[DISCUSS]Bot Plannets

[Tactitus]

Originally Posted by mist

how can you calculate 100+ values associated with ship stats... oh, wait, people did

with enough data, it's possible to work out the trends behind pretty much anything, once those are known it's possible to extrapolate what will happen next, resulting in an attempt to play the system

however, isn't that what people are doing anyway? rather than playing the system they're attempting to play the other players. players are also pretty predictable and, with a few exceptions, can be relied on to do pretty much the same thing every time you poke them. hence, why are bots which act the same such a bad thing?

It's bad in the sense that bots aren't paying to play the game.

Every hour spent programming more "realistic" bots is an hour not spent programming a better game. Granted, it may make the game more enjoyable for the few remaining players, but how is it going to attract more players? Bot planets which just initiate roids--ostensively for new players to take--will instead be harvested by the more experienced and more organized players/alliances. Bots which are "smart" enough to seriously challenge the top players/alliances won't be of any use to new players. Bots which are just as predictable and just as unpredictable as humans are just a non-paying substitute for the real players we all wish the game still had. Instead of simulating customers, let's instead try to attract some. Yes it's hard and we may fail, but ultimately I think we have to decide if we want to keep trying to regrow the business or just make the few remaining customers "comfortable" until the end. The term for the later btw is a going-out-of-business plan. :/

[cypher]

this idea is as old as pa prolly...

if the bots are coded well they can put this game back in business.. better then anything else... cuz so many peeps wanna play when there are enough planets around (one of them being me)

but as i said.. it has to be coded right...

can't be too easy too farm them ofc... no changing weird limits.. that's saying a 1 mil guy can farm them...and a 2 mil guy can't... stupid basically.. they would need some sort of high artificial intelligence.. and they can fight in an alliance together.. and actually attacking other gals or something. if it can be done tho i'd say it should have a way higher priority then game development. i mean instead of updating stats and stuff, use 1 round to make decent bots. so we have a huge planetbase in this game once again. 300 people blocking then, won't matter one bit. Skill would be required once more then, which is not needed these days anymore... sadly enough.

[pablissimo]

I think Tact's got a point there. As a potential customer, I'd rather start playing a game that could boast 'solid gameplay and community spirit, you'll have a blast' than that boasted 'and if it gets too quiet, we've got bots!'.

[mist]

saddly, people don't want to play unless there's so many players, it's rather a self defeating thing.

and while coding bots would mean that the person couldn't be coding something else, there's little to stop someone else coding bots while whoever's coding the game gets on with it

-mist

[Qdeathstar]

Now If i'm planet ranked 150 and the bots decide to attack me it could easierly end up with a high ship multiplier and a low eta multiplier facing me with 20k big ships incoming at me a very low eta which i dont have a chance of defending.



No, remember if you read what i said carefully, i said it would only attack the top twenty five planets.

And it will have to send atleast 1000 ships, (100 of each type) to the planet, so there would still be losses on the otherside. In addition, having to defend your planet stops you from attacking other planets.

[mist]

in that case, if you ever get in charge remind me to get to 26th and then start stockpiling resources

artificial limits will always be exploited, your suggestion is one.

-mist

[Qdeathstar]

except with people above you being attacked, you wont ever be able stay right there....

[pablissimo]

What would then be the incentive to try and get a top 25 place?

[Qdeathstar]

To win the game.... some one would be at the top 25 no matter what, and if people are saving resource.. you could bring back cargoships taht could take the res.... in addition, they'd still have roids, makeing them easy targets...

[mist]

you forget alliances... a person's planet is nowhere near as relevant as their alliance, so all they've got to do is save resources and use alliance defence. likewise, cargoships arn't going to make it thought, and last time they were in they only stole resources produced that tick iirc

-mist

[SirKillALot]

I could easily create a bot script and two. But it would demand a lot of extra space in the database as we need to record what kind of player it should be. How boring to be if the bots only did the same all the time. And when a real person go down to 10 roids or so after the gal has been blown away the player would have stopped to play. But would the bot have?

