No. When ReiserFS was included in the kernel most people felt it was the right choice to make. Suse was already including it in their kernel and I believe that Mandrake was too.
>>Vendors generally modularize everything they can.
The drivers at least could have been modularized and are a vendor bug.
>>But there are lots of important configuration choices that cannot be modularized. And that, again, is a kernel architecture problem.
I question the "lots" and the "important" but if that's the problem then that is what needs to be fixed.
If so many bits and pieces of functionality need to be approved by one person, that really is an architecture problem.
Well... Mostly Linus is just a fall guy. Not many people feel that MOSIX needs to be included into the standard kernel. I think that RML's patches show that people are more than willing to patch -p0 the kernel themselves. This is especially true for cluster people compared to desktop users.
Your concerns about the build system and configuration are real and being addressed in the 2.5 kernel. There has been tons of discussion about this and I'm surprised you seem to have missed all the articles/debate about it.
Every single machine I have (a few dekstops and a few laptops) has required recompilation, even with recent kernel installs.
That is the real problem. All the talk about how it needs to be simple to compile the kernel is really just a work around for a poor default kernel. File a bug report with your vendor.
The other poster nearly got it right. The horizon is a combination of T (the number of times a query is repeated before it dies) and N (the number of connections each computer has). The values for the first client were T=5 and N=4. I'm fairly sure that most clients must have raised the value for T.
His chart called "reachable users" describes how the horizon grows as T or N change.
I think now that there are normally over 1000 people in your horizon possibly up to 8000.
The other thing about the article is that it was written before clients started caching replies and that changes your horizon around quite a bit.
Quite frankly caching the replies probably helps but the Gnutella protocol is still awful.
I'm more impressed with Morpheus as a decentralized file sharing network. There is an open source Morpheus client called "gift."
The weird thing is that the only way to get documentation about how Morpheus works is to download the source tarball for gift and poke around in the READMEs. There is no other public documentation for it any where on the net.
Basically it sets up tons of little mini servers that index songs for up to around 300 people. Clients have a list of these servers and query them to find files. If you want to a "horizon" of 6000 computers then you only have to make 20 or 30 queries. In Gnutella (without caching) the same horizon would be 6000 queries. No one really knows what it would be with caching and it changes depending on whether it's a popular query or not.
Actually Gnutella in that case is much worse than just 6000 queries because many computers have no songs shared and are stilled searched where in Morpheus computers that don't share songs do not index. And another thing that makes Gnutella worse is that I think the replies are relayed multiple times instead of just once.
I'm not a gift developer or user myself... But I would say it was a far better way to go than Gnutella.
Seven years ago viruses came on floppy disks and they traveled slowly. Now a smart virus could infect millions of computers within an hour. This is an enormous threat and it is only through luck that no very malicious internet viruses have been written already.
While Microsoft has a shocking attitude towards security, the real problem is not their software itself. The problem is that they are a monopoly. If everyone runs the same software, even a small vulnerability can bring the entire network down.
Microsoft should be more security conscious but that really does not solve the core problem.
Unfortunately, most people do not see security as enough of a priority to deal with the cost and hassle of changing software. The only solution I can think of is to encourage people to make backups. Backups do not help when a virus destroys hardware but they are better than nothing.
Eventually, there will be a truly devastating internet virus. We have gotten lucky this far but our luck will not carry us indefinitely.
The big thing of course is that it's cheaper. WinCE and palmOS cost something like $10-$15 per unit. This adds up if you sell very many thousands of units.
Face it, most users don't even know what an OS is could care less what OS runs their PDA so long as it does what they want. This PDA is especially nice because it comes with windows apps for your desktop so the PDA and desktop can communicate.
>> OK, for one, the ubersimple install script only works for ubersimple apps and still leaves all your.o files hanging out there
It require packaging applications slightly differently. I prefer this way of packaging applications, not just for the security reasons, but because it is easier to uninstall. Changing.tar.gz files into.debs is more complicated than most people want to deal with. The.o files are easy to handle with more simlinks.
>>not to mention he forgot the -r on cp
That was a test to see if you were paying attention.:)
>>a lot of us do run apache, but some run TUX, and there are others;
I would say that easily 90% of Linux web servers use Apache. I consider that a dangerous level. Bind is also dangerously popular.
