for reading is very slight approaching 0) and meta only for the rest. Way cool. :)
I don't know what gets under my skin more - the folks saying NTFS does 'full logging' when it doesn't or the folk saying ext3 doens't when it does.
Ext3 logs both the meta-data and the file data. It's even concidered a 'bug'.
* Journaling of metadata only. Currently everything is journaled, incuding data, resulting in a performance drop as all data gets written twice.
Journaling of metadata only is supported but is not enabled. It turns out to involve several extra complications in the journaling buffer state, so I'm testing the simpler case first to get that reliable on its own.
I've been using the alpha for a while now and there is a big performace hit with full logging. I can't wait until we get a choice of full or meta-only logging. Eventually we should be able to have full logging on those rarely written to partitions like
It's an alpha product - flaming it or praising it are both a bit premature - but if we must talk about it at least we could talk about what it -acutually- does.