The face of information

What do "Big Mamie", Wooden Ships and Iron Men (the board game), and Eve Online fleet combat reify?   Lag, but not in the way you might think.

A few weeks ago I had the excellent fortune of staying overnight on the USS Massachusetts (BB-59, at her museum berth at Battleship Cove). Her keel was laid in 1939, launched on September 1941, commissioned on May 1942, decomissioned on March 1947.  She served with distinction in WWII.  Yet, she reflected a design that was limited by the London Naval Treaty of 1936 - the South Dakota class of battleship was quickly eclipsed by the Iowa class battleship later in the war.  Ultimately, of course, all BBs were then eclipsed by the aircraft carrier.

A striking point involves the considerable distance of time spanning concept to design (to the original strictures of the Second London Naval Treaty), design modification (treaty politicking), commissioning, and subsequent modifications made during the war in response to changing naval warfare environment (radar, more anti-aircraft).

I used to play Wooden Ships and Iron Men - the classic Avalon Hill naval simulation board wargame (18th and 19th century sailing vessels).  The rules were simple, but the effects of simulated wind in confounding the best laid formations and plan made a lasting impression on me.  I'm not sure, however, whether Nelson had the difficulties I had in maintaining formation in large fleets: it seemed to always degenerate into a furball of broken sail and a frenzy of ship boardings.  My difficulties I attributed to some gap separating simulation (rules) and a more continuous reality.

I vaguely recall an optional rule set for simulating command lag in large fleets of sailing ships.  These rules sought to simulate semaphor communications in fleets of those times.  From the admiral's ship the sailing commands would be issued, however it would take time for those orders to disseminate throughout the fleet.  The visual effect of propagating course change amongst all the ships was mesmerizing.  A critical element of strategy here (which I never mastered) was being able to successfully anticipate the lag effects and avoid entangling one's fleet while at the same time sheparding them to some location to achieve an advantage over an opponent. 

"Big Mamie's" design reflected pre-war warfare assumptions (battleships) and was shaped by treaty commitments (Second London Naval Treaty).  Semaphor simulation in Wooden Ships and Iron Men saw information as more dynamic - a form eroded from successive waves of process and effect.  For example, a simulated fleet would be more often than not be operating in a conflicted state (e.g. 1/2 operating under the former orders, and 1/2 operating under the latest orders).  Yet even here, as complicated as this was, one could still visualize the "shape" of the information flow as it moves across the map.

I'm sketching a longer piece on how I perceive information and strategy dynamically interrelate in Eve-Online. To illustrate some of the complexity.  Consider a not uncommon example: a 250+ person (blue team only) raid that stretches multiple systems, 4+ hours, involving (at least) four squads, 5+ text channels, voice channels, multiple (and staged) sub-goals, and yes consider the gradual (and dramatic) degeneration of communications discipline (as evening drags on) on different channels at different rates.  Yet, it somehow comes together (sorta).

This question begs the next logical step stretching from Big Mamie, to Wooden Ships and Iron Men, to Eve-Online now, to the future MMOG.

At what point will games, strategy, and information relationships with players break down?  At what point will players run stark naked mad - or optimistically, at what point will games need to face "better information" over "more information."  What will that face look like?

If you have early thoughts on this subject enter them here.

Comments

[Peter]

1.

RTS.

"If my soldiers were to begin to think, not one would remain in the ranks."-Frederick the Great

I firmly believe the future of MMO information flow will be a streamlined interface built for maximum efficiency of command, looking like a real-time strategy game.

Autonomy in the individual would be in how well they can execute orders.

The basic enginery is there already, and Blizzard's presence in the field makes me think it won't be very long until this happens.

[Ben Steeples]

2.

Ahh, the art of war. (pun, phrase coinage, and references intended)

Taking Eve Online as your example. Six months ago, all fleets were led by a single pilot as far as game mechanics go. 'Off' the commander of that fleet (by offensive action or through some unfortunate connection problem) and the fleet would most likely die.
It was not uncommon to see an overwhelming force literally annihilated by a smaller, more organised, force when this happened. The cries of despair are just as real, even if it is a virtual loss.

