Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Lyn Farel]

I have always wondered how the capsule survives everytime to the explosion of the ship it's inside...

[Jace]

Even considering other capsuleers. Most capsuleer aggressors are only after loot and killmails, careless of the technology that makes them who or what they are… let alone how to use that technology against them. There's very little reason to fear capsuleers for doing anything more than the typical. An even additional backing fact: a capsuleer also almost never likes relying on a crew, and many simply deny their crew even exist (as witnessed on the EVE Online forums).

So prospecting a capsuleer wanting to use boarding teams to seize hold of another capsuleer alive… talk about something beyond unheard of. 

But of course, you hang out with characters like ours these days, you tend to be exposed every day to statistical out-of-the-box-thinking freaks.  xD

Yeah, in my fiction it has usually been a capsuleer sending a boarding party to get the crew off of the other ship. And as far as the players that try to deny crews....well they aren't in my arcs.

[Arista Shahni]

I have always wondered how the capsule survives everytime to the explosion of the ship it's inside...

I'm pretty sure it 'ejects',  game wonkiness leaves us at 0 m/s in the middle/next to our wreck.

[Vic Van Meter]

Well, boarding actions on capsuleers might be sanctioned by at least one body I know of, CONCORD.  While it's somewhat uncertain how often capsuleer ships and NPC ships are actually boarded by parties or what kind of actual interior control they have over the situation, I bet CONCORD does it pretty regularly.

CONCORD is the only entity I know of that can make sure that pressing the red button is also lethal for the capsuleer since they can kill the signal and cut off access to the clones.  Then, I'll bet if they want to get a capsuleer alive for information, they can definitely do it.

Other than that, I think it's just one of those things that probably exists and definitely happens on an unroutine basis, but the game is 10 years old and can't handle that kind of an addition without serious, serious overhaul of the engine.  The one they use to walk us around the captain's quarters ain't gonna cut it.  Like I'm sure our ships take damage to the guns that needs to be repaired, warp drives shred a certain number of ships each year due to mechanical failures, and sometimes you need to scrape ice off the windshield.  It's just outside the scope of the actual video game.

On that note, two ideas randomly came to mind.  I wish we could have customizable captain's quarters on our larger ships so that we can have our own little pad instead of the standard station ones.  I also wish we could have a little boarding party launcher highslot thing where you can hire NPC crew.  So then once you're down to the hull of a ship you're shooting at, you can fire your boarding party.  If you can survive for some extra length of time than normally killing the ship would take and win a dice roll, the enemy capsuleer gets ejected and the ship is then flown (automatically) to a station you picked out.  You get the ship (that you would pay repairs to fix) and you'd get the entire contents of the cargo hold.  This would assume the opposing ship hasn't hired a better defense team and wins the dice roll.  We'd also have to take into account that a boarding party you house on a destroyer, no matter how expensive, probably wouldn't be able to take over a Titan so size is an issue we'd have to take into account.

To keep the enemy capsuleer from instantly self-destructing or the aggressive capsuleer from blowing up the ship, neither get to know whether it succeeds or fails until the moment the enemy capsuleer gets ejected, if at all.  So the capsuleer trying to board doesn't know whether his guys have failed and they may still be working and aboard when he decides it didn't work and blows the ship up.  The boarded capsuleer doesn't know whether or not the boarding succeeds or how long he has until he's popped out of the ship.  Or maybe the act of boarding puts a time limit on self-destructing the ship (but neither maybe knows what that is either).

I'm imagining scenarios like a capsuleer sighing at his boarding failing, then blowing up the ship, but at the last second seeing it become his (and smacking his own face).  Or both capsuleers firing boarding parties at each other, being ejected simultaneously, and ending up staring at each other in space in capsules, having swapped ships.  Mostly, though, it kind of takes piracy back to what it's kind of supposed to be.  Sinking a vessel in open waters is essentially a total loss for pirates on the high seas.  It also adds a new dynamic for people attacking fleets of larger ships in that they can hopefully try to sneakily hijack them.

