Jump to content
Mrloud15

Survival Server General Meeting. Please read for important information regarding the meeting including the time and agenda.

Recommended Posts

Also I think having 2 meetings is silly. I know you're trying to make it more available to people in other timezones, but it's going to lead to very fragmented discussions and ideas. We've had good attendance scheduling for early evening US time on a weekend. Europeans can stay up late (but not unreasonably late), and it'd be morning for aussies, right?

It is a possibility that having 2 meetings could lead to fragmented discussions and ideas. However, as far as I know having 2 meetings has not been tried before, and I think it's worth the attempt to try to hear as many opinions as we can. If it turns out that having 2 meetings is a bad idea, we won't do it next time.

  • Upvote 4

Share this post


Link to post
Share on other sites

Also I think having 2 meetings is silly. I know you're trying to make it more available to people in other timezones, but it's going to lead to very fragmented discussions and ideas. We've had good attendance scheduling for early evening US time on a weekend. Europeans can stay up late (but not unreasonably late), and it'd be morning for aussies, right?

 

Yeah and also why not one meeting at 12:00 AM EDT. Perfect timing for the whole server! Perfect for our Euro friends, perfect for Est and good for West :tongue:

 

12:00 AM EDT

17:00 Euro

9:00 West

 

3 good hours in my opinion.

  • Upvote 1

Share this post


Link to post
Share on other sites

After attending both meetings this would be my winning formula,

 

optimized version of civcraft+unique events/quests+a planned map=a lively server

  • Upvote 3

Share this post


Link to post
Share on other sites

After attending both meetings this would be my winning formula,

 

optimized version of civcraft+unique events/quests+a planned map=a lively server

Sounds good. What are the details of the optimised civcraft you want, and how would each type of player like that, and how is it to moderate? What unique events are you thinking of, that are suitable if you don't get any more players, suitable if you get a lot more, and easy to make arenas for and moderate? Got any quests written up that can be used? What's a planned map? If you mean worldpainter, anything in particular that you've seen that can be done quickly enough for the short revision lengths you want?

  • Upvote 1

Share this post


Link to post
Share on other sites

Optimizing would mean adding the kinds of tweaks discussed in the meeting so that our version of civcraft is more pvp focused and perhaps easier on newboids.

The first kind of events are server wide. Barney mentioned a server where you take damage if you go outside in the day, maybe once a month we could run something big like this for a few days. For smaller weekly events we could just mess around with mob types/numbers/strength. I'd imagine that keeping an eye on minigames and adapting the promising ones into events would keep us well supplied.

The second kind of events relay on groups (or individuals) organising themselves to complete a particular objective first, like slaying the end dragon. These should be going on all the time. Therefore, the intention behind a planned map is to increase the likelihood of you getting caught up in adventures when you're out exploring. Maybe you'll find a ruined city full of goodies at the edge of map, a ravine lined with diamonds or that the numbers of dungeons/temples have been increased and new traps added :3.

To clear some other stuff up:

It'd be impossible to please every player.

Civcraft self-moderates a lot of things.

Server wide or time dependent events work well for most server populations.

A planned map is one with some game design behind it. It might have purposeful layouts, challenging terrain, loot stashes, sunken ships, abandoned castles, secret rails, hidden mob spawners or more exciting kinds of stuff.

Civcraft revs are typically very long

I hope that somewhere in this disorganised response I've covered all your questions.

Share this post


Link to post
Share on other sites

Some feedback on the meeting!

 

It was good to be given a say, and it's especially good to bring everyone together in one place (nearly) to talk about things. Barely anyone is inclined enough to regularly check forum threads, so definitely more people were involved in discussing survival's future than would be otherwise. totemo's civcraft thing sounded pretty great, though it would be bad to rip it off completely.

 

One thing I disliked was the loose management of the meeting. If you listen back to the meetings Torn/Four_Down ran, we were interacting with everyone suggesting things, pointing out problems, trying to promote a conversation and find common ground etc etc. The beginning of this meeting seemed pretty disorganised - the admins would perennially warn people for being off-topic, but did nothing else to encourage people to stay on-topic. It didn't really feel like a dialogue - it felt slightly like we were just putting verbal suggestions in a suggestion box. An admin said something like "any suggestions for minor changes?" and I thought "well yeah, there's been hundreds, where do you want to start?".

