These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE General Discussion

 
  • Topic is locked indefinitely.
 

At what point is something an Exploit and not game Mechanics ? Bumped for 60 Minutes

First post First post First post
Author
Murk Paradox
Ministry of War
Amarr Empire
#261 - 2013-07-01 21:16:02 UTC
Kaarous Aldurald wrote:
Quote:
Oh, I expected actual instances where the ruling was completely in the bumper's favor. That's usually what someone means when they say precedent.


Well, seeing as you must not have the capabilities to look this up yourself...

http://www.minerbumping.com/

Just over a year of precisely that. The precedent was established by the New Order, in a GM response that cannot be discussed on the forums, but is given in great detail as to the specifics.




Poor example to use.

This post has been signed by Murk Paradox and no other accounts, alternate or otherwise. Any other post claiming to be this holder's is subject to being banned at the discretion of the GM Team as it would violate the TOS in regards to impersonation. Signed, Murk Paradox. In triplicate.

Murk Paradox
Ministry of War
Amarr Empire
#262 - 2013-07-01 21:18:46 UTC  |  Edited by: Murk Paradox
EDIT-Screw it, changed my mind, the formula given isn't correct enough to give a true dps rating.

The amount of time given and the dps required for the EHP listed is just under 115.

This post has been signed by Murk Paradox and no other accounts, alternate or otherwise. Any other post claiming to be this holder's is subject to being banned at the discretion of the GM Team as it would violate the TOS in regards to impersonation. Signed, Murk Paradox. In triplicate.

S Byerley
The Manhattan Engineer District
#263 - 2013-07-01 21:32:43 UTC  |  Edited by: S Byerley
James Amril-Kesh wrote:
S Byerley wrote:
Now you're just being silly; computers are quite smart, especially when analyzing something already broken down into 1's and 0's. Your inability to come up with a naive solution doesn't indicate much of anything.

Uh no, they aren't. Computers can compute, and they can do it really well. They can't come up with solutions of their own to the more broad-reaching types of problems that humans face.


That's just your flesh sack pride talking.

To put things in perspective: A full simulation of the human brain takes about an exaflop (+/- an order subject to debate). We're currently in the tens of petaflops and the exaflop projections are for ~2020. Keep in mind, that's a full simulation, fundamentally more powerful.

Now, consider how much of the brain is tied up in mundane tasks and specialization. We're so bad at numerical math, for example, because we have to do several extremely expensive conversions to get it in the right format - it's an inefficient hack.

It shouldn't be surprising that computers are already better at most applications. The specialized tasks are starting to fall as well and the limiting factor is often the algorithm rather than the hardware (A couple million years of genetic tuning gives us a pretty wicked head start as far as optimizations).

James Amril-Kesh wrote:
They can't, for example, determine intent or what would constitute harassment.


You say that like humans can. No, we use heuristics; computers are more than capable of doing the same.
Murk Paradox
Ministry of War
Amarr Empire
#264 - 2013-07-01 21:49:29 UTC
baltec1 wrote:
Ace Uoweme wrote:


If webbers were designed to pull. They're designed to hold a target still.

Thus, a very messy mechanic.


Same job, they help manoeuvre the ship.



Funnily enough, I think of bumpers as tugboats.

Tough little sausages that they are.

This post has been signed by Murk Paradox and no other accounts, alternate or otherwise. Any other post claiming to be this holder's is subject to being banned at the discretion of the GM Team as it would violate the TOS in regards to impersonation. Signed, Murk Paradox. In triplicate.

Murk Paradox
Ministry of War
Amarr Empire
#265 - 2013-07-01 22:01:33 UTC
Gallowmere Rorschach wrote:
E-2C Hawkeye wrote:

Yea I have to agree....bumping someone for an hour is not working as intended.

How so? If I smash something repeatedly with a big enough hammer, it's going to keep moving, no matter how long I swing at it. Well, until it runs into something that I don't have the force to move, like a brick wall, or a tree.



Here's my take on it. Crimewatch. Their timers. Your aggression timers. A Victim has the same timer as the aggressor. 15 minutes right? (not counting the fun LE timer =)). That timer should NOT be able to be refreshed until a certain amount of time has elapsed. Why? Because of that same timer.

