Legalized items/exp farming

[Gr3yling]

This shouldn't happen in roguelikes.  That's one of the genre's most important features - content needs to stay interesting through repeated playthroughs.  If you can't do that you shouldn't make roguelikes.

Wait, what shouldn't happen?  Events in roguelikes shouldn't have patterns that are readily apparent?  Isn't that a pretty common thing in any genre?  I mean, don't you succeed at playing roguelikes and most other types of games by recognizing patterns as you encounter the same (or very similar) situations in multiple playthroughs?

Are you saying you think roguelikes should be based more on strategy and planning rather than memorization and pattern recognition?  I can't say that's a bad idea if you are, but I think there are plenty of great and interesting roguelikes that have obvious repetitive elements.   For instance, I think the content of ADOM is very interesting, but I do think it tends to be rather similar from playthrough to playthrough. 

I don't get it, it sounds almost like you are saying good games don't have content that can be solved algorithmically.  Don't you consider rogue to be a good game?

Basic automation for simple tasks is fine.  Autofight and autoexplore don't do any harm (though better design makes them less necessary).  But if you're seriously contemplating automation for quests and other complex tasks, you need to reevaluate your design decisions.

I guess I more or less agree with that.  And, maybe the issues that we are trying to deal with using automation can be completely overcome through other means.

One thing I think is important to point out is that simulation frees the player to concentrate on other less small scale elements of the game.  So, if you wanted to greatly increase the scope of what you allowed the player to do, it helps to let them simulate parts of the game that they don’t want to micromanage.  Like in Dwarf Fortress, I think (I haven’t actually played it).

Another example where simulation makes sense is a game where you followed the PC over a long time period, and essentially raised them until they were old enough to complete the ultimate objective of the game.  You might decide how they allocated their time growing up, what kind of apprenticeship they had, early adventures they went on that kind of thing.  You really don’t have time to play through years of the PC’s life without simulation, I mean.

I guess for some reason I keep thinking that is what Xecutor had in mind, and that he was going to add more “macro” features to keep the player busy while they were simulating menial tasks.

[Gr3yling]

Well, I wasn`t entirely serious, because once you go down that route, it becomes a minefield of silliness. As in, we could also say then that we`re really playing a pen & paper game because everything else - computers et al is just automation. Hell, even the paper is - otherwise you`d have to do calculations in your head.

That's exactly what I was thinking.  Why is that silly?

Instead I prefer to separate concepts of interface & gameplay  (which is probably why these have different names too). And interfaces evolved over time from basic ones - due to being first on the block & also hardware limited - to more advanced, with some features being an obvious norm nowadays.

They have different names and they are separate entities, but they are very closely related.  As you know, it's hard to play a computer game without an interface.  And some gameplay features have become integrated into the interface for so long we take for granted that they are different things.  That's the point I was making about map drawing.

Yes, I`m one of these people who still likes to draw a map, but that`s totally different and would never work in a roguelike, for obvious reasons like randomization and general pace. Though I was thinking recently that taking away the map entirely could be an interesting option for making games much harder, raising the tension and maybe implementing new mechanics (breadcrumbs?)

So it sounds like you are saying that most roguelikes as we know them today exist because that feature was absorbed by the interface.  And instead of making them "non-games" they freed the player to concentrate on other features that could later be added. 

Optional is fine...though it could become a bit worrying then, sort of like the "optional" ability to save that removes the permadeath threat. Say there was a game that had options to automate heavy chunks of gameplay - it`d be seemingly the same and yet I would be perhaps a bit distrustful of how serious its core mechanics are.

The ability to reload saves in ADOM after death is there, isn't it?  You used to be able to copy save files.  I guess you still can?  I never try, so I don't know.

I think a better option would be for enemies that are no longer a threat to the character to run away from them.
I don`t really think it`s a problem - after all if an enemy is "trivial", you just steamroll it anyway -  but this solution is very nice, I `ve seen it already implemented in one non-RL RPG and it worked rather well.

