NARIS: The Fly on the Wall Your Mother Warned You About

[Belmaria]

NARIS - Nationstates Anti-Raiding Intelligence Service

Human Intelligence Project

NARIS maintains a human intelligence project which aims to gather information from within known raider organizations. At present, this project is our primary focus. If you would like to become a spy for NARIS, telegram Belmaria for more information.

Heuristic Detection Project

After working on a software project for over a year, members of NARIS have put together an alpha stage heuristic engine intended to be used to detect the movement of known and suspected raider nations into susceptible regions, and are in the early stages of testing this engine against actual raiders using the NationStates API and web interface.

Any software developers or statisticians that would like to help NARIS end raiding for good should telegram Belmaria for more information.

EDIT: Due to an apparent lack of interested individuals with the necessary skills to assist the project, development will take a back seat to human intelligence work until further notice.

Raider Blacklist Project

Although currently inactive, the raider blacklist may some day become relevant again, and as such, will be listed as a project of NARIS.

[Tim-Opolis]

Engine intended to be used to detect the movement of known and suspected raider nations into susceptible regions

You mean that fancy feature that is the Activity and Reports pages? Crazy how that already exists. As for whatever detection figures out what is and isn't invader, sounds fancy, I think I'll stick to my years of experience with spotting rather than some loudly proclaimed algorithm from someone I've never seen defending unless this is an alias.

and are in the early stages of testing this engine against actual raiders using the NationStates API and web interface.

Really now? And which of the regular updaters are you, as what we could really use are more numbers for the fight rather than more scripts. Afterall the best way to test your script is actually defend during update

[Belmaria]

Engine intended to be used to detect the movement of known and suspected raider nations into susceptible regions

You mean that fancy feature that is the Activity and Reports pages? Crazy how that already exists. As for whatever detection figures out what is and isn't invader, sounds fancy, I think I'll stick to my years of experience with spotting rather than some loudly proclaimed algorithm from someone I've never seen defending unless this is an alias.

and are in the early stages of testing this engine against actual raiders using the NationStates API and web interface.

Really now? And which of the regular updaters are you, as what we could really use are more numbers for the fight rather than more scripts. Afterall the best way to test your script is actually defend during update

1. Our system uses the world and region APIs, which include access to the dossier functionality.
2. Watching all of the world's relocations would be a waste of time, which is why this system is necessary. The heuristics in our engine cut out the fluff and only give us possible raiding activity to focus on, which saves a lot of time.

[Tim-Opolis]

You mean that fancy feature that is the Activity and Reports pages? Crazy how that already exists. As for whatever detection figures out what is and isn't invader, sounds fancy, I think I'll stick to my years of experience with spotting rather than some loudly proclaimed algorithm from someone I've never seen defending unless this is an alias.


Really now? And which of the regular updaters are you, as what we could really use are more numbers for the fight rather than more scripts. Afterall the best way to test your script is actually defend during update

1. Our system uses the world and region APIs, which include access to the dossier functionality.
2. Watching all of the world's relocations would be a waste of time, which is why this system is necessary. The heuristics in our engine cut out the fluff and only give us possible raiding activity to focus on, which saves a lot of time.

Let's cut the bullshit, eh lad? How much experience with R/D do you have, like actual on-field experience. Because this is me, an R/Der of 6.5 years, telling you that this does not look like an effective system. You have in no way demonstrated how it's practically better than using the combination of the activity and reports page, you haven't even demonstrated that this script even exists, you certainly don't have anybody buying its legality until a source code is posted up for analysis, and you most definitely haven't demonstrated any significant amount of R/D experience that should make any defender look at this and conclude "Sure, this looks viable".

Also, watching all of the world's relocations is not even close to a waste of time, and whatever your algorithm is, the only way it could catch all invader nations is if it literally does look at *all* the world's relocations, unless you're just cutting the very basic stuff like moves to GCRs which even then have frequently been invader in the cases of places like Balder and such.

[Belmaria]

1. Our system uses the world and region APIs, which include access to the dossier functionality.
2. Watching all of the world's relocations would be a waste of time, which is why this system is necessary. The heuristics in our engine cut out the fluff and only give us possible raiding activity to focus on, which saves a lot of time.

Let's cut the bullshit, eh lad? How much experience with R/D do you have, like actual on-field experience. Because this is me, an R/Der of 6.5 years, telling you that this does not look like an effective system. You have in no way demonstrated how it's practically better than using the combination of the activity and reports page, you haven't even demonstrated that this script even exists, you certainly don't have anybody buying its legality until a source code is posted up for analysis, and you most definitely haven't demonstrated any significant amount of R/D experience that should make any defender look at this and conclude "Sure, this looks viable".

