[Career Mode] Contract Proposals

[regex]

What I would like to see in career mode play:

TL;DR: Players should have to create a proposal for a contract which would have to be approved based on past performance (reputation?) before funding is released. Funding is used to launch craft to attempt the mission. Once the mission is completed any left-over funding is released to the general fund which can be used for mission-less flight or overflow.

A proposal is a series of objectives that comprise a contract. The player decides what objectives each mission should try to attain, with objectives available for each SOI. For instance, "Place orbiter around Duna" could be an objective as could something as simple as "Perform a suborbital Kerbin flight". Although a contract could have many stacking objectives care would have to be taken to prevent gaming the system; orbiting Kerbin should be assumed when the overall objective is landing on Eeloo. This could be handled by specific objectives preventing lesser objectives from being added. This system would allow the player to define missions as simple as a Mun landing or as complex as a grand tour of the system.

Once all of the objectives are chosen and the proposal created it would be submitted for approval. Approval should be based on past performance, perhaps with the dev-mentioned reputation stat. You would gain rep by completing contracts and certain objectives should have no rep requirement in order to allow the player to build it back up (or maybe have a rep requirement very low that would still allow for end-game failure).

An approved contract would have a budget which would be used to complete the mission. This budget would be based on the player's rep and the planned objectives, and the calculation would be as complex as required. It probably wouldn't be an easy task to balance such a system. Either way, any craft built to complete the objectives (you would specify this before launch/flight) would be bought using the budget for the contract. Once the objectives were completed any left-over money would be added to the player's general budget, which could be used for personal projects, maintenance, philantropy, hiring Kerbals, whatever. Partial completion should also have some rewards, especially early on. Repeating objectives from past contracts should have diminishing returns; this might require other methods for the player to gain money and rep in order to continue playing past a certain point.

Rep, and possibly money, should also be rewarded for achievements outside of contract play. For instance, if the player manages to land on Ike during their first launch, without even bothering to create a proposal (honestly, who's going to fund a first launch to Ike?), that should count for something within the contract system. It should also be included in the calculations for diminishing returns on future proposals. Players would naturally start a career mode game with a modicum of funds in their general budget.

[Kegereneku]

It's surprising to hear that from you, as I though you didn't want to be "obligated" to justify yourself with past performance/reputation to get the funding you want.

Anyway, to give you a constructive analysis : I liked the idea at first. It had a nice ring of making the players parts of the decision, while being actually guided in his progression.

The player would be encouraged to build complex rocket and reuse design, yet forced to actually accomplish something.

But as you said :

An approved contract would have a budget which would be used to complete the mission. This budget would be based on the player's rep and the planned objectives, and the calculation would be as complex as required. It probably wouldn't be an easy task to balance such a system.

"Not an easy task to balance" is quite an understatement, fact, I doubt it is even feasible :

- How to determine how much budget is to be given in the first place. (and shouldn't we give less money to good-player ?)

- How do you determine what objective was reached with which budget (as docking will allow modular ship)

- How does reputation influence the progression (I would recommend it as a negative-feedback balance mechanism rather than a positive one)

- How much left-over money the player can be allowed to "save" (as it can ease everything recursively)

Lastly : - How to keep player from abusing the system ? (even accidentally)

My worry is that without a negative-feedback* somewhere (in contrast to a deny-of-further-budget), players would (fair to reality) systematically overstate their objectives to get bigger budget and as soon as they can assemble anything in orbit : reach them easily, maxing-out reputation/budget in a recursive positive-feedback.

On the opposite side : a bad player would be forced to "grind" his reputation back, something we agree to be bad gameplay.

I noted that this problem is recurrent when giving money-per-launch (as opposed to budget-per-time) as soon as docking-ring are added.

example of how it can go wrong :

- One overbuilt a Mun-Lander so that it can wait in orbit and another ship can refuel it (and transfer Mun sample).

- Using far less-budget than needed, the player use this ship to do contract on Minmus.

- The player fund a Duna-lander with the BIG left-over, then, even working on a lesser Duna-orbit contract, land on Duna and get back to home : huge reward.

- Finally, using the now HUUUUGE LEFT-OVER : fly through any difficulty later.

At least the science-part of a reused-vessel wouldn't be up to date, but the problem stay : The progression is exponential rather than linear.

*the problem of negative feedback : As we all now God Manley...I mean Scott Manley, will inevitably do everything better than us with less money. If we use a negative-feedback to make the game more challenging for him (a lost battle) there's the risk of making the left-over so insignificant that a good-player would have to GRIND for private funds.

I think a solution would be :

- Good reputation : whoever fund you get cheapskate and consider you don't need that much budget.

- Bad reputation : whoever fund you still want things done and give you bigger budget.

[pizzaoverhead]

This would be a good government prestige oriented complement to the more prescriptve commercial missions as seen in Mission Controller. One is a more open and exploration based "go and be the first to do something", and the other gives a focused direction and progression, covering the two big gameplay styles.

Am I right in thinking that, as reputation is used to decide if a proposal is approved or denied, it is basically a system where reputation unlocks objectives? I'd like to see a more flexible balance system used for it: Rank each objective with an appropriate reputation, giving a low ranking to more simple manoeuvres such as achieving Kerbin orbit and a high reputation to more difficult ones such as Eve sample return. Scale the funding for a objective downwards the further its reputation is from the player's current reputation. This means that you can propose a mission consisting of achieving Kerbin orbit and Eve sample return for your first mission. The Kerbin orbit would receive a high level of funding due to having a high confidence in success, whereas the Eve sample return would be considered impossible at this level and be given almost none. Achieving the objective should give a cash reward. The funding and the reward should come from the same pool. For example, if you propose a Kerbin orbit, 75% of the pool could be offered as funding, with the remaining 25% on successful completion. A high confidence level reduces the reward offered because of the perceived simplicity in achieving the task. The pool would be regenerated with diminishing returns after being exhausted.

Interface-wise, a graphical representation of the proposal would be cool.