Maybe that would work fine.  I was opposed to the idea previously, but I can't think of any reason it wouldn't work, now.

[akeley]

Instead I prefer to separate concepts of interface & gameplay  (which is probably why these have different names too). And interfaces evolved over time from basic ones - due to being first on the block & also hardware limited - to more advanced, with some features being an obvious norm nowadays.

They have different names and they are separate entities, but they are very closely related.  As you know, it's hard to play a computer game without an interface.  And some gameplay features have become integrated into the interface for so long we take for granted that they are different things. 

Since  this is exactly what I said already, it seems we`ve hit some semantic block here. Similar with this thing about mapping - I never meant that "most roguelikes as we know them today exist because that feature was absorbed by the interface." - only that doing it on paper, while suitable and fun in some RPGs could never work in RLs because what`s the point of making a map of a level that will randomize on the next playthrough? Madness. maybe that`s why early RL dungeons were single-screen, later on automap was added if necessary.

I can`t really explain any better why there`s nothing wrong with automating interfaces but plenty when you try to do it to gameplay. And these bats are part of gameplay, not interface. Sure, we can squabble over the fact that auto pickup is also part of gameplay and yet I like to have it ON - seemingly contradicting myself - and yet to me it`s just obvious why tiny things like that are okay while large chunks like quests are not. You simply have to draw a line somewhere.

That whole pursuit of removing or automating "trivial" and "boring" things just seems dangerous to me. So what are we gonna be left with, eventually, just boss battles? Also, wouldn`t these "exciting" tasks start to be boring after a while as well?

There has been the trend in mainstream gaming over last decade, trend that left a rather sorry-looking landscape since some genres previously known for their complexity/difficulty have been "streamlined' (ie "neutered") in the name of "accessibility". And the argument was exactly the same - we don`t want to do boring, outdated things, eww! - we just want to have fun.

Okay, I`m exaggerating since I know we`re talking two different cultures and design ideas, and yet, similarities are there. You mentioned FF 12 earlier. I`m pretty sure Gambit system was used to appease those who don`t like to pause and control NPCs manually - transition to a totally action system, which is what was used in FF 13.
Now, in an action RPG it`s fine - but FF used to be strictly turn based. Long story short, it could be also said it`s tedious to control every soldier in XCOM manually, what a design palaver that was, eh?

But there`s another angle here - trying to understand why some folk are so keen on this idea I thought about what I read in other threads and generally saw watching RL scene for a while now. Seems to me a big chunk of players treat RLs like puzzles - games to be tactically dissected and ultimately winning being priority, but more in a abstract, purist way. Meanwhile I`m still more on the roleplaying side of things - I love the good tactics and challenge on offer in RLs but still need my imagination fix and so the dungeon is still a dungeon and Kobold Paladin a living entity, not just a yellow letter K that has some important attributes attached to it.

Considering this, definitions of "boring" might differ - I understand somebody perceiving the game in purely tactical way might have trouble with these things. But to me weaker creatures or more mundane tasks are  part of the gameworld. A well designed one, of course.

[Vanguard]

I don't get it, it sounds almost like you are saying good games don't have content that can be solved algorithmically.  Don't you consider rogue to be a good game?

No, I don't.  While I appreciate Rogue's contributions to game design, other games have improved on its formula to the point that it isn't worth playing anymore.

In an ideal roguelike, every choice would matter.  Every turn would be spent on something significant.  In most games that goal is unachievable, so the new goal is to come as close as you can.  Automation is a sloppy and inelegant method.  It's a crutch and it will weaken your games.  The better and more difficult method is to refine your design until those trivial choices are replaced with better ones.

That's one area where the original Rogue excels.  Through a combination of a strict food timer and offering little incentive to waste time, the game encourages you to make every turn count.

I guess my stance is that automation should be used in situations where there are no meaningful decisions to be made and no reasonable changes that could remedy that.  Autopickup is a good example.  You don't need to think about whether you want a potion of healing.  Of course you want it.  Automation just saves you a keypress.