I don't know, it sounds fun. 

[Esna Pitoojee]

Regarding capturing a capsuleer:

Why would the capsule have any architecture (physical or software) that would allow it to accept such disabling commands from anywhere in the ship? CONCORD has their own killswitch built straight into the capsule, and while someone onboard a ship certainly could sever a capsuleer's link to that ship, I'd be quite surprised if they could effectively intrude into the capsule.

[Vic Van Meter]

Regarding capturing a capsuleer:

Why would the capsule have any architecture (physical or software) that would allow it to accept such disabling commands from anywhere in the ship? CONCORD has their own killswitch built straight into the capsule, and while someone onboard a ship certainly could sever a capsuleer's link to that ship, I'd be quite surprised if they could effectively intrude into the capsule.

Sometimes, you need information.

[Morwen Lagann]

Regarding capturing a capsuleer:

Why would the capsule have any architecture (physical or software) that would allow it to accept such disabling commands from anywhere in the ship? CONCORD has their own killswitch built straight into the capsule, and while someone onboard a ship certainly could sever a capsuleer's link to that ship, I'd be quite surprised if they could effectively intrude into the capsule.

Sometimes, you need information.

Esna's not questioning the why, he's questioning the how.

And tbh he's right to do so - except in the largest classes of ships, the capsule is generally not accessible from within the ship, certainly not in the space of time between a self-destruct being activated and it going off.

[Vic Van Meter]

Regarding capturing a capsuleer:

Why would the capsule have any architecture (physical or software) that would allow it to accept such disabling commands from anywhere in the ship? CONCORD has their own killswitch built straight into the capsule, and while someone onboard a ship certainly could sever a capsuleer's link to that ship, I'd be quite surprised if they could effectively intrude into the capsule.

Sometimes, you need information.

Esna's not questioning the why, he's questioning the how.

And tbh he's right to do so - except in the largest classes of ships, the capsule is generally not accessible from within the ship, certainly not in the space of time between a self-destruct being activated and it going off.

See, that's what always confused me.  It has to be somehow accessible remotely if it's not physically.  Data has to flow to the guns and engines and whatnot scattered around the ship, at the very least.  I always wondered if the remaining crew compliment to a ship was to run things specifically to make sure the capsuleer wasn't vulnerable, but if data can go one way, a boarder could essentially send data the other way from pretty much anywhere on the ship.

Theoretically, anyway.

[Esna Pitoojee]

Physical contact is one thing and I could go on for hours about it; suffice to say I presume in most cases it would take a significant amount of cutting and ripping away support equipment to actually reach the capsule. That's my thought, though.

Regarding data: I'm not doubting there are data links from the ship running in and out of the capsule; there absolutely have to be for the data flow. My questioning was more 'why wouldn't there be several layers of significant protection between the brain burner/termination and anything receiving information from the outside'?

For instance:
- Would externally connected subsystems even share control equipment with the burner/termination pack? Given the value of getting a capsuleer home alive, I'd be very surprised if there weren't independent backups for those systems.
- Would externally connected subsystems be programmed to allow commands to be received from an external port? Not being a software science person, I would feel that this be the flimsiest wall of electronic defense - but I'd be shocked if this kind of thing didn't exist in some fashion.
- Is the external connection equipment even configured to allow commands to be transmitted in? Remember, the only people with a vested interest in maintaining an override - CONCORD - can do so from within the pod.

[Vic Van Meter]

Physical contact is one thing and I could go on for hours about it; suffice to say I presume in most cases it would take a significant amount of cutting and ripping away support equipment to actually reach the capsule. That's my thought, though.

Regarding data: I'm not doubting there are data links from the ship running in and out of the capsule; there absolutely have to be for the data flow. My questioning was more 'why wouldn't there be several layers of significant protection between the brain burner/termination and anything receiving information from the outside'?