 

So my suggestions boil down to

 

1. Try to enumerate each option on the agenda, and don't ask really open-ended questions with no qualification/context.

2. Admins: don't guard your thoughts! It's okay to have opinions on our suggestions, and it's okay to talk sensible realism down to us.

  • Upvote 6

Share this post


Link to post
Share on other sites

In a happy coincidence, the CivCraft plugin was just open-sourced yesterday!  https://github.com/netizen539/civcraft

 

That's not the same CivCraft.  The Shotbow code is more properly referred to as CivilizationCraft and it's inspired by Sid Meier's "Civilization".  The subreddit for that is r/CivilizationCraft/.

 

The CivCraft that I have been describing is an agglomeration of various mods, as described in this r/CivCraft/ post: http://www.reddit.com/r/Civcraft/comments/1emr0o/welcome_to_civcraft_20_a_basic_guide_to_the_mods/

Edited by totemo
  • Upvote 2

Share this post


Link to post
Share on other sites

Going a little bit more along the topic of CivCraft, I'm generally up to try it as it doesn't look bad at all. 
However I'm really quite against the PrisonPearl feature - I don't think there's any need for it and frankly it could get out of control. 

Share this post


Link to post
Share on other sites

However I'm really quite against the PrisonPearl feature - I don't think there's any need for it and frankly it could get out of control. 

 

Most likely it would get out of control at times.  There will be people who imprison others unjustly for fun.  There may well be a Genghis Khan or a literally Hitler.  And then there will be subreddit dramas, impassioned pleas for justice, vault raiding parties, rescues, reprisals, bounties, contracts, politicking, corruption, duplicity, manipulation and other niceties of the real world.    

 

For game balance, there needs to be a non-trivial cost of dying.  Just losing your inventory is not enough, because you can easily reduce that to a diamond pickaxe, a gold axe for chests and some baked potatoes, and then go raiding people's vaults or griefing their farms.  Some servers impose a death ban, which is similar in principle to pearling.  However, a death ban is a blunt instrument: how do you adjust it to fit the severity of the crime?  

 

Pearling is a more subtle mechanism because it allows the players to develop their own process for dealing justice, just as people do in the real world.  An upper limit on pearl durations may be in order.  As I understand it, there's also a one-alt allowance built in, so don't judge in terms of an MCBouncer-style ban.  It's a different beast and needs very careful consideration.

Edited by totemo
  • Upvote 3

Share this post


Link to post
Share on other sites

If you listen back to the meetings Torn/Four_Down ran, we were interacting with everyone suggesting things, pointing out problems, trying to promote a conversation and find common ground etc etc. The beginning of this meeting seemed pretty disorganised - the admins would perennially warn people for being off-topic, but did nothing else to encourage people to stay on-topic.

 

I didn't hear the start of this round, I entered the first one about 40 mins late (20 mins before Torn). I saw several people fighting any common ground in these meetings, thats why they were so off-topic. The previous meetings had dissenters, who were polite and quiet and thats why it didn't appear to get off topic. The dissenters in this meeting were not polite and quiet, and it took Torn something like two hours to get kicked because the admins were wanting to be inclusive (people still shouted at the admins for not being inclusive). I was probably the loudest dissenter and while I was not entirely polite I said about one rude thing per hour. I repeated one dissenting idea about five times, and once more here, only where they were an answer to the question of the moment.

 

Pearl plugins, death bans, or whatever the game allows for retribution like base camping - It depends on the person doing it not how they do it. The culture of the server means that whatever sufficient tool you give players will be abused to an extent you won't be happy with. You're coming up with ideas based on how they are used rationally, and deciding they aren't suitable because you can predict how they'll be used irrationally.

  • Upvote 1

Share this post


Link to post
Share on other sites

Pearl plugins, death bans, or whatever the game allows for retribution like base camping - It depends on the person doing it not how they do it. The culture of the server means that whatever sufficient tool you give players will be abused to an extent you won't be happy with. You're coming up with ideas based on how they are used rationally, and deciding they aren't suitable because you can predict how they'll be used irrationally.

 

This is very much something that we've been dealing with. For instance, in the meetings it was suggested we pursue some of the ideas used on the civcraft server, one of which is the citadel plugin which allows players to reinforce their builds and chests. This naturally assumes that griefing is allowed for gameplay purposes, but it allows people to grief just to cause harm and frustration. Civcraft has prisonpearl to try and deal with griefers, but I am 100% certain that it would be abused to deprive others of enjoyment of the server and the game.

