Immediate turn-offs

[Strato Incendus]

I just noticed there are some immediate turn-offs for me when facing a new level, and I wanted to ask what these are for you .

Things that make me reluctant to play a level are:

• first of all: a generic title . This just gives me the impression that the creator merely created something random without putting much effort into it. Sometimes you get surprised and a generically-titled level has a great solution. But I mainly encounter generic titles on the lower ranks, where these really just feel like filler levels.
• or a title that makes some obscure reference to something very niche, where chances of other people getting the reference are very low, compared to e.g. a pop culture reference. I make some of these occasionally, too - and with growing individualism in society, there's probably going to be less and less common knowledge to allude to in this regard . But there's a difference whether a pack is full of e.g. references to movies, songs, or well-known places - or just to, say, some random anime game instead . Too many "insider references" in level titles to me feel like repeated attempts at cracking jokes which nobody else understands.
• if the level is a no-brainer 10-of-everything level which just takes up unnecessary time to click through - rather than a challenge arising from skill restriction
• a level with a really repetitive solution, i.e. that requires 30+ of one and the same skill
• on a similar note: fiddly crowd-control where you have to keep spamming builders, or just a lot of the same skill in general, to stall lemmings until certain tasks have been completed
• levels with an unnecessarily "unergonomic" terrain shape: If you constantly need additional builders just to cross a couple more pixels, rather than the gaps being measured out for a specific number of builders; or if you have to build close to oddly shaped ceilings where the lemming is constantly turning around and coming back just to add a couple more bricks, etc.
• and another related issue is what I would call "relative pixel precision": Pixel precision is fine as long as it affects single lemmings. If a builder has to go in one specific spot, e.g. to cross a gap, allow a climber to still climb up out of said gap from the other side, and make the builder turn around or not, then fine, that's what framestepping is for! The problem arises when two or more lemmings have to be in very specific spots relative to each other, so the skill assignment becomes a game of Mikado. Pixel precision therefore becomes exponentially worse the more lemmings are required for a given maneuvre. But it's also annoying enough if just a couple of pixels decide whether a lemming slips by or not, harking back to the stalling issue.

A lot of packs I've started playing recently seem to go "no-brainer, no-brainer, no-brainer, WHAAAT?!", starting with lots of back-to-back 10-of-everything levels before stumping you right away with the first actual puzzle that shows up.

I'm exaggerating a little here, but putting lots of pointless any-way-you-want-it 10-of-everything levels first, basically creating an entire rank worth of filler levels, and starting out with actual puzzles only from rank two onward or so, is the best way of creating a sudden difficulty spike - which is usually unwanted. Think of ONML, where the entirety of the Tame rank is "whatever you want", and then Dolly Dimple (Crazy 02) just catches you off-guard.

As the Tame rank clearly demonstrates for me, these early X-of-everything levels without a tight skill restriction teach the player jack sh*t. Resource conservation, or the necessity for it in the first place, in my opinion is best taught through puzzles where the most required skills are heavily restricted or even completely absent, so you need to come up with a more creative, different approach to achieve the same goal. Once acquired this way, these skill conservation tactics can then be applied later on to levels with a larger skillset, but where still all the skills are accounted for.

Levels with an abundance of skills rarely feel rewarding to me, because whenever I solve one of these it feels like I couldn't really fail on that level to begin with. But I've talked about that sufficiently in my "case against X-of-everything levels" already .

[Gronkling]

huge levels with loads of skills that have one specific solution

and MEMES

[Proxima]

Your level design posts are always excellent food for thought, and start some discussions that are well worth having, even if I usually end up disagreeing with most of what you say

Generic titles: I agree that it's preferable for a title to have some connection to a level, its looks or its solution, but the level is a pretty small unit, and when you have to title every individual level, they can't all be outstanding. As you say, this is particularly the case in the lower ranks of a pack, where most levels don't have a specific trick to use as a "hook" for the title. So I appreciate it when a designer makes an effort to come up with good titles, but I really wouldn't describe lack of them as a "turn-off".

