The Future... Lessons from Rome in mod development and team management

[Epistolary Richard]

From this thread:
https://forums.totalwar.org/vb/showthread.php?t=70145

Lessons from Rome in mod development and team management

The RTW mod community has been something of which ancient Rome itself would have been proud. There's been scandal, strife, civil war, sabotage and smut. Glorious achievements and ignominious failures. Mighty champions and egregious villains. And somewhere amongst it all quite a few decent mods as well!

Not only did we see a larger community with Rome, we also saw modding teams themselves get larger as more and more content was generated in order to provide complete total conversions.

Given then, that the objectives of our community must be a) To release great mods, and b) To have fun - what then are the lessons from the RTW mod community that we can offer to those who will attempt to form teams and make mods for M2TW?

For as George Santayana might have said, "Those who cannot learn from history are doomed to never manage to release their mod." The floor is open for your words of wisdom.

[Epistolary Richard]

Here are a few things modders have had to say on this topic in the past:

Step 0: there is no point starting a mod if you have no idea about modding and have no motivation to learn along the way.

At the end of the day, you want a mod with good gameplay. My personal opinion is get a 'skeleton' of the mod working. Basically a map and factions, and some units. They don't have to be balanced. Then once you've done that, you can play around to your hearts content with stats and stuff to make the game fun.

If I knew this two years ago I never would have started Fourth Age: Total War. I got lucky and found the right people to run the project...

The sole reason that started this whole fiasco *laugh* is my desire to see Gondorian soldiers in the game, appearing just like in the movies. I found the idea to be very exciting and, most importantly, quite achievable. But instead I kept making more and more models and before long it’s enough for a decent sized mod.

Creating Your Team
Accept everyone who applies, whatever their experience. Every contribution has the potential to be useful eventually. This does not mean include every contribution that is made. A crappy model is still a crappy model, and it’s your job as leader to not include it into the mod. But less experienced members can be taught, and in time grow to be very valuable.

http://www.twcenter.net/forums/showthread.php?t=52350

Lets start by looking at how to assemble a team. The guiding rule here is to keep it small. Managing a team of people is a full-time job, even when all the team resides in the same building. If you're dealing with an on-line team, you can easily spend all your time managing it, and that means you won't be spending any time on making your mod. Adding more people to the team doesn't mean more work will get done. The more people you have, the more time spent managing them.

http://developer.valvesoftware.com/wiki/Making_a_MOD

If you have never been apart of a mod team before, forget trying to start your own. A common misconception is that either it is easy to make a mod, or easy to get a team together. It really needs to be said, that this could not be further from the truth. Instead join a mod, learn what they do, learn from their mistakes; get to know people in this scene. Any time spent here will be made up 10 times over if you set out to make your own. Mod DB help wanted is a great place to find what jobs are available, otherwise contact a mod team, tell them what you can do, send them some samples and ask if they have a job for you - simple!

http://mods.moddb.com/getting-started/make-mods/

[Dol Guldur]

All good advice.

One of the major issues resulting in the failure of mods to materialize is in the spreading of talent among too many mods. There is nothing wrong with people starting and working on their own mods. This cannot be forced. What could be done is to create an environment where like-minded modders and similar mods are more likely to have contact with each other and thus build relationships that will lead to a joining of minds and mods/mod ideas.

This could be accomplished by providing subforums in some way that funnel modders/would-be modders into areas by era or genre. For example, for M2TW you might have discussion/resource subforums such as Historical - Americas, Historical - Europe 1300s, Late (Romantic) Arthurian, etc.

Bringing those working on or desiring to work on mods very similar to each other into a discussion subforum is more likely to create joint ventures and even abandonment of initial mod ideas for a common idea. It would go some way to naturally unifying the isolated and many. Fewer mods there may be, but more will likely be finished.

[Epistolary Richard]

I still find some of the advice hard to reconcile with each other - some people say accept everyone in your team as even a little work helps you to your goal, others say only accept really dedicated people who you absolutely can't do without. Big teams or small teams?

