Choices are hard. Especially the ones that promise to have some sort of effect down the line later. Few and far between as they are, these choices are often put to gamers trying to masquerade as more than they are to elicit some sort of reverence, or reaction. While you can have some sort of meaningful choices now and then, truly long-term consequences are all but impossible due to the amount of effort and work required to program all permutations and variables. It’s unsustainable to provide a game like this except in precisely measured, small chunks, and for a very short amount of time.

Let’s use the template a certain Telltale game has set up. You have a choice of saving 1 person out of 2, with the shunned one dying a guaranteed death. In The Walking Dead, that person participates minimally in the next episode with the smallest of changes compared to the other person, and then unceremoniously dying in episode three – no matter which you picked. Gameplaywise, this character’s death makes absolutely no sense, but in a directorial sense, it had to occur. It’s just a shame that choice wasn’t really yours to make.

There are reasons for these kind of arbitrary limitations. Modelling a character, scripting for them and voicing said character takes time – and when a character is interchangeable and most gamers won’t see the alternative – it’s a lot of work for little gain. Furthermore, if those people are truly different, that should affect the narrative at large at almost every turn. Were this real life and you picked between me and someone athletic, I’m more of a relax at home kind of person who doesn’t leave his apartment too much out of comfort.

For the story, a mutual character we both know – me and the athletic person – is hospitalised through an accident. The other person might be out on a jog to witness it, or get news of it sooner from bystanders and such, perhaps able to interact with the person who, perhaps, is on his or her dying breath, which may or may not be a huge secret important to one, or both characters. The athletic character has a chance to come on the accident victim where I would not, and might only find out in the next day – or days. Something as simple as seeing this or hearing their last words could affect most people however, and this is my point.

Supposing the secret heard was misunderstood by one person or not shared with the other, we have so many story branches that it has become effectively impossible to keep it going without trimming the choice tree a bit. Suppose then, that you chose the athletic guy, who saw this death first-hand and learned of a secret integral to him or her. Shock or trauma would bring most people to want to grieve or process the events. In such an example – the person would be more likely to stay home, remember this and the person and alter his/her life or choices to cope for these events. But if the story is about *both* these people, it will present a myriad of complications.

If you assume that both could have seen the accident and they both knew the victim, but not each other, and you had to choose between me or the more athletic person, you are now looking at 2 people, 2 lives, which would be affected differently – and the consequences of the event drawing different paths in different lives. I might not know of a secret that might not be meant for me is the smaller issue – the actual issue is that, where this a game, you’d *have* to work on both eventualities, at once. Where a lazy guy found out, and where a fit person came across this.

You now have to model the game around the impact of this death for the athletic person – them taking time off work, interactions at a funeral, the pain of loss – but if I was the other character, I likely wouldn’t know that same day if not later. So if someone does choose to play as me for some weird reason, they’d have to model *another*, disconnected grieving process, completely different. My animations, my face, my voice – at best they could get away with the same cut scene of both people attending the funeral and shedding some tears – with absolutely no personalization, to get away with it.

Then what? A dear friend has passed – what’s next? Well, let’s say the accident was caused by sabotage, part of the secret only the athletic person is aware of as per the dying words. They have several options now -pursuing truth, cowering in fear, or attempting to progress with their life without the breaking point. And what about ‘me’? In all likelihood, mourning and returning to my life would be the answer if I had no motivation. That is why games that tell grand, choice-ridden stories often seek to have the entire *group* find out about whatever secret or factor is important for the narrative.

Humans are highly modular. We do what we feel like most of the time, based on what is best for us. So consider athletic person – they might interrupt their routine to drink at a bar for their sorrows – something they might never, or at least rarely, do. This might break their routine, cause them to stop driving at an establishment they never frequent out of sudden depressive need to drink. Simple, right? Wrong. There they can narratively encounter a new unconnected character who can offer many things – companionship, information, warnings, a threat. Another branch. Or, you know, our jogger could jog shell-shocked and wind up talking to the family of the victim to find out more, which leads to yet another branch with more characters, locations, etc.

Then you have to write for those specific choices, but if the game offers to let you play as ‘me’ too then you have to also fully scene out the ways I’d be acting differently – I might mention this person to someone and cause them to turn pale with fear as a conspiracy I’m unaware of is claiming lives now and such. The permutations are near-infinite, and to account for each and every choice in a truly meaningful way is crippling to development time, budgets, and HDD space. For each location your character *might* go to, you have to do work. For each character they *might* find or interact with, you’d need to work and plan. It very quickly becomes prohibitive to attempt anything big with such limitations.

Episodically, in small chunks, the size would be less of an issue if you could delete an earlier bit with a functional save for the newer bit. However voicing scenes, art for locations, textures for them, newer characters and such that would organically and realistically come and go is just too much. Consider some Telltale games and their branches, why they got cut off – often abruptly – to save time for the developers. This has existed in a minor form for a long time, where you are given a choice to say out of 3 identical outcomes , but the attitude in which they are delivered is different. Passive, friendly, aggressive, etc. It pretends to let you roleplay, in that you can select how ‘you’ would reply, but the outcome will be identical..because it has to be.

I believe that in the near future, it might be possible to have a “central AI” and a pile of tools for it to play with, for our amusement. An intelligence, whether simple or artificial, to take from the toy box elements and bits for us – to construct a story based in random generation and assemble the necessary components to make it somewhat believable. A basic set of ten characters, where ‘Joe’ in one story would be a hired hitman, yet a lovable pizza maker in another, and finally an alcoholic drifter in the third one. All the same character, physically,  in a new story, constructed by a machine – the big part of this would be recording an insane amount of voice acting so the modularity is preserved – or rendering it entirely text-based.

Right now in videogames, choices and storytelling is very rigid and has to come from a beginning and stop at an end, with little room in-between for variations and what you want to do.  There is little room to actually explore a character, or a world, besides what they are shown to be like – the player can possibly draw their own conclusions from dubious actions or uncertain behaviour, but in the end, everyone is something very straightforward in video game narratives – even if that straightforward might be a calamitous pit of confusing personality complexes. Smart writing exists, don’t get me wrong, but for truly meaningful world-altering choices that flow out to affect other people, I think due to many limitations we still have quite a way to go.