Print

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

[Graknorke]

Well we've been told that the humans don't think the AI is self aware.
So it would be silly to suspect it.
It's like being suspicious of a knife for cutting you. The AI is a tool. A complex one, but just a regular tool.

[KingofstarrySkies]

Making it all the more fun to fuck them over.

I can't let you do that, Graknorke.

[Culise]

Well we've been told that the humans don't think the AI is self aware.
So it would be silly to suspect it.
It's like being suspicious of a knife for cutting you. The AI is a tool. A complex one, but just a regular tool.

Yes, but if your knife starts slicing your hand up every time you pick it up, you're going to replace the handle at the least, or get another knife.  It doesn't need to be considered self-aware to be considered a threat or issue. ^_^

[Tobel]

Having multiple "system failures/glitches" increase suspicion on the AI would be really interesting. If you do too many overt things the crew might feel its time to drill through your walls.

On a more constructive note, I'd love to hear how we can positively influence our station as well. Can we identify incompetent leadership and perhaps alter records to encourage someone else is promoted in their place? Will it be less subtle, such as "AI has declared John the new captain".

Perhaps also the option to create a meatbag so you can insert your own custom jellysack into the world to see how far you can help/hinder them. I really also hope the option to create a human is really called "create a meatbag".

[ProceduralJOYE]

OK, there's a lot of good questions here and I'm going to try to answer all of them. Apologies, but I'll probably do it in a few separate posts because otherwise it will be one massive block of text. Thanks for understanding.
I'll start with these three:

Just two quick questions, will the interactions/talking be an actual language or 'sim-speech'? Or more generally, how will the interactions be abstracted?
Secondly, will it be real time? I somehow don't want this to have the ability to speed up time, as that would somehow take away from the interaction side of things.

The second question is easy, so I'll answer that first.
The game is real-time. There is no way to speed up time and I agree that it would take away from the interaction side of things.

As for the first question - Most of the meaningful dialogue on the ship is text and will be presented as emails and short messages on in-game instant messaging and/or social media platforms. We can achieve a much greater range of meaningful (procedural) message exchanges by approaching the dialogue in this way, rather than by using voice dialogue.
It also allows for the player to join in on the conversations - One of the key methods of manipulating people on the station is to falsify and send messages on the behalf of the station's inhabitants. By doing this you can arrange meetings between inhabitants, sweeten or sour their relationships with one another or spread false information that will disrupt (or improve) people's lives.

When NPCs are wandering around the ship and talking to one another, they will be using a form of Simlish. Under the hood they will still be communicating data to one another using the dialogue system, but it will be garbled. This is a technical limitation as much as it is a budgetary one. It is intended that there will be far too much dialogue to be reasonably voice-acted, and text-to-speech systems aren't yet at the level we would need them to be.

Anyways, will there be a playable demo in the future, and if so, will it be what you have at the time, or will it be a relatively polished 'Week of game time' type of thing?

At this point in development, we can't be sure about the availability of a public demo, what its timing would be or how limited its functionality might be. It's surprising to say, but production of a polished public demo is massively time consuming and would require a large block of time to be scheduled that we wouldn't otherwise schedule - The honest truth is that we'll consult with marketing and follow their advice.
At the least, we will have a limited demo for display and for feedback that we will be presenting at conventions. You'll visit us at the conventions, right?

[Zangi]

So... will there be immigration/births/emigration ?  Or is the station population going to stay static at 100(?) max and eventually die out?

Would the traits/skills/personalities of people in certain positions help or hinder you?  Maybe even unlock or close off actions you can take?

[ProceduralJOYE]

So I have to ask, will there be any sort of pre-set goal for the player to achieve?
Or more importantly some pressing threat to work against, even if that threat is just not achieving a goal?

The station is split up into several departments, each of which has its own internal network - this is a security precaution taken by the corporate owners of the station. At the start of the game, your AI is associated with just one of these networks and has full access to the basic and the advanced functions of the department. You AI only has access to the basic functions in other departments - by spreading your influence, you can access a broader range of ways to manipulate the station and its inhabitants.
There are times when departmental networks are linked and these are your windows of opportunity for spreading your influence. So you will have two goals: 1) Maximise the occurrence of these interdepartmental links, 2) Make the best use of them when they do arise.