And much more. It would be a bigger script than the whole game itself.

[Qdeathstar]

SirKillALot Did you read what i had in mind...

let me refresh...

a random number generator selects the

target cords (limited by score
number of ships from each race (so, it can have a bit from all races)

and they just attack... they cannot be attacked.

it was more than that..by a bit ... so read up a bit ;_)

[mist]

Originally Posted by Qdeathstar

a random number generator

given enough random numbers, the result isn't noticably random, as in the pa combat system.

-mist

[Qdeathstar]

whats your point .... O.o

[mist]

the randomness of bots seemed important to your post, so i was pointing out the fallacy of it

anyway, what's the point of bots that can't be attacked? might as well just deduct ships and roids off people randomly

-mist

[tauric]

If you want "big" players to be effected by the bots then let the bots attack traveling fleets at random.

Your fleet ran into a pirat fleet while traveling at their destination. You lost x ships type a x ships type b
Do you want to recall your fleet? (well balanced attackfleets are effected)
Your fleet ran into a pirat fleet while traveling at their home base. You lost x ships type a x ships type b
x roids were destroyed. Do you want to pursuit and bash their base?

All fleets could be affected, big ones small ones, defending and attacking ones.
If you are able to tell how many fleets are out, then you can probably tell how strong it is.
It may be called the x factor instead of botplanet.
A scan goes around pa every tick and every 16th planet that has send out fleet 3 will be fleet scanned and will meet an opponent of approx 50% and it will take damage.
You might even meet two piratefleets in one run.
Give them a strong homebase that aliances must seek and destroy before beying able to roid me poor noob safely :-)

[Ramihyn]

Originally Posted by mist

given enough random numbers, the result isn't noticably random, as in the pa combat system.

-mist

What do you mean? That if you just know enough random numbers, then they arent random anymore or respectively that you can "predict" them?

[Qdeathstar]

thats why i asked "and your point" hehe.

[Tactitus]

Originally Posted by Ramihyn

What do you mean? That if you just know enough random numbers, then they arent random anymore or respectively that you can "predict" them?

Basically. In large sequences of independent random events the randomness tends to "cancel out." See the Law of Large Numbers and the Empirical Law of Averages.

[Qdeathstar]

yeah, but, we arent talking thousands of numbers... we arnt talking about averages...

we are talking about 50 random numbers... being that their are an infinite amount of numbers

You say its predictible, ive ran a randome number generator to supply

1. Coods

a number for the parrellel, cluster, and galaxy (assuming all numbers <10}

total ammount of ships Assuming a number <10000 (although this will not always be the case, depending on the top score, the max number of ships will be twice the ammount the number one player has.

and an attack time 1-24

Ive did this twenty one times, 3 attacks a day.

Under that logic, you should be able to pick out the numbers ;-) GL

[Ramihyn]

Originally Posted by Tactitus

Basically. In large sequences of independent random events the randomness tends to "cancel out." See the Law of Large Numbers and the Empirical Law of Averages.

Yes fine - just a question of about how many gazillion numbers we talk about

I guess it's my turn to ask "Whats your point?" (esp. regarding the type of use of the random numbers here)

Apart from that, i dont see the point in these total random attacks either :O

[mist]

totally random attacks are largely a waste of time. asside form anything else, the odds of them working are slim, hell, the odds of them actually being able to launch on the planet they choose arn't 100%.

if there are going to be bots in the game they need to act like humans do, rather than acting randomly - otherwise pa will turn in to a "beat the bots" scenario. the best way i can see of doing this is to get "experts" in the field, alliance leaders, BCs, DCs etc to get together and tell us what they'd do, and then have the bots operate more or less along these lines of best practice.

fairly like expert systems do, when used. imagine that :P

-mist

[Heartless]

I have not read this thread 100%, however, the first page and some of the 2nd page post I read carefully.

