GPG is commonly used to sign source code tarballs such as the linux kernel. Those tarballs are mirrored across the world to hundreds of untrusted servers. With this flaw it's possible to modify signed source code (and introduce backdoors for instance). It's definitely not a theoretical problem.
Simplified it's this: Standard X servers can only draw windows to the screen (framebuffer) directly. Real translucency can not be implemented with this technology.
The X composite extension allows an X server to render windows to memory first. The composite manager is then responsible to create the final image, composed of all the windows. A composite manager may opt to render some windows translucent over others for example.
Why is the audio quality of the movie so bad? Like this their American accent makes it very hard to understand anything if you are not a native English speaker.
Netwosix will try to offer the first valid alternative to historically secure systems like the *BSD stability, performance, and support for mission-critical application deployments users can benefit from outstanding robustness, scalabilty, and reliability
Big words. If he wants stability why does this 19 year old jerk use the latest 2.6.14.5 stable kernel? Where do all those features come from? You won't get security and robustness by just repackaging software. Is Ciaglia saying he reviews all the code in his distro and fixes broken stuff? All on his own? On several platforms? Look at their site: Their "community" consists of merely two people.
Goodness, just because every kid can burn a bootable Linux CD that doesn't mean he just created a new distro!
Call this a flame. But I really don't see the point of Netwosix.
I'm planning to expand the portage tree, doing other 300/400 packages ports, with everyone security oriented.
Is this distro based on Gentoo's portage? If so, what is this distribution trying to fix? And is this guy "porting" packages all alone? He probably never figured how much work it will be to maintain his own distro.
I am not a medical expert, but I guess it is really "instant". I heard stuff like that generally produces an enormous shock to the body. Muscles contrat. That causes the blood pressure to instantly rise over a critical level which renders you at least inconcious if it doesn't even damage your brain.
Yes they are. And they are closed source. And the nVidia ones don't work well with newer kernels. And they stopped supporting them. And the nv open source driver does not have accelerated OpenGL.:-(
There are stilll many vendors that do not provide hardware documentation so that it is just very very hard if not even impossible to write driver for their hardware. Best example is nVidia and ATI graphics boards or Cannon scanners. All drivers we have for them are made from reverse engineering the hardware! Goodness, I remember the times when my Star 9-pin printer came with a book that contained a detailed description of the printer protocol. Or a OKI laser printer whose user manual documented PCL!
Yes, it is true. Being a software devloper myself I really care about what I wear. I am always really pissed of when I see those "you get a free T-Shirt" taglines that go along with every con announced. This T-Shirt thing is just too American. And Think Geek doesn't help at all. Hey, those oversized, ugly labelled Ts were fashionable in the 80ies maybe. That is now 25 years ago. Who are you gonna impress with your "Java", "Oracle", "Apache Committer", "Google Summer Of Code", "SYN/ACK" Ts anyway? Non-IT guys won't understand. IT guys already know them. Girls don't care about the label but how you look in one of these. And let me tell you: everybody looks crap in them.
1. The cow is not a wooden box. When pressure is applied the cow will bend or move its feet for instance which would instantly change the whole distribution of forces. 2. Who said that tipping was about statics anyway? Can this scientist evaluate the case when you use momentum?
For USB and Firewire that's already done. Both busses define like device classes and protocols those use. The actual devices need either no special drivers or those can be implemented in user space. Printer drivers (CUPS) are completely user space. Scanner drivers for Sane are completely user space.
In my opinion a software engineer that codes an interface without documentation should loose his degree. But this is API. The original disussion was about Specs.
I have one really good example: ACPI. It is an enourmously complex spec. And the reality does not follow the specs at all. Hardware manufactures work so closely together with Microsoft that they release the hardware when it runs Windows. No Wonder those poor Linux kernel guys never get a chance to get it working correctly. There are just too many bugs in the hardware and they end up writing work-arounds for many cases.