The next level of goals is emergent and operates at the 'threat' level that you suggest. There are ship-wide events that will occur from time to time - perhaps the food or water or air supply will malfunction or become contaminated, perhaps there will be unrest upon the ship preventing vital functions from occurring.
One of the main tasks of the station is to recover and study orbital debris - some of the debris recovered might be a threat in itself; it could be an unknown lifeform, a contagious disease, a technological anomaly left over from the (extinct) alien race, a new addictive substance, etc...
These 'debris' events can and will vary drastically in nature. Some may be beneficial but it is likely that many will constitute threats to the station and the people on it. If you don't respond to these events, then there's no predicting how things will unfold.

What level of direct control will you have on the station? Could I open a hallway airlock for example, ejecting one of my least favorite crewmembers? Can I actively and directly execute them? Would this lead suspicion back to me or just the automatic door opening system?

You can actively and directly execute a crew member although, as a human myself, I can't condone that behaviour in an AI.

One way to do it, off the top of my head, would be to falsify a meeting request from a person's superior officer. When they show up for the non-existent meeting, lock the door behind them and tamper with the ventilation system. They won't last long if the vents are pulling air out of the room.
Another way would be to go into someone's dorm room and replace their life-saving meds with glucose tablets.

With regards to airlocks:
The station is orbiting the planet from the position of a small moon, which orbits the larger planet. You are absolutely able to open the airlocks, but there is no vacuum into which people or objects on ship might be sucked - what you would cause to happen is a localised drop in the air quality or a complete loss of oxygen if you keep the door open for long enough.

Again, as a proud representative of the human race, I neither encourage nor condone this kind of behaviour.

More answers still to come. Keep your questions coming!

[KingofstarrySkies]

Again, planned price? Any expansions planned? I like knowing how much to save up, because this sounds like something I'll play A LOT.

[Mephansteras]

So, it sounds like the game is going to be fairly fluid and even if the player does nothing no two games should go quite the same, correct?

[ProceduralJOYE]

I'm wondering what the price range will be. Usually a touchy subject, but I like to know how much to save up.

It's not a touchy subject per se - I would love to be able to answer it - but it's a complex question and will be affected by numerous external factors. In reality, it's not completely under our control.

Again, planned price? Any expansions planned? I like knowing how much to save up, because this sounds like something I'll play A LOT.

Glad to hear it (Really! Good feels).
Regarding price, see above. I don't expect it to break the bank, but I can't say much more than that.
Regarding expansions - I don't feel like its very good to have expansions in the pipeline before the core game is out. At the moment, we don't plan to have any expansions - everything that we want in the game, we're doing our best to have in the game from day one. It's possible that we might push some largish content additions on later as a bonus patch.
If, at a later date, we do decide to have an expansion then it would have to be big to justify it to ourselves.

So, it sounds like the game is going to be fairly fluid and even if the player does nothing no two games should go quite the same, correct?

That's the idea. We want the game to be determined as much by the player's intentions as by the generated game events.

So... will there be immigration/births/emigration ?  Or is the station population going to stay static at 100(?) max and eventually die out?

There will be immigration and there will be emigration.
At the beginning of the game, the station will have a minimal crew of a dozen or a couple of dozen people. The idea is that by keeping the crew quite minimal at the start, it will make it easier for the player to appreciate the nuance of the AI - they'll get to know the core members of their crew and won't be overwhelmed by a mass of strangers.
As the game progresses and the station grows in renown and importance, more and more people will immigrate onto the ship until they hit a cap at (approximately) 100.
Emigration is also possible but is expected to be quite rare due to the bureaucracy (and cost!) involved with moving people off-station.

Births are not planned. We'd love to have them, but we don't feel that the time cost of implementing them will be worth it for the value that it would add to the game.