From what I have read there is just a few conclusions:
- People fear bots become farms
- People fear they are predictable
- People fear they may become unfair

Yet nobody has defined why a normal new player wouldn't become a farm. Nor why the usual pa player isn't predictable. And the unfairness has just been a follow-up from those two.

My thoughts on this are not fully detailed but everyone who reads carefully should be able to get the point.

Basically you have to see how the game is working since rounds. The new, inactive, unexperienced player is a farm unless he knows someone / gets in touch with someone (which, tbh, is very very hard since p2p as the major alliances every round are way too paranoid).
Also, the usual pa player IS predictable. He attacks where he thinks it is benefitting him. He analyzes the ship stats to decide which ships he should build. He oversleeps sometimes and sometimes he is out partying. For the single player alone it therefore comes down to 2 basic values: "skill" and "activity" which determine everything concerning basic things like tick-plans and fleet compositions.
For the "community" aspect it comes down to another very few attributes: "helpfulness" and "egoism". The more selfish players tend to send 2-3 attack fleets out more constantly and offer less defense while the more helpful players are the opposite to those, more or less.
The alliance part of the bot planets can and for the time being, as I know it is a VERY hard task to simulate politics with an AI, should be limited to defense and attacks. It is quite simple to realize defense/attack scripts, and to estimate the best available defense fleets and to write a script for attack management. Target galaxies will be picked by their ratio. Yes this might lead to "independant wave attacks" and occasionally to piggybacking of bot-bot and bot-human. But then again, it has never been different during the free times of pa. It was just that it all was distributed between a lot more planets. If you come up with 3 times as many bot planets than human planets in a round with 2500 human players, for example, it is highly unlikely that new players will get bashed to the ground constantly all round long.

It is not hard to develope the scalable AI for a bot planet. It is hard to develope the AI to simulate alliance interaction. And it is hard to decide how many bot planets should be injected.

[mist]

my main "problem" with bots is how to make them attack different targets. there's always going to be one gal, cluster, planet, whatever, with the best ratio. having all the bots attack that planet would rather suck. however, last time i was paying attention that's generally what happens with the human players as well, so no solutions to steal there.

following that theme, generally the alliances settle down in to blocks, and attack the opposing block (in recentish rounds anyway) which makes target picking tricky for bots, as they're unlikely to be in a block.

in other thoughts, your bot "stats" don't seem to allow for fear - not wanting to attack a big planet because it's likely to get defence, result in reprisals, etc

-mist

[Heartless]

The big planet problem is an issue of parameters of the target-picking function. Could as well be dependant on the skill of every bot :]

[Qdeathstar]

the bots would look for the best targets, but the strongest planets

[Heartless]

What about a brainstorming session how such bots should be designed, everyone? ;-)

[mist]

Originally Posted by Qdeathstar

the bots would look for the best targets, but the strongest planets

oxymoron

-mist

[Qdeathstar]

not really, the largest planets have a highe roid/score ratio... === best target

Most top 10 planets have a lower roid/score ratio...

but it wouldnt be based on that, but rather who ever the top 50 were, they would be the onlyones the bots (may) attack

Ive already discused simple bots, based on random numbers.... read the thread for mine if your interested (Its a long one, so it may take some time :-/)

I dont think bots have to be complex, or simulate real planets..they simply have to bring down the larger planets, and provide targets for the smallest planets.

[mist]

i've read all the threads on bots. however i'm 100% against the idea of bots who only attack the top 50, for a few reasons.

first off, you'd need an awful lot of bots if you're going to have them roidable by the weak and able to roid the strong. additionally, if the top 50 are going to be attacked by bots while the rest arn't people are going to strive to be 51st untill the last possible moment.

i'm not quite sure what you hope to achieve by adding bots to the game

-mist

[Qdeathstar]

the robots dont have to be actual entities who act real. They could be planets with roids and veiw ships that only the very week could attack, and then just have fleets from B:B:B attacking the huge planets.

You'd have to have a score <1000000 and a size <100 to attack and the courd would be B:X:X...


and striving to be 51 would be imposible, because spots would change so much... and if they tried to be place 51, they keep their score down, and dont roid so much.

