Correct me if I'm wrong, but isn't there some sort of DRM and/or trusted computing in the iPhone intended to prevent this?
A quick Google reveals almost nothing on the subject, so I could have been entirely off here. It could be entirely software-based, which makes it much easier to deal with.
The question here is whether it's a crack or a hack. I was assuming both -- some restriction had to be cracked, and then they could start to tinker with it. But maybe it was simply a hack.
You're still trying to turn it into a blanket policy. It is a limited operation which redirects only traffic to particular servers and most likely targets a particular strain of botnet which uses these servers for coordination.
Great, and what happens when the botnet author simply switches to Freenode? Or Dalnet?
It's not a blanket policy, I understand that. The problem is, every IRC network either has to be fairly totalitarian, or has to effectively function as an open relay for this sort of stuff, at least for a short amount of time. So, first it'll be this network, next it'll be a network that I care about, and eventually they might simply block 6667 -- many ISPs already block outbound port 25.
And no, the world does not need another Eggdrop.
Eggdrop doesn't do what I wanted, and there's no reasonable way to make it do what I wanted.
The control channel wasn't targetted. The redirection was used to send deactivation commands to the bots. The bots were the threat and the target of the defensive action.
Sounds like offensive action at this point.
But again, it was all done with the assumption that anything connecting to that server was one of these bots.
Is there a legitimate use for a channel that is populated by hundreds or thousands of non-human clients?
I was actually intending to write something that worked that way. It was designed to have a network of bots which cooperate on a potentially hostile IRC server to hold a channel. This is because when joining a channel that doesn't exist yet, you get to be the operator, but unless you "register" it with something like ChanServ, anyone else you delegate as operator can deop you and take the channel. The plan was to have a secure network of bots which hold ops, thus preventing any one bot/person with ops from taking over the channel -- the first offensive action against one of these bots would cause all the others to ban you.
I don't need hundreds of thousands of them to make that work. But then, where'd you pull that number from?
But again, why must a legitimate use be proven? Start by proving that a particular channel is bad. And once you've done that, send it to Efnet, give them a day or two to shut down the channel and ban everyone (or do those remove commands), before you interfere with their service.
They could disallow the kind that sits on thousands of computers and waits for commands from its master.
There's nothing illegal about that, either. In fact, there's already a legal bot of that kind on millions of computers. It's called "Windows Update".
The fact that one uses IRC and one uses HTTP is an implementation detail. Microsoft could just as easily include a "critical update" that causes all computers to DoS apple.com (or ubuntu.com), but they don't, and no one blocks update.microsoft.com assuming that they will. No one blocks *.archive.ubuntu.com, either. And for that matter, we could, in fact, design a package manager around getting updates from an IRC channel -- which would be a nice way of pushing a critical security update out to everyone instantly.
But if there was an update sent out, I really don't think hijacking those websites is the answer. The answer is to tell Microsoft (or Ubuntu) about it, wait for them to correct the problem, and block the boxes that are actually doing something illegal (like sending spam).
Your argument is a little like throwing the hands in the air and saying "What can we do? We can't disallow open relays. We'll just have to filter the spam."
That's a bit extreme, given that IRC is entirely self-contained. Again: It isn't as if efnet itself is sending the spam. Also, no one's been able to convince me that anyone so much as mentioned the issue to them -- the general procedure with something like an open relay is, you send mail to abuse@isp.net. It would also probably be easier for efnet to send those deactivation commands than for the admin to hijack DNS, setup some sort of proxy of his own...
And your argument is a bit like throwing your hands in the air and saying "We'll just have to disallow IRC servers."
If you assert your rights to control of your computer, you also assume responsibility for its actions.
Fair enough.
You know, I really, really wish this was the common attitude everywhere. For instance: I'd rather ISPs, universities, etc would forward complaints on to me, rather than disconnecting me at the first sign of a takedown notice. So yes, I'd rather the legal action, if any, be directed at me, and not the ISP.
It would also, at a stroke, eliminate the problem of botnets. If a person being DoS'd could go sue the owners of the zombie computers doing the DoSing, even if it was only for a few dollars from each zombie, it would make people think twice about using insecure OSes (or not getting security updates, or not having antispyware, or not caring to understand security). It would also offset the cost of being DoS'd.
Right now, we have this attitude of you not being responsible for whatever your computer does, as long as someone else "made" it do that. Which means no one cares whether they're part of a botnet or not, and if you're DoS'd, you have to find the controller of the botnet and sue them. Not that the controller isn't deserving of a lawsuit, or of swinging from the tallest tree around, but there would be no botnet controllers if there were no zombies to control.
What is true for you is also true for your victims. The spam, or cracking, or DoS, or whatever that your computer commits.
Visiting an IRC server is not spam, or cracking, or DoS. If I know about the malware, and am monitoring it, chances are I have it sandboxed and firewalled to the point where it can connect to the control channel, but can't really send out spam (or crack, or DoS anything). Hell, maybe I am doing this as part of a reverse-engineering process, trying to figure out who this bastard is and who he's targetting...
But I don't need a reason to justify this. I generally consider it a basic privacy right to have my Internet access un-fucked-around-with, so I'd say the burden is on you and Cox to justify that extremely selective, coercive action instead of something simpler -- like, oh, blocking the infected machines from the network until their users clean them up (and offering a free clean-up service).
Actually, if it was my network? Let them connect, but to a different, local network, from which they can only connect to a webpage which says "You must clean up this computer before you may connect. You can download spybot here..."
Yes its your pc, that's not the point. The point is that you have allowed it to become part of the problem by inadequate supervision of what its doing.
The right thing to do with such a PC is, disconnect it from the network.
The wrong thing to do is forceably remove software from it. Would you agree with their tactics if they'd been using an exploit of their own to gain access to these Windows boxes, then using Remote Desktop to download and run Spybot in order to remove the botnet? How is what they did significantly different?
If someone did this for me while I was gone, I'd be kind of annoyed and maybe a little creeped out. I most certainly would prefer they'd asked permission first, because while in all likelihood I would rather have it fixed, it's my door, and for all they know, I was the one who kicked it in, and I was doing a photo study on it, or making a home movie, or something.
It's also not just one door. They're fixing thousands, automatically. Which means it's not like my one door, where I notice it and say "Hey, thanks!" Thousands means you're much more likely to run into that one person who really didn't want you to fix it, and had a good reason not to.
I actually do have a real-world example of something like this: For the first month or so of college, I had a roommate, we'll call him J. Now, J wasn't a bad guy, and he had a stereo and a TV, which was cool, but he was also a farm kid, liked sports, etc, so while we were both in computer science, we had little in common -- and I would doubt he's still comp sci.