For instance:
- Would externally connected subsystems even share control equipment with the burner/termination pack? Given the value of getting a capsuleer home alive, I'd be very surprised if there weren't independent backups for those systems.
- Would externally connected subsystems be programmed to allow commands to be received from an external port? Not being a software science person, I would feel that this be the flimsiest wall of electronic defense - but I'd be shocked if this kind of thing didn't exist in some fashion.
- Is the external connection equipment even configured to allow commands to be transmitted in? Remember, the only people with a vested interest in maintaining an override - CONCORD - can do so from within the pod.

Well, I think there would be some significant protection for the data lines, certainly.  But if you're boarding a ship, you're not probably using the front door, so we're probably assuming you've already cut your way in.  It can't be impossible to get at the data lines, since I'd assume just from experience that you'd need to be able to get into them for maintenance.  If you're on a spaceship that contains people, detritus is going to eventually get into places you don't want it to be, if nothing else.  It'd probably be difficult, but I can't imagine the lines are that deeply buried that you couldn't get to them at all.

Protecting the capsuleer with some kind of insulating software might also be an issue, but not necessarily a big one.  Suffice it to say that a signal would have to go to a camera, for instance, take the image the camera is showing, and then send it back.  Like an online download, there's no way of knowing that all the data coming from the camera is coming from the camera if something gets in the way.  It's just a signal that probably gets translated into data the brain can understand.  There's no reason why that same data couldn't register a noise so loud in the brain that you can't think anymore or throw the brain into a seizure.

There's always the possibility that the capsule will always prevent that sort of thing if it comes down the line (like a flare compensator in a camera, for instance), but in that case I'd say electronic warfare is even MORE of a possibility.  After all, that means there's something between the capsuleer and whatever is receiving signals that's making decisions and has to be fairly universal between all existing ships.  In that case, it would be theoretically possible, if you can get into the data, to simply flood the computer with gibberish and crash it.  I guess the BSOD is something else that probably happens semi-routinely in the EVE world that they don't make us deal with for game reasons.

So on those three points:

1.  I think the brain burner is probably not only something that would be most heavily protected, but the hardest to hack since it's tied to something that can be monitored directly from the capsuleer himself (vital signs) or the capsule alone (the state of the capsule).  Ejecting the pod probably isn't as difficult, because that requires input from the ship, some parts of which are very far away, and requires activating ship equipment to eject the pod.  That's why I'm saying it would probably be more likely to eject the pod (or to keep it from ejecting if you had the time or programming to do that, since that's a good way to make sure you've got time to work on the pod if you know your capsuleer isn't going to deep fry his brain to escape somehow).  But yes, making the scanner fire to fry the capsuleer in the pod is probably not something you can do very easily from a remote part of the ship.  It's an autonomous part of the capsule.

2.  This is one of those things where you'd have to take a wild stab at how computers work in EVE.  There is probably some manner of encoding that would make it more difficult to interfere with the signal, but that's counterbalanced by EVE's need to have a capsuleer be able to use any ship they have learned to fly themselves.  Therefore, you can figure that any specific make and model of ship would interact with a capsule in a logical and predictable way.  It's one of those things that could go both ways, though.  Maybe encoding is so complicated that there's no way to decrypt the signal in time, or maybe it's as easy as hooking a portable AI in and giving it a few seconds to work out which combinations are sending what signals it knows must be heading to the capsule.  I'd imagine it's somewhere in between, that if you got into a ship and spliced in, it's probably going to take a minute or so, but it wouldn't be entirely alien.  That may also depend on if the capsuleer decides to dick with his own computers and writes his own encryption.  That might make it harder or less stable to interface with depending on the computer skill of the capsuleer, but would probably make it a lot more difficult to break in.  It's harder to break into a custom Linux system than a Windows computer, after all, just because compatibility issues means you'll know what you're getting into on a PC.

