> discovered that even though it's not obvious on the home pages, White Box and CentOS already have effectively merged already.
I think that's not true, though. IIRC a developer who first joined WBEL, later deserted and joined CentOS. The original creator of WBEL is still around.
>If you're expecting stories to get any kind of verification I think you're going to be disappointed. Hell the NY Times can't even get that done
I can understand that something can't be 100% verified. That is fine. The grandparent, however, rightly complains that the/. isn't even *trying* and I share the impression.
slashdot.org more and more feels like some fucked-up wikinewsia - you can post corrections, you can give suggestions, you can curse, but it'll always have new lame stories with tons of stupid mistakes, typos, duplicates and yesterday's (or yesteryear's) content. Even with open source projects, once you propose or create a fix, it rarely re-appears. Here, though, re-emergence of crap is the rule. What the FUCK is going on?
>Checking RHN and contacting Redhat proved that we could not get the older kernel version ANYWHERE. We basically ran on the new version after manually extracting the IBM rpm file and installing it.
I understood this the first time around - I meant to say with both CentOS and RHEL the solution is the same (as they are binary compatible). It's quite possible that it indeed was impossible to download the older kernel (although I think you can get older RPMs if you specify the version) - that's yet another reason for using CentOS (you can download all RPMs at any time from any of the mirrors around the world).
With RHN I usually suggest to use the "download RPMs" switch, then their up2date not only updates the machine but also gets those RPMs on the local HDD so that they can burn them on CD-ROMs and/or share with the whole company via NFS/Samba. You can also get the RPM via the Web (http://rhn.redhat.com/help/faq/using_rhn.pxt#63) and you can get SRPMS of the kernel package and rebuild RPM as well. You can use CentOS kernel RPM to upgrade your Red Hat kernel (that'd make it unsupported by Red Hat, though). And finally, you can install 2-3 CentOS RPMs (yum, centos-release, etc.) to perform online conversion of a RHEL box to CentOS box (some consider it waste of money if the system is still under a valid maintenance contract, though - but after that it's a different story:-)). One thing about this is that some RPMs won't install if they look for redhat-release info which can be fixed with vi editor:-), see CentOS docs or forums for details
There's nothing worse that having to waste your time with underqualified users unworthy one's attention. What a waste of time... The more time one spends with such folks, the worse one bomecomes... I'm trying to get out of that vicious circle myself.
You can do that with quite a few tools (the good ones are commercial but have better features). You could use CentOS (as RHEL) with 3rd party management tools.
>Being able to manage all of my system updates from one web page, knowing the exact status of the machines. They also have a higher level of RHN that allows even more management, like provisioning new installs.
Again, you can get better tools from companies that have been doing that longer than RH.
>If you want to run an Oracle database, you'll need an RHEL-compatible distro.
You're wrong.
If you want to run an Oracle database, you can run it on any OS that allows Oracle to run (Debian, Windows, Red Hat Enterprise Linux, CentOS, etc.)
If you want to run an Oracle database in an Oracle-supported configuration and on GNU/Linux OS, you'll have to run it on an Oracle-approved OS which is, IIRC, Red Hat Enterprise Linux, SuSE Enterprise Linux, Miracle Linux, Red Flag Linux, United Linux 1.0.
>So basically, while I might be running: >2.4.21 on RHAS 3.0 >IBM would check for the freaking sub-release >number so I had to be running: >2.4.21-27.0.2.ELsmp >I couldn't run: >2.4.21-9.ELsmp
Yeah, tey check the sub-release number - your problem is that you had a wrong kernel version.
The grandparent asked if you had such an RPM, could you still use CentOS. The answer is Yes, you could if you had the right kernel version (which you didn't have). CentOS provides same versions as RHEL so both in your (RHEL) and CentOS case, the solution would be to update to 2.4.21-27.0.2EL
> Not that I'm against variety or anything, but doesn't it make sense for these two projects to merge?
Uhmm, RHEL (Red Hat Enterprise Linux) isn't exactly a project that could merge with CentOS.
If you mean WBEL and CentOS, CentOS is bigger so I guess it'd make sense for WBEL to merge into CentOS but I don't think its existence is any kind of hindrance to CentOS (or vice versa) so if they want to be separate projects, let them be.
Lower costs don't mean shit - they just lower the entry barrier (if anything, that means more competition = harder to make money for new startups).