During that first month or so, the easiest way to make friends was to simply leave the door to your room open -- people would just wander in. Or you could walk down the hall to someone else's door and wander in. This is how I met the people next door, which was actually a bit like me and my roommate -- K was like me, was very much into tinkering with computers, was Leftist (had Michael Moore DVDs), and so on. His roommate, whose name I don't remember, but call him L, was Republican, into sports, etc.
I can't say I didn't see it coming. After all, I spent quite a bit of time in K's room, watching his Family Guy collection with him. And then I went back to my room, and played Quake 3 with him over the network -- which meant we were shouting loudly enough to hear each other from next door...
So one day, I came back from class to find J gone from my room and K finishing setting up his custom-built bunk. They had mentioned the idea for me, but effectively, they traded roommates on me without my knowledge or consent.
I was happy with the change, though -- most of my pseudo-righteous-indignation was just because it was so damned funny -- but really, it would have been nice to at least have a little warning.
It's a 1.5 ghz single-core Celeron, so you wouldn't buy it for performance -- never mind that most apps don't get much of a performance boost out of 64-bit. (Most get some tiny amount, some get something huge like 50% that makes it all worthwhile...)
It's also got 256 megs of RAM. If it had more than 2 gigs, we'd really, really want 64-bit. But it's got 256 megs and a magnetic hard drive (meaning you won't exactly want to be running more than 2 gigs worth of stuff in swap).
So, I mean, thanks for the info, but I honestly can't think of a reason I'd want it to be 64-bit.
Why is it that so many gadgets like these actively prevent you from tinkering?
I mean, I could understand if it simply voided a warranty, but I'd like to be able to make that choice for myself. I guess Apple Knows Best...</sacrasm>
Efnet wasn't blocked: Access to a small number of IRC servers was redirected.
So were the redirects forwarded? Seems to me they were blocked -- and those servers were indeed part of Efnet. The entire network may not have been taken down, but much more of it was taken down than simply a control channel.
That isn't a policy decision: It is an individual reaction to a surge in bot traffic.
If it's an individual reaction, that means either they have violated their own policy, or there is a policy in place which states that the admin can do whatever the fuck he wants to stop a surge in bot traffic. Or perhaps something more specific -- in which case, it gets closer to being a policy decision.
At a company of that size, there's really no such thing as an "individual reaction". There's authority and accountability, and policies for everything, even if all the policy says is, "you can do whatever you want". And I do think that it would make a lot more sense for there to be a policy of not randomly interfering with user traffic.
It is a defense against a malicious threat to the network infrastructure.
What makes it malicious? In fact, what makes the control channel itself malicious?
And why didn't they take a defense which doesn't involve silently redirecting an entire service? And why affect every single user, and not just the zombies?
In short, why not simply block whatever traffic the zombies were designed to actually run -- for example, outbound port 25 -- and more importantly, tell the users what they're doing? Or, why not simply block the individual users who are infected?
It's not necessarily the blocking itself that I object to. It's how it was done. Like I said: Why not block Google, if it enables a bot? How is this different?
Besides, if efnet really can't tell drone control channels from normal human chatter without being notified, then this is their wakeup call.
Unless they're prepared to monitor every single channel by using human ops (thus preventing temporary private channels), you're talking about implementing a turing test.
But again, why shouldn't they allow drone control channels? Should they have to investigate every single bot on the network to make sure it's a harmless one? They can't very well disallow all bots.
What you're asking efnet to do is a bit like asking a registrar to not allow people to register domains which will be used for phishing. Sure, you could write some logic that sort of prevents them, maybe, but if it's at all effective, it's going to have false positives. And there's no way you can afford to hire a person to do it manually. And this is a registrar -- a for-profit entity. Efnet is free.
Okay, so Pepsi didn't win your vote, but if you're human, an ad has created a positive association in your mind between a brand and an ad you liked.
Again, assuming I liked the ad.
Repetition improves memory recall. (See BonziBuddy, a brand you hated, you can still remember its name.)
Right. But again, I don't see how it's good for BonziBuddy for me to remember their name, because I won't even use their free stuff (they are spyware), and the only way I'd even mention the name to someone else is to say that they're spyware.
And hey, you and me both remember the brand name from the little camera ads;) Who're you trying to fool?
It took awhile. X-10? For awhile, I was thinking "X-Cam"... Does it really work that well subconsciously if I can't even consciously recall it very well when I'm trying?
By the way, this is another example of advertising not being effective. How the hell is X-10 differentiated from anything else? I mean, yes, I've got great recall on the subject. But suppose it was a positive subconscious connection -- I might end up buying an Xbox instead. Or maybe an airport Xtreme. Or maybe there will be an X-11 camera next to it on the shelf, by a different company...
I can assure you, AOL ads are not targeted to you. That doesn't mean they are not effective to other people.
I get that.
What I have to wonder is, wouldn't it be a MUCH more effective tactic to not annoy the hell out of people?
For example: PeoplePC is arguably no better than AOL, and from what I've heard, may be considerably worse. But their ads didn't annoy me. They didn't send me 50,000 random CDs in the mail. So for awhile, when I heard people who claim to have no other option for Internet than dialup AOL, I suggested they check out PeoplePC.
Recently, I've found out that they have spyware. But it's an example of how a good ad can be effective, whereas a bad ad can have the opposite effect.
In fact, I would imagine you are quite open to advertising that proposes a consumer of a brand that subverts against mainstream advertising.
Tricky. But I bet I'm not.
I bet I'd enjoy watching the ad much more, simply because you'd have to get creative to pull it off. But I think I've outgrown the whole conforming-to-non-conformity thing.
Maybe I'd be open to advertising of something that's associated with maturity, then. Maybe not.
The biggest problem with most ads of this type, unless it's done very subtly, is that they insult my intelligence pretty directly. Ads that try to be counterculture are usually trying so hard that it's like watching... well... that retarded seatbelt rap PSA thing. Or "butt out", from South Park.
As for AOL and Sony moving to disassociate from their own brands, you're right, it can be a double edged sword for companies.
That is all that I'm saying here. Put an ad in a videogame, and unless it's placed very well, it's going to piss me off, and discourage me from buying your product. Do enough of this, and you become like AOL and Sony -- so I actually do believe that no advertising is better than bad advertising.
Don't get me wrong, even good ads can be annoying. But I really think that ads and product placement really can be done right in a videogame. For example:
Good:
Guns. (Everyone knows what a Deagle is from Counter-Strike. I am much more likely to buy a Desert Eagle or a Night Hawk as my first pistol, even if it's completely wrong for me.)
Cars, so long as it's done right. (You generally don't need a voiceover telling you what the car is, but there should be enough detail now for you to tell just by looking at it.)