Also regarding releases, is it good to work for a long time on one big release or put together lots of small releases? The big release approach always has the challenge that they may never _reach_ a release. Small release approach though tends to have less of an impact on fans, it also makes it harder to rejig your mod as you go if your developing release to release rather than sitting down with a single design document.

And there's also leadership and decision-making - so many mod projects it appears need a single brain to keep it all together, but some seem to manage to have a more collaborative approach.

And there's also the matter of what happens when things go wrong. Countless mods are going to be developed for M2TW, many of those will fall by the wayside with their stuff never being released, never being used and basically being a big loss to the community. Even 'ongoing' projects might suddenly hit the skids as they realise that all the stuff they've previews is on the hard-drive of someone who has suddenly disappeared. This, to a large extent, is completely inevitable - however I wonder if there is anything that can be done that can allow the community to get some worth out of these projects or at least not let the creations of projects that have fallen fallow from completely being lost to the members of the team who carry on.

[wlesmana]

The size of the mod would depend on what the core members are comfortable with. Some like to work in a big group where everyone is supporting each other with a lot of discussions. Others prefer small teams where everyone knows each other. When I mention "core members", I meant members who are expected to do the bulk of the workload and must interact with most of the team. A Historian only needs to deal with other Historians/Researchers and maybe with Designers. A Modeler must interact with the Designer (and Researcher through one) and the Texturer. And his work will be judged by all members. If your Modeler doesn't like the Sound department guy criticising his work when everyone else loved it, then there are some issues the mod leader must deal with.

A big group is harder to manage but the stagnation rate is potentially smaller.

Any group needs the Final Decision Maker. Other groups who claimed to have none have one that acts like it unofficially. There's always a personality that's dominant and could convince others his way is the right way. I think it's better to point one person (usually the mod leader himself) for this position, to keep the goals focused. However, this person must also be open-minded and accepts suggestions and criticism.


I personally doubt M2TW projects would improve over RTW's. They're still amateur, non-profit mod projects and not many mod teams are like TheLordz. The same mistakes would be repeated and modders would still prefer to start their own instead of joining existing ones.

[Encaitar]

Regarding 'frequent mini-releases' vs 'few, major releases', I think it depends a bit on the type of mod. A mod that tweaks/extends the vanilla game can easily have many releases, as you really only have to get the new stuff working for it to be 'ready' (okay so it can get a bit more complex as the mod gets larger, but still). For 'conversion' mods (such as fantasy-based mods), it's harder to have mini-releases, as you can't really use the vanilla data to fill in for the stuff you haven't done.

With ME:TW I got quite frustrated with some of the reaction to our initial release. It was primarily a beta 'battle' release, and we had included the ability to load up our very early work-in-progress campaign map to 'have a look around'. Despite clearly stating on the downloads page, in the install process, in the readme, and in-game as well that it was no more than an alpha 'sandbox', we received a large amount of complaints about the campaign being 'unbalanced' or missing features (such as anything other than the core buildings!), and criticism of 'how crap our mod was' (purely based on this alpha campaign). I'm still unsure as to whether it was the right thing to do or not (including the campaign), as we did get a lot of useful feedback from people who understood what it was.

[Lord Adherbal]

first of all this is only really important for total conversions. Mods such as SPQR and Darth which start by altering the original game (but possibly evolve into something completly different) don't really need big teams because there is no predetermined point when these mods are finished or not. They just evolve until the team(leader) desides he had enough. And they often rely on independant model(lers) for their graphical departement, were every contribution is welcome but not vital.

As for big VS small team, a couple of dedicated modders of whom a lot have varied artistic skills (modelling, texturing, ...) is most important. A huge team full of "historians" and txt editors is useless and I bet it can be frustrating for the actual workers having to listen to all these guys. When I see a mod being announced in a topic with pages upon pages of historical info about factions and what units they will have I just know this mod will never make it. To actualy start a mod a single person has to start modding - not talking - and then present his work so others can decide to join the project. The project usualy stands or falls with the dedication of this single modder.