 

As one possible alternative, the sadmins have been kicking around the idea of a bounty plugin - so you place a bounty on those who have griefed you and players have an incentive to kill that griefer. A bonus is that the plugin would be used to encourage PvP even if there has been no "crime" or grief comitted. We didn't want to use one that depends on an economy plugin, but it turns out that such a non-economy plugin is difficult to find - so I decided we could write one and I'd take steps to learn commandhelper and implement it through that. It seemed straightforward - allow the player to put the bounty in a gui inventory, when the target is killed then whoever caused the most damage in the previous X seconds gains access to the reward. But each method I thought of to fulfill this function came along with the question and answer of how it would be abused.

 

  • You could just use an existing plugin like LWC to enable delivery of bounties, but without a way to confirm for certain who killed the target players would likely deceive the paying player. Alternatively, the person who put the bounty on the target might not uphold their end of the deal. It's also somewhat complicated/inconvenient to arrange a bounty in this manner to deter something that might be a regular problem.
  • If you make it so that the bounty is deposited directly in their inventory upon success, then they might instantly be killed by the target's buddies and the bounty goes to the wrong party in the end.
  • If you make it so the bounty hunter can wait to collect the bounty when they are in a safe location, they might abuse it as safe storage of valuables - defeating the purpose of allowing chests to be raided.
  • So you could set a timer on it - if they don't claim it within a day or an hour or something, then the bounty vanishes. But then, the question came of using the plugin itself as a method of instant item delivery across long distances. If I want to send bob on the other side of the map some gold, I put a bounty on joe using the gold as the reward. Bob goes next door to joe's house, kills him, and gets the delivery. No need to travel across dangerous roads.
  • So we make it so that the bounty has to be collected at a specific location, rather than it being deposited in the player's inventory, but where are these locations, and wouldn't players just set up base adjacent to them and use them as mail depots anyway?

 

etc. etc.

 

It's part of game design, I suppose. It's frustrating because it would be a trivial thing to make and implement in its basic form, but because of it being abused it has to become more cumbersome for non-abusing players. Lots of things that have been proposed in the past have been denied because of abuse potential. Hell, PvE split off from survival because of abuse of no-pvp zones.

 

It's definitely challenging to try to come up with a game setup that's fun for everyone, but won't be used by dicks to screw other people over or make the game unfair.

  • Upvote 1

Share this post


Link to post
Share on other sites

As one possible alternative, the sadmins have been kicking around the idea of a bounty plugin - so you place a bounty on those who have griefed you and players have an incentive to kill that griefer. A bonus is that the plugin would be used to encourage PvP even if there has been no "crime" or grief comitted. We didn't want to use one that depends on an economy plugin, but it turns out that such a non-economy plugin is difficult to find - so I decided we could write one and I'd take steps to learn commandhelper and implement it through that. It seemed straightforward - allow the player to put the bounty in a gui inventory, when the target is killed then whoever caused the most damage in the previous X seconds gains access to the reward. But each method I thought of to fulfill this function came along with the question and answer of how it would be abused.

 

  • You could just use an existing plugin like LWC to enable delivery of bounties, but without a way to confirm for certain who killed the target players would likely deceive the paying player. Alternatively, the person who put the bounty on the target might not uphold their end of the deal. It's also somewhat complicated/inconvenient to arrange a bounty in this manner to deter something that might be a regular problem.
  • If you make it so that the bounty is deposited directly in their inventory upon success, then they might instantly be killed by the target's buddies and the bounty goes to the wrong party in the end.
  • If you make it so the bounty hunter can wait to collect the bounty when they are in a safe location, they might abuse it as safe storage of valuables - defeating the purpose of allowing chests to be raided.
  • So you could set a timer on it - if they don't claim it within a day or an hour or something, then the bounty vanishes. But then, the question came of using the plugin itself as a method of instant item delivery across long distances. If I want to send bob on the other side of the map some gold, I put a bounty on joe using the gold as the reward. Bob goes next door to joe's house, kills him, and gets the delivery. No need to travel across dangerous roads.
  • So we make it so that the bounty has to be collected at a specific location, rather than it being deposited in the player's inventory, but where are these locations, and wouldn't players just set up base adjacent to them and use them as mail depots anyway?

 

etc. etc.

 

 

Why not set up some sort of timer, where the reward is put into the bounty "winners" inventory after a certain amount of time. That gives players the chance to flee the scene, and if they were to die, the timer wouldn't stop. That way even if they die after 30 seconds, in the next 30 seconds they would receive their award. An argument to this could be players camping the player, but if they have a solid minute to get away, that's on the players.