My purpose is to keep the top planets in line, and give the weak planets a chance..

[mist]

your purpose seems to be to make the game unplayable for those who put effort in and dull player vs computer for those who play casually, neither of which seems particularly useful

-mist

[Heartless]

Well it's pretty easily answered what bots should be able to do:
They should simulate real players, otherwise they are ... useless / unfair.

[Qdeathstar]

unplayable? no... its just that your not going to have a run-away victory... and we do need something for the casual player to do than simply be food for the Masters.....

[mist]

Originally Posted by Heartless

Well it's pretty easily answered what bots should be able to do:
They should simulate real players, otherwise they are ... useless / unfair.

give the man a cookie :P

Originally Posted by qdeathstar

unplayable? no...

say there's 5000 bots in the game and 1000 real players - for example. the bots are going to be quite crap so that newbies can attack them easilly, i think you mentioned a score less than 1m for hitting them, so say they're average of 500K score. 5000 bots at 500K score a piece is 250M score hitting the top 50 every night. ok, alliances can probably cover that, certianly once the round gets going (in the mean time, top 50 get nuked and change every night - the game is meaningless, but hey, that's not important is it?) however, untill quite late on it's going to place a significant drain on their defences, with a two fold effect. firstly "ordinary" alliance memebers arn't going to have any defence, so most attacks will be successful. while it's important that attacking is possible, i'm not convinced a huge success rate is a good thing. secondly it means that everyone's fleets are out in defence, so there is little attacking. also there's little point attacking the top 50 as they're going to have incommings anyway, so it'll screw your cap

all in all, i'd equate that to pretty unplayable.

-mist

[Qdeathstar]

Mist, i think your missing the point..

the planets attacking would not be actuall planets, they wouldnt originate from anywar. Barbians, if you will. The attakee simply gets a message "X ships coming from an unknown sector". The atackee would be chosen from a set of random number generators that are limited from the cords of the top 50 players. Then, the time is chosen again, by a random number generator. Then how many ships will be based on the ammount of ships of the attackee. Say he has 1000 ships. Use a random generator with a limit of 5 to 14. Devide that number by 10. Say the number chosen was 11. Devide that by 10, you get 1.1. Next, times that by 1000 (the ammount of ships he has) and that will be X (the ammount of ships, which would be 1100. Easily defendable. even it its 1400 ships at the max, (1.4 x 1000) it would still be defendable. But it would also result in some losses for the top players, and cause them to focus on protecting themselves.

On the other hand, Make 100 planets for the bottem planets to attack. The planets are corded as B:X:X
where B indicates you are moving into a robot sector, and X:X is the actual cords. The fleet of the robot is determined by 1/2 the attackees fleet, and the roids are determined by twice the ammount of the attackees fleet. If more than 2 planets attack a bot planet, simply add the factors about together, to get the final total.

Example:

A planet with 100ships and 7m roids, 4c roids, 2e roids, 4unititated moves wants to attack a robot. The stats for the robot become 50 ships and 14 m roids, 8 c roids, 4e roids, and 8 unit roids. Easy roids right?
Thats good, this way a planet <1million can get stronger faster.


and once the galaxy is within a relative equalibrium say there is less that 1 million defference between the top planet and the 100th planet, the 100th and the 200th, ect the bots can sleep, until the difference in score gets out of hand again, and then you can choose to bring back bots for the lower planets, or bring the attacking bots back online.

Realistic bots wont solve any problems..

[Qdeathstar]

and i didnt say it needed to be every top planet every night, perhaps 10 of the top 50... at different times throughout the day, deterimined by the RNG.

[Heartless]

Qdeathstar, I am absolutely disagreeing with your idea of only bots to take down the top 50 planets, randomly picked ofc.