3.  I think this is almost a guaranteed yes.  Just taking the aforementioned idea of a camera into play, you're talking about pouring data into the visual center of the brain, telling it what it is "seeing".    Anything going into the capsule is transmitting some kind of command to do something, and that's not even going into what it can do from outside the pod.  Certainly, commands run all over the ship to make sure the capsuleer can fly in place of a few hundred crew members, so anything automatic that takes a cue from elsewhere is fair game.  To that effect, though, not all commands and lines are created equal.  I'll assume that it's not every power cable that you can tell to fake a destruction signal to the pod to command it to eject.  I would assume that the line for that isn't hard to find, though.  Knowing everything that might happen to your ship means having lots of little sensors everywhere to say when a catastrophic hull breach is about to turn your ship inside out.

All of this is assuming that the capsuleer has a lot of external and internal control over the ship.  If a lot of it is handled by shipboard crew and most of what the capsuleer has to handle is outside the ship, I'd imagine the capsuleer would be more insulated from that kind of thing.  They'd also have a lot less control over what was happening inside the ship.  Since you can't board in-game, it's hard to say either way.  This is all speculation, of course.

[Morwen Lagann]

We know a couple things for certain about the capsule.
- The scanner is designed to go off as soon as ANY unauthorized breach of the capsule is detected. The whole process is pointless otherwise.
- There are only two situations in which a capsule can be ejected from a ship, at least as dictated by game mechanics: by the will of the capsuleer piloting it, or by the ship being destroyed. The capsule being removed from the ship when docked is a similar but separate process.
- There are only two known situations where the capsuleer can be removed from the capsule safely: when docked, or if their ship is large enough to contain what's referred to in PF as a "cross-capsule gantry". These are supposed to be pretty rare if not entirely not-present on anything smaller than a battleship, and even then it's implied they're not seen very often in the first place.

There is absolutely no reason that the capsule should treat people outside of the capsule forcibly opening it as any different from a standard hull breach and not instantly activate the scanner and kill the capsuleer within. None at all.

The neural scanner's systems have no need to be connected to anything outside the confines of the capsule, and it would be mindbogglingly retarded if they were. The only thing they need to do is monitor for threatening fractures in the capsule's hull, and that is something that doesn't need to depend on anything on the outside. Connections the capsuleer needs in order to fly the ship are not (or should not be) connected to the systems that govern the neural scanner, and as far as boarding actions go, this would make them wholly irrelevant.

Not to mention, one would think that the system would be designed such that if it were detected that it was being tampered with, the capsuleer would be notified and have the option of letting the scanner kill them and send them to a waiting medical clone anyway.

On vessels not equipped with CCGs, the capsule is known to be deliberately positioned so that it has maximum protection from everything that might want to get to it. Short of somehow vaporizing the entire ship in a manner that the capsule is taken along with it (hello smartbombs~) you are not going to be reversing a ship's self-destruct sequence by going after the capsuleer, because you simply will not get to them in time. And I don't just mean getting through the capsule to threaten the capsuleer directly, you won't be getting to the capsule itself.

And if your thought, then, is to try disabling the individual self-destruct systems, keep in mind that some of our ships are very, very large - even assuming that the mechanism isn't simply "explode the warp drive and/or power reactor", for most ships the two minutes allotted for self-destruct is going to be nowhere near enough time to disarm whatever explosives are used to do the deed.

tl;dr if you want to capture a capsuleer or prevent self-destruct you are very likely doing it wrong if you're trying to do it through boarding actions.

[Vic Van Meter]

We know a couple things for certain about the capsule.
- The scanner is designed to go off as soon as ANY unauthorized breach of the capsule is detected. The whole process is pointless otherwise.
- There are only two situations in which a capsule can be ejected from a ship, at least as dictated by game mechanics: by the will of the capsuleer piloting it, or by the ship being destroyed. The capsule being removed from the ship when docked is a similar but separate process.
- There are only two known situations where the capsuleer can be removed from the capsule safely: when docked, or if their ship is large enough to contain what's referred to in PF as a "cross-capsule gantry". These are supposed to be pretty rare if not entirely not-present on anything smaller than a battleship, and even then it's implied they're not seen very often in the first place.

There is absolutely no reason that the capsule should treat people outside of the capsule forcibly opening it as any different from a standard hull breach and not instantly activate the scanner and kill the capsuleer within. None at all.