[Kevin Granade]

How about a game that *thematically* calls for repetitive tasks?  In a survival game you might want to stock a larder to survive the winter, or craft enough ammunition for a battle.  The story you want to tell is that the player is surviving because of their own effort, is truly self-sufficient.  Are you going to simply reduce the amount of food needed for winter until it requires a reasonable number of actions to prepare?  Make ammunition appear in bundles? (no wait, isn't that automation too?)  How about literal farming?  Are these topics off-limits for a game because they're repetitive?  I'd say a better approach is to build mechanics into the game that depict the passage of time, and impose a cost for its passage.  Make time a resource of the game like any other.  For wintering in DDA, I'm seriously considering adding a set of mechanics for subsistence living that can literally be left running for months with occasional interruptions, because in a survival scenario, just holing up in a cabin over Winter is a completely valid approach, and I don't want to make the player manually play through that tedium.  The alternative of declaring by fiat that there's no such thing as Winter, or making the Winter scenario artificially more friendly to normal exploration-based play don't sit right with me.  In fact we have a large-ish group of players specifically clamoring for Winters to be harsher, specifically because they want to pit themselves against the challenge.

Personally I'm not a fan of the trope that the way to make your character stronger is to achieve goals (reaching a level of the dungeon, killing enemies, accumulating "treasure"*).  I prefer systems where skill and stat advancement comes from actually using the skills/stats in question.  Many implementations of this lead to "grinding" the skill, often with various checks that a particular use of the skill is "significant".  Instead of doing that, why not simply have the resource you're expending be the passage of time, and present ways for the player to expend that resource in a straightforward way?

In DDA, I'm planning on introducing a system where skill advancement is based directly on time spent practicing.  That practice includes time spent using the skill "in anger", but a safer way to hone your skill is to spend time on practice actions in a safe environment, at least when you can afford to do so.  Firearms are a major part of DDA gameplay, but access to them is sometimes difficult, so having the firearms and ammunition necessary to practice shooting is a goal in of itself.  Melee is simpler, because you have no shortage of melee weapons you can practice with, but of course firearms are generally going to be more effective.  Likewise, the hunger/thirst/fatigue clocks make spending time practicing things inherently costly.

*As opposed to intrinsically useful loot.

[Vanguard]

How about a game that *thematically* calls for repetitive tasks?  In a survival game you might want to stock a larder to survive the winter, or craft enough ammunition for a battle.  The story you want to tell is that the player is surviving because of their own effort, is truly self-sufficient.  Are you going to simply reduce the amount of food needed for winter until it requires a reasonable number of actions to prepare?

You should design your game around whatever possibility allows for the greatest depth and the least tedium.  Otherwise you're just weakening your game out of a misguided commitment to realism.

[Gr3yling]

You should design your game around whatever possibility allows for the greatest depth and the least tedium.  Otherwise you're just weakening your game out of a misguided commitment to realism.

Vanguard, it sure seems like you're awfully dogmatic about this topic, and I'm having a very hard time understanding your point of view because of that.  I think that even if simulation aspects shouldn't be incorporated into most games, Kevin (and to a lesser extent, I) have described situations where adding them is perfectly valid. 

Are you saying that a roguelike that takes place over the time periods that Kevin is talking about cannot ever be successful due to a "misguided commitment" to realism?  Misguided how?  Just because we disagree with you doesn't mean we are somehow misguided. 

You seem like you are saying that certain topics for roguelikes are "off limits" because they require simulation, presumably because that simulation would make the game less fun or interesting.  But I really don't see any way that you have convincingly demonstrated that this is true.  Again, I certainly accept that there are some situations where simulation is not appropriate, but your complete rejection of it in any circumstances is a bit hard for me to understand.

I'm not sure if I believe any gameplay element is intrinsically bad.  I think the success of an element depends on the type of roguelike that it is included in, which is to say, the other elements that it is combined with. 

I've said this before, but when someone brings forth a new gameplay idea, I think that the community should be supportive, rather than dismissive, of that idea.  It is very easy to find fault with such ideas, but I think what shows true creativity is to imagine ways that they would work rather than how they would fail.