If you cannot accomplish your goal in 15 minutes you shouldn't be attempting it. I know I know, there's going to be blowback about structure grinding etc... but seriously, if you start a fight with something NOT shooting back.... there should be an escape timer.

And this is why-

Eve is cold, and harsh. It should punish the incompetent.

That goes for the pirates too. This is where Crimewatch and Concord need to be on the same page.

Using the aggression timer on a victim who cannot actually make an aggression timer, so you can buy yourself more time to accomadate your inability to kill a freighter is, well, for lack of better terms, exploiting a mechanic.

Just my opinion. I think anyone should be able to kill and shoot anyone. But I also think you need to be able to do the job you pretend to do without having to use a mechanic to allow you to "safely".

At a certain point it should be considered "excessive".

Or try to convince everyone else that Eve shouldn't be a sandbox.

This post has been signed by Murk Paradox and no other accounts, alternate or otherwise. Any other post claiming to be this holder's is subject to being banned at the discretion of the GM Team as it would violate the TOS in regards to impersonation. Signed, Murk Paradox. In triplicate.

Khanh'rhh
Sparkle Motion.
#266 - 2013-07-01 22:10:04 UTC
S Byerley wrote:
In any case, credit card fraud is always the default example in data mining (and quite applicable) so here's a quick google result: http://news.byu.edu/archive12-sep-frauddetection.aspx

I refer you to my previous posts where I said you could neither:

- Show a computer model that shows intent *
- Do so without changing the conditions of the rules **
- As a bonus, your "best example" achieves 90% accuracy and the banks openly admit this merely flags the account, it doesn't act as a judge.

* - in this example, the data is it's own system and doesn't need to make external inferences. For instance - your system can't take one or two transactions and show fraud; it's just comparing raw data.
** - I have already shown that you need to demonstrate how it can distinguish between identical data sets giving different conclusions. Your model cannot do this.

Look son, you're way out of your depth on this. I've torpedoed your argument many times over and you're not even trying to argue it's possible within the terms of the EvE rules, you're just tossing noise.

The "you can show journals" line was a ploy, by the way - it demonstrates you're basically just hacking at google searches and don't really know what research is being done in the area. I've read recent developments in heuristic analysis and I can tell you we're 10's of years and a leap in computing technology away from doing what you want.
Quantum state computing is essentially a pre-requisite for the kind of pattern analysis you're looking for, by the way. Not sure what the cost of those was 40 years ago because they don't exist today.

Or, to put it in other terms - what you're asking for is several **orders of magnitude** more complex than being able to predict every share price rise and fall for the next 12 months.

You are, in every possible facet of human possibility, wrong. All you're doing by clutching at non-related web articles is showing that, not only are you wrong, but you lack the basic knowledge to understand WHY you are wrong.

We've tried to help but you don't want to know that you're wrong, so is there further point in bashing you over the head on this?

Actually I'm not done .. there's literally a 20metre section of books in my local library which, in a meta context, is basically the sum of knowledge about why you are wrong. No, not figuratively - literally.

S Byerley wrote:
Closest I could find was kernite; too lazy to search properly Ugh

http://eve.battleclinic.com/killboard/killmail.php?id=18057395


This just shows that your inability to see our contacts list is working as intended, I assume?

"Do not touch anything unnecessarily. Beware of pretty girls in dance halls and parks who may be spies, as well as bicycles, revolvers, uniforms, arms, dead horses, and men lying on roads -- they are not there accidentally." -Soviet infantry manual,

Khanh'rhh
Sparkle Motion.
#267 - 2013-07-01 22:12:30 UTC
Oh, and your article isn't even about credit card fraud .... I missed that you claimed that. Actually being able to predict whether a single transaction is fraudulent or not would be a multi-billion dollar breakthough.

I mean, if you think you can use 40 year old technology to achieve this why in the **** aren't you out doing it instead of crying about pretend spaceships in a game?

"Do not touch anything unnecessarily. Beware of pretty girls in dance halls and parks who may be spies, as well as bicycles, revolvers, uniforms, arms, dead horses, and men lying on roads -- they are not there accidentally." -Soviet infantry manual,

Khanh'rhh
Sparkle Motion.
#268 - 2013-07-01 22:18:23 UTC
You know what ... I missed a trick. It's been so long that someone was blind enough to simply say "the stats are right because the stats show it" that I forgot the most simple, most basic tenet of this type of mathematical analysis:

Correlation does not equal causation.