Places where ads would be in reality. (Billboards, in a place you'd expect a billboard. Decals/paint on a
If Google poses a threat to the network infrastructure, not because of sheer popularity but because they facilitate malicious attacks on a big scale (let's say they allow browser exploits in ads or in landing pages to ads on their site)
There's a big difference between actually serving up the exploit (like you're describing) and merely being the network (like efnet is). It's not like people go to efnet and get infected, and efnet is deliberately not doing anything about the problem. It's also not like there's a simple technical solution that efnet could deploy to prevent anyone from using them for communications.
Let me make this simpler:
Suppose you had a botnet which was controlled by inserting a random text string into a webpage. This could easily be done by whoever controls it, by, for example, making an Anonymous Coward post on Slashdot.
So, each node of the botnet, in order to get its instructions, runs a Google search on a particular magic string (something unlikely to EVER show up anywhere but a command for it), and finds this AC post on Slashdot.
Is it OK to block Slashdot, even? Ok, fine, you block Slashdot, and next time, the botnet author uses Myspace.
So, is it OK to block Google? Really? Are you really saying it's Google's responsibility to examine every search result they return for potential botnet instructions?
If it's Google's responsibility to do that, then shouldn't it also become their responsibility to censor other things, like, oh: 09 f9 11 02 9d 74 e3 5b d8 41 56 c5 63 56 88 c0
What about the Pirate Bay? Should Google block results that return illegal torrents?
Redirections for purely economic reasons are a whole other story.
So it's OK to redirect just because you think that activity might be illegal.
So it's OK for Google to censor search results in China, because, hey, that's the law there.
I know someone will cry "strawman", but the fallacy I'm really using here is "slippery slope" -- because it really does apply here. As soon as you start blocking access to something because they "facilitate attacks", you now have to start defining a policy about what things you may block and what you may not.
I challenge anyone to come up with such a policy that is specific enough that it could never be used to block something that you think shouldn't be blocked. Honestly, I think such policies are as hard as trying to legally define "pornography"...
(Speaking of which: Some ISPs do attempt to block pornography, but none of them do it without you explicitly signing up for their filter service. Some people think porn is bad, some think it's good, and some don't care -- but that is a moral judgment, not an administrative one.)
Oh, and no one's answered my other question here: Was efnet even notified? Standard procedure here isn't to simply block a service -- you at least tell them you're going to, and give them an opportunity to self-police. And efnet can and does self-police, if they're aware of the problem.
I wonder why you're lumping him in with people like RMS and Linus Torvalds (though I think Torvalds may be better at communicating with the masses than you think), and not with people like Steve Jobs and Bill Gates.
Keep in mind -- Mark Shuttleworth is not some random project leader. He's the CEO of a company. Big difference there.
I guess I was thinking of the case of an mmap'd file, but I guess processes don't function as mmaps of their executables. (Might be kind of cool if they did, though.)
There are leaf nodes which need to handle the connections.
True enough, meaning we have a problem if the spam takes up a significant amount of a particular leaf's connection. Right now, it doesn't take up much of my own bandwidth, and I doubt it takes up much of the bandwidth of the places I'm connecting to.
If it did, I'd start blacklisting IPs, or throttling them, but that's for me to do, on my own server, not for my ISP to do for me without asking me or telling me.
No, you can't. You can only handle your own filtering because all providers have effective first-defense systems in place, use big systems to receive mail, much bigger than they would have to use without all the spam, and constantly monitor email traffic so that they can intervene manually when another spam or worm avalanche comes along.
Frankly, I don't believe you here -- I'm going to have to do the research. (And by that, I mean, I'm going to have to actually call up my ISP and ask them.)
I say this because I run my own mailserver, and that is where my spamfilter is. Sure, my ISP monitors it -- they don't want me sending spam out -- but I really doubt they're preventing me from getting spam coming in, considering I get something like 500 a day.
No, their infrastructure. It's their last mile, it's their DNS server, it's their IRC server (the fake one).
That I pay to use. And if I'm paying for "internet service", I'm paying for service to the Internet, not to just the sites Cox allows.
Because more people would complain. A problem would have to be much more severe to justify redirecting Google.
What you just said is that it's perfectly fine for an ISP to block Google, or redirect people from Google to Microsoft Live Search, so long as they're prepared to deal with the complaints.
I guess we'll just have to agree to disagree on that point. Sorry, but net neutrality is more important to me than helping lazy users and admins deal with the spam/botnet "problem".
Granted. But this goes back to the ago-old argument that one should never buy a new computer, because 6 months later it will be half-priced anyway.
What I do is, I look for the sweet spot.
That is: 4 cores is an improvement over one, certainly. However, back when I bought this computer, one core running at 2 ghz or 2.4 ghz was something like twice to three times the price of one core running at 1.8 ghz.
In other words, I'm not telling you to wait 6 months, I'm telling you to buy the hardware that was cutting-edge 6 months ago, which is 80-90% of what's cutting-edge now, but half the price. It also means you can do this twice as often as you could afford to before.
Generally, I find that this just-behind-the-curve hardware is almost as good, costs much less, and is more likely to be physically more reliable, as well as have better drivers in any OS.
If I am forbidden to buy the best computer I can afford, when I need said computer in order to run Linux, then that could be considered by some to be a failure of Linux on the desktop. It is not necessarily the fault of Linux, but it is indeed a problem.
I'll agree with you there, and in your situation, it is kind of worse -- but take your soundcard. There are all kinds of high-end soundcards supported by Linux.
So, in some cases, bleeding-edge hardware really won't be supported. In other cases, it's that you can get hardware just as good at about the same price that will be much better supported, if you do the research ahead of time.
And if that research takes too much time, buy a Dell with Ubuntu and let them do the work for you.
By the way -- would it really cost $200 for you to upgrade to 64-bit, or is that the full version? Is it possible to buy just an upgrade?
I know you'll say you shouldn't have to do this, but I very rarely have Linux hardware issues anymore.
Why?
Because when I get a new machine, I research it thoroughly. If I can't be sure that everything will work at least as well as it does on my old one, I don't buy the new machine, or I buy different hardware. Generally, by the time the hardware I want is supported, it's half the price anyway.
With Ubuntu, it simply reported "sync out of range" and there was nothing that could be done. Safe mode generated the same error, and with no UI to interact with, that's the end of it.
Others have pointed out other ways of dealing with this, but I could point out -- you can force Ubuntu to give you a VGA console (boot with video=vga or something), and you can hit ctrl+alt+f1 to get to it -- and this will work on just about anything.
Then, it's going to take a bit to fix the problem, but I bet you're not the only one with that problem, and I bet there's already a fix. The best thing about Linux is, once you figure out what that fix is, you will never lose it. It's so much easier (at least for me) to back up all of the configuration for everything, and then selectively restore the fixes that worked -- on Windows, registry hacks aren't always as portable, and I really don't have an easy way to pull up the registry for a fresh Windows install side-by-side with the registry from the old one (that had all my tweaks).