Also, watching all of the world's relocations is not even close to a waste of time, and whatever your algorithm is, the only way it could catch all invader nations is if it literally does look at *all* the world's relocations, unless you're just cutting the very basic stuff like moves to GCRs which even then have frequently been invader in the cases of places like Balder and such.

I don't expect people like you to even understand, let alone support, what we're doing here. If something isn't physical, tangible, and two feet in front of your face, you probably wouldn't acknowledge its existence. We understand abstract concepts, obviously far more effectively than you do.

Regarding how our system is more effective, we look for statistical signs of raider activity. We have compiled a partial list of statistics regarding actions which may constitute raider activity, and use those statistics in conjunction with our engine to cut out unnecessary activity from our dossiers.

[Dream Killers]

Let's cut the bullshit, eh lad?

This entire thread summed up in a single line.

He has asked you for basic information that, if provided, would go far into substantiating your claims that you know enough to develop a tool that could effectively be used in this manner. He has very clearly demonstrated that he knows R/D enough to retort any base comments of not understanding how this meta-game works -- especially from the side you are going to partake in -- and yet you respond with condescending remarks that ultimately waste his time. If he is to be an ally in your fight against Raiders such as myself this is not a good way to start that relationship.

But what do I know? I am just a filthy raider, right?

[Belmaria]

Let's cut the bullshit, eh lad?

This entire thread summed up in a single line.

He has asked you for basic information that, if provided, would go far into substantiating your claims that you know enough to develop a tool that could effectively be used in this manner. He has very clearly demonstrated that he knows R/D enough to retort any base comments of not understanding how this meta-game works -- especially from the side you are going to partake in -- and yet you respond with condescending remarks that ultimately waste his time. If he is to be an ally in your fight against Raiders such as myself this is not a good way to start that relationship.

But what do I know? I am just a filthy raider, right?

Our strategy does not rely upon alliances with defenders. And regardless, I have no control over the subjective opinions of anyone.

What, specifically, do either of you want to know about this project? I haven't heard a single noteworthy question yet.

[Ever-Wandering Souls]

So you're looking to track potential sleepers more than taggers, yes?

[Belmaria]

So you're looking to track potential sleepers more than taggers, yes?

Yes. This is a way to use behavioural analysis to find new, clean puppets. It will augment our previous attempt to maintain a database of known dirty nations.

[Tim-Opolis]

I don't expect people like you to even understand, let alone support, what we're doing here. If something isn't physical, tangible, and two feet in front of your face, you probably wouldn't acknowledge its existence. We understand abstract concepts, obviously far more effectively than you do.

Regarding how our system is more effective, we look for statistical signs of raider activity. We have compiled a partial list of statistics regarding actions which may constitute raider activity, and use those statistics in conjunction with our engine to cut out unnecessary activity from our dossiers.

My, my, aren't we on a high horse. I can't understand what you're doing here? That may be because of actually explaining it to folks you're dropping buzz words as if you're some genius here to save all of us heathens in GP.

Cool, so what are these statistical signs of invader activity? You're talking a lot of talk, sure, but until people can actually see some of the work you're claiming to have done, you're just like every other person prancing into Gameplay thinking that you've got it all figured out and that we're all just godless heathens who haven't been playing the game to its full potential. Much like all of those, I expect you'll fade away, unless you prove my suspicions wrong.

Our strategy does not rely upon alliances with defenders. And regardless, I have no control over the subjective opinions of anyone.

What, specifically, do either of you want to know about this project? I haven't heard a single noteworthy question yet.

Great, but if you're in the business of protecting regions from invasion, you're kind of stuck working with the people who do the job day in and day out. Whatever this tool is, if it's real, you're going to have to sell the idea of it to the defender faction, afterall. Here's a fun fact, you're not doing that too effectively at this point in time.

So, since you apparently can't discern my questions from the stuff I'm posting, here's a convenient list.

1. What experience within R/D are you bringing to the table that makes you able to competently make a tool like this? It takes an understanding of both invader tactics and update dynamics afterall to create this tool, and you haven't demonstrated that.
2. You're using a lot of fancy terms, but it'd help if you'd make it clear for us poor godless heathens. What specifically, without any of the bullshit terminology, is this tool doing that makes it more effective than just consistent day to day spotting.
3. Can we see any evidence that this actually exists?
4. Once this releases, will you also release a source code for the script? If not, why not?
5. What in the world is THIIS, because I don't think anybody's ever heard of it.
6. What are the various things this engine is factoring to make its conclusions? Sure, you can say behavioral analysis, but what does that entail?

