I am compiling a load of ideas for the next version of Chom Isis and would like to give you all the chance to have a say. The lovely chaps, who I am endebted to, are playing through version 3.2 and the first big game I will advertise for will be of that version. I'm putting bug (and a few feature improvements) fixes into 3.2 (and 3.3) but the backend in 3.3 has some big improvements. Please feel free to keep things vague so as not to ruin any chance you might have in the existing game.
Application Programming Interface (API)
The API will allow Chom Isis to connect to other systems on the net. It will come in two main forms: Feed and Programmatic. The Feed API will allow you to receive RSS feeds on various topics, such as global information on who is winning, pawn states and message board. Then there will be private feeds only accessible to you that will report private information. What you do with these RSS feeds is up to you. The programmatic interface will use either JSON or XML (not decided yet, prolly JSON) to provide data. You will need some sort of coding ability to use this. A load of documentation will come with it. With this in place, I will be able to update some of the screens with AJAX - something I wanted to do long ago.
New Alliances Screen
Current one is fugly. It's going to change.
Disruption + Winfall jobs
Narrowly missed 3.2 release, they will offer a quick bump or drop to the happiness of the pawn.
New Front Page
A public visage to the game. But what to put on it? All the public information and perhaps the odd stat or two.
Do I remove location from the Spy job?
Pete (who's not playing) suggested that the Location intelligence nuggets are not very useful because you can do just as well by moving the pawn to where you want them and then using that knowledge. You know where the pawn is because you've moved it! Do you agree?
Location Engine
Every location will have a colour level, red to yellow to green. Jobs in red areas cost more. Green areas cost less. Good jobs positively improve the area and vice versa. Same as in Chom 2.
Pawn happiness affects cost
Happier the pawn, the more it costs to do jobs on it. Same as in Chom 2 (but with less affect).
UI Notification boxes
Little boxes to show when there are new things on Alliances, Intelligence and Auction screens.
Things I might include...
Random News Engine
Automatically generated stats and 'world events' triggered at random or when certain thresholds met. Changes to pawns, life inputs and the map.
Share nuggets in an Alliance
Have a feature to send the nuggets to an Alliance. From the Alliance screen, you can see all the nuggets shared and copy any of them to your own intelligence screen.
Something else to add to the
Something else to add to the Move Pawn / Spy thing... when you do a successful "Move Pawn" job on a pawn not belonging to you it would be rather handy if an info nugget was added to the pawn in the info screen.
I'm not sure that nuggets of info in and of themselves are worth enough in the impact to cost scale on the jobs with the current balance.
Given the assumption that the most you should spend on pissing off a pawn is a murder job, the most griefing you can cause costs 2200 (using the Don).
A single nugget of info gives you a 5% probability boost. the difference between 94 and 99% for the Murder job is about 120-150.
A Single spy job will not give you a useful nugget...it requires at least 2, both of which have to pull out the same infoset, and not return identical results. Cheapest Spy Job is 50, so minimum cost of 100, saving you 20-50 (approximately 1% of the cost).
A Move Pawn job costs about 150. This does give you a reliable nugget, however you could just have well set that money directly into the Murder Job.
Rather than suggest a re-tweak of the numbers, can I instead suggest a radically new way of using nuggets. The core issue is that info-nuggets have no inherent value other than providing a percentage cost reduction to jobs. How about, instead, executing a set job on a given pawn requires certain Pre-requisite nuggets, in keeping with the nature of the job.
An example list of suggestions;
1) Spy - no pre-requisites. Provides 1 from Owner, Life Input, Location (choice by player when setting job). Could potentially have a multi-levelled job that returns more info, or more accurate info if providing a selection, however the more intrusive the Spy, the more chance of warning in the pawn diary.
2) Move Pawn - requires Nugget "Current Location"
3) Destroy Life Input - requires nugget "Life Input"
4) Disruption/Windfall - requires nuggets "Curent Location" and "Life Input"
5) Murder - requires nuggets "Current Location", "Life Input" and "Owner"
To set the job you'd need to select at least one of each of the required nugget info-types, however at this point you would not know if they are correct or not...only when the job completes would you find out if you had been successful or not.