With UNIX (for example), an average startup would have to shell out 500K on equipment, now with Linux they need 100K. What does that mean? Nothing, because EVERYONE does the same thing, so it doesn't actually help anyone involved break even faster (as their costs are lower, their selling prices are lower, and as I said above there's more competition).
Right now Google's benefiting from it (especially if those! bozos! at! Yahoo! stil! use! Network!Apppliance!) because they're the biggest search engine of the new generation, but that won't last.
Actually what happened is that played "capitalist pigs" against each other - they made an aliance or something with 3Com (who has a cross-licensing agreement with Cisco) so they got away with it (as usual)... And I also heard that 3Com helped tem sell thru 3Com channel (not in the U.S., though).
> has performed 135.3 trillion floating point operations per second (teraFLOP/s) on the industry standard LINPACK benchmark, making it the fastest supercomputer in the world."
Did you read the fucking article?
"This performance was achieved at Lawrence Livermore National Laboratory (LLNL) at only the half-system point of the IBM BlueGene/L installation. Last November, just one-quarter of BlueGene/L topped the TOP500 List of the world's top supercomputers."
See, this is the SAME supercomputer that has already topped the list last November, so the latest record did NOT make it the fastest supercomputer in the world.
It already had been the fastest supercomputer in the world.
>Latency was noticable, but did not prevent a 20+ min conversation from taking place. It can be done.
Precisely. Had you talked over a land line instead, you could have finished the conversation in under 15 minutes instead!
Seriously, such latency is annoying. I talked over Skype last night - WLAN at home and then Skype-out to the opposite side of the globe - we pretty much had to "sync" every minute and wait 1-2 seconds before one starts talking to make sure we don't start talking at the same time. I wouldn't want to use SkypeOut for business calls.
Mobos lose money because margins are pathetic (and they're pathetic because of too much competition). And price competition is fierce not only in mobos but in virtually all components and parts that go into one (resistors, PCBs, capacitors, on-board chips, etc.). Even chipsets don't make money (at least in case of VIA - just look at their less-than-stellar results). Because of all that, mobo manufacturers are extremely cost conscious. For same reasons, there are almost no mobo factories left in Taiwan - they've all moved to China or (a few) to SE Asia. There's no way a company can afford to make mobos (at profit) in Taiwan. (I bet the factory in that article won't survive beyond 2005. In fact I was very much surprised to hear that there actually is a mobo factory in Taiwan that is still in use.)
I think IBM asked Novell to buy SuSE in order to prevent Red Hat from becoming a sole provider of enterprise Linux (IBM couldn't have done that directly so they had to act thru a proxy). Aparently that wasn't enough as (it seems to me) SuSE is not doing much better than before. They have solid but not very exciting products (think Benz).
Even Red Hat is slowing down (fewer new subscriptions (128K vs 132K in the same quarter last year, IIRC)). I wonder if the whole model of selling support and services is as scalable as Red Hat and Novell would want us to believe.
>There doesn't seem to be a shorter way to write 2004 that (sic!) to write 2004.
There doesn't seem to be a better way to write than than to write than... Joking aside, it doesn't matter what is what - what matters is that everyone understands 2k4 to mean 2004 (and by the same logic Windows 2003 is usually written as W2K3).
>MySQL has had replication for a long time already.
Yeah, but how reliable (read: it sucks) is it?
From TFA: "I believe it will change how MySQL is perceived in the market," said Axmark, who then added that he thought this release would make MySQL an option for at least ten times as many users as before."
It says in the article that mySQL has a 40% share in the open source DB market. If they're gonna get 10x as many users, that'd be 400%, so I guess they plan to make further gains in the commercial market. If open source DBs have a 10% share (means shit, anyway), then mySQL plans to shoot from a 4% to a 40% market share. I just don't see that happening.
About the market share: what does that mean - 40%? Of what? By number of copies in use, by data, by number of paying customers...?
>Maybe I'm old, but Netscape stored its bookmarks in an HTML file you could regularly FTP up to your homepage, or something similar.
You can sync/upload them from Firefox using these plugins. They work "as seen on TV" but I stopped using them after I realized that a) I rarely use my bookmarks b) what the hell - if my PC crashes I'll find those pages if I really need them
>Why do I keep seeing all these people expecting me to put my eggs in their basket?
