when you burn your money you will realise how stupid it is and stop doing it
Realise how stupid it is? Perhaps. Stop doing it? Naah. Just look at what some people do with their money. Gambling, drugs, collecting strange items ( paying a gazillion bucks for a piece of toilet paper because Elvis wiped his ass with it some umpty years ago ). Money wants to be spent on stupid things. I could make up some statistics to prove it, but I'm busy picking the most expensive PC case for my 2000$ motherboard.
I don't think the whole affair is about "the right to cheat". To me it is about poorly implemented anti-cheating measures.
Contradicts this:
It's simply impossible to create a waterproof
client.... which appears in the same paragraph. If a waterproof client in impossible to create, then "poorly implemented anti-cheating measures" have nothing to do with the failure of the programmers to keep out the cheaters.
I don't see a contradiction here. I am just saying that client-side security is wrong. If you want a secure solution - sorry, you have to implement it on the server side or just trust your users not to use cheating software.
Why should I have to spend thousands of dollars on a lock (or thousands of dollars developing a lock) to keep out scum? A lock should only be required to keep out casual thieves - the "accidental" thieves.
Why you should? Because scum exists. But this is not the point - EQ creators are doing something different - they had a simple lock that probably worked for the casual cheaters. Now they are creating a very complicated lock that's still in the wrong place - on the client side.
But the hardcore thieves, and the hardcore cheaters, are human excrement. No excuse exists for their behavior.
OK. I might have sounded like I am excusing their behavior. I am not. But there's a reason to separate cheating from publishing cheating programs. I view the later as pointing out: "Your security does not work. It will not work this way, ever.". If the EQ creators get this clue (and move their security to the server side) it will ultimately work against cheaters. I am aware that moving everything to the server might be impossible - if this is so, then EQ should be honest with their users and admit it instead of sustaining the illusion that they can stop cheaters.
I don't think the whole affair is about "the right to cheat". To me it is about poorly implemented anti-cheating measures. It's simply impossible to create a waterproof client.
Process table scanning? I can rename the cheating program, change its size, whatever. Memory scanning? A bit tougher to beat - but I can run the client under VMWare, Bochs or any other virtualization software, this way it only sees what's inside the virtual box. Another option is using a mutation engine (virus writers use them to fool virus scanners). Besides, the client can only scan for known popular cheating software, not for any custom code. Code obfuscation? I don't know of any successful attempt at deterring hackers by obfuscating code.
To put it short (for the 1000-th time in this thread) - if you don't want the user to know it, don't send it to the client. So my sympathetic view of the cheaters comes not from my sympathetic view of cheating, but from my attitude towards client-side security.
Ok, I'm probably going to get modded down for saying this
Next time I see this line I'm gonna mod down without reading the rest of the post. C'mon people, karma is not the dick size. Well, ok, it is the dick size (mine is excellent), but still whining about supposed modding down is annoying.
"I don't think even Oracle had the drop column feature until Oracle 8i or something."
I must be missing something here. I have to wonder why they would not put such a useful feature in the first release. *shrug
It's a bit hard to implement cleanly. When you have the records placed in blocks and you drop a column, you make each record a few bytes smaller, therefore getting a nasty fragmentation - lots of 'holes' too small to fit a record in. So my guess is they preferred to leave it out instead of having it implemented poorly.
Would you pay £4 ($6) per half-megabyte for GPRS in the UK?
I wouldn't. I don't.
Well, I could use it for text-only browsing and online board games - with a decent protocol even a long game of chess would fit into a kilobyte. On the other hand I somehow doubt that text browsing and board games are the killer apps:-)
I would GLADLY allow my face to be used after my death, except for the fact that my entire head will be resting in a liquid-nitrogen filled dewar soon after my death.
Me too, I'd just stipulate that the guy who gets my face scares some of the people I don't like by knocking at their windows at night, heh heh.
actually, due to natural selection, the human race is getting prettier.. There's been legitimate research to back this up, I was watching about it on TLC.
Basically, good looking people get laid more than the ugly ones.
I find this counterintuitive. Of course, good looking people get laid more than the ugly ones, but that's exactly the reason I'd expect them to be less willing to spoil their fun with having kids/wives/husbands.
Article: 'avoid explorer'. Note that i'm not saying they are right.
gTsiros' advice: 'be careful where you browse'.
My response, rephrased:
THIS IS JUST PLAIN FUCKING STUPID. IF YOU BELIEVE THAT I.E. IS SECURE, KEEP USING IT. IF NOT, DROP IT. USING INSECURE BROWSER IS STUPID, EVEN IF YOU THINK THAT YOU ARE ONLY BROWSING 'SAFE' SITES.