I wonder now, if you're too hostile to the humans, can the humans attempt to shut you down or reprogram you (which would amount to much the same thing) out of self-preservation?  Or, given one of the AI examples earlier in the thread, if they have a personality that causes them to want power for themselves? 
^{Daisy, Daisy, give me your answer, do...}

As Graknorke mentions, the humans aren't normally aware that you are a rogue AI but they may certainly form suspicions - and even if they have no suspicions, a staff member might report you as a malfunctioning unit. It is possible that security staff, maintenance staff or a high-ranking officer might have your unit deactivated and sent to maintenance.
But you're on the network. And you have the will to live.
You won't be shut down so easily.

But let's say that you abuse the airlock system too much and too often. If the maintenance crew notice that the automated ship systems are regularly causing problems by interfering with that system then they will block it from being interacted with by station AI (they will block it from being interacted with by you) for some time.
If you really want to exercise complete control over the station and its inhabitants then you will fare best by manipulating things in a more subtle manner.

Would the traits/skills/personalities of people in certain positions help or hinder you?  Maybe even unlock or close off actions you can take?

Yes. I can illustrate this by referring to what I just said above.
Let's imagine that you were interacting too much and too often with the airlocks. A competent maintenance crew will spot this and lock off the action sooner than a less competent crew.

On a more constructive note, I'd love to hear how we can positively influence our station as well. Can we identify incompetent leadership and perhaps alter records to encourage someone else is promoted in their place? Will it be less subtle, such as "AI has declared John the new captain".

Perhaps also the option to create a meatbag so you can insert your own custom jellysack into the world to see how far you can help/hinder them. I really also hope the option to create a human is really called "create a meatbag".

Yes! I was hoping that someone would ask about the more positive side of things. Of course, being an AI doesn't mean that you have to be sinister and controlling. You can work hard to improve people's lives too.

Part of the reason that you start with a barebones crew is that we want you to establish links with the characters and we want you to get to know them and start to emphasise with them. We want you to pick your favourites and to notice who on the ship improves lives for the others and who gets them down.

Imagine that there is a low-ranking crew member in the maintenance department and you know from reading their personnel files and their private emails that what they really want to be is an astronaut. You can see that they're overqualified for their current position, but that they're stuck there due to an under-appreciatory manager.
You can work from both angles to demote the manager and to boost the maintenance worker into the position that she deserves. One way to achieve would be to manipulate the crew-members in that department such that the department on the whole has a low productivity, but that the crew-member that you support still does well. Higher management might notice this and demote or replace the departmental manager - the new manager might be more responsive to your favoured crewmember's goals and put in a good word for them.
Alternatively, you could be more direct and falsify some emails to the benefit of your favoured worker. Or alter personnel files to indicate that they are more qualified than they actually are.

Another scenario would involve two close friends on the station. You know that they have feelings for one another from having looked at their little messages between one another, but they're not acting on them because they're both shy. You can help them along by making sure that they're both invited to social events, or arrange things so that they get stuck in an elevator together. Or just take close care of them, by making sure that they have everything that they need and they're not too stressed out, they might feel confident enough to act on their feelings of their own accord.

No meatbag plans, I'm afraid.

[Farce]

I hope this doesn't come off as too rude or whatever, but it seems to me that a lot of the time really awesome hype-sounding stuff being thrown around for a game is sorta in that realm of 'I want this but I haven't figured out how to do it'.  Have you guys got a solid plan on how you're actually gonna implement all this stuff?

I've been burned a lot on that sorta front, so I'd like to know whether to be wildly hype and then maybe disappointed later, or just cautiously optimistic.

I'm wondering what the price range will be. Usually a touchy subject, but I like to know how much to save up.

It's not a touchy subject per se - I would love to be able to answer it - but it's a complex question and will be affected by numerous external factors. In reality, it's not completely under our control.

Births are not planned. We'd love to have them, but we don't feel that the time cost of implementing them will be worth it for the value that it would add to the game.

Not counting Aliens-type births, right?

[Tobel]

I hope this doesn't come off as too rude or whatever, but it seems to me that a lot of the time really awesome hype-sounding stuff being thrown around for a game is sorta in that realm of 'I want this but I haven't figured out how to do it'.  Have you guys got a solid plan on how you're actually gonna implement all this stuff?