[Gr3yling]

Since  this is exactly what I said already, it seems we`ve hit some semantic block here.

Okay, I see what you mean there.  I agree that my post said almost the same thing yours did. I did read your post, somehow I just missed that. 

I guess we are examining exactly the same information and coming to different conclusions.  Which is fine.

Similar with this thing about mapping - I never meant that "most roguelikes as we know them today exist because that feature was absorbed by the interface." - only that doing it on paper, while suitable and fun in some RPGs could never work in RLs because what`s the point of making a map of a level that will randomize on the next playthrough? Madness. maybe that`s why early RL dungeons were single-screen, later on automap was added if necessary.

Okay, I do see what you are saying.  That makes sense. 

I can`t really explain any better why there`s nothing wrong with automating interfaces but plenty when you try to do it to gameplay. And these bats are part of gameplay, not interface. Sure, we can squabble over the fact that auto pickup is also part of gameplay and yet I like to have it ON - seemingly contradicting myself - and yet to me it`s just obvious why tiny things like that are okay while large chunks like quests are not. You simply have to draw a line somewhere.

Akeley, what do you think should be done in games that have a large scope, like dwarf fortress, for instance?  Aren't you going to have to simulate some parts of that game given how much the player would have to micro-manage otherwise?  Or do you not consider games of that scope to be a true roguelikes?

Also, I'd still like to hear what you think should be done in the case of an adventure that spanned a large portion of the PC's life, where time constraints would make it difficult not to simulate gameplay.

I really feel like there is more of what I would call simulation in just about all roguelikes than you are arguing.  I mean, in ADOM, you don't choose individual types of melee attacks like you can in ToME.  Isn't that a simulation feature?  Doesn’t any game where you don’t micromanage every action the PC takes incorporate some simulation aspects?

That whole pursuit of removing or automating "trivial" and "boring" things just seems dangerous to me. So what are we gonna be left with, eventually, just boss battles? Also, wouldn`t these "exciting" tasks start to be boring after a while as well?

There has been the trend in mainstream gaming over last decade, trend that left a rather sorry-looking landscape since some genres previously known for their complexity/difficulty have been "streamlined' (ie "neutered") in the name of "accessibility". And the argument was exactly the same - we don`t want to do boring, outdated things, eww! - we just want to have fun.

I actually don't have a big problem with the bats.  I kind of thought everyone else did, and I also wanted to give Xecutor's idea a chance.  So, I understand where you are coming from.

But I wonder some about the implications of the kind of relativistic argument that you are making.  It sounds like you are saying that we only find things in games fun because there are other, less fun, things are there as a contrast.  Well, does that mean that we should intentionally add “bad” elements in order to make the player appreciate other parts of the game more?

In fairness, I suspect this isn’t what you had in mind, so I’d like to hear more about this idea.

Okay, I`m exaggerating since I know we`re talking two different cultures and design ideas, and yet, similarities are there. You mentioned FF 12 earlier. I`m pretty sure Gambit system was used to appease those who don`t like to pause and control NPCs manually - transition to a totally action system, which is what was used in FF 13.

I think that’s a reasonable conclusion. 

Now, in an action RPG it`s fine - but FF used to be strictly turn based. Long story short, it could be also said it`s tedious to control every soldier in XCOM manually, what a design palaver that was, eh?

I see what you are saying here too.

But there`s another angle here - trying to understand why some folk are so keen on this idea I thought about what I read in other threads and generally saw watching RL scene for a while now. Seems to me a big chunk of players treat RLs like puzzles - games to be tactically dissected and ultimately winning being priority, but more in a abstract, purist way. Meanwhile I`m still more on the roleplaying side of things - I love the good tactics and challenge on offer in RLs but still need my imagination fix and so the dungeon is still a dungeon and Kobold Paladin a living entity, not just a yellow letter K that has some important attributes attached to it.