Working on ChivalryTW as team leader often caused frustration, with lack of help, new members disappearing after having spent a lot of time to introduce them to 3D modelling, having to listen to too many oppinions etc...
Working for the Lordz has been a much more pleasant experience. It's actualy a rather small team (and it was near dead when I joined) but there have always been other hard workers in the team to support and encourage eachother. I'm the do-it-all who combined all the material and turns it into playable betas, Duke John is expert in coming up with innovations no one would ever think about without smoking a lot of pot, Hoggy makes (superb) models faster then anyone else, Zimoa takes care of all the PR, management and whining fan(atic)s, and all the other Lordz work on varied other stuff such as animations and maps, at a less contant pace but it still makes up a large part of the workload. And most importantly we have a very clear aim of what we're trying to achieve and everyone seems to disagree about the decisions made (and those who don't are being ignored :P).

The fascination for the Napoleonic period - and the fact we're quite unique at that - is also a large factor of what makes the team so succesfull. The result is that anyone who considers making a Napoleonic themed mod is most likely to join us nstead of starting his own mod. If that was true for all other possible historical eras/themes the modding community would be a lot more succesfull.

[alpaca]

Well, the thread title is what we learned from R:TW.
I personally all but started modding with Rome, and not even immediately after it was released, so I learned quite a lot from this community and this game.
The first unpleasant thing to learn was that developers sometimes don't support mods very well, a fate that may very well await us with M2TW, apart from sometimes dedicated individuals who try to dig up some info but are often shackled by their contracts.

I also learned that preparation is crucial for a mod (or any larger project whatsoever). If you just sit down and start working, you can get a few interesting and worthwhile ideas, but some time you have to arrange it all in a fashion and discuss it with other people in order to improve it and actually realize it.

During my time with Chivalry, I learned that it is extremely important for a mod team (as has been mentioned before) to have at least one very motivated and dedicated individual - ideally the mod leader - who just keeps working all day and drags the others into his surge. Just having one such person can make all the difference for less motivated members (like myself usually, that is to say I was not the ideal cast for the mod leader job and Chiv is better-off now that DaVinci took over).

The size of the team is not overly important in my opinion. You can have success with a small team of dedicated individuals who each contribute a lot to the mod, and you can have equal results coming from a large team - even when each member only stays on for a while and doesn't do all that much work individually - if you have a "brain" at the top who is good at assembling all this stuff. I guess it all boils down to what the core team (most mods have something like this) feels more comfortable with, so both approaches are probably valid in their own way, and they both have advantages and disadvantages.
The major advantage of the large team approach is that you minimize the number of really dedicated people, so that you run a smaller risk that one of them will mysteriously disappear. On the other hand it may sometimes be difficult to make decisions on controversial matters because there can be any number of opinions (I recently read a funny little anecdote where a journalist stated that "As in each room where only journalists are present there were distinctly more opinions than people"), but at least you have all the opinions there and can make a decision between them instead of everyone having the same idea about the shape of the mod - which will mostly culminate in a better result but sometimes in civil war.

The number of releases, too, depends mainly on the feelings of the core team.
If you can motivate your team well enough to keep working until a large milestone is met than I reckon from a PR point of view a small number of releases is preferable. The reasons for this are that you can build up hype and thus get a much larger impact when you make a release, and that you won't spend as much time with release-associated tasks like writing read-mes, searching for small bugs, etc.
On the other hand, you might have problems with narcissist team members (me again) who like to see their work out there and being enjoyed. If that is the case, it may be advisable to make a large number of releases, e.g. one every month. This has the added advantage that you have external milestones for which the psychological threshold to break is much higher than for internal ones.


