Can YOU imagine being on the end of the phone, trying to help somebody recover a server and every few minutes find that they were running with experimental kernel patches, or ancient/buggy software, or that a fault seemed to be caused by a random frob off SourceForge that you'd never heard of?
No, I can't imagine Hell. I can, however, imagine a company claiming to support Linux at least making an attempt to support kernels that are fully official and straight from Linus. This is what Hans was complaining about: that some distros don't support official releases of the kernel because they don't like this or that feature.
Sure, it's all well and good to borrow stock with the intent of selling it, buying it back at a lower price, and returning it to its original owner. But you still have to find someone to sell it to.
Remember that although this worm could have executed any code it wanted on all of those hosts, it had no malicious payload
Let's think of a worst-case scenario, here...
The worm had a program to propagate itself in a space of 376 bytes. It had up to, what, 1500 bytes to carry whatever program it wished? Let's say it used those 1500 bytes to set up a program that would listen on a particular TCP port for instructions from the author's computer. Then, rather than propagating itself as fast as possible, it sends out a packet every few minutes, gradually and insidiously infecting all MSSQL servers on the Internet.
The 1100 extra bytes are used to write a program to disk, and then launch it. This program listens for connections on some high port, or perhaps just listens for UDP packets of a certain description (since it knows the firewall lets those through). At first, it simply catches all worm packets and records the IP addresses, so that it knows what other hosts are infected.
The author's computer listens for these packets, and makes a similar list of infected hosts. Then, when the time is ripe, he starts sending additional instructions to those hosts.
The hosts receive the new instructions, modify their program based on the contents, and then echo the packet out to the hosts in their lists. The author numbers the instruction packets, and the hosts make a note of which ones they've received and ignore repeats. That way, once all infected hosts are updated, the patches stop flying around.
One of the first instructions to be sent out is to make the program launch at boot time. Then, the infected computers are sent instructions to stop propagating themselves. They're sent instructions to report back to the original source. The author looks at the hosts, sends out special non-propagating instructions to military hosts to send him their data. He sends out instructions to hosts that may have access to credit card databases to send him the numbers and expiration dates. He gathers whatever other information he deems useful.
Then, he sends out an instruction for all hosts to delete all data from all databases.
How difficult would it be to write the initial program for that? How difficult to make those patches, and make them work? My guess is, someone who knows assembly well could pull it off. It may take a fair amount of time and patience, but the amount of money to be made is pretty considerable and could make it worthwhile. Hey, if I were going to write a malicious worm, that's how I would go about it.
But the most pertinent question is, how many MSSQL servers are still out there, unpatched, vulnerable, serving critical data?
To reiterate my original post. First you decide what your requirements for Good Enough are, then you choose the OS that provides you with that level of support. What can you possibly argue with there?
Well, yes. Your original post was that WinCE is Good Enough (TM). The problem is that you said that it's Good Enough for laser eye surgery. You didn't say it's Good Enough for "most applications" or for DVD players or other non-mission-critical devices; you specifically mentioned moving lenses around for eye surgery and said that WinCE or Linux would be Good Enough for that.
Sorry, but they aren't. According to the article, QNX is. That's what sets it apart. Well, that and the technical details of how it achieves that.
No, ridiculous name-calling doesn't make you look smart. The grandparent post was pointing out that a major part of the original poster's argument was...ridiculous name-calling.
Earlier Republicans were supporters of anti-trust laws.
Trouble is, early Republicans weren't Republicans by today's definition. Today's Republican Party barely tolerates equal rights for minorities. Contrast that with Abraham Lincoln, who was a "Republican".
In response to the summary, I'd say the first step is to ship computers with some sort of programming language built-in, but the fact is that programming is a complex thing these days and there's no way to just make it simpler so that kids can learn it easily. If you want to learn to write real programs, you'll just have to commit some time and effort to it. That's why I decided to change my major to physics.
Slashdot Mirror
You'd think for a book about clustering, they'd be able to at least distribute the chapters evenly.
Unfortunately for Microsoft, this is one area where it will take real innovation to usurp the top player.
You can't make Windows somehow incompatible with Google to force Windows users to use Microsoft's search engine. Google will find a way around it.
It ain't gonna happen.
You're part of the problem.
So am I.
Deal with it.
But unless you know which checksums represent valid code and which don't, the number given in the grandparent post is the relevant one.
And you don't know which checksums represent valid code until you've written out and checksummed every valid line of C code.
It's like that Freddy vs. Jason movie that's coming out. Except I think a long, tedious legal proceeding will be a lot more interesting.
Can YOU imagine being on the end of the phone, trying to help somebody recover a server and every few minutes find that they were running with experimental kernel patches, or ancient/buggy software, or that a fault seemed to be caused by a random frob off SourceForge that you'd never heard of?
No, I can't imagine Hell. I can, however, imagine a company claiming to support Linux at least making an attempt to support kernels that are fully official and straight from Linus. This is what Hans was complaining about: that some distros don't support official releases of the kernel because they don't like this or that feature.