Edited December 6, 2013 by pizzaoverhead

[tetryds]

But then how would you procedurally generate a graphical representation for each mission? (edit: It has been stated that the game is not meant to have an end)

It would either be a too simple representation, or too complex to make.

Maybe just a map track of where you should go would be enough?

[pizzaoverhead]

But then how would you procedurally generate a graphical representation for each mission? (edit: It has been stated that the game is not meant to have an end)

It would either be a too simple representation, or too complex to make.

Maybe just a map track of where you should go would be enough?

Symbolism. A coloured circle with a name in it for each planet, a circle around it for an orbit, arcs connecting different states. Nothing accurate, just something you can stick numbers on. The picture below would be too much detail. Graphical representation is not required for this idea to function, it would just be a bonus feature. You could build it just fine using drag and drop or selecting each objective from a menu to add to a list.

[tetryds]

But wouldn't that be telling me how to make my own missions?

"Land at Mün, or Land at Minmus then land at Mün then come back" seems a lot more Kerbal-ish

Maybe a "Land here" area at the map vision, or "orbit at 100k", but not telling me when or how to make my burns, what if i want to make a direct insertion to Mün?

[LethalDose]

The original idea of "contract proposals" is interesting, but it's implementation would be a total nightmare: not only in design, but in getting it appropriately balanced as well.

Can anyone name another game where a system like this has been applied? I certainly can't think of one.

Typically in a game, you get rewarded on the backend of a mission/task, meaning that you don't state your goals and get resources a priori, you do what you want to do, then get rewarded when you're successful. I think it's way more likely that we'll see contracts implemented as relatively random events (occur every xmin to xmax days) that take the form of "Complete Objective [obj] by [time limit] to receive [bonus]". It's simple to design and balance and it's a system that gamers can immediately recognize.

This is the way games have done it for the last 30 years, and I'm guessing SQUAD is gonna follow this paradigm. I think they should. I'm not opposed to innovation, but the game has to be approachable to a wide audience.

Edited December 7, 2013 by LethalDose

[boomerdog2000]

I noted that this problem is recurrent when giving money-per-launch (as opposed to budget-per-time) as soon as docking-ring are added.

I recognize your other points however, if someone can reuse a lander for multiple missions by docking it to a mothership in orbit, is that a bad thing? That is a fairly complex mission, that requires forethought and planning. Not to mention, as you said, the vessel will not be up-to-date. Engines, fuel, science equipment, solar panels, a lot of these would probably be swapped out as the design grew.

Compared to the Apollo program, reusing a lander already in space is incredibly efficient and I feel we should not punish the player for doing that. Plus the techs required for these sort of dockings and reuse are still a little ways down the tech-tree requiring the player to do more typical missions beforehand.

[regex]

It's surprising to hear that from you, as I though you didn't want to be "obligated" to justify yourself with past performance/reputation to get the funding you want.

You had no idea what I wanted before or what I thought was acceptable game-induced constraint.

- How do you determine what objective was reached with which budget (as docking will allow modular ship)

There's probably a way somewhere in the code. I imagine that each ship will be built with a "flag" that indicates what contract it was built for (if any), so a tug that had a mission of "Orbit Duna" could bring a lander with a mission of "Land on Ike". Both could be used to complete their respective contracts. There would probably be a need to restrict the number of concurrent contracts accepted and being used at any one given time.

You could also have "restrictions" as well as objectives, stuff like "in a single launch" or "without help" that would increase the rewards.

- How does reputation influence the progression (I would recommend it as a negative-feedback balance mechanism rather than a positive one)

- How much left-over money the player can be allowed to "save" (as it can ease everything recursively)

I think reputation should be an "unlock mechanism" for future objectives, and it should "relock" objectives that aren't usable by the current reputation. There is nothing wrong with letting the player do their own thing using their general fund which is why I suggested letting the player get rewards for completing objectives outside of a contract. Of course, the player should be allowed to save as much money in their general fund as possible.

example of how it can go wrong :

- One overbuilt a Mun-Lander so that it can wait in orbit and another ship can refuel it (and transfer Mun sample).

- Using far less-budget than needed, the player use this ship to do contract on Minmus.

- The player fund a Duna-lander with the BIG left-over, then, even working on a lesser Duna-orbit contract, land on Duna and get back to home : huge reward.

- Finally, using the now HUUUUGE LEFT-OVER : fly through any difficulty later.

At least the science-part of a reused-vessel wouldn't be up to date, but the problem stay : The progression is exponential rather than linear.

Sounds like a smart player. As far as reusing ships I don't see any problem with that but, IMO, they should be unflagged and thus unable to complete objectives for future contracts. One way to allow reusing is to have "refurbish fee" that simply flags the ship as usable by the contract. You could use the contract's budget for this so your Ike lander could be reused for a future Pol mission (for instance). This gives incentive to create robust, reusable craft.

*the problem of negative feedback : As we all now God Manley...I mean Scott Manley, will inevitably do everything better than us with less money. If we use a negative-feedback to make the game more challenging for him (a lost battle) there's the risk of making the left-over so insignificant that a good-player would have to GRIND for private funds.

No one cares about what Scott Manley does or how well he does it unless they've got a massive inferiority complex. If a player can amass a small fortune in their general fund, more power to them, it gives them the freedom to do whatever they want. I feel like the system is also flexible enough to allow an advanced player to not have to ever grind since they can make proposals that fit into their current plans; this removes the randomness of the game generating five dumb "missions" and then the player choosing from one.

- Good reputation : whoever fund you get cheapskate and consider you don't need that much budget.

- Bad reputation : whoever fund you still want things done and give you bigger budget.

That's an interesting twist, but you'd have to be careful not to reward good(bad) play, where an advanced player overbuilds in order to secure more funding for future missions.

This would be a good government prestige oriented complement to the more prescriptve commercial missions as seen in Mission Controller. One is a more open and exploration based "go and be the first to do something", and the other gives a focused direction and progression, covering the two big gameplay styles.

I heartily agree. This is more of a sandbox approach since I hate the idea of the game doling out "missions" to me, but other people obviously enjoy that sort of play; there's no reason the two can't live together.

