Well, trusting is the cheapest solution. But otherwise, give them Remote Desktop Access to a PC (GotoMyPc, Netviewer, CoPilot) and then you can monitor. And of course, it is a good idea to structure your network to not overly trust systems and users, so if they administer one server, they should not be able to get to other, more important services.
However if you have sensitive IT, for simple services it might be good to actually DIY - I mean if you Monitor them and you are qualified to understand what they do... you can also do it.
On a multi-user server which is concurrently used there is no such thing as "middle of the page". However you are right, for single user applications it might make sense to hint the GC. I guess this is where RT-Java will come to an rescue.
(And well, malloc heap reorganisaton in C is not that pleasant, also).
Well, the JbossMQ in 4.x may be rocking, in 3.x the cluster support, redundancy and routing is pretty weak compared to ther JMS solutions (commercial and non commercial).
Well, I think there wont be a single Framework. Because for one we do have this Gnome/KDE conflict and second, various distributions judge the DRM and copyright problems differently. At least for non-streamed content integration like wincodecs or alterative decoders will not be present officially in a vendor supported player. Especially not when it is backed by Real Networks. -- www.eckes.org
And why is it a virtual office, of you use physical computers? If it is the work place at home, call it Soho, or call it the workplace of a telecommuter, but I dont see what the virtual here is, besides a disturbing buzz-word.
And by the way, did I miss the content of this article? It is just listening some well known web sites. Where is the news?
Well Well, the fight between UN/CEFACT and OASIS was obvious a long time ago. Ever since OASIS started to work on UBL, they stepped on the foot of CEFACT, by working on message content definitions.
So it is no wonder, that CEFACT is reorienting themself. This is not only becuase of the personal changes. But of course those also have some meaning.
The UN work is not realy money driven, but highly dependend on individual work. I know that a lot of ppl ware working at CEFACT or OASIS TCs very enthusiastically, and it is not a cheap thing for a company to sponsor chairs, editors or contributors.
BTW: I was personally very impressed on the quick and quality work the ebXML open process had produced. The work realy slowed down, after it moved to OASIS TCs and to other CEFACT Working Groups. And I had the impression, that the CEFACT was the more modern and agile body to carry on the work. OASIS has pretty strict attendence rules for paying members - you have to torture yourself to be part of the community.
JavaVMs had exploits in the past, so I do not see ground for your claim that they are the most safe VMs. I know simulations of turing engines which are far more secure and less exploitable.
Remember Brown Orifice in netscape VMs?
Some more: http://www.securitytracker.com/archives/target/130 1.html
There is a problem with Trademark Laws. In some countries you have to actually defend every abuse of your Trademark, or otherwise you wont get it granted anymore.
If there is a public display of that Logo on a high profile site, it has to be lawfully attacked. But on the other hand, I can think of a easy regulation for allowing the PCI Logo to devices which are actually PCI devices.
No the idea is simple. If the image of your disk is not a 1:1 copy of the bits, but understands the structure of the files and the file system (like a tar backup would), you can restore it on any sized disk, as long as the data fits in the free space.
Re:Only if it's the same size disk
on
Ghost for Unix
·
· Score: 1
I suggest you have a look at
partimage.org, it understands about filesystems (UFS, HFS, HPFS, XFS, JFS, NTFS, FAT, ext2, Reiser) and is therefoe better for resizing and efficient dumping of filesystems.
The clause in question is included in most commercially seeded open source licenses. The MPL was one of the fiest with it. The idea is, that the company who opens up the source code still has the right to use it for properitary closed source products. And in addition to that, all patches to the free and open version can be used by the company.
In fact if you publish source under the GPL you are free to use that source in another commercial product and do not need to open up the source. But as soon as you accept paptches, you cant include those patches in your closed product, cause they will be under the GPL and you do not have the copyright.
Thats why you see licenses like the MPL. In fact it means a company treats its source base against patches. This can be "fair" or "unfair" to the community, but it is OSI compliant, since it does not restrict the open branch of a product. It only grants additional rights to the original licensor.
virtual punch cards (not real ones) are acutlly in use on some old mainframes. You copy those card jobs to tapes and read from a more stable media. But still it is 80col card jobs. JCL or some other stuff.
Used to work with Siemens BS1000 Systems that way, 10 years back in time.
Well, with Windows XP it is more or less a mainstream OS already. Of course the applications are harder to port than with Linux Apps. But of the other site, Microsoft Shops are expected to jump on that bandwagon since they expect a lot of update business. So they will port it natively to 64bit (or rewqrite it in.NET Technology).
Same is true for Java Applications, they soon will run under 64Bit. So I guess there is not much time untile W32OW64 is not needed anymore:)
Currently the code quality of 64bit Linux is pretty bad due to gcc problems, but once those are ruled out (i.e. gcc3.0 is stable enough to compile everything) this wont be a big issue.
Actually the swirl is a heavyly discussed marketing instrument for debian to rule the world, you may look at... for other market leaders facilidating the swirl to impress buyers.
Slashdot Mirror
Well, trusting is the cheapest solution. But otherwise, give them Remote Desktop Access to a PC (GotoMyPc, Netviewer, CoPilot) and then you can monitor. And of course, it is a good idea to structure your network to not overly trust systems and users, so if they administer one server, they should not be able to get to other, more important services.
However if you have sensitive IT, for simple services it might be good to actually DIY - I mean if you Monitor them and you are qualified to understand what they do... you can also do it.