I was surprised you did not mention exim as an email server because it is the default email server on debian. I considered not putting email servers on my list.
Linux is not immune to internet worms, or have you forgotten the Ramen worm? Imagine if the worm had used new vulnerabilities instead of old ones. Ramen was not as widespread as "ILOVEYOU", code red, etc, etc, etc, but it is still a threat.
Viruses and the internet.
on
Linux Virus Alert
·
· Score: 4, Interesting
I remember when slashdot first talked about the RST trojan. That time Qualys did an abysmal job reporting on the virus. (Read the comments on the article.)
The good thing is that apparently there was not a single case where this virus infected anyones computer except for the anonymous person who reported it to Qualys. This new virus is at least three times more dangerous because three different groups have seen it.:P
The most difficulty part with this type of virus is getting people to run it as root. The easiest way would be to install the virus through a Makefile which are often run as root. This is one reason I think the standard tar.gz install should be:
#-----
zcat foo.tar.gz | tar -xv
if source
cd foo/
./configure
make
fi
cd..
su
cp foo/usr/local/tar/
ln -s/usr/local/bin/foo/usr/local/tar/foo/foo
#-----
Makefiles are too complex for most people to read but a script that installed things my way would only be 5 lines executed as root and thus easy to audit.
(Normal.debs would install normally because debian developers are trusted.)
On a completely unrelated topic, this virus can't spread very well. Linux users download packages from central repositories but they don't share ordinary binaries amongst themselves. The virus only infects elf excecutable files where in Windows it could infect emails and.doc files and all kinds of stuff that should be data but instead is executable.
These days, the only dangerous way to spread a virus is through an internet worm. Linux is vulnerable to worms because almost everyone uses the same kernel, webserver, dns, and email server. If we could diversify these things, it would make Linux less vulnerable to worms.
I know people are going to say that Linux is already more secure than Microsoft. That's true but it's because Microsoft does not care about security or threats to the internet. A truly malicious virus could cost billions of dollars in lost hardware and take out the American phone system for weeks.
I'm fairly sure Micheal Hall is aware of the three debian distributions...
I would bet that he reads over 50 linux articles every day including Saturday and Sunday. He basically posts every single article on linuxtoday.com.
I was pretty glad when debian started the testing distribution. After the testing distro was created there wasn't as much need for Progeny.
I use unstable myself, but I realize that a lot of people aren't comfortable updating as many programs as I do as often as I do. These people would be more happy with frequent stable releases.
Btw. I don't understand why Slashdot puts the extra space in URL... Is that supposed to protect someone from accidentally highlighting the URL and then middle clicking in mozilla and being miraculously transported to the page?
Yeah. You're right. I'm actually aware of the gimp-print project. But I wasn't thinking about printing, but more using CYMK directly throughout the program. I'm no expert but I think that you lose some quality converting images from CYMK to RGB and back?
I did some looking around and it turns out that CMYK color space is probably going to be supported in 2.0. I thought that it wasn't.
To me the stuff in gegl looks like it grew out of a need to work around patent issues with various color algorithms. It lets people build proprietary plugins for thing that are illegal to do in open source.
That's very frustrating.
I wish they didn't allow people to patent algorithms and we could go back to the way things were before.
But you are right. Aparently they found a way to work around all the patents in that area.
>>I've heard this same, untrue statement for each windowing system Microsoft has ever released, including WINDOWS M.E.!
Believe it this time. Probably Linux uptimes are still better but XP really is aceptably stable.
Your probably right that I am too optomistic about Linux on the desktop. That is the curse of the computer scientist. It's going to happen some time but you're right that it could take a long time.
Reasonable estimates put Linux web browsing at 1%. Of course Linux users are going to go to LowEndMac just the same as they are more likely to surf something like OsOpinion.
OS X runs on Unix, that's true. But it doesn't really apeal to the same market as Linux. It doesn't have any more games that Linux does. I personally like OS X, but most of my classmates still mock anything to do with Macingtosh, so OS X is not l33t. Finally, Apple has expensive hardware.
Windows XP is a far far bigger threat to Linux on the desktop. Face it, Windows 9x operating systems were utter crap. They were the biggest reason to use Linux ever. With Windows XP, Microsoft has finally created a operating system that doesn't fall over every three minutes.
