webdiplomacy.net

So one player lost their last SC in Spring and couldn't regain any in Fall, which means they're going to lose their remaining unit and are out of the game. Understandably but annoyingly, said player never comes back to disband their unit. Our phase length is two days. Are the rest of us going to have to wait that long twiddling our thumbs simply for the timer to run out and the game to automatically disband their last unit?

Yes

I'm sure this isn't the first time this has been asked. Is there a summary of previous responses from admins/developers on this (e.g., impossible to implement, or lower priority, etc)?

I looked at the FAQ (Yes, I actually read it!) and didn't find anything there related to this question.

No, this is too advanced for the FAQ, I think. We're updating it however, so we'll keep it in mind.

Foxcastle , I presume what you're asking is whether we can have the site automatically disband a unit when there's no retreat available. I don't know how difficult that is, but it's low priority.

If you're asking whether we can skip a player's retreat/build when they have 0 SC, cannot retreat into a centre and cannot retreat into the same province as another player.. That's hard to program. I believe Jmo once said it would require a major overhaul of the game code.

Would it be possible to allow different time limits for retreat and build phases? Setting them to be shorter than the normal negotiation/order phase would reduce the onerous wait in the case where an eliminated player never bothers to come back.

That would be technically possible. We dislike adding too many options to the game creation screen though (as well making game selection difficult), so in the past we've decided not to offer this option. If there's a lot of demand from the community, we may revisit this.

This is a necessary "feature". I and others who once belonged to the Judge pbem community ser the 72/24/24 deadline scheme as natural... It was not complicated for the newbies of the 90's, and there were thousands of them back then as well.

Claesar wrote: ↑

Mon Sep 10, 2018 6:16 am

If you're asking whether we can skip a player's retreat/build when they have 0 SC, cannot retreat into a centre and cannot retreat into the same province as another player.. That's hard to program. I believe Jmo once said it would require a major overhaul of the game code.

Not a programmer, but I feel like that can't be true. The game code already must determine the valid retreat options to list in the dropdown menu. It should be a simple addition to automatically select 'Disband' when that's the only available option.

Laszlosaurus wrote: ↑

Mon Sep 10, 2018 4:18 pm

Claesar wrote: ↑

Mon Sep 10, 2018 6:16 am

If you're asking whether we can skip a player's retreat/build when they have 0 SC, cannot retreat into a centre and cannot retreat into the same province as another player.. That's hard to program. I believe Jmo once said it would require a major overhaul of the game code.

Not a programmer, but I feel like that can't be true. The game code already must determine the valid retreat options to list in the dropdown menu. It should be a simple addition to automatically select 'Disband' when that's the only available option.

The game already does that. It's not what I said though.

Not a programmer

Your example is a bit like saying "Shouldn't it be easy to turn my car stereo on every time the heated seats are on? We already have a light on the dashboard for heated seats, and the stereo has an on/off button - can't we just connect them together".

I am a programmer:

Unfortunately, webdip's adjudicator was not written with forced options in mind. The part of the code that determines what options you have is totally unrelated to the part of the code that moves turns forward. It knows about "no orders" and "some orders" - but extending it to know about "each order has only one valid entry" would be a fairly major rewrite.

We dislike adding too many options to the game creation screen though

This is a really important point - analysis I ran a while back on finished games suggested that having more options for game types lead to fewer games being played overall. This is possibly supported by the low number of judge games being played, too - but I'm sure there are lots of reasons there.

You could put the option for varying retreat and build phases into the advanced options. By default, length of all phases would be the same, just like before. But if someone bothers to go in advanced options, they'll be able to independently set different phase lengths.

Yes, providing the options wouldn't be the absolute worst. People can probably click through some more options.