That's the philosophy of open source, I guess. If you believe your bookmarks are better than some average Joe's... I don't know... I wouldn't share mine either. Fuck them.
The whole idea of "layers" (h/w, OS, apps) is so backward.
I hope by 2015 it'll all become seamless (brain-Google, brain-Gmail, etc. interfaces) and that Windows and Linux will become OS of the old generation.
It's hard for me to understand why anyone would want to see any of the following (examples): a) BIOS setup screen(s) b) boot dialogue c) login screen d) Start menu (or its equivalents) e) dial-up (xDSL or other) dialog box f) traditional interfaces and menus (File:: Save As...) h) Internet/Networking setup i) privacy and security options j) OS
It will be absolutely absurd if any of these things existed in 2015. Why would anyone have to deal with any of that or even know what OS (if any) the computer runs on?
We'll by then have 100 times (or more) more computing power at our disposal. If we still use today's interfaces to access it, that would be pathetic.
Slashdot Mirror
> discovered that even though it's not obvious on the home pages, White Box and CentOS already have effectively merged already.
I think that's not true, though. IIRC a developer who first joined WBEL, later deserted and joined CentOS. The original creator of WBEL is still around.
>If you're expecting stories to get any kind of verification I think you're going to be disappointed. Hell the NY Times can't even get that done
/. isn't even *trying* and I share the impression.
I can understand that something can't be 100% verified. That is fine.
The grandparent, however, rightly complains that the
slashdot.org more and more feels like some fucked-up wikinewsia - you can post corrections, you can give suggestions, you can curse, but it'll always have new lame stories with tons of stupid mistakes, typos, duplicates and yesterday's (or yesteryear's) content.
Even with open source projects, once you propose or create a fix, it rarely re-appears. Here, though, re-emergence of crap is the rule.
What the FUCK is going on?
>Checking RHN and contacting Redhat proved that we could not get the older kernel version ANYWHERE. We basically ran on the new version after manually extracting the IBM rpm file and installing it.
:-)). One thing about this is that some RPMs won't install if they look for redhat-release info which can be fixed with vi editor :-), see CentOS docs or forums for details
h p?id=13
I understood this the first time around - I meant to say with both CentOS and RHEL the solution is the same (as they are binary compatible).
It's quite possible that it indeed was impossible to download the older kernel (although I think you can get older RPMs if you specify the version) - that's yet another reason for using CentOS (you can download all RPMs at any time from any of the mirrors around the world).
With RHN I usually suggest to use the "download RPMs" switch, then their up2date not only updates the machine but also gets those RPMs on the local HDD so that they can burn them on CD-ROMs and/or share with the whole company via NFS/Samba.
You can also get the RPM via the Web (http://rhn.redhat.com/help/faq/using_rhn.pxt#63) and you can get SRPMS of the kernel package and rebuild RPM as well.
You can use CentOS kernel RPM to upgrade your Red Hat kernel (that'd make it unsupported by Red Hat, though).
And finally, you can install 2-3 CentOS RPMs (yum, centos-release, etc.) to perform online conversion of a RHEL box to CentOS box (some consider it waste of money if the system is still under a valid maintenance contract, though - but after that it's a different story
You can see all CentOS RPMs at
http://www.centos.org/modules/tinycontent/index.p
(the current RHEL 3.0 Update 4 corresponds to CentOS 3.4)
> I worked my ass off to get out of support.
There's nothing worse that having to waste your time with underqualified users unworthy one's attention.
What a waste of time... The more time one spends with such folks, the worse one bomecomes...
I'm trying to get out of that vicious circle myself.
>rhn is a big one.
l ogies/management/ads/default.mspx
You can do that with quite a few tools (the good ones are commercial but have better features). You could use CentOS (as RHEL) with 3rd party management tools.
>Being able to manage all of my system updates from one web page, knowing the exact status of the machines. They also have a higher level of RHN that allows even more management, like provisioning new installs.
Again, you can get better tools from companies that have been doing that longer than RH.
1. Novell
http://www.novell.com/products/zenworks/
2. Microsoft
http://www.microsoft.com/windowsserver2003/techno
3. Altiris, Opsware, etc.
>If you want to run an Oracle database, you'll need an RHEL-compatible distro.
You're wrong.