Am I right in thinking that, as reputation is used to decide if a proposal is approved or denied, it is basically a system where reputation unlocks objectives? I'd like to see a more flexible balance system used for it: Rank each objective with an appropriate reputation, giving a low ranking to more simple manoeuvres such as achieving Kerbin orbit and a high reputation to more difficult ones such as Eve sample return. Scale the funding for a objective downwards the further its reputation is from the player's current reputation. This means that you can propose a mission consisting of achieving Kerbin orbit and Eve sample return for your first mission. The Kerbin orbit would receive a high level of funding due to having a high confidence in success, whereas the Eve sample return would be considered impossible at this level and be given almost none. Achieving the objective should give a cash reward. The funding and the reward should come from the same pool. For example, if you propose a Kerbin orbit, 75% of the pool could be offered as funding, with the remaining 25% on successful completion. A high confidence level reduces the reward offered because of the perceived simplicity in achieving the task. The pool would be regenerated with diminishing returns after being exhausted.

That's a great balancing idea. I kind of wrote the proposal idea in my head over a week or two and put it on paper right before bed last night. There's obviously a lot I hadn't thought of but I felt like it was solid enough to put out there.

Interface-wise, a graphical representation of the proposal would be cool.

I love this, it would be awesome.

The original idea of "contract proposals" is interesting, but it's implementation would be a total nightmare: not only in design, but in getting it appropriately balanced as well.

Design, no, balance, yes. The GUI could be a simple left/right column with a nested collapseable tree of SOIs and tiers of objectives that could be dragged back and forth. Outsource objective ideas from the community and cherry-pick the good ones. Certainly there's some back-end craft work to be done to handle Kegereneku's cases above, but I think that would be fairly trivial compared to balancing this system.

Typically in a game, you get rewarded on the backend of a mission/task, meaning that you don't state your goals and get resources a priori, you do what you want to do, then get rewarded when you're successful. I think it's way more likely that we'll see contracts implemented as relatively random events (occur every xmin to xmax days) that take the form of "Complete Objective [obj] by [time limit] to receive [bonus]. It's simple to design and balance and it's a system that gamers can immediately recognize.

This is the way games have done it for the last 30 years, and I'm guessing SQUAD is gonna follow this paradigm. I think they should. I'm not opposed to innovation, but the game has to be approachable to a wide audience.

There's no reason the two systems can't live together. The reason I'd like to see this is that I think KSP deserves far better than "the last 30 years" of gameplay. "Do missions" is so damn droll at this point; I'm sick of it. I want to direct the future of my space program, not have some dumb talking head dole out random missions.

[Kegereneku]

I recognize your other points however, if someone can reuse a lander for multiple missions by docking it to a mothership in orbit, is that a bad thing? That is a fairly complex mission, that requires forethought and planning. Not to mention, as you said, the vessel will not be up-to-date. Engines, fuel, science equipment, solar panels, a lot of these would probably be swapped out as the design grew.

Compared to the Apollo program, reusing a lander already in space is incredibly efficient and I feel we should not punish the player for doing that. Plus the techs required for these sort of dockings and reuse are still a little ways down the tech-tree requiring the player to do more typical missions beforehand.

I'm not saying this is a bad things of course. I'm all up for reusing a good old tug for years or assembling mothership (two things we know epic stories are made of). I just noted that balancing a money-per-launch system become a problem as soon as docking ring are unlocked, and myself I would like docking ring to be unlockable very early without it unbalancing the entire budget/economy system, something which doesn't look possible in the mentioned case of money-per-launch.

There's probably a way somewhere in the code. I imagine that each ship will be built with a "flag" that indicates what contract it was built for (if any), so a tug that had a mission of "Orbit Duna" could bring a lander with a mission of "Land on Ike". Both could be used to complete their respective contracts. There would probably be a need to restrict the number of concurrent contracts accepted and being used at any one given time.

You could also have "restrictions" as well as objectives, stuff like "in a single launch" or "without help" that would increase the rewards.

You are indeed surprising me, you are now suggesting more railroading than I've ever suggested.

I raise you a few things to take into consideration :

- Orbital refueling or orbital-tug aren't lander but may be reused many time. Even if they may end-up inefficient at some point in the future.

- There should never be a reason to discourage "help" from already launched spacecraft.

- Restricting the number of launch should be based on budget-reason rather than arbitrary restriction.

Plus managing flagged ship for unique-mission doesn't sound very fun, extremely linear even. Are you really DEATH-TO-RAILROADING Regex ?

I think reputation should be an "unlock mechanism" for future objectives, and it should "relock" objectives that aren't usable by the current reputation. There is nothing wrong with letting the player do their own thing using their general fund which is why I suggested letting the player get rewards for completing objectives outside of a contract. Of course, the player should be allowed to save as much money in their general fund as possible.

Ah... so if I understand right, you wouldn't get objective concerning Duna unless you gained reputation through say... an historical reenactment of Mercury then Apollo ? (don't mind the personal joke)

Still, I don't see it solve being able to land/return from Duna using a modular/overmade rocket regardless of your objectives (and de facto unlocking the reputation). I know our world encourage overachieving but I think it will get very confusing if the objective don't actually matter. It get us back to my point about players lying (as do politician) to fund their space program.

Still not as confusing as managing flagged ship in a complex infrastructure though.

I feel like the system is also flexible enough to allow an advanced player to not have to ever grind since they can make proposals that fit into their current plans; this removes the randomness of the game generating five dumb "missions" and then the player choosing from one.

I don't think it's a wise move to try to make player the source of theirs owns achievements.

For starter, what's the point of succeeding at something if you knew right from the start that you could make it ? (bringing us to my point about determining how much budget you needed)

Then, as far I know most players actually look up to Career-mode giving them a structured progression, so there's no reasons to think those mission will necessarily be "dumb". Which bring me to my last point : I don't imagine your proposal letting us generate "smarter" missions out of fail-safe balance-restricted choice.

You may have forgot that Game-designer work hard to make better mission / game mechanic that most of us would come up with.