[Belmaria]

I don't expect people like you to even understand, let alone support, what we're doing here. If something isn't physical, tangible, and two feet in front of your face, you probably wouldn't acknowledge its existence. We understand abstract concepts, obviously far more effectively than you do.

Regarding how our system is more effective, we look for statistical signs of raider activity. We have compiled a partial list of statistics regarding actions which may constitute raider activity, and use those statistics in conjunction with our engine to cut out unnecessary activity from our dossiers.

My, my, aren't we on a high horse. I can't understand what you're doing here? That may be because of actually explaining it to folks you're dropping buzz words as if you're some genius here to save all of us heathens in GP.

Cool, so what are these statistical signs of invader activity? You're talking a lot of talk, sure, but until people can actually see some of the work you're claiming to have done, you're just like every other person prancing into Gameplay thinking that you've got it all figured out and that we're all just godless heathens who haven't been playing the game to its full potential. Much like all of those, I expect you'll fade away, unless you prove my suspicions wrong.

Our strategy does not rely upon alliances with defenders. And regardless, I have no control over the subjective opinions of anyone.

What, specifically, do either of you want to know about this project? I haven't heard a single noteworthy question yet.

Great, but if you're in the business of protecting regions from invasion, you're kind of stuck working with the people who do the job day in and day out. Whatever this tool is, if it's real, you're going to have to sell the idea of it to the defender faction, afterall. Here's a fun fact, you're not doing that too effectively at this point in time.

So, since you apparently can't discern my questions from the stuff I'm posting, here's a convenient list.

1. What experience within R/D are you bringing to the table that makes you able to competently make a tool like this? It takes an understanding of both invader tactics and update dynamics afterall to create this tool, and you haven't demonstrated that.
2. You're using a lot of fancy terms, but it'd help if you'd make it clear for us poor godless heathens. What specifically, without any of the bullshit terminology, is this tool doing that makes it more effective than just consistent day to day spotting.
3. Can we see any evidence that this actually exists?
4. Once this releases, will you also release a source code for the script? If not, why not?
5. What in the world is THIIS, because I don't think anybody's ever heard of it.
6. What are the various things this engine is factoring to make its conclusions? Sure, you can say behavioral analysis, but what does that entail?

1. I don't have any experience in defense or raiding. You don't have to have experience in a field to understand it.
2. It's not my fault your vocabulary is tiny. It sounds like you're getting pissy because someone showed up with a larger cranium than yours, and you don't want to be out shined. Nonetheless, this system is more efficient at handling the huge amount of data on the activity page than just blindly trying to decode the raw data on your own, or even with a team. It coordinates statistical probabilities with the data available through the API, and "spits out" what it thinks are the most likely instances of raids about to take place. Think of it as an advance warning system. It looks for signs of raids and gives us a chance to prevent them before they have a chance to take place.
3. We'll release information once it's out of testing.
4. We intend to release the source once we've finished it.
5. THIIS was a failed attempt at a proactive intelligence service. We originally sought to decode raw data and provide advance warning to regions we believed were going to be raided. That proved too time consuming, so we suspended our original efforts to focus on this system.
6. At this time, we can't go over the specifics, but some of the things we look at include population size of the region, size and speed of the movement of nations into a region, known affiliations, WA statuses, and (starting in today's test) the executive power of a region's WA delegate.

[Khronion Denral]

What the hell, I'll bite.

How do you deal with the 20-second delay in the API, which has limited the accuracy of real-time software tools? If you can't respond faster than 20 seconds, you're still too slow to detect well-triggered R/D actions.

If you have a way around that 20-second delay, you better be ready to defend its compliance with the NationStates terms of service, because everyone involved in R/D is keenly aware of the potential for rule-breaking scripts.

And even then, it sounds like you're trying to solve a problem that doesn't exist. The Grey Wardens have already used in-game filters to reliably filter out non-raider activity from the in-game activity feeds, which are not subject to the same restrictions placed on the API. If a team of humans can reliably detect invader movements during an update, why bother with a software tool?

[Belmaria]

What the hell, I'll bite.