However, the recent expansion Revelations has allowed for a hierarchical fleet structure; with squads, wings, and respective commanders. This has imposed a more militaristic structure on fleets, much like those which seem to naturally form within alliances/guilds of MMOG's. Fleet commanders are left to macro-manage the 250+ pilots; whilst wing and squad commanders are left to look after much smaller numbers. It enables many more players to feel as if they can 'change the game', as they are put in a position of [relative] power. And if the head of the fleet is suddenly lost, one of the many sub-commanders is able to be promoted into his place.

So, although the lines of communication become longer through increased player numbers, and complexity of the system; the natural militaristic style that evolves within online gaming seems to adapt.

Incidentally, I used to play Man O' War, which your wikipedia link refers to. That included none of the intricacies of communications lag and tactical warfare, just get your ship in range and shoot. So it seems that in this case, online gaming has improved on the table-top version.

[Ace Albion]

3.

All the people I talk to about EVE who know their onions, say that the imposed fleet structure is basically ignored, that commanders still command the whole fleet directly on teamspeak, and whomever has the most advantageous (in-game) leadership skills (aka gang-bonuses) are appointed as group/squad/etc leaders according to what gives the best boost to ship stats.

In summary, business as usual.

[Lacero]

4.

I think that's partly because the people who have the skill and knowledge to run a fleet only know how to do it in the old way. It's also not obvious that the new way is any better, so nobody seems to be trying to change. At least not publicly.

[Ben Steeples]

5.

Using the old fleet structure has it's advantages; namely much less client lag.

However, the onion owners I know prefer to use the fleet system. I know that one major alliance that been in the news recently, has wings specifically for support, and wings specifically for the objective. Thus although a single pilot is 'in charge', the wing commanders have say on what their group should be doing at any one time.

Holding my own onions for a moment, my alliance has had tremendous success in locking down 3-4 systems using a single fleet; allowing individual wings to go off to each system, whilst still having an experienced leader 'in charge' should anything go wrong.

I agree though, one person remains ultimately in charge either way. However the fleet system gives more autonomy to those who wish to sub-group their fleet; particularly as many alliances have more than one person who is capable of 'command'. It's all down to macromanagement of fleets, knowing where you are in relation to the enemy and the objective; versus micromanagement of what half a dozen pilots are doing at any one particular moment.

As you say though. There is the overwhelming advantage to just stick the most skilled (not experienced) pilots into the important roles and leave it at that, due to the bonuses they might offer.

[Tripp]

6.

Interesting topic. My guess is that the degree to which people take a game/a raid/an event and group communication seriously, they will put energy into making communication a priority. Common sense, I think.

Will it evolve and get easier in time as technologies and structures improve? Sure. But the ceiling will always be how seriously people care to be.

Think of contemporary military and the importance put on communications. Lots of energy and research gets put into making this better, lots of training goes into getting people to do it better.

In games, where people are playing for pleasure, it's rare to see much organization/communication for any length of time. This is true in an action (like a raid) or through things like guild websites. Most WoW players, I think, prefer just to PLAY the game, not talk/write/stratagize about it. Where groups do "take it seriously" they are seen by most others as "hard core" or even "fanatics."

Good TECHNOLOGY for communication in MMOs exists. Good discipline in using it is very rare.

[Erillion]

7.

There is a significant difference between the Age of Ironclads and the Age of EVE. Its the use of Teamspeak and more or less instant out-of-game communication of fleet commander orders down to the last pilot.

Yes, EVE has the new wing and squadron command options and I have acted as wing and squadron commander myself. But its mostly used to group support, capital, sniper, recon and close-combat ships into appropriate units BEFORE combat starts. To place them were they are most effective !! before !! the fighting starts. To jump as a unit to the system were the battle will take place.

Fleet commander has multiple com channels for his cloaking recon ships, his capital ships and the actual fighting fleet. But when the enemy fleet jumps in (or an enemy Titan is about to Doomsday your fleet with a screen blanking blast) he does not use the chain of command to transfer his commands down to the chain. He gives his orders via Teamspeak, Ventrillo or EVE in-game-voice-chat DIRECTLY to the whole fleet. And if he did his homework he is usually only calling primary and secondary targets for concentrated-fire-takedown, as the fleet better be in optimal positions BEFORE combat starts, as lag and fog-of-war will make maneuvering difficult once the volleys start flying back and forth. Thats why TS hacking and faking FC commands is so much frowned upon in EVE as one of the worst kinds of metagaming. And unfortunately its used a lot in EVE.