Of course, Windows XP, isn't going to stop Linux on the desktop because Linux is cheaper. In the next couple years I expect more and more coorporations to use Linux on the desktop to save money.
>>Alright, maybe I posted a little too soon, but shouldn't "flamebait" be attracting flaming responses?
It's because the good people of slashdot recognized your feeble attempt at attracting flames and did not deign to give you the pleasure.
Slashdotters are known world wide as the most discerning and reasonable of all people. Please don't consider a personal afront that you were not flamed. Lesser mortals most certainly would have.
I guess you were right after all. I don't know how I missed it at first.
And that Nerd Literalist trap! Hoooweee! Was that ever creative or what?!? I would have never expected something like that from a complete moron such as yourself.
Evidently, you didn't read my comment. Those examples you mention weren't designed. If you claim that they were then you are an idiot but I won't bother to mock you.
As you didn't read my post, I don't feel I have to respond any further to yours.
Alan Cooper says that programmers can't design good programs but he is wrong. Programmers can learn to design programs as well as anyone else. The problem is that no one else is particularly good at designing programs either.
I remember a part in Alan Cooper's book _The Inmates are Running the Asylum_ where he was complaining about stupid car door remote controls. He had various gripes but at the end he said something like, "The other thing that annoys me is that sometimes other people's remotes accidentally unlock your door. There should be another button to put the car into secure mode where that won't happen." Um... HELLO??? A option for security? There shouldn't even be a choice about something as basic as that. When the car locks it should darn well be secure.
The problem is that Alan Cooper didn't think about what he was designing. He just threw the option for "added security" on at the end without thinking about it.
Most programmers don't think about it either.
In real life though customers and bosses often end up designing the user interface and they are just as bad. Worse even...
In fact, I wouldn't call this designing because there is no actual design process involved, the boss will just draw something up off the top of her head and have the programmer write it.
The biggest UI problem with applications today is not that they were poorly designed but that they weren't designed at all.
Part of the problem is education. Most CS graduates have never had a single UI course.
In the game you have all these really wacky characters that get together and save the world. In the movie you have a bunch of scientists and stuff.
I love the scenery in the game: the quaint villages, the jungles, and the towns. But the movie was mostly indoors in a big city.
It's almost as if they tried to make the movie appeal to a more adult audience but missed the target.
The vilian in the movie was not evil enough but merely misguided.
But mostly I wanted to see more colors. More old fashioned clothing mixed with futuristic clothing. More kids running around causing havoc. More jokes.
Slashdot Mirror
No. When ReiserFS was included in the kernel most people felt it was the right choice to make. Suse was already including it in their kernel and I believe that Mandrake was too.
>>Vendors generally modularize everything they can.
The drivers at least could have been modularized and are a vendor bug.
>>But there are lots of important configuration choices that cannot be modularized. And that, again, is a kernel architecture problem.
I question the "lots" and the "important" but if that's the problem then that is what needs to be fixed.
Well... Mostly Linus is just a fall guy. Not many people feel that MOSIX needs to be included into the standard kernel. I think that RML's patches show that people are more than willing to patch -p0 the kernel themselves. This is especially true for cluster people compared to desktop users.
Your concerns about the build system and configuration are real and being addressed in the 2.5 kernel. There has been tons of discussion about this and I'm surprised you seem to have missed all the articles/debate about it.
Every single machine I have (a few dekstops and a few laptops) has required recompilation, even with recent kernel installs.
That is the real problem. All the talk about how it needs to be simple to compile the kernel is really just a work around for a poor default kernel. File a bug report with your vendor.
His chart called "reachable users" describes how the horizon grows as T or N change.
I think now that there are normally over 1000 people in your horizon possibly up to 8000.
The other thing about the article is that it was written before clients started caching replies and that changes your horizon around quite a bit.
Quite frankly caching the replies probably helps but the Gnutella protocol is still awful.
I'm more impressed with Morpheus as a decentralized file sharing network. There is an open source Morpheus client called "gift."
The weird thing is that the only way to get documentation about how Morpheus works is to download the source tarball for gift and poke around in the READMEs. There is no other public documentation for it any where on the net.