Actually, the less area a chip takes up, the less likely it is to have a defect in it. It yields more chips per wafer because their are more chips on a wafer and each individual chip is less likely to have a defect.
The thinner the paths, the more likely it is for them to become underetched - the acid solution gets under the mask and breaks the paths.
Manufacturing processes change quite frequently. Although a.13 Micron process will mean that these companies will be able to yield more chips per wafer, the pricing model on high end graphics cards has remained static over the past few years.
What? I think that.13 micron isn't about more chips per wafer - in fact it yields probably less chips per wafer - the thinner your tracks are, the lower success rate. As far as I know it's all about power consumption and clock rate - the smaller your stuff, the less power needed and the faster it can run without overheating.
In the same spirit as the transitioning of the underperforming Xabre chip to 0.13u, I would also like to announce that I am beginning work on a 0.13u port of the Riva128, ATI Rage Pro, and the timeless Trident ViRGE. These chips will still perform like 3 legged dogs, but the fact that they are on a 0.13u process seems to be newsworthy these days.
Actually, I'd buy and old underperforming chip in 0.13u technology if it lets me get rid of the noisy fan.
instead of abandoning IE, which is a decent web browser, be careful (not paranoid, but like anyone who's been on/. for more than...5 minutes won't click on a goatse.cx link) about where you actually browse.
Yeah right, my browser is buggy, therefore I should limit the way I use it, preferably to pages created by me (notepad.exe is the best) and stored safely on local disk of my computer that is disconnected from the network. Any other bright ideas?
This seems like something that could be pretty easily defeated with an OCR library... may have been already.
Nope. The letters consist of many dots of random size, are a bit blurred, sometimes a grid is added. As far as I know this defeats all OCRs, at least those available now.
You can't get away with x users * average bandwidth = required shared bandwidth.
I agree, especially if you average over 24h, including night hours. But this is not my point. I have just said that the (x users * maximal burst bandwidth) estimate is too high when applied to browsing. My estimate is that downloading/viewing time ratio is way below 1/10 even for a single user during peak hours. Of course it does not mean that you can just divide the (x users * maximal burst) by 10, but the more users you have the more (x users * maximal burst * idle ratio)/(actual use) approaches 1. Of course this breaks when users stop acting independently - for example all rush to seek news on an important event at 7:00 PM. But for 'normal' peak hours they act independently enough. Of course this only applies to good old healthy browsing.
That's just wrong. Primetime traffic sets the lower limit of the bandwidth a network provider has to install. When everybody and their mother want to surf at high speed at the same time, then that too puts a limit to overselling.
I don't think so. Most users actually spend more time watching/reading the pages they download than actually downloading them, so there definitely is potential for oversubscribing. So even if I and each of my 100 neighbors get a bazillion Mbit/s transfer while browsing, it does not mean that the provider has to put 100 bazillion Mbit pipe for us - even if we all sit at our computers, probably only a small fraction of us are actually generating transfer at the given moment. Of course I only mean browsing, not downloading ISOs, streaming media, p2p, whatever.
Just because horizontal gene transfer is theoretically possible doesn't mean that it's likely. They are still looking for evidence of horizontal gene transfer between bacterial genomes in nature and, with the exception of plasmid transfer which plants don't really have, haven't found it.
As far as I know there is some indirect evidence of horizontal gene transfer coming from phylogenetics. Scientists have found very similar pieces of DNA (introns) in different species that are phylogenetically far away. Moreover, their closer relatives do not have such pieces. One can consider several explanation of this phenomene:
1. Their common ancestor had this piece of DNA. This is not improbable, but we are talking about pieces of DNA that aren't expressed (that's what introns are), so there is little or no evolutionary pressure on them, so one must explain why the closer relatives dumped them.
2. They same piece of DNA evolved independently in both species. Not very probable, see 1 - no pressure on introns, so convergence not very probable.
3. Horizontal gene transfer occured.
4. The phylogenetical tree is wrong. Well, it seems that however we arrange the tree, some introns do stick out.
Of course, this is not an established scientific proof, but it seems that scientists find more and more of evidence supporting point 3. By the way, crossbreeding is not the only possibility of HGT - viruses are capable of transferring genetic material (at least their own) between organisms, so they may be responsible for HGT. Again, google for 'introns' 'horizontal gene transfer'.