If you want to run an Oracle database, you can run it on any OS that allows Oracle to run (Debian, Windows, Red Hat Enterprise Linux, CentOS, etc.)
If you want to run an Oracle database in an Oracle-supported configuration and on GNU/Linux OS, you'll have to run it on an Oracle-approved OS which is, IIRC, Red Hat Enterprise Linux, SuSE Enterprise Linux, Miracle Linux, Red Flag Linux, United Linux 1.0.
>So basically, while I might be running:
>2.4.21 on RHAS 3.0
>IBM would check for the freaking sub-release >number so I had to be running:
>2.4.21-27.0.2.ELsmp
>I couldn't run:
>2.4.21-9.ELsmp
Yeah, tey check the sub-release number - your problem is that you had a wrong kernel version.
The grandparent asked if you had such an RPM, could you still use CentOS. The answer is Yes, you could if you had the right kernel version (which you didn't have). CentOS provides same versions as RHEL so both in your (RHEL) and CentOS case, the solution would be to update to 2.4.21-27.0.2EL
This package was released by CentOS (see the first announcement here):
http://www.linuxcompatible.org/story40375.html
I believe they're not 100% the same.
It is unlikely that the way trademarked info was taken out, the way repositories (up2date, yum) work and other details are 100% the same.
I don't know for WBEL, but for CentOS IIRC up2date is a wrapper for yum. Maybe WBEL even doesn't come with yum.
> Not that I'm against variety or anything, but doesn't it make sense for these two projects to merge?
Uhmm, RHEL (Red Hat Enterprise Linux) isn't exactly a project that could merge with CentOS.
If you mean WBEL and CentOS, CentOS is bigger so I guess it'd make sense for WBEL to merge into CentOS but I don't think its existence is any kind of hindrance to CentOS (or vice versa) so if they want to be separate projects, let them be.
Lower costs don't mean shit - they just lower the entry barrier (if anything, that means more competition = harder to make money for new startups).
With UNIX (for example), an average startup would have to shell out 500K on equipment, now with Linux they need 100K. What does that mean? Nothing, because EVERYONE does the same thing, so it doesn't actually help anyone involved break even faster (as their costs are lower, their selling prices are lower, and as I said above there's more competition).
Right now Google's benefiting from it (especially if those! bozos! at! Yahoo! stil! use! Network!Apppliance!) because they're the biggest search engine of the new generation, but that won't last.
Actually what happened is that played "capitalist pigs" against each other - they made an aliance or something with 3Com (who has a cross-licensing agreement with Cisco) so they got away with it (as usual)...
And I also heard that 3Com helped tem sell thru 3Com channel (not in the U.S., though).
> has performed 135.3 trillion floating point operations per second (teraFLOP/s) on the industry standard LINPACK benchmark, making it the fastest supercomputer in the world."
Did you read the fucking article?
"This performance was achieved at Lawrence Livermore National Laboratory (LLNL) at only the half-system point of the IBM BlueGene/L installation. Last November, just one-quarter of BlueGene/L topped the TOP500 List of the world's top supercomputers."
See, this is the SAME supercomputer that has already topped the list last November, so the latest record did NOT make it the fastest supercomputer in the world.
It already had been the fastest supercomputer in the world.
>Latency was noticable, but did not prevent a 20+ min conversation from taking place. It can be done.
Precisely. Had you talked over a land line instead, you could have finished the conversation in under 15 minutes instead!
Seriously, such latency is annoying. I talked over Skype last night - WLAN at home and then Skype-out to the opposite side of the globe - we pretty much had to "sync" every minute and wait 1-2 seconds before one starts talking to make sure we don't start talking at the same time. I wouldn't want to use SkypeOut for business calls.
What the fuck are you blathering about?
Mobos lose money because margins are pathetic (and they're pathetic because of too much competition). And price competition is fierce not only in mobos but in virtually all components and parts that go into one (resistors, PCBs, capacitors, on-board chips, etc.). Even chipsets don't make money (at least in case of VIA - just look at their less-than-stellar results).
Because of all that, mobo manufacturers are extremely cost conscious.
For same reasons, there are almost no mobo factories left in Taiwan - they've all moved to China or (a few) to SE Asia. There's no way a company can afford to make mobos (at profit) in Taiwan. (I bet the factory in that article won't survive beyond 2005. In fact I was very much surprised to hear that there actually is a mobo factory in Taiwan that is still in use.)