/!\ btw : "Scott Manley" is merely a metaphor for "extremely-experienced-player" who WILL push the balance-mechanism to their limits or rather outside just by playing normally. (Speedrun should remain something requiring hard effort)

That's an interesting twist, but you'd have to be careful not to reward good(bad) play, where an advanced player overbuilds in order to secure more funding for future missions.

Indeed, but now that I read your precision about objective-unlocking reputation. I think the reputation-part of you idea may work with Fixed-Budget(time) + Prestige money based on whether or not you recovered parts and those not-so-expendable-Kerbonaut.

Although I still disagree about the feasibility and interest of making-up missions yourself.

[regex]

- Orbital refueling or orbital-tug aren't lander but may be reused many time. Even if they may end-up inefficient at some point in the future.

Depending on how resources end up working, this will either be an issue or a non-issue. Refueling craft probably wouldn't have a specific contract (unless your grand tour mission had enough funding to buy a tanker or two) so they would be paid for out of the general fund, which ties in with the idea of funding overflow. Either way, it would reward smart infrastructure planning.

- There should never be a reason to discourage "help" from already launched spacecraft.

That was only an example, but sure.

- Restricting the number of launch should be based on budget-reason rather than arbitrary restriction.

You're misinterpreting the idea, I never suggested restricting the number of launches.

Plus managing flagged ship for unique-mission doesn't sound very fun, extremely linear even.

It's a way around reusing the same craft for multiple contracts for free. Much like in the real world, some repair, maintenance, updating, and refurbishing may be necessary before a craft is considered "mission-worthy" again. You can justify that as the funding organization requesting certain updates before "trusting" the old craft to perform its duty (which could actually be a fun mechanic in itself, where the more trustworthy a craft becomes, the less you have to pay to reuse it for a contract.)

Ah... so if I understand right, you wouldn't get objective concerning Duna unless you gained reputation through say... an historical reenactment of Mercury then Apollo ? (don't mind the personal joke)

Yes. However, because the player has a general fund they are never prevented from launching their own craft and doing whatever the hell they want, they just might need more money.

Still, I don't see it solve being able to land/return from Duna using a modular/overmade rocket regardless of your objectives (and de facto unlocking the reputation). I know our world encourage overachieving but I think it will get very confusing if the objective don't actually matter. It get us back to my point about players lying (as do politician) to fund their space program.

I don't consider that an issue in the slightest. Once again, I suggest slightly reduced rewards for "sandbox achievements", maybe because the funding organization considers you a "loose cannon" or something.

Still not as confusing as managing flagged ship in a complex infrastructure though.

Not at all. Each craft would have a current contract, or none. To make it simple, a craft cannot be assigned to a new contract until its current one is finished. You could even have that as a sort or filter option in the space center.

For starter, what's the point of succeeding at something if you knew right from the start that you could make it ? (bringing us to my point about determining how much budget you needed)

You say "I can do this" and either fail or succeed. If you stick to safe, proven missions, you're not only letting yourself down but you're eventually running out of funding/reputation because you refuse to advance. An interesting mechanic along those lines would be to reduce reputation instead of gain it for repeat objectives (or just not award it), especially if you've done nothing else in-between. Spamming the Mun with contracts shouldn't allow you to amass a hoard or a track-record of exploration.

Then, as far I know most players actually look up to Career-mode giving them a structured progression, so there's no reasons to think those mission will necessarily be "dumb". Which bring me to my last point : I don't imagine your proposal letting us generate "smarter" missions out of fail-safe balance-restricted choice.

There's no reason this system couldn't provide generated missions or even have a "random contract" button that took your current rep into account. In fact, it could even be smart about it and suggest missions for you based on previous objectives you've completed or failed. Whether the player accepts those suggestions or not is entirely up to them. The existence of a general fund allows the player additional freedom to make their own choices. And, once again, this system could easily live next to something like a derpy, cookie-cutter, "ferry the random Kerbal around" sort of mission generator. They could even share code.

[Kegereneku]

Okay, first, let's bring everything back together. Answering point by point have the nasty habit of making us forget the global view.

You can sum up my message by "You are turning perfectly normal way of playing into gamebreaking problem"

I do understand that you want player to have personal funds to use as they damn please, the points I made in my earlier post is that your suggested way of doing so would skyrocket "general fund" up to ridiculous height where it make fun out of the system. A system that is only meaningful until docking-ring is unlocked (or require crazy constrain) isn't what I would call a good gameplay, but you did get my point about player lying to get funds.

You are "restricting the number of launch".

Or more correctly, you are (trying) to restrict the number of spacecraft authorized to do / help achieve an objective to avoid budget abuse, which end up doing the very same things even if the "general fund" is fixed and always available. I don't remember you telling what would happen for a player who fail to reach his objectives and waste the budget. Does he need to reset from an earlier save ? Does he "abandon mission" and lose reputation point (or my variant the problematic negative-feedback) ? Does he just propose a new budget and get penalty if he just refuel the earlier spaceship ?

I can see a Lose-Lose situation if you NEED a space infrastructure but never had the budget to built it.

I can see a Lose-Lose situation if you do build a space infrastructure but cannot use it because, either it's made with fund from ongoing objective or you lack the budget to finish the job.

And I still see a crazy situation where you just make-up mission objective to get budget. Ironically this is mostly how work the real world, if we want to be really true to reality we would need to use the entire budget to get the same one next year and get budget cut before going to Duna.

By the way, you remembered me a critique about continuity :

"We must stay able to retry/save a mission and continue playing without being forced to reset to when you got the budget, and live with any Kessler syndrome collection you started."

Lastly, about the basic idea of "User-generated Mission" through "Contract proposal".

It sound to me we agreed that it will do fundamentally the very same thing than ""a derpy, cookie-cutter, "ferry the random Kerbal around" sort of mission generator"" except that it require even more code to keep the player from scamming the contractor. (Basically you play against an ECONOMIST AI and the goal is to fraud).

I don't think there any way to balance this, since it's based on unbalancing the outcome.