Basically it sets up tons of little mini servers that index songs for up to around 300 people. Clients have a list of these servers and query them to find files. If you want to a "horizon" of 6000 computers then you only have to make 20 or 30 queries. In Gnutella (without caching) the same horizon would be 6000 queries. No one really knows what it would be with caching and it changes depending on whether it's a popular query or not.
Actually Gnutella in that case is much worse than just 6000 queries because many computers have no songs shared and are stilled searched where in Morpheus computers that don't share songs do not index. And another thing that makes Gnutella worse is that I think the replies are relayed multiple times instead of just once.
I'm not a gift developer or user myself... But I would say it was a far better way to go than Gnutella.
While Microsoft has a shocking attitude towards security, the real problem is not their software itself. The problem is that they are a monopoly. If everyone runs the same software, even a small vulnerability can bring the entire network down.
Microsoft should be more security conscious but that really does not solve the core problem.
Unfortunately, most people do not see security as enough of a priority to deal with the cost and hassle of changing software. The only solution I can think of is to encourage people to make backups. Backups do not help when a virus destroys hardware but they are better than nothing.
Eventually, there will be a truly devastating internet virus. We have gotten lucky this far but our luck will not carry us indefinitely.
One way to have your own way is to buy a radeon. It has an open source driver. I would buy that before I baught a geforce.
But then again, I really like to get my own way.
Face it, most users don't even know what an OS is could care less what OS runs their PDA so long as it does what they want. This PDA is especially nice because it comes with windows apps for your desktop so the PDA and desktop can communicate.
It require packaging applications slightly differently. I prefer this way of packaging applications, not just for the security reasons, but because it is easier to uninstall. Changing
>>not to mention he forgot the -r on cp
That was a test to see if you were paying attention.
>>a lot of us do run apache, but some run TUX, and there are others;
I would say that easily 90% of Linux web servers use Apache. I consider that a dangerous level. Bind is also dangerously popular.
I was surprised you did not mention exim as an email server because it is the default email server on debian. I considered not putting email servers on my list.
Linux is not immune to internet worms, or have you forgotten the Ramen worm? Imagine if the worm had used new vulnerabilities instead of old ones. Ramen was not as widespread as "ILOVEYOU", code red, etc, etc, etc, but it is still a threat.
The good thing is that apparently there was not a single case where this virus infected anyones computer except for the anonymous person who reported it to Qualys. This new virus is at least three times more dangerous because three different groups have seen it.
The most difficulty part with this type of virus is getting people to run it as root. The easiest way would be to install the virus through a Makefile which are often run as root. This is one reason I think the standard tar.gz install should be:
#-----
zcat foo.tar.gz | tar -xv
if source
cd foo/
make
fi
cd
su
cp foo
ln -s
#-----
Makefiles are too complex for most people to read but a script that installed things my way would only be 5 lines executed as root and thus easy to audit.
(Normal
On a completely unrelated topic, this virus can't spread very well. Linux users download packages from central repositories but they don't share ordinary binaries amongst themselves. The virus only infects elf excecutable files where in Windows it could infect emails and
These days, the only dangerous way to spread a virus is through an internet worm. Linux is vulnerable to worms because almost everyone uses the same kernel, webserver, dns, and email server. If we could diversify these things, it would make Linux less vulnerable to worms.
I know people are going to say that Linux is already more secure than Microsoft. That's true but it's because Microsoft does not care about security or threats to the internet. A truly malicious virus could cost billions of dollars in lost hardware and take out the American phone system for weeks.
People don't want to switch from mp3. People don't have hardware that plays
The same comments were made for ogg and were moderated up to +5 for the last ogg article.
You are not as oppressed as you believe.
I'm fairly sure Micheal Hall is aware of the three debian distributions...
I would bet that he reads over 50 linux articles every day including Saturday and Sunday. He basically posts every single article on linuxtoday.com.
I was pretty glad when debian started the testing distribution. After the testing distro was created there wasn't as much need for Progeny.
I use unstable myself, but I realize that a lot of people aren't comfortable updating as many programs as I do as often as I do. These people would be more happy with frequent stable releases.
Can I encourage you to grow a thicker skin?
Wrong article. This was meant to go with the dave jones article but I guess I had too many windows open.