[mist]

Originally Posted by Qdeathstar

and once the galaxy is within a relative equalibrium say there is less that 1 million defference between the top planet and the 100th planet, the 100th and the 200th, ect the bots can sleep, until the difference in score gets out of hand again, and then you can choose to bring back bots for the lower planets, or bring the attacking bots back online.

so, you want there to be no real difference between every planet in the game? you want to keep those who put a lot of effort in to the game to the same size as those who're useless? how about we just pick a random winner, and have people attack random people? any other suggestions for making the game crap?

Originally Posted by Qdeathstar

Realistic bots wont solve any problems..

if you could have 100K players again, would you? the general concensus seems to be "yes". realistic bots are about as close as it's going to get, therefore it would seem they're the best way to solve the problem. then there's things like them providing more targets so human newbies don't get hit all the time, making blocking less evident as the blocks will form a smaller percentage of the universe etc etc.

-mist

[Ramihyn]

Originally Posted by Qdeathstar

Realistic bots wont solve any problems..

AFAIK the single point most (serious) ppl agree about, is that PA needs more players. Why would 5.000 more players be great, but 5.000 realistic bots behaving like them, not solve any problems?

Care to explain? Or are you against enlarging the planet base in the game?

[Ramihyn]

Originally Posted by Qdeathstar

Example:

A planet with 100ships and 7m roids, 4c roids, 2e roids, 4unititated moves wants to attack a robot. The stats for the robot become 50 ships and 14 m roids, 8 c roids, 4e roids, and 8 unit roids. Easy roids right?
Thats good, this way a planet <1million can get stronger faster.

That would be the perfect base (bottom stage) for pyramid farming! This could even result in exponential pyramid farming (for a while) at the cost of a pod each rock *joy*

[Qdeathstar]

No because you forgot about the roid cap limit< which stats that bigger planets can attack the bottom ones

[Qdeathstar]

Mist, first my keyboard is ****ing up so if it randomly goes into caps, sorry. anyway, there would actualy be a 10million differencen betweeen the top planets and the bottom planets while they are off, when they are turned on, they are only to limit the growth of the big planets, not to decimate them. they might attack 4 top fifty planetets every night, so that means a planet gets 10 days where it wont be a target, to grow... they arent going to kil all you ships, in fact, when they attack, you might even gain score, depending on how well you defend. they would be a limiting factor... not ment to permantly destroy anyone, just limit there grwoth. and, in a perfect world, bots which act real would be good, but i dont know if our coders will ever be up to the challeng, in addiction to coding for new rounds, doing all that for free. another thing is that for real bots to have an effect, there has to be a lot of them, increasing server load. those bots will have the same bandwith needs, ect as a player would, except you et no revenue from them. another thing, is if they are real planets, they will be an alliance, or a power block, simply because they act as real players. my way, they simply limit growth, while allowing the very bottom, who cant be attacked by the top to get the roids they need. my way would also limit top players ability to attack, becasue once in a while they would need to defend. but it wouldnt be every night. i think the main reason you people are against it is because you often finish in the top fifty, and oh no, dont do anything that would limit our ability to show off the fact that we can rap0r anyone... plz ive never finished in the top 500 individually.. so i think i would have some perspective about the view from the bottom players...unlike you.

[mist]

Originally Posted by Qdeathstar

Mist, first my keyboard is ****ing up so if it randomly goes into caps, sorry. anyway, there would actualy be a 10million differencen betweeen the top planets and the bottom planets while they are off

when planets reach several hundred million score (people still that that big right?) that's going to be nothing, relativly. you might want to work on that

Originally Posted by Qdeathstar

in fact, when they attack, you might even gain score, depending on how well you defend.

so, how would they be helping to keep the top players in check?

Originally Posted by Qdeathstar

in a perfect world, bots which act real would be good but i dont know if our coders will ever be up to the challeng, in addiction to coding for new rounds

my major concern, so gotta agree with you there. but assuming bots are going to be added at all i think it's worth putting in the effort to make them decent.