A comment to what Dol Guldur wrote:
I agree with you to some extent. However, most people who favor starting a new mod over joining an existing one do this for a very good reason: they can't agree with the mod policy and/or the content, or there exist personal animosities between them and at least one important member of the team they don't want to join.
If the reason they want to make a new mod is the former, then it is a completely valid position. If it is the latter, you will probably end up with two pretty similar mods striving to contrast from each other, which is mostly contra-productive, and probably one of those mods will wither and die (I guess this would usually be the one that gets less public attention).

[Duke John]

Duke John is expert in coming up with innovations no one would ever think about without smoking a lot of pot

To be clear, I do not smoke pot

[Monkwarrior]

I agree with Adherbal. I've worked only in very small teams (2-3 people), but with members highly motivated. The people have changed, but the team is still there. The best way to convince others to join the team is to show that you are really able to make it alone, although in very long term, and your ideas are worthy to be tested.

Nobody had to encourage me about modding. I feel well doing it, better than playing (I haven’t finished a single RTW campaign and even only started three or four). And I think this is the only way it is possible to do. Nowadays I can model and skinning a little bit (not an expert with 3ds max and photoshop, but it works), making maps, coding, scripting, etc, etc, etc. It is only question of having interest to learn.

I think that the coordinator of the project must be a self-supporting person, able to continue the project even alone, and must be the reference for the rest of the team.

Any help is welcome, but only if this helps you to save time.

[Dol Guldur]

This exchange of views is very interesting and important. I'm not sure that I have any firm conclusions in my mind from it as of yet (other than knowing that Duke John does not smoke pot), but perhaps that will come with more posts (the firm conclusions, not Duke John smoking pot).

I do not think I disagree with any of it.

But what are the lessons and how do we tackle it this time around to increase the likelihood of more mods being finished?

[Epistolary Richard]

Any group needs the Final Decision Maker. Other groups who claimed to have none have one that acts like it unofficially. There's always a personality that's dominant and could convince others his way is the right way. I think it's better to point one person (usually the mod leader himself) for this position, to keep the goals focused. However, this person must also be open-minded and accepts suggestions and criticism.

I'd agree with this - decisions need to be made and people have to not have their noses put out of joint doing it. I'd be slightly wider than that though, I don't think a team needs someone who makes the Final Decisions (though it's undoubtedly efficient), rather they just need a process by which decisions are made (e.g. vote, decision by coordinator of that area etc. etc.) which all the team members are aware of.

With ME:TW I got quite frustrated with some of the reaction to our initial release. It was primarily a beta 'battle' release, and we had included the ability to load up our very early work-in-progress campaign map to 'have a look around'. Despite clearly stating on the downloads page, in the install process, in the readme, and in-game as well that it was no more than an alpha 'sandbox', we received a large amount of complaints about the campaign being 'unbalanced' or missing features (such as anything other than the core buildings!), and criticism of 'how crap our mod was' (purely based on this alpha campaign). I'm still unsure as to whether it was the right thing to do or not (including the campaign), as we did get a lot of useful feedback from people who understood what it was.

Hmmm... an interesting tale. IIRC Chivalry got something a little similar with their custom_battle release. Looking back on it do you think there was anything else that you could have done to avoid this issue - or do you think that it was inevitable.

everyone seems to disagree about the decisions made (and those who don't are being ignored :P).

Shurely shome mistake - ed

The fascination for the Napoleonic period - and the fact we're quite unique at that - is also a large factor of what makes the team so succesfull. The result is that anyone who considers making a Napoleonic themed mod is most likely to join us nstead of starting his own mod. If that was true for all other possible historical eras/themes the modding community would be a lot more succesfull.

The Lordz do certainly enjoy a brand-dominance in that area. Caused I think partly by the quality of the work produced (after all, how would most people match it?), partly by the sheer difficulty of converting to that period and also their reputation built up over the course of releases.

Periods and settings closer to the original game will have greater will be more likely to have modders who believe they can go it alone.