Okay, I think I really do understand your point of view pretty well now.  You feel like you can’t experience being your character as vividly if there are simulational aspects, right?  That actually makes complete sense to me.  I actually hadn’t thought of that before you just mentioned it.

You want to really *live* the process of killing of those 10 bats, dismembering them, and stuffing the bat wings in your pack! 

No, seriously, I do get what you are saying.  That makes sense.

[Vanguard]

Are you saying that a roguelike that takes place over the time periods that Kevin is talking about cannot ever be successful due to a "misguided commitment" to realism?  Misguided how?  Just because we disagree with you doesn't mean we are somehow misguided.

It's misguided because decisions are being made to support realism, with little regard for whether they would make for compelling gameplay.  Realism itself is misguided because DDA has already gone beyond the pale of implausibility.

Simulations aren't inherently bad.  All video games could be seen as simulations.  My point is that there is a difference between simulations designed to be mechanically compelling, simulations designed to support realism (your character must sit in a cave for 3 months because it's cold outside), and simulations designed to cover up bad design (kill 20 bat quests).

If you find a way to make your auto-hibernate or auto-grind systems mechanically interesting in their own right, go ahead and implement them.

[Gr3yling]

It's misguided because decisions are being made to support realism, with little regard for whether they would make for compelling gameplay.  Realism itself is misguided because DDA has already gone beyond the pale of implausibility.

Okay.  I see what you are saying.  I haven't played DDA, so it's hard for me to comment much about how well or poorly the design elements we are talking about would fit.  All I can say is that what Kevin has said so far does make sense to me.  Although, since I haven't actually played the game, it's hard to know if those ideas would work as well as it seems like they would.

So, are you saying that, in general, it's not reasonable to have a realistic system for character advancement even in a fantasy/sci fi game?  I guess I've never thought about those ideas being contradictory, but I guess you could argue that they are.  I'll have to think more about that.

Simulations aren't inherently bad.  All video games could be seen as simulations.  My point is that there is a difference between simulations designed to be mechanically compelling, simulations designed to support realism (your character must sit in a cave for 3 months because it's cold outside), and simulations designed to cover up bad design (kill 20 bat quests).

If you find a way to make your auto-hibernate or auto-grind systems mechanically interesting in their own right, go ahead and implement them.

All that sounds reasonable to me.  I agree.

[akeley]

Akeley, what do you think should be done in games that have a large scope, like dwarf fortress, for instance?  Aren't you going to have to simulate some parts of that game given how much the player would have to micro-manage otherwise?  Or do you not consider games of that scope to be a true roguelikes?

Also, I'd still like to hear what you think should be done in the case of an adventure that spanned a large portion of the PC's life, where time constraints would make it difficult not to simulate gameplay.

I really feel like there is more of what I would call simulation in just about all roguelikes than you are arguing.  I mean, in ADOM, you don't choose individual types of melee attacks like you can in ToME.  Isn't that a simulation feature?  Doesn’t any game where you don’t micromanage every action the PC takes incorporate some simulation aspects?

While still interesting, this discussion is getting difficult - reason being we start talking about really diverse games, even if it`s all kept within the RL genre. And so,
trying to define what DF is can be a risky territory - vide dedicated thread on this board. But its sim part is definitely a different beast than "normal" RL like Crawl, its sort of hard to compare. In general I`m not against automation in certain genres or particular games - but only when it makes sense and is sort of genre-specific,  lets say in
4X strategy, while in squad based tactics I want to have total control.

As for that ADOM vs TOME example, I see it as a different design decision more than a step towards automation. Not every game has to feature detailed combat, some are more simplistic and still work very well. But I also feel we`re straying bit too much from the OP here, which is what I originally objected to - that cave example. Micromanaging and combat systems in games seem a different subject to me than that (perhaps wrongly, but that`s how I feel).

It sounds like you are saying that we only find things in games fun because there are other, less fun, things are there as a contrast.  Well, does that mean that we should intentionally add “bad” elements in order to make the player appreciate other parts of the game more?

In fairness, I suspect this isn’t what you had in mind, so I’d like to hear more about this idea.