Originally Posted by Qdeathstar

increasing server load.

true, but allegedly not one of the big worries. with pia running on less servers with more players, pa should have the load spare

Originally Posted by Qdeathstar

those bots will have the same bandwith needs, ect as a player would

not true, bots run on one of the pa machines would only need internal bandwidth - which to all intents and purposes is free.

Originally Posted by Qdeathstar

they will be an alliance, or a power block, simply because they act as real players

certainly a possibility. however if the bots are like players they'll have to be like all players, ranging from the 1337 to the inept. while the top end bots may power block, i doubt all of them would - asside from anything else the top end bots would want to attack the lower end bots too much to block with them :P

Originally Posted by Qdeathstar

my way would also limit top players ability to attack, becasue once in a while they would need to defend

if the bots are any good (although not too good - something that i'd imagine would be quite easy to do) they should give the top players some competition. particularly in block wars. i doubt bots would block with human players - rather tricky to conduct diplomacy with them after all. therfore you'd end up with 3 power blocks, rather than just the two. this would help to keep things fresh, as the two power blocks have to worry about who's winning - if humans concentrate on crushing humans the bots might get ahead... if they ally with the bots then odds are they'll have to attack other humans for decent roids soon enough. this competition at the top should keep the game intersting, and hopefully decrease the problems for those who're smaller as the top will have something else to think about. in addition to smaller (crapper) bot planets providing more targets so that the weight is taken off the lower end players a little

Originally Posted by Qdeathstar

i think i would have some perspective about the view from the bottom players...unlike you.

you know where i've finished in every round then, what alliances i've been in etc etc then...

-mist

[Ramihyn]

Originally Posted by Qdeathstar

No because you forgot about the roid cap limit< which stats that bigger planets can attack the bottom ones

I guess you mean the attack limit by score?

That is why you do pyramid farming. Moving rocks from smaller to bigger planets step by step. The bottom stage produces the rocks and that suggestion would be perfect for the bottom stage. If i need 200 E rocks, i would init 50 on my farm which then hits 4 bot planets which each would have 100 E rocks (which makes the roid initing for farms way cheaper or respetively means you can move way more rocks at once from each farm). Then a smaller member hits my farm and thereby the rocks move upwards till they reach the big guy who needed them. Ofc as with any farming, you have to disguise it a bit.

[Ramihyn]

Originally Posted by Qdeathstar

and, in a perfect world, bots which act real would be good, but i dont know if our coders will ever be up to the challeng, in addiction to coding for new rounds, doing all that for free.

Why should the same ppl do that on top of their other (also free) work?

Implement a socket interface to control a single planet and document the interface/commands. Set up a seperate server where ppl can apply for a login/pwd and give every account access to 10 planets. Create a few example sources in perl, VB, C(++) and spread the word. I'm sure you will have some ppl write bot's and you can see which ones are probably useable. The planets could be marked in some way and lets say 300 human players could signup and play own planets too. Ofc cut the startup crap and just give the bots a certain set of res/con, roids, ships to work with as a start - same for players and probably have a ticktime of 1 minute so the development cycles are fast.

You will (hopefully) end up with a handfull of useable scripts/programs, experience and can consider using one or two of them in a real game.

Originally Posted by Qdeathstar

another thing is that for real bots to have an effect, there has to be a lot of them, increasing server load.

Nowadays and for this task, processing time shouldnt be the problem

[Ramihyn]

Originally Posted by Qdeathstar

plz ive never finished in the top 500 individually.. so i think i would have some perspective about the view from the bottom players...unlike you.

Oh please - most of my planets each round, had also low scores

[mist]

Originally Posted by Ramihyn

Why should the same ppl do that on top of their other (also free) work?

Implement a socket interface to control a single planet and document the interface/commands. Set up a seperate server where ppl can apply for a login/pwd and give every account access to 10 planets. Create a few example sources in perl, VB, C(++) and spread the word. I'm sure you will have some ppl write bot's and you can see which ones are probably useable. The planets could be marked in some way and lets say 300 human players could signup and play own planets too. Ofc cut the startup crap and just give the bots a certain set of res/con, roids, ships to work with as a start - same for players and probably have a ticktime of 1 minute so the development cycles are fast.