What is this bullshit?
For a moment I thought that was a story submitted by Roland Piquepaille.
The editor, go fuck yourself.
Come on...
It's just history repeating:
http://www.linux-mandrake.com/en/mdkfuture.php3
Selling some extra workstations won't help much - is just a drop in the sea.
I think IBM asked Novell to buy SuSE in order to prevent Red Hat from becoming a sole provider of enterprise Linux (IBM couldn't have done that directly so they had to act thru a proxy).
Aparently that wasn't enough as (it seems to me) SuSE is not doing much better than before. They have solid but not very exciting products (think Benz).
Even Red Hat is slowing down (fewer new subscriptions (128K vs 132K in the same quarter last year, IIRC)). I wonder if the whole model of selling support and services is as scalable as Red Hat and Novell would want us to believe.
>There doesn't seem to be a shorter way to write 2004 that (sic!) to write 2004.
There doesn't seem to be a better way to write than than to write than...
Joking aside, it doesn't matter what is what - what matters is that everyone understands 2k4 to mean 2004 (and by the same logic Windows 2003 is usually written as W2K3).
Does the fact that they're OSS-based make them immune to rules?
Shall we let some spammers go wild just because they might be using sendmail?
I say ban their ass.
> I want something that just works.
Sounds like CentOS would be good choice for you (just works, slower upgrade cycle, free) or indeed SuSE (it's simply a quality distro)
>MySQL has had replication for a long time already.
Yeah, but how reliable (read: it sucks) is it?
From TFA:
"I believe it will change how MySQL is perceived in the market," said Axmark, who then added that he thought this release would make MySQL an option for at least ten times as many users as before."
It says in the article that mySQL has a 40% share in the open source DB market. If they're gonna get 10x as many users, that'd be 400%, so I guess they plan to make further gains in the commercial market.
If open source DBs have a 10% share (means shit, anyway), then mySQL plans to shoot from a 4% to a 40% market share. I just don't see that happening.
About the market share: what does that mean - 40%? Of what? By number of copies in use, by data, by number of paying customers...?
>Maybe I'm old, but Netscape stored its bookmarks in an HTML file you could regularly FTP up to your homepage, or something similar.
r profil es.php?id=22
You can sync/upload them from Firefox using these plugins. They work "as seen on TV" but I stopped using them after I realized that
a) I rarely use my bookmarks
b) what the hell - if my PC crashes I'll find those pages if I really need them
1.
https://addons.mozilla.org/extensions/autho
2.
http://syncmarks.mozdev.org/
>Why do I keep seeing all these people expecting me to put my eggs in their basket?
That's the philosophy of open source, I guess. If you believe your bookmarks are better than some average Joe's... I don't know... I wouldn't share mine either. Fuck them.
>Not everyone in the world is lazy as americans,
Well, Brazil... I mean, you can't say it's a nation of dilligent book readers either.
> I guess this means North Korea will have to use Microsoft?
Hey - when someone terrorizes the U.S., it's only fair to respond in kind!
The whole idea of "layers" (h/w, OS, apps) is so backward.
:: Save As...)
I hope by 2015 it'll all become seamless (brain-Google, brain-Gmail, etc. interfaces) and that Windows and Linux will become OS of the old generation.
It's hard for me to understand why anyone would want to see any of the following (examples):
a) BIOS setup screen(s)
b) boot dialogue
c) login screen
d) Start menu (or its equivalents)
e) dial-up (xDSL or other) dialog box
f) traditional interfaces and menus (File
h) Internet/Networking setup
i) privacy and security options
j) OS
It will be absolutely absurd if any of these things existed in 2015. Why would anyone have to deal with any of that or even know what OS (if any) the computer runs on?
We'll by then have 100 times (or more) more computing power at our disposal. If we still use today's interfaces to access it, that would be pathetic.
>Wonder if Microsoft licenced it or just ripped it off on the basis of "it's cheaper to litigate until exhaustion"
;-)
No, that's patented by IBM (2004, an invention made during the SCO lawsuit).
>BTW, UI patents what a stupid thing...
Of course - if they were enforced we wouldn't have Evolution and a bunch of other proggies with copied UI...
> Microsoft's algorithms will suck so bad that nobody will ever dream this search tool will be useful in any way
What's it to you? You're not using Windows anyway?