The neural scanner's systems have no need to be connected to anything outside the confines of the capsule, and it would be mindbogglingly retarded if they were. The only thing they need to do is monitor for threatening fractures in the capsule's hull, and that is something that doesn't need to depend on anything on the outside. Connections the capsuleer needs in order to fly the ship are not (or should not be) connected to the systems that govern the neural scanner, and as far as boarding actions go, this would make them wholly irrelevant.

Not to mention, one would think that the system would be designed such that if it were detected that it was being tampered with, the capsuleer would be notified and have the option of letting the scanner kill them and send them to a waiting medical clone anyway.

On vessels not equipped with CCGs, the capsule is known to be deliberately positioned so that it has maximum protection from everything that might want to get to it. Short of somehow vaporizing the entire ship in a manner that the capsule is taken along with it (hello smartbombs~) you are not going to be reversing a ship's self-destruct sequence by going after the capsuleer, because you simply will not get to them in time. And I don't just mean getting through the capsule to threaten the capsuleer directly, you won't be getting to the capsule itself.

And if your thought, then, is to try disabling the individual self-destruct systems, keep in mind that some of our ships are very, very large - even assuming that the mechanism isn't simply "explode the warp drive and/or power reactor", for most ships the two minutes allotted for self-destruct is going to be nowhere near enough time to disarm whatever explosives are used to do the deed.

tl;dr if you want to capture a capsuleer or prevent self-destruct you are very likely doing it wrong if you're trying to do it through boarding actions.

The simple answer is that anything that talks to the ship or receives information from the ship in the capsule is absolutely connected.  It has to be.  Otherwise, we wouldn't be able to fly the ship, fire the weapons, or even order a cup of coffee.  That's the problem with the argument that the capsule is completely insulated that I can see.  If it was, the ship would be a brick and the capsule would be a very expensive cargo hold.  Information has to be flowing in and out.

To know that the ship is going to be destroyed or in the case of a hull breech, the ship has to send a signal to the capsule saying so.  If everything is hardwired to the capsule directly, that means any sensor that is responsible for relaying that information can send a false signal to eject the pod.  For a capsuleer to see what is outside of the ship, he would need a real-time link to a camera on the ship constantly feeding information back and forth, which would be a fairly direct link to the capsuleer's brain.  The capsule wouldn't necessarily know if information from the camera is being subverted, it just feeds the information.

To get the capsuleer out (which I'm assuming has to be possible since the Blood Raiders are said to particularly enjoy capturing live capsuleers for their blood) is probably a difficult process that involves deactivating the ship's self-destruct, deactivating the neural scanner, and preventing the capsule from ejecting.  That's probably not easy to do, regardless of how much the Sani Sabik are supposed to like extracting capsuleer clones.  The ways I can see it happening would be:

1.  The CPU of the ship would need to be overridden so that the eject feature will not work.  I imagine the CPU has to sit between the capsuleer and the rest of the ship to process information, so it's probably accessible from most places that have any kind of data flowing through them.  Not all data lines are created equally, though, so you'd need something that's moving a lot of information you can subvert.  Whether you completely crash the CPU or just disable the trigger, you'd absolutely have to be able to do that to keep the capsuleer from ejecting the pod.

2.  The capsuleer would need to be made unable to activate self-destruct.  One would assume this would be a similar problem from #1, keeping signals from leaving the pod.  Crashing the CPU might be a temporary solution, but crashing the capsuleer would probably be a more direct and nastier solution.  If you can send a signal up the datalines that knocks the capsuleer unconscious (assuming you can do it without tripping the scanner) you'd buy some time.  It would probably be easier to simply invade the CPU and either snap the link the capsule has with the ship or disable the line that sends commands to destroy the ship.