You will (hopefully) end up with a handfull of useable scripts/programs, experience and can consider using one or two of them in a real game.

Nowadays and for this task, processing time shouldnt be the problem

problems with this cunning plan are severalfold.

firstly: pa has said it'll never have this kind of interface... i keep asking for it so people can use clients
secondly: jolt won't let people have access to servers, remember all the fuss pateam had getting access?
thirdly: rather than creating examples it'd be easier just to write the things
fouthly: 10 bots wouldn't really be enough in order to demonstrate alliance features etc
fifthly: this would involve running one of the bots in a real game - which means trusting people to have no back doors, for the bot to be fair etc etc. can't see that happening.
sixthly: people, evidently, have different ideas about what bots should do

there are probably more problems, but i can't remember them at the momet

-mist

[Qdeathstar]

i dont think this will solve many of PAs problems anyway..

NEXT QUESTION (topic) heh.

[Ramihyn]

Originally Posted by mist

firstly: pa has said it'll never have this kind of interface... i keep asking for it so people can use clients

I know that that is the current point of view, though the reasons for this are partly just wild myths. I know that it is wasted time to discuss these things in regards to PA - i have extensive experience of pointless suggestions regarding PA and know that it is mostly wasted time. However while at one point you realise that you can talk for weeks and months about simple things for PA, you notice that they are actually and in a short timeframe done in other "places".

At the moment the first clone will use bot's to simulate a large universe, you will look totally different at these discussions (and then if PA still exists, i expect them to hastily try to catch up after a heated discussion on these boards). The sad thing is that PA lost the drive to lead the competition a while ago IMHO. The common mindset seems to be to come up with reasons why something cant be done instead of trying to find ways how to do it. Stagnation at its best.

There is more energy wasted to discuss some things here, then it actually needs to just do them.

Originally Posted by mist

secondly: jolt won't let people have access to servers, remember all the fuss pateam had getting access?

Who needs access to servers from Jolt for this?

Either nobody (enough ppl have servers on their own) or only the same person who already has access and uploads new code versions.

Originally Posted by mist

thirdly: rather than creating examples it'd be easier just to write the things

Heheh NO WAY :P

I mean examples are like 5 line code snippets to demonstrate how a function is used and the service accessed. Nothing more.

Originally Posted by mist

fouthly: 10 bots wouldn't really be enough in order to demonstrate alliance features etc

It would take a while till ppl have good bots running for single planets and smaller groups. 10 planets would be enough for 2 attack groups, or 6 attacking planets and 4 defenders/farms/scanners for example.

IMHO ppl should first prove that they can actually control 1 to 5 planets before you allow them to control 50 or 100. Because 50 or 100 can easily kill (and thereby ruin) many other planets by doing stupid things like launching all fleet at one or two targets. That doesnt take any effort and giving everyone 50 to 100 in the beginning, would likely lead to a very disturbed universe where you regularly have a script newbie ruin some planets, defeating a usable test environment.

Originally Posted by mist

fifthly: this would involve running one of the bots in a real game - which means trusting people to have no back doors, for the bot to be fair etc etc. can't see that happening.

Ofc not. Any bot which would be actually used in a commercial or PA environment, would ofc have to be delivered in source and with some descriptions. If someone creates such a dirty code that it isnt possible to see if potential backdoors are in it, i wouldnt use the code. If somebody cant code clean enough so other ppl can understand it, it's probably not worth bothering with anyway.

And in the industry you dont trust ppl - you do NDA's and contracts

Originally Posted by mist

sixthly: people, evidently, have different ideas about what bots should do

Thats partly a reason why a "open" contest where ppl with different ideas participate, everyone can see what works and what doesnt (and ideas like "lets randomly do some random stuff" will soon not be discussed anymore .

You may find that the best bots are from somebody who never talked here or are way less complicated then those we sometimes talk about.