BEGINNINGS AND ENDINGS: HOW TO WRITE THE HARDEST PARTS OF YOUR STORY.

Depends on the narrative, but generally I go for either establishing the status quo or showing the inciting incident. Usually the former, since it allows the audience to catch up with the protag and what their life is like before you start throwing craziness around. The latter is more for times when it feels too slow to build up to things, and I can better introduce characters via action.

A beginning should help the audience understand who the protag is and what kind of story to expect. (Not necessarily what kind of story they're getting, if you're planning a twist.) You don't really need to establish things like worldbuilding up front- keep it minimal at first, so that your audience can ease into the story and get attached to the characters, rather than trying to juggle a bunch of proper nouns that don't apply to the immediate situation.

Endings... fuck, man, I don't even know. I just kind of wing it. Tie up whatever loose ends are important to the main story, then ease out... somehow.

Or just cut to a black screen with poorly translated white text, that seems to have done well in the past. :V

ETA: Wait, I do have some advice on endings!

Assuming you established the status quo previously, show how all the stuff that's changed over the course of the story is doing with its new status quo. Same with the characters- demonstrate how they're different after all the events of the narrative, and what they're up to. Apparently people care about characters or something?

I'm sure we both meant the 'you' as in 'general populace', not you specifically piano~ ;p

One time I had a whole story planned out pretty thoroughly up until just before the final battle, where I just had written "The rest of the story probably will go through so many changes that whatever gets written here will end up being dumb, so I'll figure this out later, once I have a better feel for the characters and the story arcs." Like to the point that I didn't even know which side would win. I wanted the ending to fit really well with the rest of the story's primary struggle, and at that point, I wasn't totally sure what that parts would end up getting emphasised and what parts would end up being minor.

Predictably I never finished that story.

Don't use that method. I thought there was some logic to it, in my head, but it was a trap. Figure that shit out first, before anything else in the game. Before you even decide who the main character is.

Beginn with the ending, figure out all the crucial characters and events that lead to it, (main plot points) fill out some of the gaps directly, while still leaving some room for originality as you go, and finally, let the story start in medias res.

Or at least thats how I did it... still no idea if it'll work out in the end. ;P

I usually have a scene i start with. Usually it ends up being the beginning, sometimes I need to put things in front of it. I tend to have a vague idea of an ending, which often changes.
It's not difficult to keep your audience interested during these two parts, you need to come up with quality meat to fill the space between them. That is crazy hard to do. If you know your ending, you can end up predictable easily. That's why I focus more on characters and environments than the actual story.

When it comes to writing stories, you need to understand the kind of story you want to tell before diving in to write it. Having a general overview of the basics is crucial to making it coherent. Stories have conflict; without it, they aren't really stories. The beginning establishes that conflict and the ending resolves it. This may sound obvious, but understanding it conceptually is important for defining your task. There's a certain logic to writing that makes it both completely clear and wonderfully unpredictable.

See, in a way, the beginning and ending should be the most obvious parts of the story to plan for. How does the conflict arise, and how does it eventually conclude? These are the two points that everything else falls between, but it's the middle stuff that makes it complicated. I'm sure you've heard that characters may often write themselves, and this is where the logic becomes unpredictable. If the actions of your characters make sense for them personally, things may take a turn you didn't foresee when they act on their own (because they did what made sense to them instead of following the script). These surprises can be fun for writers and their audience because new developments are included that neither may have seen coming. However, because the beginning and end are both established events, these happenstances don't impact the overall story very much. One way or another, for better or worse, it must all come back together.

So what's important to a beginning? A bunch of the stuff mentioned so far is good to keep in mind. You want to engage the audience and keep their interest piqued. You need to establish the circumstances of the story, including physical things like time and place, as well as concepts like tone and theme, and there are many ways of doing this. There are some basic things you can assume your audience will understand on their own, but be careful with how you present new concepts or unique aspects of your story. Everything needs a frame of reference in order to make sense, and some methods are much classier than others. "Show, don't tell" is important. Infodumps are boring. Long strings of nonsense fantasy names will confuse people. It may hurt, but keep lore on a need-to-know basis. You can stuff an in-game library with your Silmarillions if you have to. The quicker you can establish the conflict and characters involved, the better. I wish people these days had a longer attention span, but instant gratification will never stop being a thing. -_-

As for endings, that depends on a lot more factors. Again, the conclusion itself should be fairly obvious: how does the conflict cease? If you want player choices to matter, multiple endings can solve this conundrum for you. Causality is such a fun thing to play with; you should never be afraid of it (but respect that it can be hard to use well). Endings are a little more elegant than beginnings, because all they really need is closure. The audience will be satisfied with the ending if they feel their questions were answered and the conflict has truly ceased. Even in the case of a "bad" ending, the conflict could be resolved by the hero's defeat, meaning they can no longer pursue their efforts. It's also important to recognize what DOESN'T belong in an ending. This gets more difficult with extensive plot threads, but you need to have a clear idea of the story's main conflict and how it is resolved. Since games often depend on how much effort the player is willing to sink in, non-crucial plot threads should resolve on their own arcs, independent of the story's main conclusion. If you have a big cast, not every character needs their last little quip to tie things up. Consider also that not everything needs to reach epic-scale just for coolness' sake. Your final boss might not really need three forms to feel compelling. It's tempting to drag things out, but don't. Elegance lies in brevity. You don't want your ending to turn into an endurance run, or your audience will just wish it was over.

There are way too many microcosms for me to get into here, and this post is already dragging on. I hope some folks find this advice helpful!

Firmly in team "I have the opposite problem" here. I always write the beginning and ending first, with a rough and flexible outline as to how that beginning gets to that ending. For me it's a question of "This is how my characters start, and this is where I want them to end up." And figuring out all of the infinite possibilities between point A and point B can be very daunting indeed. And because of that I never stick myself to following a rigid plan; even when I'm eventing, sometimes I'll just throw the original idea out the door and make up something on the spot.

Remember that things can, and probably will, change.