In context - data showing someone was being bumped for X times over Y locations does not, and cannot, tell you why.

We need to know the why, because we punish based on the why.

By definition, no model which tells you what happened can tell you why.

Just adding to the pile of "you're so wrong, so very very very measurably wrong" I have got going here.

"Do not touch anything unnecessarily. Beware of pretty girls in dance halls and parks who may be spies, as well as bicycles, revolvers, uniforms, arms, dead horses, and men lying on roads -- they are not there accidentally." -Soviet infantry manual,

Cipher Jones
The Thomas Edwards Taco Tuesday All Stars
#269 - 2013-07-01 22:26:24 UTC
Quote:
He hasn't made an effort to move to another location, and they weren't following him around, so no.


Literally impossible to bump someone more than once without following them.

internet spaceships

are serious business sir.

and don't forget it

Gallowmere Rorschach
The Scope
Gallente Federation
#270 - 2013-07-01 22:28:41 UTC
Cipher Jones wrote:
Quote:
He hasn't made an effort to move to another location, and they weren't following him around, so no.


Literally impossible to bump someone more than once without following them.

Touché salesman, touché.
Aiwha
School of Applied Knowledge
Caldari State
#271 - 2013-07-01 22:57:11 UTC
Did GSF lose their ships each time they ganked? If they did, then that's the way the cookie crumbles, scout out gates for traps next time, blah blah blah. It happens all the time.


If they didn't lose their ships with each gank, then that's an exploit.



And bump tackle isn't an exploit. You had an hour to hire a merc group, there's a whole channel of reputable merc's who would have easily dispatched a half dozen gank BC's for a few hundred mil.

Sanity is fun leaving the body.

Vaju Enki
Secular Wisdom
#272 - 2013-07-01 23:06:11 UTC
Themeparkers don't understand EvE Online.

The Tears Must Flow

Rich Uncle PennyBags
EVE Online Monopoly
#273 - 2013-07-01 23:07:37 UTC
As far as I understand, the line is thus:

If they bump for:
Ransom, Tactical advantage, waiting for backup, any other reason with an end goal.
It's A-O.K.

If they do it for hours on end without any recognizable goal or reason other than to cause trouble and be dicks, you have a case.

S Byerley
The Manhattan Engineer District
#274 - 2013-07-01 23:17:21 UTC
Wow, haven't seen this much incoherent rambling in a while; ah well, here we go

Khanh'rhh wrote:

- Show a computer model that shows intent *
- Do so without changing the conditions of the rules **
- As a bonus, your "best example" achieves 90% accuracy and the banks openly admit this merely flags the account, it doesn't act as a judge.


What is it with you guys and "intent"? It's not relevant and it's not adequately defined so I guess if it's a requirement of your.... challenge?, I can just skip the rest.

Khanh'rhh wrote:
The "you can show journals" line was a ploy, by the way - it demonstrates you're basically just hacking at google searches and don't really know what research is being done in the area. I've read recent developments in heuristic analysis and I can tell you we're 10's of years and a leap in computing technology away from doing what you want.


I never said I could "show journals". I mean, I could, but it's not worth digging for a reference you won't be able to understand anyway. I don't keep track of them because I don't publish in KDD. Where exactly are you reading about "heuristic analysis" btw? I can't say I've ever heard anyone use that phrase; or are you just mashing buzz words together?

Khanh'rhh wrote:
Quantum state computing is essentially a pre-requisite for the kind of pattern analysis you're looking for, by the way. Not sure what the cost of those was 40 years ago because they don't exist today.

Or, to put it in other terms - what you're asking for is several **orders of magnitude** more complex than being able to predict every share price rise and fall for the next 12 months.


Woah there buddy; slow down. First of all, wtf kind of "pattern analysis" have you decided this is? Second, orders of magnitude are easy; heck, we throw them out in most of complexity analysis. Third, I alluded to old *techniques* because the algorithms haven't evolved all that much.

ROFL@ quantum computing (this is how I know you're just being silly). If you know what does and doesn't fall under the BQP complexity class, a lot of very smart people would like to talk to you.