For my part, the worst problem that I had with Ubuntu was trying to install on a machine with about 64 megs of RAM. But there's always the alternate installer, and other people make other rescue disks that work. Still, I am going to suggest that the next round of Ubuntu install CDs have the option to boot without X.
Likewise, when I tried Ubuntu on a laptop, it recognized the wireless card and then refused to use it. (It just doesn't work - trying to set the WEP key does nothing, it just says "activating device" and then returns to not working.)
Probably something else you had to install.
Let me guess -- it's a Broadcom. For those, you have to download firmware from the Windows drivers, because we can't legally redistribute that. It's legal for you to download it, just not legal for us to put on the install CD.
when it fails, there's no way to get help. Your options are basically to whine on forums, and then get completely useless advice like editing configuration files on a read-only CD with an OS that doesn't display a UI.
Try IRC -- at least that way, when there's a misunderstanding like this, it gets cleared up quicker.
In the first place, ctrl+alt+f1 will give you a UI, it's just a commandline one. In the second place, the CD is read-only, but the OS isn't running off the CD. It's running off something called "UnionFS" which combines the CD and RAM -- so any changes you make to the CD will actually be stored in RAM, and anything you don't change will be read from the CD.
With Windows, there's a support number you can call, or you can take it to a local computer store, or ask for help among the massive number of Windows users - in short, you're not stuck with snobs on forums who think you should be able to hand-edit configuration files without being able to see anything on the screen.
That hasn't been my experience.
My experience is, the support number generally wants you to have a support contract. Canonical sells Ubuntu support contracts too, so that buys you nothing.
Local computer store -- you win here. But asking for help among the massive number of Windows users is as likely to make your problem worse as not -- those snobs on the forums generally know what they're doing. And they won't charge you for it -- I swear, half the time I Google for the answer to a simple Windows question, I land on "Experts Exchange", which will happily show me the question (so they get the Google hit), but wants to charge me to see the answer.
And again, if you're willing to pay for support, you may as well buy commercial support from Canonical. $250/year will buy you 9-5 desktop support. Or buy a Dell with Ubuntu pre-installed, and these installation issues go away -- AND you get support (from Canonical) bundled, the way you would with a Windows computer from Dell.
ONE SIGN ON. ONE KEY staying with the user session, whether they open a shell, click on an app in KDE or Gnome, SSH or NFS to another machine or disk. One sign on.
I'm not entirely sure you need Kerberos and such for this.
For example: I sign on once to my desktop, and one more time to my KDE wallet -- and I could skip the second step, actually, by removing its password. I can then ssh anywhere I want -- I have the key already, and it's not encrypted on-disk. I can login to any website, and Konqueror uses that KDE wallet to remember the passwords.
The problem is there's NO LEADERSHIP.
Linus, RMS, Mark Shuttleworth, and quite a few others would like do disagree with you.
In particular, there doesn't have to be leadership governing every single project, so long as there's leadership governing a distro -- which can then fix every other project any way it wants.
But Linux works specifically because there's no permanent leadership. If Mark Shuttleworth screws up, and Ubuntu fails, we can go to Debian, or Gentoo, or Slackware, or LinuxFromScratch, or Arch Linux, or... need I go on?
A lot of people would argue that all these distros are part of the problem, but they really are not. Ubuntu is good enough, and so is Kubuntu, Xubuntu, Edubuntu, etc. And Ubuntu is also an example of why multiple distros is a good thing -- Ubuntu happened because Debian wasn't good enough, so they forked Debian. If Ubuntu falls, we could use one of the other distros, or we could simply fork Ubuntu.
Linux can be used by people afraid of their computer. It just won't perform as well as when it's run by people who tweak their computer.
This is true of any OS. The OS can do a lot for you, and perhaps it could even figure out how to make sure your video never skips, even at the expense of other background tasks. But it can't do your thinking for you.
What I don't get is why they think it's effective. For example:
Misplaced quality belief... when you see Tide vs No-Name which makes better detergent?
If I don't care (and I don't), then whichever is cheaper. If I do care, then whichever actually performs better. If I don't have the time to test myself, I go read Consumer's Reports.
I'm unusual, I know, but I really have to wonder if anyone actually consciously says "Oh, that's Tide, it must be better."
Product recognition... which brands do you recognize... this leads to comfort and purchases when there are two competing products.
Not really.
I use one of several pretty much no-name local ISPs. Between them and, say, Mediacom, the only reason I'm thinking about Mediacom is that it does actually look faster, and not much more expensive.
Would advertising influence this purchase at all? Actually, yes, it does. I will never fucking buy AOL, ever. There is nothing they can do to make up for their crappy software -- and worse, their massive campaign of install CDs. So in that case, the fact that I recognize AOL leads to comfort -- that no matter what the competition is, at least they aren't AOL. (Unless they're Earthlink.)
Scaling up everyone's bandwidth only makes the problem worse
Then you're not listening. I don't mean simply scaling up the last mile, I mean scaling up everyone's bandwidth a lot, and scaling up the trunks even more.
The only way this "makes the problem worse" is if you fail to use decent email filters, or fail to actually provide enough bandwidth for people to saturate it -- or if botnets are being used for DoS attacks, and there are even ways of dealing with that now.
email is already heavily filtered at all levels (and at great cost)
Irrelevant. As a user, I can do my own email filtering pretty much for free, and I do. Gmail also does pretty decent email filtering, and I figure Google can afford to do that right, for the users who can't or won't learn to run their own filters.
users are resistant to education and will continue to run every attachment that promises them free money, cheap viagra or porn
Then that is their loss. I shouldn't have to suffer because someone else is insecure about the size of their penis or bank account.
It's like anything else. Car users are resistant to education, but if you don't change the oil, you're going to experience engine problems. No one tries to sell a car with a service plan that includes a bunch of oil-changing ninjas who will track your car down wherever it is once a month and change the oil. And no one bitches that changing the oil and occasionally getting a tune-up is "too complicated".
This is about mitigating an immediate problem by exerting control over the infrastructure.
MY infrastructure, not yours to control. Or that's the way it should be.
If your service is being used to remote control big bot networks and you fail to do something about that, prepare to become collateral damage.
So where's the evidence that anyone even tried to contact efnet about this? I'm sure they'd happily ban whoever was involved, maybe even add those "bot.remove" commands or whatever in order to shut down the botnet. But that's efnet's decision to make, not Cox's.
Would you feel the same way if your ISP blocked Slashdot, because a botnet was being controlled through anonymous coward trolls? What if they blocked Google, because botnets were using it to search for a particular (made up) term? Why is it OK to block an IRC network, but not Google?