In you place I would consider either :

- Restricting the reward to a fixed amount of money regardless of what was spent (there's a upper limit of course). (you would however unlock "reputation" for being cost-efficient)

- or not offering budget by flight, but offering a huuuge "private fund" if you succeed using your fixed-budget*. Money that can be used to go over-budget.

- or making the choice of the player inconsequential, ex : you are still asked to do something, you still get the same progression, you just get to chose what sound more fun right know.

*fixed-budget : by the way, are we clear bout what I mean by that ? I have a doubt.

I don't consider that an issue in the slightest. Once again, I suggest slightly reduced rewards for "sandbox achievements", maybe because the funding organization considers you a "loose cannon" or something.

Your suggestion make it an issue, we both agree it shouldn't.

Not at all. Each craft would have a current contract, or none. To make it simple, a craft cannot be assigned to a new contract until its current one is finished. You could even have that as a sort or filter option in the space center.

Oh boy ! You have nooo idea how complex my space infrastructure could be. Between the Kerbin and Munar refueling-station, the mobile base traveling through biome waiting for a lander-module to transfer sample/fuel... from a thruster-less science-updated module brought to the surface after being brought in orbit by an 10 years old orbital-tug... not counting eventual satellite refueling service.

I tell you GOOD LUCK trying to make a Fair, Consistent and Easy-to-Use contract proposal system that reward/acknowledge my work in a satisfying way.

You say "I can do this" and either fail or succeed. If you stick to safe, proven missions, you're not only letting yourself down but you're eventually running out of funding/reputation because you refuse to advance. An interesting mechanic along those lines would be to reduce reputation instead of gain it for repeat objectives (or just not award it), especially if you've done nothing else in-between. Spamming the Mun with contracts shouldn't allow you to amass a hoard or a track-record of exploration.

I don't want to be dramatic (I do) but that's a wonderfully horrible way to make a game. We know modern game are way easier than "3 Lives Left" games of the past but making progress slowly harder and harder... because you didn't do thing "right" is just being cruel. Worst than simply loosing in a strategy game (because it's days of effort undone rather than hours).

Of course, drama aside, I'm pretty sure you have a LOT of alternate solution to that, I'm just putting thing into perspective.

Edited December 7, 2013 by Kegereneku

[LethalDose]

First: I agree with everything Kegereneku said above, especially the first line: "You are turning perfectly normal way of playing into gamebreaking problem"

Second: Beyond gameplay issues, I don't think you really understand the complexity your implementation would require:

1. Tracking of previous missions, both those done on "contract mode" (is that what you called it?) and off
   
2. Algorithms for evaluation of previous missions, on multiple dimensions you've mentioned including success and fame
   
3. User interface to propose user-designed missions.
   
4. Algorithms to evaluate user-designed missions based on the proposal and conditioned on multiple previous performance parameters.
   
5. Adjusting the algorithms to avoid exploits.
   
6. Balancing this multivariable system
   

And probably other issues I haven't thought of. Your proposal, in addition to statements like this:

This budget would be based on the player's rep and the planned objectives, and the calculation would be as complex as required.

indicate that you've either failed to comprehend the complexity needed for your proposal, or have and simply don't care. The algorithms needed to produce the evaluations you're expecting would be incredibly complex and the time needed for their development would require removal of other features from the game.

Third: In this thread, you've said that there should be monetary rewards for missions off "Contract mode", so what would this add? You're basically adding a game of "HORSE" to KSP (saying what you're going to do, and then doing it, instead of just doing it).

Finally, this statement:

There's no reason the two systems can't live together. The reason I'd like to see this is that I think KSP deserves far better than "the last 30 years" of gameplay. "Do missions" is so damn droll at this point; I'm sick of it. I want to direct the future of my space program, not have some dumb talking head dole out random missions.

indicates that you're basically asking for innovation for the sake of innovation. Jim Sterling has a piece on this, that heartily agree with, viewable here:

[WARNING: PROFANITY]

http://www.escapistmagazine.com/videos/view/jimquisition/7005-Innovation-Gamings-Snake-Oil

Your proposal vastly increases the complexity of coding and playing the game, but adds very little and not nearly enough to justify its inclusion.

Edited December 7, 2013 by LethalDose

[regex]

Okay, first, let's bring everything back together. Answering point by point have the nasty habit of making us forget the global view.

Sure, let's do that. How do all your concerns about my system not apply to a standard "do missions" sort of system?

You say that a general fund would skyrocket if an experienced player could work frugally. How does this differ from a "do missions" sort of system? I don't see the difference. My system gives you reward up front to build your craft (to do the job, remember that money doesn't hit the general fund until the contract is completed) and rewards your frugality/efficiency by giving you the rest upon completion. That means that your starting budget can be quite small, maybe enough to hire an astronaut or something.

You are worried about having to authorize craft/crew to perform a job but can't see an issue with that under a "do missions" sort of system?

You are worried about negative feedback regarding failing an objective but you can't see how that would be a balancing issue under a "do missions" sort of system? How else do you create a "reputation" system as SQUAD is proposing? Maybe you could set time limits on contract completion. You could also check to see if any craft were assigned to the contract and if none were it fails.

I can see a Lose-Lose situation if you NEED a space infrastructure but never had the budget to built it.

Since you get some of the reward up-front to build a craft to complete the contract, there will almost never be a situation where you have run out of money. You don't need space infrastructure to go anywhere in the solar system right now, and I doubt you would under career mode either since you can always do a manned flight; RT2 style networks aren't needed for Kerbals.

I can see a Lose-Lose situation if you do build a space infrastructure but cannot use it because, either it's made with fund from ongoing objective or you lack the budget to finish the job.

Then you could build space infrastructure from your general fund. The idea behind that is that the player has freedom to expand and build as they see fit. It's not really any different under a "do missions" sort of reward system since it all gets piled together. This system just front-loads some of the reward in an attempt at realism.

And I still see a crazy situation where you just make-up mission objective to get budget.

Okay, well, that's where a negative feedback system comes into play. A player should have incentives to, and be rewarded for, being honest and completing the mission. Failed contracts and rep could also reduce the front-loading of the rewards to a certain extent. Honestly, there has to be some way to "fail" at your career otherwise it's just sandbox mode with some fake nods toward "limits".