 

The biggest problem I could see with this plug in is clan mates killing each other to get the bounty. Man this whole "players abusing the system" thing is really annoying.

  • Upvote 1

Share this post


Link to post
Share on other sites

Why not set up some sort of timer, where the reward is put into the bounty "winners" inventory after a certain amount of time. That gives players the chance to flee the scene, and if they were to die, the timer wouldn't stop. That way even if they die after 30 seconds, in the next 30 seconds they would receive their award. An argument to this could be players camping the player, but if they have a solid minute to get away, that's on the players.

 

The biggest problem I could see with this plug in is clan mates killing each other to get the bounty. Man this whole "players abusing the system" thing is really annoying.

 

Initially I was thinking about it depositing items into the hunter's inventory directly, but my concern was that their inventory might be full at that time (or the aforementioned death immediately after target execution). My thought had been to give them an option to claim the bounty in the allotted time, but I see what you're saying - it would be a set timer that will try to put the reward in their inventory on timeout regardless of their interaction with it. I think that's a workable solution too.

 

But both methods are still susceptible to the system being used to safely deliver resources over long distances. Maybe it can have some kind of cap on how much you can put on a bounty, and some kind of cooldown on how often you can use it. hmm. But yeah, clanmates/friends of the target collecting bounties would not really be in the spirit of it. So what I'm thinking is that there should probably be some penalty beyond just death...

 

BUT the problem there becomes the same sort of one that exists with prisonpearl. Make it too powerful of a tool to be used to abuse or harass beyond what is sportsman-like, and it will most likely be abused. Totemo was making the point that providing mechanisms that influence gameplay or conduct is much more effective than trying to socially engineer behavior, so if we made something that made a good effort at being balanced and coupled that with moderation policies against abuse... it still would likely be an imperfect setup.

 

Still ideas to be ironed out if that citadel plugin is to be used - which is definitely not a set in stone plan btw.

 

edit: oh i know, maybe you can optionally define a blacklist along with your bounty. So if you know that a target has buddies who will basically share the reward with them, then you can add their names maybe so that they can't claim it? hmm...

Share this post


Link to post
Share on other sites

Initially I was thinking about it depositing items into the hunter's inventory directly, but my concern was that their inventory might be full at that time (or the aforementioned death immediately after target execution). My thought had been to give them an option to claim the bounty in the allotted time, but I see what you're saying - it would be a set timer that will try to put the reward in their inventory on timeout regardless of their interaction with it. I think that's a workable solution too.

 

But both methods are still susceptible to the system being used to safely deliver resources over long distances. Maybe it can have some kind of cap on how much you can put on a bounty, and some kind of cooldown on how often you can use it. hmm. But yeah, clanmates/friends of the target collecting bounties would not really be in the spirit of it. So what I'm thinking is that there should probably be some penalty beyond just death...

 

BUT the problem there becomes the same sort of one that exists with prisonpearl. Make it too powerful of a tool to be used to abuse or harass beyond what is sportsman-like, and it will most likely be abused. Totemo was making the point that providing mechanisms that influence gameplay or conduct is much more effective than trying to socially engineer behavior, so if we made something that made a good effort at being balanced and coupled that with moderation policies against abuse... it still would likely be an imperfect setup.

 

Still ideas to be ironed out if that citadel plugin is to be used - which is definitely not a set in stone plan btw.

 

edit: oh i know, maybe you can optionally define a blacklist along with your bounty. So if you know that a target has buddies who will basically share the reward with them, then you can add their names maybe so that they can't claim it? hmm...

 

I love the idea of a bounty system. On paper it adds to pvp and the thought process of the players as in "do I really want to grief this and get a target on my back..". I don't think it would be too powerful of a tool, because it would be much harder to get things that players would want, such as diamonds. It seems that prisonpearl would just require enderpearls. I actually feel it would be harder, especially if players are in diamond, to get a bounty worth going after. Why would I fight a guy in diamond for the reward of just 2 diamonds? And if you have enough diamonds, why wouldn't you just make a suit and go after the griefer yourself? The more I think about it, the more crazy that I would pay diamonds just to see someone die once. I feel without an economy based system, it wouldn't really work.In an economy based system, maybe the player wouldn't have to have tons of diamonds to be considered "rich" (for example a shop owner that sells basic gear for 'gold'). 