Obscure references: if you don't get the reference, how do you know the level is making one at all? I guess that if you don't get it, it can feel like just a generic title (like many in the original game -- "Careless clicking costs lives" and "You've lost that Lemming feeling" could go with any levels). In my own levels, I at least try to make the title have some connection with the level as well as being a reference, e.g. "Seven Pillars of Lixdom" obviously matches the level terrain; "Beauty School Dropouts" has a splat fall under the hatch; "The Hotel in Hell" and "The Ersatz Elevator" are both built in horizontal layers like the floors of a building. But most of all I take issue with the idea that there's a set of books, songs and games that are "pop" enough to make reference to, and if I like different things, I'm not in your cool club. Why shouldn't I have fun as well?

* * *

10-of-everything levels: We had a topic on this, and you completely ignored my posts. I agree, the Tame levels don't teach the player the skills they need for solving more complex or restricted puzzles. But the Fun levels do. I gave a breakdown in this post. Most Tame levels have one to three obstacles, each of which is overcome separately by a single skill; in other words, Tame 2-20 teach you nothing more about putting a solution together beyond what you already learned in Tame 1. The Fun levels present much more varied obstacles, requiring much more varied solutions, and you end up learning a lot about how to overcome different types of situation.

I grant that one specific thing the Fun levels don't teach is conserving resources or dealing with the lack of specific skills. But because the Fun levels lay a good groundwork in how to get through the situations the game presents, when you reach the first levels that cut down the available resources, it's a relatively small leap -- whereas, as you said, the leap from Tame to Dolly Dimple is huge.

There is a separate discussion that could be had here. After the topic about using talismans to make packs easier, I had a thought that I wanted to put into a post, but never got round to doing so. Icho says "That's the way of video games for me: Start out easy then get harder and harder until the player has mastered the game." But we're not making a game; we're making expansion packs for a game that already exists. Everyone who plays our packs is at least intermediate level already, and most are experts. Why are we still clinging onto the idea that every pack should have difficulty ranks and that the first should be beginner level?

I don't have an answer to that. But it does feel that most of your gripe with early ranks is that they are too basic for you, even though these ranks are aimed at players at a lower level. Of course, this topic is "what are your turn-offs", so that's fair enough. But lumping all 10-of-everything levels together and calling them all "filler" -- that's unwarranted.

* * *

After that lengthy reply, I realise I should post about what my own turn-offs are, but that will need more thought, and I want to post this so I don't lose it. I'll reply to the actual question another time

[IchoTolot]

Icho says "That's the way of video games for me: Start out easy then get harder and harder until the player has mastered the game." But we're not making a game; we're making expansion packs for a game that already exists. Everyone who plays our packs is at least intermediate level already, and most are experts. Why are we still clinging onto the idea that every pack should have difficulty ranks and that the first should be beginner level?

I disagree with that. My own packs are intended to be games that can stand alone. Technically there are expansions, but even a new player should be able to get into it and be able to solve at least a part of it. They should be able to learn the game even if they only have my pack.

Of course there are packs out there that are built with the expansion idea in mind, but I intent my work to be able to stand alone even if a player only has my pack at their disposal they should still get a decent learning curve and enjoyment out of it and maybe even become a master in the end.

For the general topic:

My turn-offs:

- Hidden stuff -  may it be traps, objects or secret levels
- Glitch/bug levels
- Disjoint unions
- Extremely fiddly levels
- Unnessesary tight timers.
- Any kind of gimmicks or timed bomber stuff.
- Tiny levels where everything feels really cramped together.
- Bare levels with near to no decoration - a clear theme or style can also make up for missing decoration.
- Extremely huge levels with tons of skills.

And generally everything I just solve in a few minutes can be a nice/good level, but won't get stuck in my head. Let's just say you get a thumbs up, but no LOTY/contest vote from me.

[Proxima]

...but you didn't disagree with anything I said.

Sure, you can design your packs that way. I was doing that too, until I abandoned my pack. My point is that there's really no reason why every pack has to be like that. In other communities, such as DROD, this is not the case. I'm really curious now about why it's become the case in the present-day Lemmings community, and why we never really noticed before now how odd this is.