- Restricting the reward to a fixed amount of money regardless of what was spent (there's a upper limit of course). (you would however unlock "reputation" for being cost-efficient)

That's actually a pretty good idea.

Oh boy ! You have nooo idea how complex my space infrastructure could be. Between the Kerbin and Munar refueling-station, the mobile base traveling through biome waiting for a lander-module to transfer sample/fuel... from a thruster-less science-updated module brought to the surface after being brought in orbit by an 10 years old orbital-tug... not counting eventual satellite refueling service.

I tell you GOOD LUCK trying to make a Fair, Consistent and Easy-to-Use contract proposal system that reward/acknowledge my work in a satisfying way.

How exactly did all that infrastructure get there under "do missions" sort of play? Does it not exist? Do you propose a strictly rigid sort of gameplay where you only ever launch for a specific mission?

I don't want to be dramatic (I do) but that's a wonderfully horrible way to make a game. We know modern game are way easier than "3 Lives Left" games of the past but making progress slowly harder and harder... because you didn't do thing "right" is just being cruel.

You're making a mountain out of a molehill. All of the balancing concerns you bring up are also valid under "do missions" sort of play under KSP unless it is strictly rigid one mission/one launch/no personal flights, which I highly doubt we'll see.

All monetary reward systems under KSP will have to be balanced against your concerns.

Tracking of previous missions, both those done on "contract mode" (is that what you called it?) and off

Is this not happening under a standard "do missions" system? Is there no way to track your accomplishments, or what science you've done where?

Algorithms for evaluation of previous missions, on multiple dimensions you've mentioned including success and fame

We're talking two numbers plus completed objectives, you're over-stating the problem.

User interface to propose user-designed missions.

Actually pretty simple in the grand scheme of things. Would a normal "do missions" system not need a GUI or do you think this is infinitely more complex? I do front-end work as a day job and this isn't exactly a tough GUI problem to tackle.

E: We've seen something similar to this in the Elder Scrolls games since at least Daggerfall (I can't remember Arena that well at this point). Interestingly, it's become less complex over time. Within KSP you can use elements of the "assign pilots before launch" screen to make it happen; you need to add additional collapsible layers to the list to handle objectives versus SOIs (categories), but that's pretty doable. In fact, you could even eliminate the need for a tree system by dragging a category/SOI onto the contract and having it bring up a window where you choose objectives within that category/SOI.

Algorithms to evaluate user-designed missions based on the proposal and conditioned on multiple previous performance parameters.

Again, given a list of previously completed objectives you can construct a fairly simple algorithm to solve the problem. You're over-stating the problem.

Adjusting the algorithms to avoid exploits.

Balancing this multivariable system

Sounds like game balancing, which we're going to have under a "do missions" system anyway.

indicate that you've either failed to comprehend the complexity needed for your proposal, or have and simply don't care. The algorithms needed to produce the evaluations you're expecting would be incredibly complex and the time needed for their development would require removal of other features from the game.

No, balancing is the problem here, not back-end programming.

Third: In this thread, you've said that there should be monetary rewards for missions off "Contract mode", so what would this add? You're basically adding a game of "HORSE" to KSP (saying what you're going to do, and then doing it, instead of just doing it).

Honestly I'd just prefer if the game reacted to the way I wanted to play and rewarded me for my accomplishments appropriately while maintaining challenge. Now that is a monumental coding challenge. My proposal here is just a twist on "do missions".

indicates that you're basically asking for innovation for the sake of innovation

Yes, because "ferry around Derp Kerman" is downright boring. Let us design and direct our space program.

E: You know, the more I think about it the more I think you're over-blowing my idea here. It's really just "choose your own mission" with some front-loading of rewards; pizzaoverhead got the idea immediately. It's not exactly complex or particularly innovative, it just puts more choice and control in the player's hands, which can be a bit frightening to some people.

Edited December 7, 2013 by regex
I can't grammar/spelling.

[LethalDose]

E: You know, the more I think about it the more I think you're over-blowing my idea here. It's really just "choose your own mission" with some front-loading of rewards; pizzaoverhead got the idea immediately. It's not exactly complex or particularly innovative, it just puts more choice and control in the player's hands, which can be a bit frightening to some people.

Yeah, at this point you're just contradicting yourself:

"It's not [...] innovative" vs "I want something different than the last 30 yrs of gameplay" ...

"It's not exactly complex", but you've stated ideas that require goal achievement and performance evaluation all through this thread.

Also, "It's not exactly complex" vs "The calculations are as complex as they need to be". I mean, you can try to hand wave about the complexity, but can't do that and be surprised when someone calls you out on it, or back-pedal and try to say its not complicated at all.

You want the game to track achievements, and keep track of which were "mission goals" and which weren't, but then say this doesn't take any back-end coding. Same issue with mission and proposal evaluations. Yeah, this takes back-end coding. And it's complicated. And it would be waste of the dev's time.

If this system is just based on one or two metrics, then you're just introducing a rep grind: "If I want to unlock contract mode for Eve, I have to increase my rep to 300", and "Grindy play is bad" has been posted all over these forums.

And if what you really want is a system that reacts to the way that you play, and what you've proposed doesn't achieve that goal, then why are you wasting our time?

I'll even play your game and pretend theres no back-end coding required (which is absurd on it's face, but whatever), and still show this is a bad idea:

What's frightening to me is removal of player agency, and your proposal, which locks players into stated mission profiles a priori, removes player agency. There's nothing about this system would increase choice, because as it stands, we already fly the missions we want, and no contracts system is every going to prevent that. If your justification for this system is "placing choice and control in player hands", then the system's justification is predicated on either:

a) There being a current system in place that restricts what missions we fly

OR

There being a proposed system implementation that would restrict what missions we fly

Anyone who thinks there's a case for "a" is hallucinating, and for "b" is paranoid. I can't imagine that "contracts" would ever be mandatory to begin with, and statements that attempt to draw an equivalency between being presented with optional random missions and loss of agency are equally paranoid. So this...