Greetings
Bernd
On a multi-user server which is concurrently used there is no such thing as "middle of the page". However you are right, for single user applications it might make sense to hint the GC. I guess this is where RT-Java will come to an rescue.
(And well, malloc heap reorganisaton in C is not that pleasant, also).
Gruss
Bernd
Well, the JbossMQ in 4.x may be rocking, in 3.x the cluster support, redundancy and routing is pretty weak compared to ther JMS solutions (commercial and non commercial).
Ever since the FWTK offered a semi free toolset, the community failed to develop real free simple, stable and secure application level proxies.
There are some more now, but most have discovered bugs due to missing deffensive programming.
That was one of the reasons I started freefire.org, even when the mailing list currently is not used.
--
www.eckes.org
Well, I think there wont be a single Framework. Because for one we do have this Gnome/KDE conflict and second, various distributions judge the DRM and copyright problems differently. At least for non-streamed content integration like wincodecs or alterative decoders will not be present officially in a vendor supported player. Especially not when it is backed by Real Networks.
--
www.eckes.org
I wonder what the point of this article is?
And why is it a virtual office, of you use physical computers? If it is the work place at home, call it Soho, or call it the workplace of a telecommuter, but I dont see what the virtual here is, besides a disturbing buzz-word.
And by the way, did I miss the content of this article? It is just listening some well known web sites. Where is the news?
Well Well, the fight between UN/CEFACT and OASIS was obvious a long time ago. Ever since OASIS started to work on UBL, they stepped on the foot of CEFACT, by working on message content definitions.
So it is no wonder, that CEFACT is reorienting themself. This is not only becuase of the personal changes. But of course those also have some meaning.
The UN work is not realy money driven, but highly dependend on individual work. I know that a lot of ppl ware working at CEFACT or OASIS TCs very enthusiastically, and it is not a cheap thing for a company to sponsor chairs, editors or contributors.
BTW: I was personally very impressed on the quick and quality work the ebXML open process had produced. The work realy slowed down, after it moved to OASIS TCs and to other CEFACT Working Groups. And I had the impression, that the CEFACT was the more modern and agile body to carry on the work. OASIS has pretty strict attendence rules for paying members - you have to torture yourself to be part of the community.
Greetings
Bernd
> This Linux exploit had ALREADY been patched.
After it was known for month, and just too late discovered to be shipped with 2.4.22. Now, where is the difference?
JavaVMs had exploits in the past, so I do not see ground for your claim that they are the most safe VMs. I know simulations of turing engines which are far more secure and less exploitable.
0 1.html
Remember Brown Orifice in netscape VMs?
Some more: http://www.securitytracker.com/archives/target/13
There is a problem with Trademark Laws. In some countries you have to actually defend every abuse of your Trademark, or otherwise you wont get it granted anymore.
If there is a public display of that Logo on a high profile site, it has to be lawfully attacked. But on the other hand, I can think of a easy regulation for allowing the PCI Logo to devices which are actually PCI devices.
very funny :)
No the idea is simple. If the image of your disk is not a 1:1 copy of the bits, but understands the structure of the files and the file system (like a tar backup would), you can restore it on any sized disk, as long as the data fits in the free space.
I suggest you have a look at partimage.org, it understands about filesystems (UFS, HFS, HPFS, XFS, JFS, NTFS, FAT, ext2, Reiser) and is therefoe better for resizing and efficient dumping of filesystems.
Well, I understand "handling a file system" more like:
- Beeing able to resize it
- Beeing able to only save used blocks
- Possibly beeing able to partially restore it
- Possible beeing able to relocate it
All this is not supported by the "dd" method. Therefore I prefer tools like partimage, see http://www.partimage.org.
Of course this does not support all Filesystems, but it will be expanded more and more.
The clause in question is included in most commercially seeded open source licenses. The MPL was one of the fiest with it. The idea is, that the company who opens up the source code still has the right to use it for properitary closed source products. And in addition to that, all patches to the free and open version can be used by the company.
In fact if you publish source under the GPL you are free to use that source in another commercial product and do not need to open up the source. But as soon as you accept paptches, you cant include those patches in your closed product, cause they will be under the GPL and you do not have the copyright.
Thats why you see licenses like the MPL. In fact it means a company treats its source base against patches. This can be "fair" or "unfair" to the community, but it is OSI compliant, since it does not restrict the open branch of a product. It only grants additional rights to the original licensor.
Greetings
Bernd
virtual punch cards (not real ones) are acutlly in use on some old mainframes. You copy those card jobs to tapes and read from a more stable media. But still it is 80col card jobs. JCL or some other stuff.
Used to work with Siemens BS1000 Systems that way, 10 years back in time.
Same is true for Java Applications, they soon will run under 64Bit. So I guess there is not much time untile W32OW64 is not needed anymore :)
Currently the code quality of 64bit Linux is pretty bad due to gcc problems, but once those are ruled out (i.e. gcc3.0 is stable enough to compile everything) this wont be a big issue.
Actually the swirl is a heavyly discussed marketing instrument for debian to rule the world, you may look at ... for other market leaders facilidating the swirl to impress buyers.
Critique
I just wonder what exactly will be sold by debian :) Perhaps a good cup of tea to make the waiting more interesting?
Anyway, I do recommend to read "Debian weekly"
Debian Weekly
Greetings :)
Bernd
(Debian Developer, I admit