So this is apparently just another site in the Microsoft Wide Web where they use IE and WINS and IIS. Any plans they will put that on a separate physical network soon?
Well, MySQL at least *has* some collation support. The bad thing is how it is used in SQL: SELECT X FROM T ORDER BY X COLLATE collation_name; This seems nice at first. The catch is the collation names. Imagine you write an application for international customers. A person from Paris would like a french collation, while a person from Israel definitely needs a different collation. So the collation to use strongly depends on the locale of the person who is logged on. Usually people use the ISO country and language codes (en-US, fr-CA etc.) in their applications to define a locale. Those codes are also used in Java's Locale class for instance. Now the MySQL collation names are like this: latin1_german1_ci, latin1_spanish_ci, etc. This does not exactly match with anything of the above. And it makes it harder than necessary for developers to select the right collation for a certain locale. You need some kind of mapping table. Moreover you can not be sure that every MySQL installation has support for all the collations you use, because they can be selected at compile time.
When you compare this with the way Oracle handles that MySQL looks pretty bad. For me Oracle is the only DB that handles collation well-enough from a developer's point of view.
And hey, what about time zones? When I store a time and date in a DB, it must be well-defined in what TZ the data is returned. Oracle for instance does quite a bad job regarding that.
I don't really have a favourite among the two. I use a DB mostly through EJB containers. So all those nifty features are of no use because the DB layer won't use them. They don't do much more than the standard SQL.
Slashdot Mirror
More precisely, a browser is free to render invalid pages ANY way it wants - as the standard doesn't say how to render non-standards :-)
Mirrordot
There are GPG signatures too (don't know if they use MD5 or SHA-1): ftp://mirror.switch.ch/mirror/kernel/linux/kernel/ v2.6/linux-2.6.15.6.tar.bz2.sign
But as pointed out earlier the bug only affects "inline" signatures.
GPG is commonly used to sign source code tarballs such as the linux kernel. Those tarballs are mirrored across the world to hundreds of untrusted servers. With this flaw it's possible to modify signed source code (and introduce backdoors for instance). It's definitely not a theoretical problem.
You can do that with the base64 URL scheme. See here.
./ed so I couldn't check...
It is supported by Firefox already. Is that what they use? Site is
Simplified it's this: Standard X servers can only draw windows to the screen (framebuffer) directly. Real translucency can not be implemented with this technology.
The X composite extension allows an X server to render windows to memory first. The composite manager is then responsible to create the final image, composed of all the windows. A composite manager may opt to render some windows translucent over others for example.
Wouldn't a single journaling filesystem transaction be considered three independant writes?
No. A single transaction comes from a single thread. So the IO scheduler has no freedom here. It consists of these operations:
1. write redo log
2. write
3. clear redo log
They must occur in exactly this order. There are flush operations involved as well but I am not an expert here.
The IO scheduler should not matter as they are only important when multiple processes access the disk.
Why is the audio quality of the movie so bad? Like this their American accent makes it very hard to understand anything if you are not a native English speaker.
Netwosix will try to offer the first valid alternative to historically secure systems like the *BSD
stability, performance, and support for mission-critical application deployments
users can benefit from outstanding robustness, scalabilty, and reliability
Big words. If he wants stability why does this 19 year old jerk use the latest 2.6.14.5 stable kernel? Where do all those features come from? You won't get security and robustness by just repackaging software. Is Ciaglia saying he reviews all the code in his distro and fixes broken stuff? All on his own? On several platforms? Look at their site: Their "community" consists of merely two people.
Goodness, just because every kid can burn a bootable Linux CD that doesn't mean he just created a new distro!
Call this a flame. But I really don't see the point of Netwosix.
I'm planning to expand the portage tree, doing other 300/400 packages ports, with everyone security oriented.
Is this distro based on Gentoo's portage? If so, what is this distribution trying to fix? And is this guy "porting" packages all alone? He probably never figured how much work it will be to maintain his own distro.