PS. Feel free to point out oversimplifications in the above post.
The oft-repeated refrain that "transgenic DNA is just like ordinary DNA" is false. Transgenic DNA is in many respects optimised for horizontal gene transfer. It is designed to cross species barriers and to jump into genomes, and it has homologies to the DNA of many species and their genetic parasites (plasmids, transposons and viruses), thereby enhancing recombination with all of them [2].
If they've already made some devices, don't you think they've already thought of that?
No.
This only means that traffic generated by one of those devices fits into the Wi-Fi bandwith (I assume that the devices actually work). This does not mean that it leaves room for anything else.
Of course, it does not mean that it does not. The article states:
The computer sends the data needed to create the icons and pictures for display using Wi-Fi, a wireless communication standard typically used to network different computers. The smart display picks up the data using a built-in Wi-Fi receiver and creates the images as needed.
So I admit that the 1024x768 x bpp x FPS estimate is probably was probably too high - sound like they don't transmit every frame blindly. But the amount of data needed to re-create a fast changing window (any kind of animation) can still be huge.
I hope they're using some sort of compression. The typical 1024x768 pixels x 8 bit per pixel (let's be generous) x 16 frames per second (ditto) gives... well, a really huge number of bits per second:-).
Listen, just because the entire world of 6 billion people is motivated by money, it doesn't mean that the few thousand of us here at Slashdot have to be as well.
When we're all dead, people will remember us for the kind deeds we did while we were walking the streets and talking the talk. The little league team you coached, volunteering at a Mormon church, and all those bake sales for the PTA will be what you were best known for. Contract #189533 for $1,730.39 will not be relevant and no one will care how much money you made.
Go give blowjobs to the homeless. They will remember you for that.
Slashdot Mirror
I don't think the whole affair is about "the right to cheat". To me it is about poorly implemented anti-cheating measures. It's simply impossible to create a waterproof client.
Process table scanning? I can rename the cheating program, change its size, whatever. Memory scanning? A bit tougher to beat - but I can run the client under VMWare, Bochs or any other virtualization software, this way it only sees what's inside the virtual box. Another option is using a mutation engine (virus writers use them to fool virus scanners). Besides, the client can only scan for known popular cheating software, not for any custom code. Code obfuscation? I don't know of any successful attempt at deterring hackers by obfuscating code.
To put it short (for the 1000-th time in this thread) - if you don't want the user to know it, don't send it to the client. So my sympathetic view of the cheaters comes not from my sympathetic view of cheating, but from my attitude towards client-side security.
Article: 'avoid explorer'. Note that i'm not saying they are right.
gTsiros' advice: 'be careful where you browse'.
My response, rephrased:
THIS IS JUST PLAIN FUCKING STUPID. IF YOU BELIEVE THAT I.E. IS SECURE, KEEP USING IT. IF NOT, DROP IT. USING INSECURE BROWSER IS STUPID, EVEN IF YOU THINK THAT YOU ARE ONLY BROWSING 'SAFE' SITES.
Thank you for your attention.
8 bits should be enough for everyone.
1. Their common ancestor had this piece of DNA. This is not improbable, but we are talking about pieces of DNA that aren't expressed (that's what introns are), so there is little or no evolutionary pressure on them, so one must explain why the closer relatives dumped them.
2. They same piece of DNA evolved independently in both species. Not very probable, see 1 - no pressure on introns, so convergence not very probable.
3. Horizontal gene transfer occured.
4. The phylogenetical tree is wrong. Well, it seems that however we arrange the tree, some introns do stick out.
Of course, this is not an established scientific proof, but it seems that scientists find more and more of evidence supporting point 3. By the way, crossbreeding is not the only possibility of HGT - viruses are capable of transferring genetic material (at least their own) between organisms, so they may be responsible for HGT. Again, google for 'introns' 'horizontal gene transfer'.
PS. Feel free to point out oversimplifications in the above post.
This only means that traffic generated by one of those devices fits into the Wi-Fi bandwith (I assume that the devices actually work). This does not mean that it leaves room for anything else. Of course, it does not mean that it does not. The article states: So I admit that the 1024x768 x bpp x FPS estimate is probably was probably too high - sound like they don't transmit every frame blindly. But the amount of data needed to re-create a fast changing window (any kind of animation) can still be huge.
I hope they're using some sort of compression. The typical 1024x768 pixels x 8 bit per pixel (let's be generous) x 16 frames per second (ditto) gives... well, a really huge number of bits per second :-).
See telnet://blinkenlights.nl