Khanh'rhh wrote:
Oh, and your article isn't even about credit card fraud .... I missed that you claimed that. Actually being able to predict whether a single transaction is fraudulent or not would be a multi-billion dollar breakthough.


Technically, I never said it was. Credit cards are very much the default example though; not sure why you think no one is doing it.

Khanh'rhh wrote:
You know what ... I missed a trick. It's been so long that someone was blind enough to simply say "the stats are right because the stats show it" that I forgot the most simple, most basic tenet of this type of mathematical analysis:

Correlation does not equal causation.

In context - data showing someone was being bumped for X times over Y locations does not, and cannot, tell you why.

We need to know the why, because we punish based on the why.


Who said anything about causation? The statement you've evidently lost in all your rambling was that data mining could be used to "mimic human judgement" with some degree of accuracy.
Tiber Ibis
The Paratwa Ka
#275 - 2013-07-01 23:49:50 UTC
Kaarous Aldurald wrote:
What I said was, that it requires coordination and planning and preparedness, and that those are rewarded in EVE, which is why they get kills and are working as intended. I also said it was hilariously easy to avoid most gatecamps, and that someone would have to be rather ignorant of how the game works to claim otherwise.

There is nothing difficult about bumping a freighter. This technique is basically shooting fish in a barrel with little to no risk. The only skill required is to be prepared with enough ships to pull off the gank.

Bumping is a lame mechanic, most people agree with this, so by defending you are simply making you look silly. Particularly as you are trying to indicate that it is some sort of elite pvp skill.
Bischopt
Sebiestor Tribe
Minmatar Republic
#276 - 2013-07-02 00:15:58 UTC
I cannot believe the sound your mouse is making.

Is it a tap dancing shoe?
Thugnificent Gangstalicio
Doomheim
#277 - 2013-07-02 00:16:18 UTC  |  Edited by: Thugnificent Gangstalicio
They're bumping him for a reason. It's no exploit, it's simply buying time to set up gank squad.
I love your response in local "I'm frapsing this so you'll be banned".
EVE players pride themselves as being above the WoW players where GMs hold your hand. This guy is obviously not EVE material.
EDIT: Wow, you're trying to stop your login-warp. You're bad.
Diomedes Calypso
Aetolian Armada
#278 - 2013-07-02 00:34:09 UTC  |  Edited by: Diomedes Calypso
This thread has definitely saved me a freighter and cargo. I rarely used it before I left but the change in the scan thing made me wonder if it was better than my orca. I'm not sure why I used it the last week or two..... it seems to have a specialty roll for only shipping things like t1 ships and ore.

That is still a purpose.. I'll just never put modules or components in a freighter after I've learned all this.

I'll act according to the premises below ... mostly # 1 ... the others are all superficial explanations.


1) Capital Ships by definition require a support fleet . (true in many eras of naval warfare )

1a) The size of support needed has to do with the potential engagements that might be encountered... no exact size of support fleet will be optimum for all voyages (you don’t want to draw attention either)
1b) Hi -sec attacks will use different mechanics and require support trained in specific ways ...
1c) there is nothing inherently wrong with a game design that makes all areas of space dangerous due to potential criminal assaults . It makes it more costly to move trade items … more trips in smaller ships or more lost cargo in unprotected giant ships


2) Knowing that your freighter will always be at risk, even with a support fleet “be an un-enticing target is your second Prime Factor in planning industrial logistics.

2a) Not all goods are supposed to be shipped by freighter and its use is primarily worth the risk reward for only cumbersome loads with any additional modules left behind. Transport ships should be used to ship modules… with faction stuff snuck away on unlikely and difficult targets and multiple trips.. Balance doesn’t mean that one ship can be used to ship all types of cargo. What alternatives a player have to move gear are part of the balance calculation.

2b) there should be no presumption that "balance" means 1 player is balanced vs 1 other player. This game is designed as a multiplayer experience and while there may be areas for solo play the use of capital ships does not need to be one of those solo areas. Even with the orca , not only should I have forward and in system scouts but I should have the in system scout on the ship to web after the gate jump


Accept the facts on the ground - developers are aware and made the call. The situation leaves players alternatives.