I recall there was even a chap once who was trying to do a whole Medieval RTW mod off his own back for a short while (and no, that's not referring to you Adherbal )

But what are the lessons and how do we tackle it this time around to increase the likelihood of more mods being finished?

I think the mere exercise of pooling these experiences and opinions and then recording them for M2TW modders to read helps in this matter.

More proactively, I think people would generally agree that every modder should have his chance with a 'Forge' thread to show what he can do - even if his initial pitch leaves the public lukewarm.

Modding forums can help highlight the those mod teams that seem to be heading in the right direction with things.

The idea of project debriefs (or post-mortems) has also been posited - though this would remain a controvesial tapic.

Any more thoughts?

[Duke John]

I don't think a team needs someone who makes the Final Decisions (though it's undoubtedly efficient), rather they just need a process by which decisions are made (e.g. vote, decision by coordinator of that area etc. etc.) which all the team members are aware of.

It works IMO much better if a single person makes the final calls. Although if the group is large enough you can have specializations with persons making the decisions in their area of expertise. However you must have the right persons for that, it's a bad manager if he is not able to defend his decisions against all counter arguments and if he cannot bring arguments on his own besides "I like that better."

[Lord Adherbal]

I recall there was even a chap once who was trying to do a whole Medieval RTW mod off his own back for a short while

I think I know who you're talking about - forgot his name though - he first joined ChivTW as a modeller but because we had most (european) models done already he left to basicly do the same mod on his own. It looked good but obviously died because he must've underestimated the workload, and IMO it was quite pointless trying to compete with ChivTW which was already in a far more advanced state.

It's a perfect example of how many people want to be their own boss and do what they like, and how this behaviour is bad for the modding community.

If it was up to me I would let all TW fansites disencourage/banish mod projects about themes that could be/are included in existing mods. I guess you can all be happy that it is not up to me

[Ciaran]

Indeed, and if only for my very deep conviction that variety and competition (though I don´t actually believe that mods compete against each other) is beneficial.

I do acknowledge that a great number of mods never makes it to release, but according to an intervie with a CDV product manager that I´ve read, the number of commercial projects that fail is around 80%. It´s a matter of fact, and maybe that is one of the lessons learned. You can´t force people into cooperation, you can only encourage it. If people leave or don´t join a team, there tends to be a reason for that.

[ozz]

Any group needs the Final Decision Maker. Other groups who claimed to have none have one that acts like it unofficially. There's always a personality that's dominant and could convince others his way is the right way. I think it's better to point one person (usually the mod leader himself) for this position, to keep the goals focused. However, this person must also be open-minded and accepts suggestions and criticism.

It works IMO much better if a single person makes the final calls. Although if the group is large enough you can have specializations with persons making the decisions in their area of expertise. However you must have the right persons for that, it's a bad manager if he is not able to defend his decisions against all counter arguments and if he cannot bring arguments on his own besides "I like that better."

From my experience in working on a team without an official decision maker, I tend to agree with wlesmana & Duke John on this. I think a team-leader or a final decision maker is best, as long as he is open-minded and is able to accept suggestions and criticism.

You would be surprised how often "civil wars" erupt in a "democratic" team for the dumbest reasons. Halting progress to a standstill many times because every member has his own vision of the mod or simply because of personal likes or dislikes. In the long run, the only casualty is the mod itself and the release date being pushed back once again, disappointing the eager fans and putting even more tension on the team.

You can have leaders in certain areas, but they must be VERY co-operative and open-minded to keep the mod moving while also keeping the rest of the team motivated and not losing focus on the end-goal. As to accepting suggestions and criticism, I think that boils down to the individual personalities of the team members... while it can be a pleasure and fun working with some, it can be hell working with others...

[khelvan]

Hi guys, ER asked me to drop by and give some thoughts.

I have to stress that there is no one right way or wrong way to organize and manage a team. This depends on the person and the team members, (and in a business setting, the organization at large).