Share this post


Link to post
Share on other sites

You're coming up with ideas based on how they are used rationally, and deciding they aren't suitable because you can predict how they'll be used irrationally.

 

No, Toby this is all practical experience of the Civ server.  Remember, we have a working example to study and draw conclusions from.  All of the ins and outs of pearling and rescues are well documented on their subreddit.  You can go in game or on their mumble server and talk to people.  There is no need to infer anything.  You can simply ask people what has happened for the past year or more because those players are still around.

 

 

The culture of the server means that whatever sufficient tool you give players will be abused to an extent you won't be happy with.

 

As a former admin and moderator, I am well aware that players will always attempt to push the boundaries.  I'll tell you one other thing: it is very wearing to have to be somebody else's self control.  That's what the moderation task is.  Being somebody else's self control.  You've come on here with dire and non-specific warnings that if players have to take responsibility for themselves then all hell will break loose.  My observation of Civ and what the old timers say about it is that, initially, there will be chaos, and then gradually the system will move into balance of its own accord.  I know this because great towns and railways exist; because there are 120+ players on at peak time on a map that's over a year old; because those players keep coming back.  None of that would exist if equilibrium had not been reached.

 

All of the drama surrounding pearling drives forward a game narrative.  The bad guys walk in (sometimes the bad guys are the alts of the good guys), make a mess and unite other players in a common goal of justice, restoration and retribution.  Entire towns go on vault raiding parties because someone kidnapped the mayor or whatnot.  Parts are played, there are winners and losers.  Survivors get to tell stories.

 

So, I am in favour of a game mechanism that is already demonstrably working (pearling) rather than inventing some other way of balancing the cost of grief.  I will note one further point about balancing PrisonPearl: the plugin has built-in provision for a daily cost to imprisoning another player.  On Civ, the imprisonment cost is 8 coal per day per pearl.  Just a reminder: ores are relatively scarce on that map, by design.  The cost is configurable: both in the material type and the quantity so by adjusting the the cost the admins can indirectly influence the "justice" system and its abusers.

Edited by totemo
  • Upvote 4

Share this post


Link to post
Share on other sites

As one possible alternative, the sadmins have been kicking around the idea of a bounty plugin - so you place a bounty on those who have griefed you and players have an incentive to kill that griefer.

 

I don't get where the griefing deterrent comes in.  Dying is low cost in vanilla Minecraft.  You drop what you're carrying and respawn at your bed.  (Read: secret base.)  That is the root of the problem and any mechanism that you come up with that does not address that will fail.

 

If you only want proof that somebody killed somebody else, have players drop heads (for example), with lore on them that states the providence (i.e. who did the killing).  These are inherently valuable items (worth the bounty) so of course people will trade them and that means that the killer does not necessarily hold the item, but the lore will tell the story.  But if you do this, then the bad guy can farm these proof items himself with his own alt.  And this should tell you that the value of a bounty hinges on whether you trust the bounty hunter, and whether the bad guy is actually punished in some meaningful way, by being removed from the game for a period of time.

Edited by totemo
  • Upvote 1

Share this post


Link to post
Share on other sites

Necromancing this thread to point out that Dumbo's creative meeting was really good, and a blueprint to be followed in future survival meetings.

 

Dumbo + others took the time to examine and respond to peoples ideas, and gave people specific questions rather than "what do you think about x?". It felt like a meeting rather than a survey.

 

Props to Dumbo and hopefully the next survival meeting will be improved.

  • Upvote 4

Share this post


Link to post
Share on other sites

Bump. Can we just get a yes or no already instead of leaving threads to stagnate in future? 

Share this post


Link to post
Share on other sites

Bump. Can we just get a yes or no already instead of leaving threads to stagnate in future? 

 

Sorry, you're no longer a contributing member of our community ;D

 

But really, would be nice to talk about S.. and the fact that it rarely has over 4 players on...

  • Upvote 2

Share this post


Link to post
Share on other sites

Bump. Can we just get a yes or no already instead of leaving threads to stagnate in future? 

 

Sure. A yes or no to what question?

Share this post


Link to post
Share on other sites

Sure. A yes or no to what question?

Argh fuck, I just realized I goofed and posted in the wrong thread, feel free to laugh at me. 

Share this post


Link to post
Share on other sites
Closing.
 
Here are the recordings of the meetings.
 

 
 
 

 
 
 
Thanks to slide23 for recording and uploading the second meeting, and kitcatbar for providing a copy of the mumble chat.

 

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...