Nah, you`re right this isn`t of course what I meant, at least not so literally. Problem is - and it keeps cropping up in these discussion - my definitions of bad, trivial or boring might vary from other folks`n regarding this subject. As I said in other threads, if I like a game - and there`s many elements influencing that - then those fetch/kill quests (and we lived with this template since RPG beginnings, really) aren`t a problem.

Example (regarding long-spanning adventures too) - I was so immersed in the Fallout 3 gameworld (yeah, I can hear folk sniggering already ;) that I`ve never used the Fast Travel option (meaning literally hundreds of in-game hours). Meanwhile in other games, yes, very much.

Guess, again it boils down to the OP`s very example - I consider  combat and questing to be essential part of any RPG. I can be very forgiving on other fronts, but if these elements suck - or are automated -  I probably won`t enjoy such title.

But, this is makes more sense talking about  normal RPGs -  I struggle to actually employ this line of thought regarding roguelikes. I just can`t think of "boring" encounters in those I played. In Crawl tension`s so high it`s completely out of a question. If I see a "weak" monster, I think "food!" or "what is it followed by?", "any loot?" and so on. Also, once you`ve became strong enough or discovered particular monsta`s Achilles heel, the monster might be a lesser threat but that`s sort of a reward for playing and also a natural flow of things which validates the gameworld and exemplifies your character`s growth.

...or something ;) I like these discussion, but admittedly I`m not a game developer and so might lack a particular incisiveness and drive to completely dissect these issues. I`d sort of like to leave some questions unanswered too perhaps, in order to retain some of the kid-like magic videogaming still holds for me..

Okay, I think I really do understand your point of view pretty well now.  You feel like you can’t experience being your character as vividly if there are simulational aspects, right?  That actually makes complete sense to me.  I actually hadn’t thought of that before you just mentioned it.

You want to really *live* the process of killing of those 10 bats, dismembering them, and stuffing the bat wings in your pack! 

Hmph, well I don`t exactly light candles, keep a styrofoam shield at hand and dress in pyjamas druid robes while playing, but it seems you got that bit ;) My "roleplayin" is somewhere in-between "totally abstract" and "get LARP`d!" styles, probably closer to the former. Kinda vague feeling, especially in RLs. But yeah, if I started automating these vital (for me) parts of gameplay something probably would be lost.

Ultimately, it might be also that thing where not all games are for all people - and there`s nothing wrong with it. Maybe games with "boring quests" style should be avoided by folks who look for something different, and vice versa, instead of trying to enforce a change to the style. Don`t know, really...but I do like both and just sort of alternate between them, depending on mood.

[Xecutor]

You guys took example with bats too literally.
So, let's complicate it a little.
Let's say the character have a quest to bring 1 bat's wing, but in perfect condition.
Sharpshooter can hit the bat in the eye, and thus have unharmed wing.
Sorcerer must avoid elemental spells, but can simply put the bat to sleep and gather the wing.
Stealth character can steal up to the bat unnoticed and cut the wing, still unnoticed.
But brute barbarian with his huge club can only hope that sooner or later he will land a hit in the head and won't turn entire bat into a bloody lump.

[Kevin Granade]

You should design your game around whatever possibility allows for the greatest depth and the least tedium.  Otherwise you're just weakening your game out of a misguided commitment to realism.

So thematics and verisimilitude aren't to be considered?  Sure there are plenty of games where the theme is just a convenient hook for the gameplay to be built around, but that's a pretty narrow vision of what defines a game.

It's misguided because decisions are being made to support realism, with little regard for whether they would make for compelling gameplay.

I put a lot of thought into how changes to DDA will affect gameplay.  Characterizing it as "with little regard" is uncalled for.

Realism itself is misguided because DDA has already gone beyond the pale of implausibility.

Non sequitur.  Just because some elements are "implausible" doesn't mean all elements of the game should abandon any verisimilitude.
The essence of e.g. science fiction is to make certain "unrealistic" changes, and explore how that changes reality.  I fully accept that DDA is off in "space opera" territory rather than anything resembling hard scifi*, but you don't just change the fundamentals of reality because they're inconvenient.  Sure, there are games where you do, but you're stating it's a hard and fast rule for all games, which is overreaching to say the least.