This would fit thematically...after all, if you want to forcibly move someone to a new home, you'd need to know where they currently live. To fuck over someone's hobby, you need to know what their hobby is.
This would also provide options both offensively and defensively. By constantly moving your pawns you would make them much harder to attack (but at a cost to yourself fiscally, and to their happiness). Offensively the requirement for multiple valid nuggets to perform high-impact jobs would encourage knowledge-sharing.
I agree that a successful
I agree that a successful move job should create a nugget.
The big rethink
I do like this system. It isn't fitting, though, because you don't need to know where the pawn lives, the Operative needs to know and for the Operative to know, Chom Isis needs to. Which it always does. You could argue either way, though.
I think the usefulness of nuggets is dimished if you're a lone wolf. Those people who have won big time in the past have been those that have collaborated up until the very last moment. There isn't a nice way of sharing nuggets at the moment but I do intend to have a way to do that (originally, they could have been bought and sold). A load of you pool the information you've gathered (perhaps using Alliances or meta-game methods) to bring down the cost of jobs. I know people have done that over email/phone before (particularly pawn owners). I'm not stopping people from being lone wolfs but it's always going to be harder. If you have a load of players with cash knocking pawns over at random, you're more likely to be out first.
I also don't like the idea of restricting people's use of the job system. You've lost your last pawn. You have a big chunk of cash. You're not in the end game. You wanna murder someone as a final hurrah. By forcing them to do a load of spy jobs first, you're just putting a barrier against it. Just extending it. I think you should just be allowed to randomly murder a pawn because you don't like their name or face.
The other reservation that I have is that it would require a considerable amount of recoding, something I'm not considering at the moment. Expanding, contracting is ok but not rewriting.
Keep the ideas coming, they're forcing me to think!
If Chom knows all the info
If Chom knows all the info about the pawn, and the operatives, how does you knowing improve the operatives chances of completion? It would indicate that the operatives are perfect, and choose to kill or not based on a dice roll. I had always assumed that the thematic point of nugets was additional information handed onto the operative to improve his chances, not to improve his dice roll to see if he can be arsed to do your job or not...
The problem is that as it stands the cost reduction from Spy jobs is not worthwhile.
Alternative suggestion...a sort of mix between the above tiered jobs logic, and the current percentile based logic. This also incorporates <99% chance jobs (which I'm going to take a wild stab in the dark, and guess that the vast majority of jobs are set to 99% probability)
As above, jobs have a "nugget requirement" (0-3 nuggets required).
If a job is set, and a nugget is incorrect, then the percentage chance of completion is dropped down to the maximum allowalbe level for remaining nuggets.
For example;
Sets 3 nuggets on the pawn (location, owner, life input)
The location is incorrect. When completing, the job now only has a 75% chance of completion.
Sets only 2 nuggets on pawn (location, life input). Maximum completion chance is now only 75%.
Player reduces cost to 1700 (I made this up) to match the 75% (he doesn't have to, but the extra money would be wasted)
The life input is incorrect. When completing, the job now only has a 50% chance of completion.
This would allow players to continue doing random murder attempts, however at a lower chance as they are un-prepared. It would make the "last gasp" murder spree more comedic, as the player would be spreading their random murders about, as they cannot do 1 big hit, but rather 2-3 lower percentage chances. It encourages nugget sharing/info gathering to improve attack chances. It still allows some defensive attempts as per the first design option by allowing you to move pawns around, and thus make nuggets more lightly to be incorrect (effectively defending the pawn from murder by 25%).
What are we trying to
What are we trying to achieve here? Make nuggets more useful? Why would you spy if the nuggets you get don't help you in murdering pawns?
The solution to most software problems is not to pile on complexity but to go back to the root and find out why it's not solving the problem you're trying to solve. Although your new solution does solve all my points, it's now devilishly complicated. The interaction between nuggets and jobs is complex enough as it is. I am drawn to the last hurrah jobs at 25% for being ill-prepared. That is funny.
The problem
What nuggets are trying to do is two fold:
1. Have a place where you can collect information on who owns what so you can select your murder such that you can get a player out of the game.