Quit whining and understand that a freighter only has a situational roll and there is NO MIN/MAX correct decision in terms of which ship to transport which equipment and what risk reward factor for your time that you should bear.

No MIN/MAx... the importance of gageing the personalities and motivations of other peoples behavior or making social bonds to make you more safe. .. OCD and aspberger guys are doomed to mental anguish ! that anguish over a dilema with no one answer is probably the real story.

.

Khanh'rhh
Sparkle Motion.
#279 - 2013-07-02 00:48:51 UTC
S Byerley wrote:
What is it with you guys and "intent"? It's not relevant and it's not adequately defined so I guess if it's a requirement of your.... challenge?, I can just skip the rest

The rules are, that it's against the rules when the intent is to greif - it's not against the rules when the intent isn't to grief. Since the actual rule we're talking about enforcing wavers on someone's interpretation of someone else's intent - it is more than central to the argument; it is the entire argument. Showing that the intent was to grief and not valid gameplay is what makes it against the rules. Maybe you will lightbulb here and go "oh ****, yeah, I was thinking of something else" but you will likely instead grit your teeth and continue to argue black is blue.

A computer algorithm cannot determine intent, ergo it cannot rule on a matter of intent. This was QED about 10 pages back but you're so wilfully belligerent that it's just not sinking in.
Quote:
I never said I could "show journals". I mean, I could, but it's not worth digging for a reference you won't be able to understand anyway

I've demonstrated an ability to grasp the subject at a level far exceeding yours. You'll get no-where trying the "I know so much I can't tell you!" line - try me. Link anything you want. Any source. Anything at all that shows that causation can be determined mathematically by correlation.

I mean, I have asked three times for a single tangible piece of evidence that computer models can accurately determine intent, and you've failed 4 times in coming up with anything.
Quote:
Where exactly are you reading about "heuristic analysis" btw? I can't say I've ever heard anyone use that phrase; or are you just mashing buzz words together?

Kind of - insofar as you simply **can't** do what you want with computer technology as it exists so you would need to construct some manner of pseudo-pattern recognition to get around your inability to measure the data you require. It should be noted this is merely my "best guess" at how one would try to achieve something which is essentially not possible.
The current leading edge in this area is a kind-of multi-tiered pattern analysis, which is many steps below what you need to model the actual why of the origin of the data. Again, with your data-mining to measure causality approach, this is vis-a-vis to intent.
Quote:
If you know what does and doesn't fall under the BQP complexity class, a lot of very smart people would like to talk to you

I don't. No-one does. It's a potential application of a technology not yet invented. Which is why you're reaching so far it's laughable.
Quote:
Second, orders of magnitude are easy

Orders of magnitude past a task we can't perform with current technology is not "easy" - what are you smoking?

The rest of your post just goes on to miss the basic fact that, I think we both know, is staring you in the face as much as it is me: you're still suggesting that a correlational analysis of server data can determine the causation.

If you can show a working model of this, there is literally a nobel prize in it for you.

I'll keep on asking for you to show a single citation that shows that any of this ~~amazing computer analysis~~ is possible.
I have no idea why people like you insist on dragging a discussion way past the point that everyone knows you're wrong. I guess it's just the problem of internet posting (that a literal sequence amplifies the "who posted on the subject last" problem) but GODDAMN man, there's only so many ways you can say nothing of any citable value and not look like a complete fool.

That was pages ago, you do look like a complete fool.

"Do not touch anything unnecessarily. Beware of pretty girls in dance halls and parks who may be spies, as well as bicycles, revolvers, uniforms, arms, dead horses, and men lying on roads -- they are not there accidentally." -Soviet infantry manual,

Jonah Gravenstein
Machiavellian Space Bastards
#280 - 2013-07-02 01:11:25 UTC  |  Edited by: Jonah Gravenstein
Straight Tinfoil : Red Frog Freight, its subsidiaries, and rival Push Industries, have pooled their resources to mount an aggressive marketing campaign, in association with MiniLuv and other freighter inspectors, to increase their revenues and respective market shares Straight

OP, its not an exploit until CCP say it is, if you want their word on it, petition your loss.

This paid advertisment was brought to you by Red Frog, PushX and MiniLuv.Twisted

In the beginning there was nothing, which exploded.

New Player FAQ

Feyd's Survival Pack