How do you deal with the 20-second delay in the API, which has limited the accuracy of real-time software tools? If you can't respond faster than 20 seconds, you're still too slow to detect well-triggered R/D actions.

If you have a way around that 20-second delay, you better be ready to defend its compliance with the NationStates terms of service, because everyone involved in R/D is keenly aware of the potential for rule-breaking scripts.

And even then, it sounds like you're trying to solve a problem that doesn't exist. The Grey Wardens have already used in-game filters to reliably filter out non-raider activity from the in-game activity feeds, which are not subject to the same restrictions placed on the API. If a team of humans can reliably detect invader movements during an update, why bother with a software tool?

The defenders are a team of single-shot riflemen, and we have created a machine gun. That's why.

Regarding the delay in the API, it's not something we can work around from a technical standpoint. We are working on a browser plugin that allows multiple users to use our engine, which will decrease the effective wait time.

[Khronion Denral]

The defenders are a team of single-shot riflemen, and we have created a machine gun. That's why.

Regarding the delay in the API, it's not something we can work around from a technical standpoint. We are working on a browser plugin that allows multiple users to use our engine, which will decrease the effective wait time.

Have you seen what The Grey Wardens are doing? They are by no means a "team of single-shot riflemen." Anyhow, I'm still not convinced you'll be able to detect sudden activity that occurs within 20 seconds of a region updating without breaking the game's rules.

Nonetheless, I'm interested to see what you come up with. I assure you that when you release the tool, everyone will be clamoring to understand how it works. This community has every reason to be skeptical tools that claim to change the playing field, and I assure you any tools will be thoroughly analyzed for full compliance with all aspects of the game's rules.

[Tim-Opolis]

1. I don't have any experience in defense or raiding. You don't have to have experience in a field to understand it.
2. It's not my fault your vocabulary is tiny. It sounds like you're getting pissy because someone showed up with a larger cranium than yours, and you don't want to be out shined. Nonetheless, this system is more efficient at handling the huge amount of data on the activity page than just blindly trying to decode the raw data on your own, or even with a team. It coordinates statistical probabilities with the data available through the API, and "spits out" what it thinks are the most likely instances of raids about to take place. Think of it as an advance warning system. It looks for signs of raids and gives us a chance to prevent them before they have a chance to take place.
3. We'll release information once it's out of testing.
4. We intend to release the source once we've finished it.
5. THIIS was a failed attempt at a proactive intelligence service. We originally sought to decode raw data and provide advance warning to regions we believed were going to be raided. That proved too time consuming, so we suspended our original efforts to focus on this system.
6. At this time, we can't go over the specifics, but some of the things we look at include population size of the region, size and speed of the movement of nations into a region, known affiliations, WA statuses, and (starting in today's test) the executive power of a region's WA delegate.

1. Sorry to break it to you, but with R/D you really do have to have on-field experience. Your charge and proclamation in The Mountains to the East earlier today says all that needs to be said.
2. How nice. It's a real shame the majority of raids, however, happen in split-second update instances, where the prevention of them is more about ensuring that we get enough people inside the region within the 5 second or so window we have to stop the invasion. Sure, your tool could stop non-update moves, but we do that easily enough at an almost casual pace without any issues.
3. Cool
4. Splendid
5. I can see how it failed, then.
6. Fair play, I guess I'll hold out to see if it actually exists.

The defenders are a team of single-shot riflemen, and we have created a machine gun. That's why.

Regarding the delay in the API, it's not something we can work around from a technical standpoint. We are working on a browser plugin that allows multiple users to use our engine, which will decrease the effective wait time.

Sure, but if we're single-shot riflemen, we're actually hitting the target and deploying, whereas your apparent lack of interest in defending means you're just firing blanks.

Will this decrease the effective wait time to something faster than the 1-2 seconds between refreshes that Defenders are using with great effect on the activity and reports pages? Doesn't seem too likely, mate.

[We Are Not the NSA]

1. I don't have any experience in defense or raiding.

Well, that explains why you're endorsing a raider point in The Mountains to the East.

Edit: And guess what? I just reported both the nations involved, because he is multying:

From the RMB of the Pacific:

Hi. I'm Henington but I'd like to use this nation to be apart of the Pacific, it seems like this region got a shiny condemnation too.

[RiderSyl]

You're the same person behind 'THIIS'? I actually remember that disaster.

This new thingy of yours, if it's not a complete failure, will likely break the script rules.
Do not want.
Will not want.

[Evil Wolf]

This new thingy of yours, if it's not a complete failure, will likely break the script rules.