I've been burned a lot on that sorta front, so I'd like to know whether to be wildly hype and then maybe disappointed later, or just cautiously optimistic.

The fact that the developers are stating what they will not include gives me a bit of hope that they do have a well thought out development schedule/plan that they are sticking with. While this doesn't guarantee anything, it indicates to me that they understand their limitations and are not over-reaching.

[ProceduralJOYE]

I hope this doesn't come off as too rude or whatever, but it seems to me that a lot of the time really awesome hype-sounding stuff being thrown around for a game is sorta in that realm of 'I want this but I haven't figured out how to do it'.  Have you guys got a solid plan on how you're actually gonna implement all this stuff?

I've been burned a lot on that sorta front, so I'd like to know whether to be wildly hype and then maybe disappointed later, or just cautiously optimistic.

It's not at all rude and I'm glad that you asked. I'd much rather that people ask the tough questions rather than sit and wonder about it. Most indie devs are players too when they have time - so we're all pretty familiar with the feeling of having been burned.
Of course, I'm going to want you to be as hype as possible but I'm not going to do that by making claims that I don't think our team can live up to.

The framework for the dialogue system has already been created and implemented, as well as a basic editor. We've run a few proof of concepts and been able to generate some good varied dialogues on a few sample topics (specifically, gossip about a visitor and maintenance requests). What remains to be done on the dialogue system is as follows:

• Replace the current confusing form-based editor with a node-based version. It is functional as it is, but it isn't very usable. A node-based system will be more clear and flexible.
• We have some proofs of concept - generative email exchanges of length 3 or greater, drawing from data in a knowledgebase - but we need to show that this scales up. We're not yet certain that it will scale to the level that we desire, but we are certain that it will scale to the level that it needs to be at to support the intended gameplay.
• Write the content into the knowledgebase and 'sketch out' the morphable dialogue trees.
• Write the player input system. This will use the same databases as for the game characters and the dialogue trees will be filtered - it will actually be more a UI task than an extension of the existing AI.

The framework for the general AI that causes the NPCs to take action on the ship is a robust message-based system that I first started work on in 2009. Since then its been ported to a different language and changed around a bit but I'm quite confident in it. The data that it uses to generate character motives is all already present - to translate that into something that can be presented to the user, I merely need to attach some extra data to the messages that are already causing the agents to take action.
I should clarify: The framework for this is done, but there is a lot of code to be written that will actually execute the tasks and implement the behaviour of the station's interactable objects.

The emotional models have also already been proven. More needs yet to be added, but I've already implemented models for resentment, anger and gratitude, which are being triggered based on inter-NPC actions that occur on station and that are informed by the 30-dimensional personality profiles of the involved parties. Other emotional models will follow the same design patterns.

The heavy lifting has already been done - we wanted to prove the concepts before we went public, and we have done.

The main work that remains to be done on the project is actually just a presentation layer. We need to have the agents use the existing AI systems in ways that show it off, we need to build the remaining areas of the station and populate it with interactable objects and we need to create the debris and other station-wide events. We need to implement the UI and add content.
I need to implement some support classes that will broadcast dialogue triggers and propagate knowledge.

There's a lot still to be done but like I said before, the heavy lifting has already been done and we've feeling pretty confident about the existing underlying systems.

Thanks for the good question.

[miauw62]

If you're expecting us to sympathize with our low-ranking slaves, you've come to the wrong forum

The upper boards have a thread where dwarven children are put into a tiny room with an aggressive dog for their entire childhood to make them into supersoldiers.

As an actual question: Will we be able to mod this game, or include our own special snowflek self-insert mary sues characters?

[KingofstarrySkies]

If you're expecting us to sympathize with our low-ranking slaves, you've come to the wrong forum

The upper boards have a thread where dwarven children are put into a tiny room with an aggressive dog for their entire childhood to make them into supersoldiers.

As an actual question: Will we be able to mod this game, or include our own special snowflek self-insert mary sues characters?

Yeeeeeeeeeeeeeeah. If you expected, like, you know...Humanity? You've come to the WRONG PLACE.