If you say you support Linux, then support Linux.
Mmm, total havoc. Yes, that's a better way to do it. No less feasible, obviously, so the question remains, was my original proposition feasible?
Sure, it's all well and good to borrow stock with the intent of selling it, buying it back at a lower price, and returning it to its original owner. But you still have to find someone to sell it to.
I think you can pretty much delete the rest of that post, and have it remain accurate.
Darl McBride: "The System 5 source code, that is really the area that gives us incredible rights...
Yeah, I think "incredible" is the right word in this situation.
One of the first posts was a very astute translation:
"Fuck off."
Does that mean that if terrorists also use Unix, SCO is just as liable as IBM for providing arms to terrorists?
Maybe we can twist the War on Terror to our benefit after all.
Remember that although this worm could have executed any code it wanted on all of those hosts, it had no malicious payload
Let's think of a worst-case scenario, here...
The worm had a program to propagate itself in a space of 376 bytes. It had up to, what, 1500 bytes to carry whatever program it wished? Let's say it used those 1500 bytes to set up a program that would listen on a particular TCP port for instructions from the author's computer. Then, rather than propagating itself as fast as possible, it sends out a packet every few minutes, gradually and insidiously infecting all MSSQL servers on the Internet.
The 1100 extra bytes are used to write a program to disk, and then launch it. This program listens for connections on some high port, or perhaps just listens for UDP packets of a certain description (since it knows the firewall lets those through). At first, it simply catches all worm packets and records the IP addresses, so that it knows what other hosts are infected.
The author's computer listens for these packets, and makes a similar list of infected hosts. Then, when the time is ripe, he starts sending additional instructions to those hosts.
The hosts receive the new instructions, modify their program based on the contents, and then echo the packet out to the hosts in their lists. The author numbers the instruction packets, and the hosts make a note of which ones they've received and ignore repeats. That way, once all infected hosts are updated, the patches stop flying around.
One of the first instructions to be sent out is to make the program launch at boot time. Then, the infected computers are sent instructions to stop propagating themselves. They're sent instructions to report back to the original source. The author looks at the hosts, sends out special non-propagating instructions to military hosts to send him their data. He sends out instructions to hosts that may have access to credit card databases to send him the numbers and expiration dates. He gathers whatever other information he deems useful.
Then, he sends out an instruction for all hosts to delete all data from all databases.
How difficult would it be to write the initial program for that? How difficult to make those patches, and make them work? My guess is, someone who knows assembly well could pull it off. It may take a fair amount of time and patience, but the amount of money to be made is pretty considerable and could make it worthwhile. Hey, if I were going to write a malicious worm, that's how I would go about it.
But the most pertinent question is, how many MSSQL servers are still out there, unpatched, vulnerable, serving critical data?
Recently I got pulled over for speeding. The cop said, "Don't you know the speed limit is 55 miles per hour?"
I said, "Yeah, but I wasn't planning to be out that long."
Apologies to Steven Wright.
Yeah, I have to take a lot of drugs too.
Wait, what's that about worms?
At least you didn't mistake him for Ballmer.
Is it? I've always said "fuck-nut", but I actually thought "fuck-not" was a good insult, in its own way.
To reiterate my original post. First you decide what your requirements for Good Enough are, then you choose the OS that provides you with that level of support. What can you possibly argue with there?
Well, yes. Your original post was that WinCE is Good Enough (TM). The problem is that you said that it's Good Enough for laser eye surgery. You didn't say it's Good Enough for "most applications" or for DVD players or other non-mission-critical devices; you specifically mentioned moving lenses around for eye surgery and said that WinCE or Linux would be Good Enough for that.
Sorry, but they aren't. According to the article, QNX is. That's what sets it apart. Well, that and the technical details of how it achieves that.
No, ridiculous name-calling doesn't make you look smart. The grandparent post was pointing out that a major part of the original poster's argument was...ridiculous name-calling.
Thankyou, thankyou. I'll be here all week.
Actually, I don't think stocks are usually traded once a company goes out of business.
"Supposing I was an idiot, and supposing I was a Microsoft executive...but I repeat myself."
With apologies to Mark Twain.
I'm firmly opposed to affirmative action. I support equal rights.
Re-evaluate your argument accordingly.
Earlier Republicans were supporters of anti-trust laws.
Trouble is, early Republicans weren't Republicans by today's definition. Today's Republican Party barely tolerates equal rights for minorities. Contrast that with Abraham Lincoln, who was a "Republican".
Politics change; only the name stays the same.
Yeah, and just leave it there, to see what'll happen.
No comments yet and it's already /.ed. Nice.
In response to the summary, I'd say the first step is to ship computers with some sort of programming language built-in, but the fact is that programming is a complex thing these days and there's no way to just make it simpler so that kids can learn it easily. If you want to learn to write real programs, you'll just have to commit some time and effort to it. That's why I decided to change my major to physics.