I want to direct the future of my space program, not have some dumb talking head dole out random missions.

would require a LOT of supposition to be even a little bit valid. These two points have little to nothing to do with each other.

The only thing that makes me feel better about this idea is that it'd be such a PITA that I can't imagine the devs would ever try to implement it to begin with. Again, done with this.

Edited December 7, 2013 by LethalDose

[boomerdog2000]

I fail to see the problem here. The basis of what regex seems to be suggesting is similar to what normal space programs do.

"We plan to do x with y ship."

"Very good, we will give you z money to do this. Keep in mind though if you screw up, we will be less inclined to give you money in the future."

Doesn't seem all that complicated, the main issues seem to be balancing, which will never cease being an issue no matter what game you're doing.

[regex]

Also, "It's not exactly complex" vs "The calculations are as complex as they need to be". I mean, you can try to hand wave about the complexity, but can't do that and be surprised when someone calls you out on it, or back-pedal and try to say its not complicated at all.

You can make it as complex or as simple as you want. It doesn't exactly require a genius to program a system to total up rep gained, see what you've already done before and how many times, and adjust rewards accordingly. Fine-tuning and balancing the system is what will take time.

doesn't take any back-end coding

I've never said that; please re-read and comprehend my posts before putting words in my mouth. My point is that the programming for this feature is trivial versus the balancing, which we've all agreed is an issue. But then, any sort of mission-for-money/rep system is going to require balancing.

If this system is just based on one or two metrics, then you're just introducing a rep grind: "If I want to unlock contract mode for Eve, I have to increase my rep to 300", and "Grindy play is bad" has been posted all over these forums.

There is nothing stopping you from just flying to Eve under this system using your general fund.

And if what you really want is a system that reacts to the way that you play, and what you've proposed doesn't achieve that goal, then why are you wasting our time?

Because what I'd actually like to see for KSP is pretty nebulous and a total nightmare to program. This feature is literally "choose your own mission" pretending to be a governmental proposal system. It's a lot more free-form than "Here are your choices today: Ferry Derp Kerman around, Land on the Mun, Fly to Eve" which is why you think it would be a nightmare to code.

pretend theres no back-end coding required (which is absurd on it's face, but whatever)

You are correct, it is absurd to suggest that this will require no back-end coding, which is why I have never suggested that. Once again, I ask that you read and comprehend my posts before insulting my intelligence.

What's frightening to me is removal of player agency, and your proposal, which locks players into stated mission profiles a priori, removes player agency. There's nothing about this system would increase choice, because as it stands, we already fly the missions we want, and no contracts system is every going to prevent that.

If I already fly the missions that I want to fly then why do I need a monetary or reputation system under career mode to begin with? I might as well just play sandbox mode and call it good since I don't need to pay for my rockets or crew.

What I am suggesting here is a system where the player proposes missions that they'd like to do, which sidesteps the current paradigm of "dole out missions" based on some other metric. It also front-loads some reward so that the player doesn't immediately pay for things out of pocket, but also gives them a general reward so that they can fly free-form missions.

As far as player agency, under a normal mission system you are locked "into stated mission profiles" which removes player agency. In fact, you might even be locked into doing certain missions in a certain order. Under this system you can create a mission that suits your own goals and get rewarded for it, which increases player agency.

I can't imagine that "contracts" would ever be mandatory to begin with

I can't exactly either, but how else are we supposed to gain money and reputation within career mode? Does it just appear out of thin air?

statements that attempt to draw an equivalency between being presented with optional random missions and loss of agency are equally paranoid.

Nowhere in my suggestion is anything that prevents you from flying whatever mission you want outside of the proposal system, aside from money to pay for your craft. That is how I would expect a contract/mission system to work under KSP. A normal amount of doing things to gain money/rep is expected under career mode; I am simply suggesting a system that places more choice and control in the hands of the player.

Edited December 7, 2013 by regex

[LethalDose]

I fail to see the problem here. The basis of what regex seems to be suggesting is similar to what normal space programs do.

"We plan to do x with y ship."

"Very good, we will give you z money to do this. Keep in mind though if you screw up, we will be less inclined to give you money in the future."

Oh, I don't think there's any debate about this at all.

The problem here is not "does this appropriately model the process", the problem is difficulty of modeling the procedure in a way that can be implemented and encourage fun game play.

Simply put, this system fails on both latter counts, despite the "nuh-uh" counter-arguments presented.

Where implementation gets hairy is your second sentence. "What normal space programs do" has humans making decisions based on data from lots of disparate sources. This is something that humans are really good at and computers are really bad at. The decision algorithms that would be needed for this would make the system:

1. Exploitable, due to simplicity
   
2. Irrelevant, due to simplicity
   
3. Grindy, due to simplicity
   
4. Have an extended release due to extended development due to complexity
   

There simply ins't a magical middle ground that avoids these issues: To develop something that wasn't grindy, irrelevant or exploitable would require too much dev effort.

Besides, justification of game systems based on realism is... risky. Game design decisions should improve game play. The OP attempts to address that issue and justify the system by claiming it would increase player choice and control, but fails. Like I said in my previous post, the problem that this system proposes to solve (loss of choice/agency about missions choices) simply doesn't exist in 0.22, and it's difficult to imagine changes to the system in any future patch that would create this problem.

The OP also tries to draw an equivalency between opposition to this overly-complex idea and being frightened by increased player control, which, frankly, I find to be very offensive and demeaning. I may be blunt, but I'd never imply that someone was scared of good design.

As for ...

"Doesn't seem all that complicated, the main issues seem to be balancing, which will never cease being an issue no matter what game you're doing."

might seem true on it's surface, it's much MUCH easier to balance well designed systems compared to needless complex ones. Using overly-complicated systems where simpler ones are perfectly sufficient only makes balancing games more difficult. And as stated above, the complexity is never justified.

Edited December 7, 2013 by LethalDose

[LethalDose]

And fine, he claimed the back-end coding is trivial, not absent.

The statement is slightly less, but still incredibly, wrong.

[tetryds]

Why can't we just pick missions, fly, make money, waste money?