Also I'd like a direct answer to the question of how to handle scenarios that are intrinsically time-intensive, is your answer really "just don't do that", or "make them shorter until they're fun"?

*Having said that, there's a sub-mode that's a straightforward "survive the zombie apocalypse" scenario rather than all the wacky scifi and alien themes.

[miki151]

You should design your game around whatever possibility allows for the greatest depth and the least tedium.  Otherwise you're just weakening your game out of a misguided commitment to realism.

Not every game has to be a logical puzzle. There are many ways to amuse the player. Sure, one is to give them difficult decisions to make. But you can also tell them a story (GTA), let them be creative (Simcity) or let them make their own story (Dwarf Fortress, I guess, although I haven't played it).

[Vanguard]

Okay.  I see what you are saying.  I haven't played DDA, so it's hard for me to comment much about how well or poorly the design elements we are talking about would fit.  All I can say is that what Kevin has said so far does make sense to me.  Although, since I haven't actually played the game, it's hard to know if those ideas would work as well as it seems like they would.

So, are you saying that, in general, it's not reasonable to have a realistic system for character advancement even in a fantasy/sci fi game?  I guess I've never thought about those ideas being contradictory, but I guess you could argue that they are.  I'll have to think more about that.

Believability is a worthwhile goal in any kind of fiction.  I really hate fantasy armor with those gigantic shoulder pads and chainmail bikinis and all that.  It doesn't add anything to the mechanical part of the game, and it makes the other parts impossible to take seriously.  Just because there are space ships and wizards doesn't mean people should stop behaving like human beings.

With that said, it's important to realize that other things may take priority over believability depending on what sort of game you are making.  If your game is heavily mechanics-focused (as most roguelikes are) then realism ought to take a back seat to increasing its depth and cutting away at tedium.  Now, if your game is intended to be a piece of interactive fiction or a lifelike simulation rather than a "gamist" game, it is probably worthwhile to allow a bit of tedium or shallowness to attain greater believability.

Cataclysm is not any good as a piece of interactive fiction or a remotely lifelike simulation.  Cataclysm is a game where you can duct tape three cars together to make an invincible supercar and macgyver up some nuclear hand grenades out of stuff you find lying around.  I'm not saying that as criticism.  Those are good features.  Cataclysm is a better game for having them but they are not realistic in the least.  I'm not sure why the DDA development team feels that player characters ought to be just as susceptible to cold weather as real people when they are clearly superhuman in so many other ways.

So thematics and verisimilitude aren't to be considered?  Sure there are plenty of games where the theme is just a convenient hook for the gameplay to be built around, but that's a pretty narrow vision of what defines a game.

I put a lot of thought into how changes to DDA will affect gameplay.  Characterizing it as "with little regard" is uncalled for.

Also I'd like a direct answer to the question of how to handle scenarios that are intrinsically time-intensive, is your answer really "just don't do that", or "make them shorter until they're fun"?

The second one.  Though not necessarily shortened, just improved in some way until it's fun (or deep or challenging or whatever).  Nothing should ever be added to a mechanics-focused that makes it worse as a mechanics-focused game, even if it's an improvement in other ways.  Making the queen weaker than a knight results in Chess becoming a more realistic but overall much worse game.

If you believe that your new rules for winter survival will make DDA deeper or more fun, then I have misunderstood you.

The reason why I said your change was being made with little regard for gameplay was because of your own words.  You said that the game called for repetition due to thematic reasons and even described it as "tedium."  I wouldn't expect you to couch the idea in those terms if your goal was a more mechanically solid game.  It sounds like you're planning to add tedious and unfun parts to the game for realism's sake, and them letting the player skip those parts through automation.

Maybe you communicated poorly or maybe I misread, but it sounds like you want to want to add a bunch in a bunch of bloat knowing that it won't be fun.