Right, except as I keep saying, I don't see how pissing people off makes them want to buy your product -- or a different one with your brand.
Which means if I was looking to buy an "emergent" business, I'd look for one which both has attention and doesn't annoy people to the point where they boycott my brand.
For example: If you're a geek, would you knowingly buy ANYTHING to do with AOL, if you had any other choice? What about BonziBuddy? Seems to me that putting the AOL brand on anything is guaranteed to get as many people boycotting it as there will be people checking it out. Sony is moving in a similar direction.
I can see how simply making someone aware of a brand might help if they don't immediately buy your product, but presumably the intent is for them to eventually buy something because they saw that ad, or at least to make it seem likely enough that someone else will buy your company -- in the hopes that people will eventually buy something because they saw your ad.
Maybe I'm just unusual, though. I was something like 14 or 15 when Pepsi did those annoying ads with the little Pepsi girl. I deliberately avoided buying Pepsi (though I wasn't smart enough to avoid other drinks made by the same company) for years until I finally decided that Pepsi actually does taste better than Coke.
You wonder why Pepsi still has to advertise, anyway, given that every fast food place has Pepsi or Coke, not to mention the random vending machines all over the place -- it's not like you're going to find anyone who isn't aware of the brand. I would much rather they charge less, or buy better ingredients (Coke tastes so much better in countries where they use real sugar instead of high fructose corn syrup), than see yet another ad that doesn't even have anything to do with soda.
When I want to use a program which I opened ten minutes ago but now find it's been swapped, I'm annoyed because it's slowing me down.
And when I want to use a program which I opened ten minutes ago, but now find that a large chunk of the code (libraries, etc) has expired from the disk cache, it slows me down as much as if it had been swapped.
The OS feels a great deal faster when you have enough RAM. Also, swap can be used for other things -- software suspend (hibernate), for example.
I haven't been on a plane with in-air radio or video in years. The last dozen flights haven't even had power plugs to run my laptop past its less-than-stellar battery life.
That's odd -- recently took a trip to Peru, coach, and only the very last flight (a 25 minute hop, really) lacked an in-flight movie. I also do tend to go for laptops with plenty of battery life (and extend it by ripping DVDs ahead of time so I don't have to spin up the optical drive), but that's just me...
Anyway, the odd thing is, some flights had a big screen overhead, and some had little screens in the seatback, but none had power plugs, that I could see.
Slashdot Mirror
Correct me if I'm wrong, but isn't there some sort of DRM and/or trusted computing in the iPhone intended to prevent this?
A quick Google reveals almost nothing on the subject, so I could have been entirely off here. It could be entirely software-based, which makes it much easier to deal with.
The question here is whether it's a crack or a hack. I was assuming both -- some restriction had to be cracked, and then they could start to tinker with it. But maybe it was simply a hack.
Great, and what happens when the botnet author simply switches to Freenode? Or Dalnet?
It's not a blanket policy, I understand that. The problem is, every IRC network either has to be fairly totalitarian, or has to effectively function as an open relay for this sort of stuff, at least for a short amount of time. So, first it'll be this network, next it'll be a network that I care about, and eventually they might simply block 6667 -- many ISPs already block outbound port 25.
Eggdrop doesn't do what I wanted, and there's no reasonable way to make it do what I wanted.
I loved all of the Puma Man lyrics they came up with. My favorite:
"Puma man, he flies like a moron..."
Sounds like offensive action at this point.
But again, it was all done with the assumption that anything connecting to that server was one of these bots.
I was actually intending to write something that worked that way. It was designed to have a network of bots which cooperate on a potentially hostile IRC server to hold a channel. This is because when joining a channel that doesn't exist yet, you get to be the operator, but unless you "register" it with something like ChanServ, anyone else you delegate as operator can deop you and take the channel. The plan was to have a secure network of bots which hold ops, thus preventing any one bot/person with ops from taking over the channel -- the first offensive action against one of these bots would cause all the others to ban you.
I don't need hundreds of thousands of them to make that work. But then, where'd you pull that number from?
But again, why must a legitimate use be proven? Start by proving that a particular channel is bad. And once you've done that, send it to Efnet, give them a day or two to shut down the channel and ban everyone (or do those remove commands), before you interfere with their service.
There's nothing illegal about that, either. In fact, there's already a legal bot of that kind on millions of computers. It's called "Windows Update".
The fact that one uses IRC and one uses HTTP is an implementation detail. Microsoft could just as easily include a "critical update" that causes all computers to DoS apple.com (or ubuntu.com), but they don't, and no one blocks update.microsoft.com assuming that they will. No one blocks *.archive.ubuntu.com, either. And for that matter, we could, in fact, design a package manager around getting updates from an IRC channel -- which would be a nice way of pushing a critical security update out to everyone instantly.
But if there was an update sent out, I really don't think hijacking those websites is the answer. The answer is to tell Microsoft (or Ubuntu) about it, wait for them to correct the problem, and block the boxes that are actually doing something illegal (like sending spam).
That's a bit extreme, given that IRC is entirely self-contained. Again: It isn't as if efnet itself is sending the spam. Also, no one's been able to convince me that anyone so much as mentioned the issue to them -- the general procedure with something like an open relay is, you send mail to abuse@isp.net. It would also probably be easier for efnet to send those deactivation commands than for the admin to hijack DNS, setup some sort of proxy of his own...
And your argument is a bit like throwing your hands in the air and saying "We'll just have to disallow IRC servers."
Fair enough.
You know, I really, really wish this was the common attitude everywhere. For instance: I'd rather ISPs, universities, etc would forward complaints on to me, rather than disconnecting me at the first sign of a takedown notice. So yes, I'd rather the legal action, if any, be directed at me, and not the ISP.
It would also, at a stroke, eliminate the problem of botnets. If a person being DoS'd could go sue the owners of the zombie computers doing the DoSing, even if it was only for a few dollars from each zombie, it would make people think twice about using insecure OSes (or not getting security updates, or not having antispyware, or not caring to understand security). It would also offset the cost of being DoS'd.
Right now, we have this attitude of you not being responsible for whatever your computer does, as long as someone else "made" it do that. Which means no one cares whether they're part of a botnet or not, and if you're DoS'd, you have to find the controller of the botnet and sue them. Not that the controller isn't deserving of a lawsuit, or of swinging from the tallest tree around, but there would be no botnet controllers if there were no zombies to control.
Visiting an IRC server is not spam, or cracking, or DoS. If I know about the malware, and am monitoring it, chances are I have it sandboxed and firewalled to the point where it can connect to the control channel, but can't really send out spam (or crack, or DoS anything). Hell, maybe I am doing this as part of a reverse-engineering process, trying to figure out who this bastard is and who he's targetting...