The MS product
- is only available for Windows
- displays ads
- has a bad user interface (you can't even rename contacts)
I am not a medical expert, but I guess it is really "instant". I heard stuff like that generally produces an enormous shock to the body. Muscles contrat. That causes the blood pressure to instantly rise over a critical level which renders you at least inconcious if it doesn't even damage your brain.
Yes they are. And they are closed source. And the nVidia ones don't work well with newer kernels. And they stopped supporting them. And the nv open source driver does not have accelerated OpenGL. :-(
There are stilll many vendors that do not provide hardware documentation so that it is just very very hard if not even impossible to write driver for their hardware. Best example is nVidia and ATI graphics boards or Cannon scanners. All drivers we have for them are made from reverse engineering the hardware! Goodness, I remember the times when my Star 9-pin printer came with a book that contained a detailed description of the printer protocol. Or a OKI laser printer whose user manual documented PCL!
Coral here
Yes, it is true. Being a software devloper myself I really care about what I wear. I am always really pissed of when I see those "you get a free T-Shirt" taglines that go along with every con announced. This T-Shirt thing is just too American. And Think Geek doesn't help at all. Hey, those oversized, ugly labelled Ts were fashionable in the 80ies maybe. That is now 25 years ago. Who are you gonna impress with your "Java", "Oracle", "Apache Committer", "Google Summer Of Code", "SYN/ACK" Ts anyway? Non-IT guys won't understand. IT guys already know them. Girls don't care about the label but how you look in one of these. And let me tell you: everybody looks crap in them.
1. The cow is not a wooden box. When pressure is applied the cow will bend or move its feet for instance which would instantly change the whole distribution of forces.
2. Who said that tipping was about statics anyway? Can this scientist evaluate the case when you use momentum?
For USB and Firewire that's already done. Both busses define like device classes and protocols those use. The actual devices need either no special drivers or those can be implemented in user space. Printer drivers (CUPS) are completely user space. Scanner drivers for Sane are completely user space.
In my opinion a software engineer that codes an interface without documentation should loose his degree. But this is API. The original disussion was about Specs.
I have one really good example: ACPI. It is an enourmously complex spec. And the reality does not follow the specs at all. Hardware manufactures work so closely together with Microsoft that they release the hardware when it runs Windows. No Wonder those poor Linux kernel guys never get a chance to get it working correctly. There are just too many bugs in the hardware and they end up writing work-arounds for many cases.
Any other examples? USB anybody?
It is common practice among ISPs to enable DNS wildcards for subdomains by default. hoststar.ch is doing that for instance.
So this is apparently just another site in the Microsoft Wide Web where they use IE and WINS and IIS. Any plans they will put that on a separate physical network soon?
You can not fix broken brains with new laws.
Well, MySQL at least *has* some collation support. The bad thing is how it is used in SQL:
SELECT X FROM T ORDER BY X COLLATE collation_name;
This seems nice at first. The catch is the collation names. Imagine you write an application for international customers. A person from Paris would like a french collation, while a person from Israel definitely needs a different collation. So the collation to use strongly depends on the locale of the person who is logged on. Usually people use the ISO country and language codes (en-US, fr-CA etc.) in their applications to define a locale. Those codes are also used in Java's Locale class for instance. Now the MySQL collation names are like this: latin1_german1_ci, latin1_spanish_ci, etc.
This does not exactly match with anything of the above. And it makes it harder than necessary for developers to select the right collation for a certain locale. You need some kind of mapping table. Moreover you can not be sure that every MySQL installation has support for all the collations you use, because they can be selected at compile time.
When you compare this with the way Oracle handles that MySQL looks pretty bad. For me Oracle is the only DB that handles collation well-enough from a developer's point of view.
And hey, what about time zones? When I store a time and date in a DB, it must be well-defined in what TZ the data is returned. Oracle for instance does quite a bad job regarding that.
I don't really have a favourite among the two. I use a DB mostly through EJB containers. So all those nifty features are of no use because the DB layer won't use them. They don't do much more than the standard SQL.