I think the main problem lies with finding a new game. You could be looking at something like:
"No in-game messaging, Anonymous players, Draw-Size Scoring, Hidden draw votes, Half-Length Retreats/Builds, Reliability Rating 90%"

What I'm trying to say is, the more narrow you make things (i.e. the more options you provide) the harder it becomes to bring people together to actually play a game.
Imagine you'd like to play a medium-speed game because you can commit to 1 move per day. Makes sense, you come home from work and enter some moves. Ah, but the retreats are 12h/phase, so it doesn't fit in your daily schedule - you can't join that game

I see. Good point!

Terminus Est wrote: ↑

Tue Sep 11, 2018 10:41 pm

I see. Good point!

Your point is very valid as well though. And we've had this request before. At this time we cannot offer a satisfactory solution.

Thanks.

The timer, at least, has ticked down on our game and we can get back to backstabbing.

A_Tin_Can wrote: ↑

Tue Sep 11, 2018 8:48 am

Unfortunately, webdip's adjudicator was not written with forced options in mind. The part of the code that determines what options you have is totally unrelated to the part of the code that moves turns forward. It knows about "no orders" and "some orders" - but extending it to know about "each order has only one valid entry" would be a fairly major rewrite.

Would it be possible to just automatically ready all the moves of players that only have one option?

Mercy wrote: ↑

Wed Sep 12, 2018 6:56 am

A_Tin_Can wrote: ↑

Tue Sep 11, 2018 8:48 am

Unfortunately, webdip's adjudicator was not written with forced options in mind. The part of the code that determines what options you have is totally unrelated to the part of the code that moves turns forward. It knows about "no orders" and "some orders" - but extending it to know about "each order has only one valid entry" would be a fairly major rewrite.

Would it be possible to just automatically ready all the moves of players that only have one option?

If I rephrase the text that ATC wrote to how I understand it:
We have three seperate pieces of code (relevant to the question). One that determines the options a unit has (moving to adjacent provinces etc.), another that realises whether a player has 0 or 1+ options during a phase, and one that determines whether the phase can move on. These three pieces of code do not communicate and to achieve this would require a major overhaul of the engine.

In other words, the website does not know whether you have 1 or 15 options.

tl;dr No, that is not possible. Unless I wildly misunderstand ATC.

Claesar wrote: ↑

Tue Sep 11, 2018 3:04 pm

Yes, providing the options wouldn't be the absolute worst. People can probably click through some more options.

I think the main problem lies with finding a new game. You could be looking at something like:
"No in-game messaging, Anonymous players, Draw-Size Scoring, Hidden draw votes, Half-Length Retreats/Builds, Reliability Rating 90%"

What I'm trying to say is, the more narrow you make things (i.e. the more options you provide) the harder it becomes to bring people together to actually play a game.
Imagine you'd like to play a medium-speed game because you can commit to 1 move per day. Makes sense, you come home from work and enter some moves. Ah, but the retreats are 12h/phase, so it doesn't fit in your daily schedule - you can't join that game

I find this argument dubious.

Today there are 29 (!!) options for game speeds. In certain spots they are really granular (5, 7, 10, 15, 20, 30 minutes; 22 hours, 1 day, 1 day 1 hour, 1 day 2 hours). Then there are 9 other game parameters each with a number of options. You already have a ton of variations, many of which are probably very rarely used in combination.

Perhaps you don't want to add 29 more speeds to accommodate the half-length build/retreat variations for every permutation, but you could add a single boolean advanced option that turns half-length build/retreat phases on or off. You could maybe restrict that option to games greater than 1 day in length.

You could even compromise by adding a half-length build/retreat speed option for the 2 or 3 most popular speeds (1 day - 12 hour build/retreat & 2 days - 1 day build/retreat) upping the speed options from 29 to 31 or 32 which would be close to negligible. You could remove a few rarely used options (22 hours, 1 days, 1 hour and 1 days, 2 hours ?!) to maintain the status quo.

There's lots of solutions here. If having too many options was a major hindrance you wouldn't have 29 game speeds broken up into weird denominations.