In my experience, the more people are being compensated, the more they are willing to follow a single person's will. That compensation may be financial, for "real" jobs, or in the form of recognition, or even in the form of having their ideas used and valued, for non-paying jobs. Perhaps it is in seeing the realization of a purpose or cause, or end goal.

For modding I think that the more "power" that individuals have in a mod team, the more they feel valued, and the more engaged they are in doing their part. EB has never had a single brain at the top. We have had several "leaders," but not in the sense of creative control. Rather, we act as moderators, or very much like project managers in business whose subject matter experts are the people on the team. The manager just makes sure that all ideas are heard, that decisions are made efficiently, and that disputes are mediated.

Size matters - a big team is of course much more difficult to manage than a small team. However, ability and reliability are the primary factors in team size. There are times when a small number of people have a great deal of skill and dedication; these can be the core component of a team, and may be all that are needed to complete smaller mods. Or just one, for the smallest.

The numbers will likely have to increase as the size and complexity of the mod grows. Like business organizations, tasks will tend to become more specialized as the complexity (and team size) grows, and thus will need more management. Unlike someone said above, I think there is a place for many historians, if the mod is focused on history. But this has to be balanced with people who can do the work, in a mod where these tasks are broken out, otherwise lots of research will be done but the mod will never get going.

As a leader and/or the public face of a mod team, I can guarantee you that staying emotionally detached will be incredibly difficult. This is where I failed most as a "leader." Emotional attachment is sure to be high for all people on the team; there is no financial compensation, so people are dedicated for other reasons. As a public figure, you can't really afford to be engaged in this way, otherwise you will make poor decisions. For instance, I did not have a job for quite a long time, and dealt with a very serious medical condition, and I channeled my frustration and emotions into EB. This led me to state things emotionally in a way I would never do normally. Especially in a business environment, I would never have said things like that. Keeping your emotions in check is very important. This can lead to people leaving the mod, as well, and I guarantee it will if you get angry/snap at people easily.

A release schedule generally depends on the mod and the mod team. Here we consider individuals' goals. For instance, if one of the goals of the mod team is to be exposed to game companies so that the members can earn jobs in the industry, it is in their best interest to release only versions that could be considered polished. Likewise if the mod is hurting for members, the team has to do a very good job of keeping the fans involved, through distributing information, or through more quick releases. Releasing too quickly will be a problem in that case, though, as fans don't like to have to restart their campaigns often. It really depends on what your needs and goals are.

Finding some balance between introducing information to the public but not committing yourself to anything is very difficult. In general, if you care at all about pleasing the fans, never, ever, ever give a release date or make any promise of any kind at all. Even in the real software development industry where people are paid to do this work and can be held accountable by someone external to them release dates are often missed. Not missing the dates leads to great disappointment and even more heartache for all involved. Keep expectations low and you have a much greater chance of exceeding them and pleasing the fans.

However, keeping the fans involved is important to most mod teams. Regular information posts with teasers can help keep interest up. I am sure there must be some way to keep the fans informed, perhaps with some sort of constantly-updated post that contains percentages done for various areas of the mod toward the next release, but I don't know how to do it.

Potential members must be fully informed that if they can't meet whatever deliverables they are tasked with that they have to be able and willing to hand over the work they have done and keep everyone informed. In fact in a situation like this making sure team members' work is transparent is very important. Team members should keep the team updated at all times as to the status of their tasks. There should be no reason to hide what work is being done from the team.

I don't really have advice on how to make and stick with decisions. I guess this is just something that you learn through experience. There should always been room for improvement, but there is no need for endless debate about things where no new evidence is presented. Change control is important; you don't want the scope of the project to keep creeping, either.

Stay as light-hearted as possible. Learn what makes the team members happy, and try to give it to them. Make sure everyone stays civil. Jerks and people who are simply not nice to others do not make for a happy or productive team. The skill that they bring is not worth the problems with other people. Personality is important. Some people just do not play well with others.