Yup, these days I would be extremely skeptical of any script that does not post it's code for all to see.

If I were a Defender, I would be wary of any such "miracle scripts" lest another Predator-like event occur, but this time effecting Defenders. After all, the Moderation has made it clear that ignorance of a script's illegality does not make you exempt from the harshest of punishments.

[Dream Killers]

1. I don't have any experience in defense or raiding. You don't have to have experience in a field to understand it.

http://www.nationstates.net/page=rmb/postid=20255977
http://www.nationstates.net/page=rmb/postid=20258698
http://www.nationstates.net/page=rmb/postid=20258955
http://www.nationstates.net/page=rmb/postid=20259042
http://www.nationstates.net/page=rmb/postid=20259100

Right.
(screenshots for posterity http://imgur.com/a/SnlBN)

The defenders are a team of single-shot riflemen, and we have created a machine gun. That's why.

I don't know about you, but when someone who is talking about NationStates scripts and tools refers to what they are supposedly developing as a "machine gun" it does not sound good.

[Queen Yuno]

Yikes.
Before you submit any script, GHR it with moderators. Just in case.

The prospect of having to use more critical thinking when creating clean puppets...
I'm not sure how a software can detect clean puppet activity? Besides common sense with a new nation suddenly moving into a discreet region o.o

[Ayvari]

1. I don't have any experience in defense or raiding. You don't have to have experience in a field to understand it.

My automatic response: Yes, you do. If you think otherwise, you are very silly. You can make a study of it and think you have it down, but you will absolutely never fully understand it until you get on the field and get yourself some experience.

[Raionitu]

So if I understand this right, your thing would not be aimed for tag runs, it would be for finding sleepers and predicting occupations before they happen, correct?
And that would be based on the 'patterns' that raider sleepers usually follow?
Also, what would you do one you detect a suspected raider sleeper, supposing my above assumptions are true?

[Direwolf Defender]

So if I understand this right, your thing would not be aimed for tag runs, it would be for finding sleepers and predicting occupations before they happen, correct?
And that would be based on the 'patterns' that raider sleepers usually follow?
Also, what would you do one you detect a suspected raider sleeper, supposing my above assumptions are true?

Based on his own patterns? Move in and endorse the sleeper, most likely.

[Sedgistan]

2. It's not my fault your vocabulary is tiny. It sounds like you're getting pissy because someone showed up with a larger cranium than yours, and you don't want to be out shined.

*** Warned for flamebaiting ***

[Ever-Wandering Souls]

I mean, for all the crap you guys are giving him, he does have a point. You're looking at it from an active defense of the region at update perspective, when manually watching has a certain advantage. He's looking at it from other perspective- watching the feed 24/7. It's not inconceivable to be able to mechanically track potential poorly made sleepers. I'd be interested to see a numbers crunch along these lines. If I'm understanding right, it's a mixture of tracking the likelihood of a region to be a sleeper target (size, founderless, executive delegacy) and matching that with risk factors like activity, movement, etc. ...So, like, a puppet that's made, and moves directly to a UCR that rates high on the risk factors list, and is largely inactive while there might trigger a certain risk rating.

My question would run more along the lines of how much chaff is in there, is there too much to be useful.

I think, if I'm right, that this is not only possible but could benifit from input on the updater side. There are factors that could be vital that you, as a non-updater, might not think of. For example, not all sleepers are in the wa while sleeping, but WA activation and a lot of endorsing others could be a risk factor. One cool way this could be executed is as a willing update watcher - I.e., Tim sees invaders massing, knows a raid is on for tonight, and can pull this up to see a handful of regions that currently have suspicious nations in them, that are in the wa (needed to be jumped on), maybe even endo gaps, narrow down using endo gap and how many raiders there are, and ideally have a list of four or five regions - small enough that they could preemptitively defend them, with the number of potential targets decreasing as update goes on. Even without the nation tracking bit, something like that could be useful even for raids without a sleeper - a quick list of large, founderless, regions within a specific endorsement range, in update order. Your faster switchers may be able to jump in and out of twenty or thirty such regions, under the right limitations.

I think you're wrong to dismiss this entirely. I see potential, if executed right. Is it more effective than leaks and straight following? I'm not sure. But it could be worth trying. Maybe you should both play nice and realize you could help each other, instead of alternating between attacking the op as not a real defender and attacking their intelligence back? I mean, come on guys, how often do you get someone who actively wants to help defending and is putting in work towards it in their own way?