But I don't need a reason to justify this. I generally consider it a basic privacy right to have my Internet access un-fucked-around-with, so I'd say the burden is on you and Cox to justify that extremely selective, coercive action instead of something simpler -- like, oh, blocking the infected machines from the network until their users clean them up (and offering a free clean-up service).
Actually, if it was my network? Let them connect, but to a different, local network, from which they can only connect to a webpage which says "You must clean up this computer before you may connect. You can download spybot here..."
The right thing to do with such a PC is, disconnect it from the network.
The wrong thing to do is forceably remove software from it. Would you agree with their tactics if they'd been using an exploit of their own to gain access to these Windows boxes, then using Remote Desktop to download and run Spybot in order to remove the botnet? How is what they did significantly different?
If someone did this for me while I was gone, I'd be kind of annoyed and maybe a little creeped out. I most certainly would prefer they'd asked permission first, because while in all likelihood I would rather have it fixed, it's my door, and for all they know, I was the one who kicked it in, and I was doing a photo study on it, or making a home movie, or something.
It's also not just one door. They're fixing thousands, automatically. Which means it's not like my one door, where I notice it and say "Hey, thanks!" Thousands means you're much more likely to run into that one person who really didn't want you to fix it, and had a good reason not to.
I actually do have a real-world example of something like this: For the first month or so of college, I had a roommate, we'll call him J. Now, J wasn't a bad guy, and he had a stereo and a TV, which was cool, but he was also a farm kid, liked sports, etc, so while we were both in computer science, we had little in common -- and I would doubt he's still comp sci.
During that first month or so, the easiest way to make friends was to simply leave the door to your room open -- people would just wander in. Or you could walk down the hall to someone else's door and wander in. This is how I met the people next door, which was actually a bit like me and my roommate -- K was like me, was very much into tinkering with computers, was Leftist (had Michael Moore DVDs), and so on. His roommate, whose name I don't remember, but call him L, was Republican, into sports, etc.
I can't say I didn't see it coming. After all, I spent quite a bit of time in K's room, watching his Family Guy collection with him. And then I went back to my room, and played Quake 3 with him over the network -- which meant we were shouting loudly enough to hear each other from next door...
So one day, I came back from class to find J gone from my room and K finishing setting up his custom-built bunk. They had mentioned the idea for me, but effectively, they traded roommates on me without my knowledge or consent.
I was happy with the change, though -- most of my pseudo-righteous-indignation was just because it was so damned funny -- but really, it would have been nice to at least have a little warning.
Just in case you were wondering, we don't care.
It's a 1.5 ghz single-core Celeron, so you wouldn't buy it for performance -- never mind that most apps don't get much of a performance boost out of 64-bit. (Most get some tiny amount, some get something huge like 50% that makes it all worthwhile...)
It's also got 256 megs of RAM. If it had more than 2 gigs, we'd really, really want 64-bit. But it's got 256 megs and a magnetic hard drive (meaning you won't exactly want to be running more than 2 gigs worth of stuff in swap).
So, I mean, thanks for the info, but I honestly can't think of a reason I'd want it to be 64-bit.
Why is it that so many gadgets like these actively prevent you from tinkering?
I mean, I could understand if it simply voided a warranty, but I'd like to be able to make that choice for myself. I guess Apple Knows Best...</sacrasm>
So were the redirects forwarded? Seems to me they were blocked -- and those servers were indeed part of Efnet. The entire network may not have been taken down, but much more of it was taken down than simply a control channel.
If it's an individual reaction, that means either they have violated their own policy, or there is a policy in place which states that the admin can do whatever the fuck he wants to stop a surge in bot traffic. Or perhaps something more specific -- in which case, it gets closer to being a policy decision.
At a company of that size, there's really no such thing as an "individual reaction". There's authority and accountability, and policies for everything, even if all the policy says is, "you can do whatever you want". And I do think that it would make a lot more sense for there to be a policy of not randomly interfering with user traffic.
What makes it malicious? In fact, what makes the control channel itself malicious?
And why didn't they take a defense which doesn't involve silently redirecting an entire service? And why affect every single user, and not just the zombies?
In short, why not simply block whatever traffic the zombies were designed to actually run -- for example, outbound port 25 -- and more importantly, tell the users what they're doing? Or, why not simply block the individual users who are infected?
It's not necessarily the blocking itself that I object to. It's how it was done. Like I said: Why not block Google, if it enables a bot? How is this different?
Unless they're prepared to monitor every single channel by using human ops (thus preventing temporary private channels), you're talking about implementing a turing test.
But again, why shouldn't they allow drone control channels? Should they have to investigate every single bot on the network to make sure it's a harmless one? They can't very well disallow all bots.
What you're asking efnet to do is a bit like asking a registrar to not allow people to register domains which will be used for phishing. Sure, you could write some logic that sort of prevents them, maybe, but if it's at all effective, it's going to have false positives. And there's no way you can afford to hire a person to do it manually. And this is a registrar -- a for-profit entity. Efnet is free.
Again, assuming I liked the ad.
Right. But again, I don't see how it's good for BonziBuddy for me to remember their name, because I won't even use their free stuff (they are spyware), and the only way I'd even mention the name to someone else is to say that they're spyware.
It took awhile. X-10? For awhile, I was thinking "X-Cam"... Does it really work that well subconsciously if I can't even consciously recall it very well when I'm trying?
By the way, this is another example of advertising not being effective. How the hell is X-10 differentiated from anything else? I mean, yes, I've got great recall on the subject. But suppose it was a positive subconscious connection -- I might end up buying an Xbox instead. Or maybe an airport Xtreme. Or maybe there will be an X-11 camera next to it on the shelf, by a different company...
I get that.
What I have to wonder is, wouldn't it be a MUCH more effective tactic to not annoy the hell out of people?
For example: PeoplePC is arguably no better than AOL, and from what I've heard, may be considerably worse. But their ads didn't annoy me. They didn't send me 50,000 random CDs in the mail. So for awhile, when I heard people who claim to have no other option for Internet than dialup AOL, I suggested they check out PeoplePC.
Recently, I've found out that they have spyware. But it's an example of how a good ad can be effective, whereas a bad ad can have the opposite effect.
Tricky. But I bet I'm not.
I bet I'd enjoy watching the ad much more, simply because you'd have to get creative to pull it off. But I think I've outgrown the whole conforming-to-non-conformity thing.
Maybe I'd be open to advertising of something that's associated with maturity, then. Maybe not.
The biggest problem with most ads of this type, unless it's done very subtly, is that they insult my intelligence pretty directly. Ads that try to be counterculture are usually trying so hard that it's like watching... well... that retarded seatbelt rap PSA thing. Or "butt out", from South Park.