3.  The capsule should be kept from firing the scanner or self-destructing within the ship.  That's the hard part, since I'm assuming that the scanner and the capsule's internal self destruct are all autonomous parts of the capsule (both have to work when the capsule is ejected).  It's also fair to assume, as you said, that an external breach that the capsule recognizes will automatically fire the scanner, as it probably doesn't know the difference between someone snapping the lid off and a ship blowing the lid off.  Still, if you had access to the capsule (and the capsuleer was either incapacitated or unaware of your presence), the capsule is a relatively ubiquitous technology.  As hardened as it probably is, it HAS to be standardized to work with so many ships from so many different races.  It also has to be able to be periodically serviced, so everything can't be completely inaccessible, otherwise you'd be randomly blowing up whole ships because a capsule's fan failed due to lint buildup.

That doesn't make the capsule the easy part to get into physically, but you could buy a lot of time by disabling the capsuleer.  You could feasibly invade the capsule with a computer code any time an executable command or file is opened inside, so any code running into the capsule, any videos or audio files they listen to, anything that feeds information into the capsuleer's brain directly, can be corrupted or contain malicious software.  If the capsule itself can discern between harmful and helpful information to keep the capsuleer safe from this kind of intrusion, like I said, that makes it even more susceptible to electronic interference and hackers, because they don't just have to mess with the capsuleer, just the program that monitors him.  Anything that gets to decide for the capsuleer what is and is not information can be hacked to cut the capsuleer off cold from his computers.

I would say it's possible, but not necessarily very common.  That's a lot that has to all go just right to catch a capsuleer.  Hence I don't see it happening often because of sheer difficulty and when it does happen it's done by extremely well-experienced boarding parties.  I'd say the majority of boarding actions are probably trying to forcibly eject the capsuleer to keep the ship and cargo in one piece.  That seems a lot easier to do.

[V. Gesakaarin]

If you're going to enumerate and penetrate a data network you need to know the protocols that define how that data is exchanged. I don't think there's been any hint that anyone knows the specifics of the man-machine interfacing that governs the capsule and the ship aside from the Jove that designed it, and CONCORD that manages it. I think the control that CONCORD is able to exert over capsuleers is due to the fact only they and the Jove know exactly how the systems work and maintain propriety knowledge over it.

That's aside the fact that the actual physical architecture of a capsule capable ship exists in a world where FTL routers exist, so it might be said to potentially be extremely intricate and also easily compartmentalized since you could just have localized networks linked to an internal FTL router system direct to the pod with multiple redundancies that can be isolated from each other or remapped as needed.

So even if someone overcomes the hurdle of somehow finding out how the Jove manage their information networks and protocols, and whatever else CONCORD has on hand to ensure that they and only they can control capsuleers through their knowledge of it, then there's the fact you're dealing with a network that isn't static but can potentially change on the fly.

[Vic Van Meter]

If you're going to enumerate and penetrate a data network you need to know the protocols that define how that data is exchanged. I don't think there's been any hint that anyone knows the specifics of the man-machine interfacing that governs the capsule and the ship aside from the Jove that designed it, and CONCORD that manages it. I think the control that CONCORD is able to exert over capsuleers is due to the fact only they and the Jove know exactly how the systems work and maintain propriety knowledge over it.

That's aside the fact that the actual physical architecture of a capsule capable ship exists in a world where FTL routers exist, so it might be said to potentially be extremely intricate and also easily compartmentalized since you could just have localized networks linked to an internal FTL router system direct to the pod with multiple redundancies that can be isolated from each other or remapped as needed.

So even if someone overcomes the hurdle of somehow finding out how the Jove manage their information networks and protocols, and whatever else CONCORD has on hand to ensure that they and only they can control capsuleers through their knowledge of it, then there's the fact you're dealing with a network that isn't static but can potentially change on the fly.

It can't be that variable or mysterious, since they have to interface ship technology with it.  Especially since the capsule would be common technology that doesn't seem to change from ship to ship, you'd think the opposite, that the routines would vary only in a somewhat superficial sense.  If people really had no idea how a capsule processes data, you wouldn't be able to design a ship for it.  It would be like trying to design GPUs and RAM for a motherboard you've never seen nor understand precisely how it works.  If there's anything mysterious about capsule tech, it probably isn't the data flow.