About the use of FC orders that appear in "Overview": Yes, it would be a non-fakable way of giving out commands. But due to the layout of the "Overview" section there is hardly enough space to see and understand it, as the right side of the screen in a fleet battle has to handle the sensor data of enemy ships (LOTS of them, now try to find the primary target :-), the drone data from your ship (if any), the fleet, the wing, the squadron channel, and THEN the command channel ... try to find that. Especially when the overview does not update properly due to lag , as it does so often. Or is bugged as is also often the case.

And here you see the real difference between successful corporations in EVE and mediocre ones. The good ones have strict com discipline. This results in a more efficient way of fighting and better results in PvP and fleet battles.

Like in a real life military force : Discipline is winning battles. Another fact why I consider EVE to be closest to real life conflicts of all the MMORPGs i know.

Have fun

Erillion

PS:

Make that
a) logistics
b) information
c) discipline
d) experience
that is winning battles.

[Nicolas Badano]

8.

I believe that the first step for a proper analysis would be to separate information from communication. What do we mean by both of them?

If information is raw data, hard facts (numbers and the like), then I think that Blizzard have hit the right spot by allowing add-ons. At the end of the day, it´s a personal decision wether you want your screen filled with bars and statistics or not. This allows for two (or more) games into one: the more "casual", play-for-fun approach, where you don´t need to care about anything but clicking and selecting the appropiate spell, and the "power-gamer" style, goal-driven, and very instrumental. Truth is that for different people the game means different things, and by allowing players to customize what they want to know, you acknowledge for all these different meanings. Beside, add-ons allow a vast in-between the "casual" and "power-gamer" approaches, making it very easy on the player to play the game as he or she likes. As long as you don´t mess it up by over-doing it (I´m looking at you, Second Life), player customization is a very magical word, and I strongly believe that WOW and EVE are among the best examples of it. I would even say that this is the face of "better information".

Now, communication is, as I see it, a different thing. I am not regarding it as a way of exchanging information between individuals, but as a way of constructing meanings and building worlds. As I think that this is off-topic for this thread, I will just set it aside for the moment and not talk about communication, although I do feel that understanding its meaning in MMOs is a key for future developments.

Finally, regarding battle organization and the good ol´ Sun Tzu, the assumption here is (if I understand correctly) that we could, in some way, come up with an "ideal" kind of organizational model, for the players to fight better. We are starting from the the hypothesis that the current one is flawed, or could be improved upon. The fact that, as Ace Albion says, the new, stiff hierarchic model included in Revelations has not replaced the traditional, bread-and-butter teamspeak, speaks of a player-built model that works better than any possible "packaged" solution that could be provided by the developers. Actually, it allows personal characteristics like discipline and experience to play a big role in the outcome of the battle, making it less game-y and much more real. Acknowledging that, I could suggest thinking not about new models, but about new and improved tools to communicate (less lag, easier switching between groups, clearer voice, etc) inspired in the use players make of them. The key for a successful development is not providing the solutions, but the tools for the players to craft their own solutions for any given situation.

Well, that´s it for now... I got really dizzy, you know. :D

[Peter Weijnen]

9.

"The key for a successful development is not providing the solutions, but the tools for the players to craft their own solutions for any given situation"

I'd say that that is entirely correct. Instead of creating a tool that limits the ability, developers like CCP tend to somehow always end up providing a tool that supports player groups from developing new abilities and ways to communicate or coordinate.

Some other tools (tagging enemies on the overview for one, once hailed as a welcome improvement), end up being totally ignored. The incorporated voice chat CCP has now lauched stands a fair chance of ending in the same way, because client stability and communications stability cannot be tied together due to the neccesity to coordinate effectively in real time even when not directly playing.

Yesterday evening saw the biggest battle in the history of Eve, and maybe even the largest in MMO history in total, with up to about 2 thousand players involved in multiple battles raging at the same time.

And with it, the relatively new revelations-patch fleet communications/organisations system was used in new ways again. One thing (and this is the clincher that stops things from breaking down in this case) that one immediately notices about recent developments in tactical organisation and communications in Eve Online is the increased tendency to actively compartimentalise information and communications. In the past, this used to mostly be counterproductive and used seldomly during combat. Now that we are talking thousands instead of merely a hundred or so involved, this is rapidly changing into a situation where units get formed, and communications are layered rather then only disciplined but still a simple system of having everyone in one teamspeak channel.

[Ben Steeples]

10.