That is all that I'm saying here. Put an ad in a videogame, and unless it's placed very well, it's going to piss me off, and discourage me from buying your product. Do enough of this, and you become like AOL and Sony -- so I actually do believe that no advertising is better than bad advertising.
Don't get me wrong, even good ads can be annoying. But I really think that ads and product placement really can be done right in a videogame. For example:
Good:
There's a big difference between actually serving up the exploit (like you're describing) and merely being the network (like efnet is). It's not like people go to efnet and get infected, and efnet is deliberately not doing anything about the problem. It's also not like there's a simple technical solution that efnet could deploy to prevent anyone from using them for communications.
Let me make this simpler:
Suppose you had a botnet which was controlled by inserting a random text string into a webpage. This could easily be done by whoever controls it, by, for example, making an Anonymous Coward post on Slashdot.
So, each node of the botnet, in order to get its instructions, runs a Google search on a particular magic string (something unlikely to EVER show up anywhere but a command for it), and finds this AC post on Slashdot.
Is it OK to block Slashdot, even? Ok, fine, you block Slashdot, and next time, the botnet author uses Myspace.
So, is it OK to block Google? Really? Are you really saying it's Google's responsibility to examine every search result they return for potential botnet instructions?
If it's Google's responsibility to do that, then shouldn't it also become their responsibility to censor other things, like, oh: 09 f9 11 02 9d 74 e3 5b d8 41 56 c5 63 56 88 c0
What about the Pirate Bay? Should Google block results that return illegal torrents?
So it's OK to redirect just because you think that activity might be illegal.
So it's OK for Google to censor search results in China, because, hey, that's the law there.
I know someone will cry "strawman", but the fallacy I'm really using here is "slippery slope" -- because it really does apply here. As soon as you start blocking access to something because they "facilitate attacks", you now have to start defining a policy about what things you may block and what you may not.
I challenge anyone to come up with such a policy that is specific enough that it could never be used to block something that you think shouldn't be blocked. Honestly, I think such policies are as hard as trying to legally define "pornography"...
(Speaking of which: Some ISPs do attempt to block pornography, but none of them do it without you explicitly signing up for their filter service. Some people think porn is bad, some think it's good, and some don't care -- but that is a moral judgment, not an administrative one.)
Oh, and no one's answered my other question here: Was efnet even notified? Standard procedure here isn't to simply block a service -- you at least tell them you're going to, and give them an opportunity to self-police. And efnet can and does self-police, if they're aware of the problem.
Mark Shuttleworth? Really?
I wonder why you're lumping him in with people like RMS and Linus Torvalds (though I think Torvalds may be better at communicating with the masses than you think), and not with people like Steve Jobs and Bill Gates.
Keep in mind -- Mark Shuttleworth is not some random project leader. He's the CEO of a company. Big difference there.
Probably right.
I guess I was thinking of the case of an mmap'd file, but I guess processes don't function as mmaps of their executables. (Might be kind of cool if they did, though.)
True enough, meaning we have a problem if the spam takes up a significant amount of a particular leaf's connection. Right now, it doesn't take up much of my own bandwidth, and I doubt it takes up much of the bandwidth of the places I'm connecting to.
If it did, I'd start blacklisting IPs, or throttling them, but that's for me to do, on my own server, not for my ISP to do for me without asking me or telling me.
Frankly, I don't believe you here -- I'm going to have to do the research. (And by that, I mean, I'm going to have to actually call up my ISP and ask them.)
I say this because I run my own mailserver, and that is where my spamfilter is. Sure, my ISP monitors it -- they don't want me sending spam out -- but I really doubt they're preventing me from getting spam coming in, considering I get something like 500 a day.
That I pay to use. And if I'm paying for "internet service", I'm paying for service to the Internet, not to just the sites Cox allows.
What you just said is that it's perfectly fine for an ISP to block Google, or redirect people from Google to Microsoft Live Search, so long as they're prepared to deal with the complaints.
I guess we'll just have to agree to disagree on that point. Sorry, but net neutrality is more important to me than helping lazy users and admins deal with the spam/botnet "problem".
What I do is, I look for the sweet spot.
That is: 4 cores is an improvement over one, certainly. However, back when I bought this computer, one core running at 2 ghz or 2.4 ghz was something like twice to three times the price of one core running at 1.8 ghz.
In other words, I'm not telling you to wait 6 months, I'm telling you to buy the hardware that was cutting-edge 6 months ago, which is 80-90% of what's cutting-edge now, but half the price. It also means you can do this twice as often as you could afford to before.
Generally, I find that this just-behind-the-curve hardware is almost as good, costs much less, and is more likely to be physically more reliable, as well as have better drivers in any OS.
I'll agree with you there, and in your situation, it is kind of worse -- but take your soundcard. There are all kinds of high-end soundcards supported by Linux.
So, in some cases, bleeding-edge hardware really won't be supported. In other cases, it's that you can get hardware just as good at about the same price that will be much better supported, if you do the research ahead of time.
And if that research takes too much time, buy a Dell with Ubuntu and let them do the work for you.
By the way -- would it really cost $200 for you to upgrade to 64-bit, or is that the full version? Is it possible to buy just an upgrade?
I know you'll say you shouldn't have to do this, but I very rarely have Linux hardware issues anymore.
Why?
Because when I get a new machine, I research it thoroughly. If I can't be sure that everything will work at least as well as it does on my old one, I don't buy the new machine, or I buy different hardware. Generally, by the time the hardware I want is supported, it's half the price anyway.
Others have pointed out other ways of dealing with this, but I could point out -- you can force Ubuntu to give you a VGA console (boot with video=vga or something), and you can hit ctrl+alt+f1 to get to it -- and this will work on just about anything.
Then, it's going to take a bit to fix the problem, but I bet you're not the only one with that problem, and I bet there's already a fix. The best thing about Linux is, once you figure out what that fix is, you will never lose it. It's so much easier (at least for me) to back up all of the configuration for everything, and then selectively restore the fixes that worked -- on Windows, registry hacks aren't always as portable, and I really don't have an easy way to pull up the registry for a fresh Windows install side-by-side with the registry from the old one (that had all my tweaks).
For my part, the worst problem that I had with Ubuntu was trying to install on a machine with about 64 megs of RAM. But there's always the alternate installer, and other people make other rescue disks that work. Still, I am going to suggest that the next round of Ubuntu install CDs have the option to boot without X.
Probably something else you had to install.
Let me guess -- it's a Broadcom. For those, you have to download firmware from the Windows drivers, because we can't legally redistribute that. It's legal for you to download it, just not legal for us to put on the install CD.
Try IRC -- at least that way, when there's a misunderstanding like this, it gets cleared up quicker.