2. Reduce the cost of a job.
Another solution (by tuning)
What if nuggets changed the job by 10%? Or more? Or a little less?
Player sets a Murder on Pawn A for 1700, giving a 70% completion chanceSets 3 nuggets on the pawn (location, owner, life input) = +30%
If they are all correct, you get 99%, if not you get 40%
Does that then give them the power without forcing them to use the nuggets? Furthermore, I can do this without having to change much code.
The problem as I see it is
The problem as I see it is that Spy jobs are not fit for purpose in their current format, and under-used for the amount of time you have spent integrating them into the game. It's not really a software problem, but more of a useage and thematic concern. If you look at it from a "reduce complexity" concept then you will eventually reduce the game to "press here to win". With software you eliminate complexity, with a game you apply it...
I know that in previous versions of Chom spy jobs have not been used at all, and having played with the current version I can't help but think that the same will occur again, as they have not really changed. I know some of your early thoughts were along the lines of nuggets/bona fides being a type of currency within the game, however there were issues with value/validity.
The current process for nuggets and spy jobs effectively does the following ;
Convert Resources + Time to Nugget(s) (Spy Job)
Convert Nuggets to Resources (other jobs)
As it stands most players will simply skip the conversion, and go straight to
Resources to (other jobs)
It's faster, more reliable, and requires far less micro-management. In effect the players are removing the complexity for you, simply by bypassing the process you've coded. A tweak to the conversion rates from Nuggets to Resources my mitigate it somewhat, however tweaking ina game like this is a tricky proposition, and you will probably find that either no-one uses Spy Jobs, or everyone uses them...hitting the sweet spot will be extremely tricky.
I'm trying to work out a way whereby Nuggets become separate to resources, and therefore a viable commidity in their own right within the game, and thus worth the effort you've put into them so far. Throughout Chom I see a lot of "potentials" that are never utilised, as they are trumped via other processes (other examples include low-percentage jobs, as mentioned above...Alliances are another. Their main purpose seems to be resource generation, hence why I'm about to suggest a link between Spy Jobs and Alliances).
So onto another suggestion. Spy Jobs can only be initiated by an Alliance (or 2 or more players). The nuggets then goto all players within the alliance (or perhaps sit in the Alliance screen, and current active members can utilize them). Spy Jobs can only be set with Alliance generated money (so cost of the Spy Job becomes less important, and you don't have to worry about offsetting Spy Job Cost + Time against the value of the reduction of the attack jobs).
This would promote alliances, and allow information sharing within the alliance. It improves the Spy Job's cost/benefit ratio. Larger alliances would be able to generate more info...
I think the key here is to
I think the key here is to make nuggets easily shareable via an alliance and to attempt to find that tuning sweet spot or get near to it. Mechanically, nuggets are about turning resources back into resources. True. However in game terms, they also provide information for tactics. If you are suspicious that a single player has loads of pawns, some spying can help you. Perhaps I might add some 'note' nuggets where the system generates some generic stats, like the number of life inputs it has or the average number of spots the deviant has. Or even what jobs are being - or have been done on the pawn. It's not mechanical but it might help make decisions between two pawns.
Having lots of them makes you better. Having alliances setting jobs is a minefield, which would take a huge amount of code effort but having nuggets shared through an alliance is very do-able.
Alliances aren't used because this is a beta. You watch 20 players together, alliances will form and will become powerful. They've worked in the past and they'll work again - with more players. There is more expansion in the things they can do but the core idea is sound.
So, should I increase the nugget bonus to 10%?
Try it...not sure how much
Try it...not sure how much testing I can do on it now really, as I'm pretty much out of the game (was never really in it).
I'm not sure that a percentage is the right way of applying it as a discount, however without the full job pricing model it's not really possible to work out the value of a Spy Job. You could work on the assumption that 1 Nugget = Cost of 2 Spy Jobs, and that Reduction should be (Average Spy Job x 2) + x%, and apply it as a flat rate, which would make Spys more viable for non-murder jobs (useful when Disruption enters the system). I'd provisionally put x at about 175%.