Possibilities of missions which can be easily procedurally (or manually, like Skyrim) generated, you pick what you can do, you get what you can get.

Aim too high and lose cash if fail, aim too low and progress slows down.

Then put a little dificult bar based on progress or unlockable sections of missions based on science to not let newcommers pick "Go Eeloo and come back" as the first mission, and that's it.

Because capitalism rules.

[regex]

Why can't we just pick missions, fly, make money, waste money?

Why not? You could easily use this system to procedurally generate missions, that was suggested earlier. It could live alongside a system where you pick some random "story" and do it.

Possibilities of missions which can be easily procedurally (or manually, like Skyrim) generated, you pick what you can do, you get what you can get.

Yep, and this proposal just attempts to put the procedural controls in the player's hands.

[regex]

And fine, he claimed the back-end coding is trivial, not absent.

The statement is slightly less, but still incredibly, wrong.

Nothing else to say? I've sufficiently assuaged your concerns about player agency? You accept that players will have to make money somehow in career mode and that there will be balancing involved one way or the other?

Well, since there doesn't seem to be "any debate about this at all" I ask that you take your insults and run along. Please leave this thread to those who want to discuss the suggestion.

[Kegereneku]

This post took way longer than I though.

Sure, let's do that. How do all your concerns about my system not apply to a standard "do missions" sort of system?

For starter we have to define what you meant by "do mission".

/!\ For clarity what I meant by Fixed-Budget, it should answer most of your question :

- Every month (example) you are given a fixed amount of money to spend for ANYTHING. Mission or not.

I interpreted "Do missions" as : mission-based progression. But alone it's not enough to make sense of it. As the current BETA-Career demonstrated you can have a technological progression without any mission, but right now it's not interesting enough.

The big question have always been "how to keep the player's progression interesting".

Science-point are in finite amount, progressively harder to reach and can't be lost. A dream-like parameter for progression.

The problem is budget and what to make its driving parameter(s) based on players' actions. (Mission, Reputation, Prestige, Time...)

* User-Generated-mission is not very different from Game-generated-mission in term of balance. But whether or not it's the only way to get budget is important.

* Money-per-flight or Fixed-budget or both, change everything. I consider money-per-flight doomed, but Fixed-budget isn't perfect either as docking theoretically allow you to transform money into useful-asset as time goes on. At least the size/cost of rocket launched per month stay stable.

* Reputation-based progression is an interesting mechanism depending on how it's done.

I don't have an proposal myself, but I think linking it to budget would work better than to what mission we trust you with. As you might be able to explore Duna regardless of your reputation.

/!\ Now to be clear : As of now you described a system where the only way to get money is "Do mission" (more precisely make up) with flawed parameters/restrictions, one had no negative-feedback, others are impossible to work with.

Since you get some of the reward up-front to build a craft to complete the contract, there will almost never be a situation where you have run out of money.

[...]

Then you could build space infrastructure from your general fund. The idea behind that is that the player has freedom to expand and build as they see fit. It's not really any different under a "do missions" sort of reward system since it all gets piled together. This system just front-loads some of the reward in an attempt at realism.

Put aside that it mean you could pretend to fail mission repeatedly to grind money, I assumed that you necessarily got 100% of the Money required to accomplish the task and kept the change in you general fund. It doesn't change my criticism.

The problem is that you get a recursive loop where a 100% success rate get you exponential amount of money, and suggested restriction leading to absurd situation like : "You can't refuel/save the Kerbun from the earlier attempt or you don't get as much money. (Ps: let those looser die !)"

As for realism... NASA is government funded, SpaceX have investor +subventions, both can "get paid" based on "offer & demand" or at least get things done. Both can fail/bankrupt but very differently. Two difficult things to simulate.

You don't need space infrastructure to go anywhere in the solar system right now

You know that it is because right "now" budget is infinite, number of part infinite, science easily accessible and the tech-tree is I hope a place-holder.

To get sample from several part biome/planet of the Joolian system (and further Gaz Giant) efficiently you better need an infrastructure, especially if we get some day that stock-ISRU we dream of. Plus Nuclear Thrusters won't be free forever.

Honestly, there has to be some way to "fail" at your career otherwise it's just sandbox mode with some fake nods toward "limits".

"Failing to progress is better than regressing" (I felt like quoting Antichamber here)

I don't mind making it possible to go "bankrupt" in career mode, but since it will be hard enough to do a lot of science on distant planet in a game that never end, it may be a bad move to "punish".

The Negative Feedback I talked about is more to keep budget from escalating out of control.

Edited December 7, 2013 by Kegereneku

[tetryds]

Why not? You could easily use this system to procedurally generate missions, that was suggested earlier. It could live alongside a system where you pick some random "story" and do it.

Yep, and this proposal just attempts to put the procedural controls in the player's hands.

That just makes no sense.

To track the player experience you need very long data collecting.

That is mostly used on race games, and you probably noticed that the first race is usually too easy, and then if you win by several seconds of difference the first races are going to be harder.

You have the impression that you got better within 10 or 15 races, but its actually that the AI adjusted itself to match your skill and gamestyle.

The codding to do that are within the range of the doable in race games, but in a game like KSP that just doesn't fit.

Basically because there are way too many variables, infinitelly more than a race game.

You know what you abble to do, a list of missions to choose from, just like the missions mod i've played, is the best and most simple choice

There is absolutelly no reason to make it complicated, risk failure and increase balancing and programming effort exponentially for something that is not needed.

Remember, K.I.S.S, Squad is not just developing this game because it's cool to give players a nice game, but for money.

The better working and most simple way to implement a feature, the better.

You can quote me on this: You are "trying to use aircraft engineering to make a toy" here, and there is a name for that: overkill.

[Kegereneku]

To complement the Analogy with Race-Game you might want to learn about "Rubber band AI"

http://tvtropes.org/pmwiki/pmwiki.php/Main/RubberBandAI

And a game using an AI to "create a story" is Rimworld, I don't know what it's worth though. But "story telling AI" are NOT an easy task.

http://rimworldgame.com/

Edited December 7, 2013 by Kegereneku