In the first place, ctrl+alt+f1 will give you a UI, it's just a commandline one. In the second place, the CD is read-only, but the OS isn't running off the CD. It's running off something called "UnionFS" which combines the CD and RAM -- so any changes you make to the CD will actually be stored in RAM, and anything you don't change will be read from the CD.
That hasn't been my experience.
My experience is, the support number generally wants you to have a support contract. Canonical sells Ubuntu support contracts too, so that buys you nothing.
Local computer store -- you win here. But asking for help among the massive number of Windows users is as likely to make your problem worse as not -- those snobs on the forums generally know what they're doing. And they won't charge you for it -- I swear, half the time I Google for the answer to a simple Windows question, I land on "Experts Exchange", which will happily show me the question (so they get the Google hit), but wants to charge me to see the answer.
And again, if you're willing to pay for support, you may as well buy commercial support from Canonical. $250/year will buy you 9-5 desktop support. Or buy a Dell with Ubuntu pre-installed, and these installation issues go away -- AND you get support (from Canonical) bundled, the way you would with a Windows computer from Dell.
I'm not entirely sure you need Kerberos and such for this.
For example: I sign on once to my desktop, and one more time to my KDE wallet -- and I could skip the second step, actually, by removing its password. I can then ssh anywhere I want -- I have the key already, and it's not encrypted on-disk. I can login to any website, and Konqueror uses that KDE wallet to remember the passwords.
Linus, RMS, Mark Shuttleworth, and quite a few others would like do disagree with you.
In particular, there doesn't have to be leadership governing every single project, so long as there's leadership governing a distro -- which can then fix every other project any way it wants.
But Linux works specifically because there's no permanent leadership. If Mark Shuttleworth screws up, and Ubuntu fails, we can go to Debian, or Gentoo, or Slackware, or LinuxFromScratch, or Arch Linux, or... need I go on?
A lot of people would argue that all these distros are part of the problem, but they really are not. Ubuntu is good enough, and so is Kubuntu, Xubuntu, Edubuntu, etc. And Ubuntu is also an example of why multiple distros is a good thing -- Ubuntu happened because Debian wasn't good enough, so they forked Debian. If Ubuntu falls, we could use one of the other distros, or we could simply fork Ubuntu.
Linux can be used by people afraid of their computer. It just won't perform as well as when it's run by people who tweak their computer.
This is true of any OS. The OS can do a lot for you, and perhaps it could even figure out how to make sure your video never skips, even at the expense of other background tasks. But it can't do your thinking for you.
What I don't get is why they think it's effective. For example:
If I don't care (and I don't), then whichever is cheaper. If I do care, then whichever actually performs better. If I don't have the time to test myself, I go read Consumer's Reports.
I'm unusual, I know, but I really have to wonder if anyone actually consciously says "Oh, that's Tide, it must be better."
Not really.
I use one of several pretty much no-name local ISPs. Between them and, say, Mediacom, the only reason I'm thinking about Mediacom is that it does actually look faster, and not much more expensive.
Would advertising influence this purchase at all? Actually, yes, it does. I will never fucking buy AOL, ever. There is nothing they can do to make up for their crappy software -- and worse, their massive campaign of install CDs. So in that case, the fact that I recognize AOL leads to comfort -- that no matter what the competition is, at least they aren't AOL. (Unless they're Earthlink.)
Then you're not listening. I don't mean simply scaling up the last mile, I mean scaling up everyone's bandwidth a lot, and scaling up the trunks even more.
The only way this "makes the problem worse" is if you fail to use decent email filters, or fail to actually provide enough bandwidth for people to saturate it -- or if botnets are being used for DoS attacks, and there are even ways of dealing with that now.
Irrelevant. As a user, I can do my own email filtering pretty much for free, and I do. Gmail also does pretty decent email filtering, and I figure Google can afford to do that right, for the users who can't or won't learn to run their own filters.
Then that is their loss. I shouldn't have to suffer because someone else is insecure about the size of their penis or bank account.
It's like anything else. Car users are resistant to education, but if you don't change the oil, you're going to experience engine problems. No one tries to sell a car with a service plan that includes a bunch of oil-changing ninjas who will track your car down wherever it is once a month and change the oil. And no one bitches that changing the oil and occasionally getting a tune-up is "too complicated".
MY infrastructure, not yours to control. Or that's the way it should be.
So where's the evidence that anyone even tried to contact efnet about this? I'm sure they'd happily ban whoever was involved, maybe even add those "bot.remove" commands or whatever in order to shut down the botnet. But that's efnet's decision to make, not Cox's.
Would you feel the same way if your ISP blocked Slashdot, because a botnet was being controlled through anonymous coward trolls? What if they blocked Google, because botnets were using it to search for a particular (made up) term? Why is it OK to block an IRC network, but not Google?
Right, except as I keep saying, I don't see how pissing people off makes them want to buy your product -- or a different one with your brand.
Which means if I was looking to buy an "emergent" business, I'd look for one which both has attention and doesn't annoy people to the point where they boycott my brand.
For example: If you're a geek, would you knowingly buy ANYTHING to do with AOL, if you had any other choice? What about BonziBuddy? Seems to me that putting the AOL brand on anything is guaranteed to get as many people boycotting it as there will be people checking it out. Sony is moving in a similar direction.
I can see how simply making someone aware of a brand might help if they don't immediately buy your product, but presumably the intent is for them to eventually buy something because they saw that ad, or at least to make it seem likely enough that someone else will buy your company -- in the hopes that people will eventually buy something because they saw your ad.
Maybe I'm just unusual, though. I was something like 14 or 15 when Pepsi did those annoying ads with the little Pepsi girl. I deliberately avoided buying Pepsi (though I wasn't smart enough to avoid other drinks made by the same company) for years until I finally decided that Pepsi actually does taste better than Coke.
You wonder why Pepsi still has to advertise, anyway, given that every fast food place has Pepsi or Coke, not to mention the random vending machines all over the place -- it's not like you're going to find anyone who isn't aware of the brand. I would much rather they charge less, or buy better ingredients (Coke tastes so much better in countries where they use real sugar instead of high fructose corn syrup), than see yet another ad that doesn't even have anything to do with soda.
And when I want to use a program which I opened ten minutes ago, but now find that a large chunk of the code (libraries, etc) has expired from the disk cache, it slows me down as much as if it had been swapped.
The OS feels a great deal faster when you have enough RAM. Also, swap can be used for other things -- software suspend (hibernate), for example.
That's odd -- recently took a trip to Peru, coach, and only the very last flight (a 25 minute hop, really) lacked an in-flight movie. I also do tend to go for laptops with plenty of battery life (and extend it by ripping DVDs ahead of time so I don't have to spin up the optical drive), but that's just me...
Anyway, the odd thing is, some flights had a big screen overhead, and some had little screens in the seatback, but none had power plugs, that I could see.