If there's anything that's different, it might be on the software side.  While not every capsuleer is going to run his own custom operational software, a few probably have that know-how.  It's probably a bit less reliable, especially for big and complex operations, but could happen.  Otherwise, I'd imagine that if we all have a common piece of technology that we need to fly, sense, and operate all existing starships, the actual software and hardware architecture is fairly ubiquitous.

A capsuleer probably would be able to fight back against computer infection better than a common computer jock just because they would instantly know what's going wrong and wouldn't need to worry about typing up code to try to counter it.  That said, capsuleers seem to be a lot like real people, in that they know how to restart and handle some basic stuff in their operating system, but most have no idea what all the acronyms in their BIOS mean and even fewer would be able to custom-write software on the fly.

[Esna Pitoojee]

If you're going to enumerate and penetrate a data network you need to know the protocols that define how that data is exchanged. I don't think there's been any hint that anyone knows the specifics of the man-machine interfacing that governs the capsule and the ship aside from the Jove that designed it, and CONCORD that manages it. I think the control that CONCORD is able to exert over capsuleers is due to the fact only they and the Jove know exactly how the systems work and maintain propriety knowledge over it.

That's aside the fact that the actual physical architecture of a capsule capable ship exists in a world where FTL routers exist, so it might be said to potentially be extremely intricate and also easily compartmentalized since you could just have localized networks linked to an internal FTL router system direct to the pod with multiple redundancies that can be isolated from each other or remapped as needed.

So even if someone overcomes the hurdle of somehow finding out how the Jove manage their information networks and protocols, and whatever else CONCORD has on hand to ensure that they and only they can control capsuleers through their knowledge of it, then there's the fact you're dealing with a network that isn't static but can potentially change on the fly.

The question of the source of the pirate factions' capsuleers is a tricky one here. Even starting to try to answer this gets into a whole bunch of different questions about NPC capsuleers in general, which I don't want to derail the thread with. Suffice to say that NPCs must be able on some level to acquire and operate capsules with CONCORD restrictions lifted, which we as PCs do not have access to for whatever handwavium reason.

3.  The capsule should be kept from firing the scanner or self-destructing within the ship.  That's the hard part, since I'm assuming that the scanner and the capsule's internal self destruct are all autonomous parts of the capsule (both have to work when the capsule is ejected).  It's also fair to assume, as you said, that an external breach that the capsule recognizes will automatically fire the scanner, as it probably doesn't know the difference between someone snapping the lid off and a ship blowing the lid off.  Still, if you had access to the capsule (and the capsuleer was either incapacitated or unaware of your presence), the capsule is a relatively ubiquitous technology.  As hardened as it probably is, it HAS to be standardized to work with so many ships from so many different races.  It also has to be able to be periodically serviced, so everything can't be completely inaccessible, otherwise you'd be randomly blowing up whole ships because a capsule's fan failed due to lint buildup.

I have always been under the presumption that the systems which monitor the capsule for breach wouldn't be near the surface or outer shell of the capsule; otherwise we'd be getting capsuleers being flashed every time a micrometeorite slipped through their shields or they accidentally rammed another ship (and we sure do ram stuff a lot...).

Rather, I've always thought they would line the interior wall of the hydrostatic void, and thus be accessible for maintenance from the interior of the capsule. From here, they could monitor for signs of imminent breach - anomalous temperature spikes, deformation of known lines, etc - and still be relatively safe from external damage (intentional or accidental).

Also, another thing to remember - the brain scanner must be capable of firing when the capsuleer is unaware of what is going on around them. We know this for the simple reason that if our ships are destroyed in a fight and we eject, but are blown up the very moment we start ejecting (hello smartbombs again), the capsule will still get us back to the vats safely. We can literally go from "well, this the last of my ship's structure..." straight to "waking up aching in a vat of goo", even if we have no time to be consciously aware of the capsule being damaged and destroyed.