Actually, the battle that involved 2,000 players was a bit moot. CCP had to place a limit of 700 pilots in the objective system, leaving the remaining 1,300 or so just hanging around.

Still, it's very cool to see that many people online at once with the same objective in mind.

[Peter Weijnen]

11.

That's actually not correct Ben. There were never more then 450 or so pilots in it. It's irrelevant however, there were a total of at least 2000 pilots involved in one large encounter between two main camps, and everyone in it was in detailed communications within their side using multiple media.

The encounter ddoesn't matter, the scale and complexity of the cummunications involved in it and it's wake do.

[Endie]

12.

One interesting implementation of lag in an online game is in the version of the Traveller strategic fleet combat game "Fifth Frontier War" run on the Traveller mailing list.

In the space-based game Traveller, information can only travel with starships at a rate of one "jump" (of one to six hexes) per week. Where the board-game allows everyone perfect knowledge of immediate moves, that is lost.

So on the TML, one person acts as referee, and calculates when each player's movement or attack is known by the opposing player. Thus, the game becomes a more "realistic" one, with each player operating with the knowledge that the unit it sees in one place could in fact be in an increasingly broad circle of uncertainty. Scouting in particular becomes vital, as does the use of other players as sub-admirals, able to react more quickly from nearer the front lines.

[Mike Bond]

13.

I'd imagine that the future of information presentation to gamers diverges into several clusters of games as (in each world) the game

designers and their inhabitants together discover and negotiate whether their activities are closer to war or to sport

on a sliding scale. Intentions of the game designers re war/sport may be one thing, but emergent culture of competition may be just

as important.

The limits placed on information presentation will (I imagine) be drawn from the fundamental nature of the game. An example

separation of war/sport characteristics within a single game is shown below (taken from my studies of online FPS tactical shooters,
such as Joint Operations, BF2 and Armed Assault):

SPORT-LIKE
* rules acceptable play
* attention to fairness
* scheduling of events
* stateless, non-persistent games
* amateur participants
* spies and traitors unwelcome

WAR-LIKE
* no global information
* practiced over varied terrain, not a pitch or a course
* victory often more important than the means
* limited supervision and structure: no referee, no half time
* innovation generally encouraged
* rules don’t effect establishment of game physics

The sport-like camp will eventually conclude that information presentation should be normalised, and either adopt a fixed UI,

or (as World of Wacraft has recently started to do) impose constraints on scripting in order to limit UI design to a restricted

subset of the overall (infinite) UI space.

Incidentally, it is very much the case in online tactical shooters that visual display is never fiddled with. Instead incredible

pains are taken to maximise the inter-person information bandwidth and latency of teamspeak. This includes general TS discipline,

specific language subsets for reporting, experimental comparisons of broadcast commanding (commander to all) versus relay commanding

(commander to squad leaders), and even dedicated squad radio men (so as to filter command level radio messages for the local squad

leaders). But all of this know-how you carry in your head, not in UI code.

The sport-like camp will frown on other information collection and presentation that breaks their ground rules -- such as presenting

intel from spies, or collecting attendance information in order to schedule events when key enemy players may not be available.

Eventually, as the gameplay technology stabilises and matures, so will the information presentation, and then it can be standardised

and absorbed into the rules of the sport.

Meanwhile, the war-like camp will not resist the ever-increasing tide of information, in itself, but become smarter about

filtering it and routing it to the right people. This will become less and less about the HUD, and more about procedures and

discipline. But at the moment, virtual worlds like to maintain the illusion that you can "be anyone" and "do anything". I think as

online war gets more sophisticated, it will become more and more stark humans carry many very important different traits and
abilities that suit them to particular roles in the overall war machine. None of these can be replaced by stat bonuses, gear or

template heirarchical structures for war management.

And I would tentatively predict that we will see the information presentation for war-like gamers to also (just like the sport camp)

eventually plateau and normalise. But it will do this around the typical levels of complexity of command and control that we see in

the real life Army -- representing the best that humans can achieve in presenting information and making decisions on it. This

stabilisation would of course oscillate a bit (maybe be like a chaotic "strange attractor"), but not make any fundamental advances.

Unless humans themselves advance in information processing capability through HCI advances, and generations of growing up and living

in virtual worlds, but thats another story...

Mike

[Mike Bond]

14.

oops sorry I totally hashed up the formatting... it looked OK in the preview :-/ If an admin sees it, delete and I'll repost.