Dave Jones and Arnaldo Carvalho de Melo gave a talk on irc about it earlier this month. You can find the transcript at http://umeet.uninet.edu/umeet2001/talk/15-12-2001
Btw. I don't understand why Slashdot puts the extra space in URL... Is that supposed to protect someone from accidentally highlighting the URL and then middle clicking in mozilla and being miraculously transported to the page?
I'm just curious where you got that information?
I did some looking around and it turns out that CMYK color space is probably going to be supported in 2.0. I thought that it wasn't.
To me the stuff in gegl looks like it grew out of a need to work around patent issues with various color algorithms. It lets people build proprietary plugins for thing that are illegal to do in open source.
That's very frustrating.
I wish they didn't allow people to patent algorithms and we could go back to the way things were before.
But you are right. Aparently they found a way to work around all the patents in that area.
That stuff is patented and thus illegal.
Instead of complaining to developers complain to your congressman because they are the only people that can change it.
>>I've heard this same, untrue statement for each windowing system Microsoft has ever released, including WINDOWS M.E.!
Believe it this time. Probably Linux uptimes are still better but XP really is aceptably stable.
Your probably right that I am too optomistic about Linux on the desktop. That is the curse of the computer scientist. It's going to happen some time but you're right that it could take a long time.
OS X runs on Unix, that's true. But it doesn't really apeal to the same market as Linux. It doesn't have any more games that Linux does. I personally like OS X, but most of my classmates still mock anything to do with Macingtosh, so OS X is not l33t. Finally, Apple has expensive hardware.
Windows XP is a far far bigger threat to Linux on the desktop. Face it, Windows 9x operating systems were utter crap. They were the biggest reason to use Linux ever. With Windows XP, Microsoft has finally created a operating system that doesn't fall over every three minutes.
Of course, Windows XP, isn't going to stop Linux on the desktop because Linux is cheaper. In the next couple years I expect more and more coorporations to use Linux on the desktop to save money.
It's because the good people of slashdot recognized your feeble attempt at attracting flames and did not deign to give you the pleasure.
Slashdotters are known world wide as the most discerning and reasonable of all people. Please don't consider a personal afront that you were not flamed. Lesser mortals most certainly would have.
But I can't say that because Quake II isn't that old yet.
Instead I'll just say a big, "Thank you," to ID. It's darn, darn cool to open the source on this.
I guess you were right after all. I don't know how I missed it at first.
And that Nerd Literalist trap! Hoooweee! Was that ever creative or what?!? I would have never expected something like that from a complete moron such as yourself.
As you didn't read my post, I don't feel I have to respond any further to yours.
Alan Cooper says that programmers can't design good programs but he is wrong. Programmers can learn to design programs as well as anyone else. The problem is that no one else is particularly good at designing programs either.
I remember a part in Alan Cooper's book _The Inmates are Running the Asylum_ where he was complaining about stupid car door remote controls. He had various gripes but at the end he said something like, "The other thing that annoys me is that sometimes other people's remotes accidentally unlock your door. There should be another button to put the car into secure mode where that won't happen." Um... HELLO??? A option for security? There shouldn't even be a choice about something as basic as that. When the car locks it should darn well be secure.
The problem is that Alan Cooper didn't think about what he was designing. He just threw the option for "added security" on at the end without thinking about it.
Most programmers don't think about it either.
In real life though customers and bosses often end up designing the user interface and they are just as bad. Worse even...
In fact, I wouldn't call this designing because there is no actual design process involved, the boss will just draw something up off the top of her head and have the programmer write it.
The biggest UI problem with applications today is not that they were poorly designed but that they weren't designed at all.
Part of the problem is education. Most CS graduates have never had a single UI course.
In the game you have all these really wacky characters that get together and save the world. In the movie you have a bunch of scientists and stuff.
I love the scenery in the game: the quaint villages, the jungles, and the towns. But the movie was mostly indoors in a big city.
It's almost as if they tried to make the movie appeal to a more adult audience but missed the target.
The vilian in the movie was not evil enough but merely misguided.
But mostly I wanted to see more colors. More old fashioned clothing mixed with futuristic clothing. More kids running around causing havoc. More jokes.
(Not the person who reported the problem or made the decision to put the docs in non-free, but the guy who gets to go through and fix it.)