[IchoTolot]

...but you didn't disagree with anything I said.

Of course. For me I don't create an expansion I create a game and for that I also cling to the idea of sorting by difficulty ranks.

My point is that there's really no reason why every pack has to be like that.

Not every pack has to be like that yes, but this way has proven itself over time. + I won't call it odd.

[namida]

It seems to have been something that crept in over time. Back in the days of DOS L1 / Lemmix, the usual form of a pack was a DAT file containing (usually) 10 levels, sometimes a few more or less. There were a few packs even back then that did indeed sort into ranks. However, I suspect the impact of Lemmini (primarily designed to allow playing the official games, so followed their structure) and NeoLemmix (initially didn't get much attention so ended up just following my preferences until it got a bit more popular; by which point the pack > rank > level structure was already well ingrained - although new-formats actually loosens this a bit) was what changed the status quo towards that which we see now.

[nin10doadict]

Kinda funny because I see myself creating a lot of levels that are doing just these things people seem to hate.
I am trying to earn a reputation as "that jerk that makes the troll levels" though. Why, I'm not even sure myself. Perhaps so it will make my levels that aren't hot garbage look better by comparison. Don't get me wrong, I know troll levels are awful. That's why I tend to limit them to one or two in a pack, and try to be creative about how I do them. (Misdirected effort at its finest ) The Troll rank in Casualemmings was just me getting pent-up trash out of my system.
I do tend to make levels that are smaller and feature only a few obstacles, with fairly small skillsets. They just feel far less overwhelming, and it's not like it's impossible to make hard levels with these conditions. Plus, they're easier to decorate. I usually don't like leaving too much empty space in the level because that feels sloppy.

[Nessy]

The first thing that caught my eye in this thread was "Generic Titles" because I personally take level titles very seriously (let's just say that the fact that "All New World Of Lemmings" doesn't have level titles like in Lemmings 1 was a HUGE turn-off for me personally). So let me go into detail about it, shall I? I take them so seriously that sometimes I have levels that don't get a name assigned to them until weeks after they are created. I'll even be honest and say that sometimes I still don't have a title so I come up with something random. I can admit that. This is the case even with my later rank levels. I feel like level titles can be just as important as everything else about a level. Level titles are sometimes the first thing that a person may see about a level. I tend to always get hooked on a level just be someone telling me the name of it even BEFORE seeing any other type of preview of the level. Not always, but I have experienced this at one point.

What the level title is or means is also very subjective in my mind. I don't really mind if a level title makes an obscure reference and in fact I prefer it as learning the meaning later on down the line one way or another gives the level a whole new meaning in my eyes. Even if a person doesn't understand the reference they won't really lose anything through it and it will just be a regular title for them. In fact, when I come across a title that seems to be a reference and I never figure it out, I would still think that the title is so unique and it would stand out more than a reference to something more popular. Of course, I am NOT saying that is a bad thing. I'm just saying that a level with a title referring to something I have never heard would stand out more in my head than a title that references popular expressions let's say.

Another thing that I disagree with is that level titles need to always relate to the level somehow, but this is a very gray area for me. A more proper way of explaining this is to just say that I basically don't like making my level titles relate to the level directly. For example, I'll use one of my own levels as an example: "Six Days Without An Accident". On the surface the title has NOTHING to do with the level, but under the surface? The "Six Days" part is 100% random, but the level deals with climbers that can easily climb over obstacles and die. This is the "Accidents" the level is referring to. I think it stands out more decent without having to relate it to the level too literally. Too literally would be naming the level "Climber Accidents" or something. Which one do you think has more of a punch?

But anyway getting back on topic I don't really have any turn-offs when it comes to other people's levels. I know that probably isn't the best attitude sometimes, but I honestly haven't encountered anything that has made me go as far as to post it here.

[Strato Incendus]

Your level design posts are always excellent food for thought, and start some discussions that are well worth having, even if I usually end up disagreeing with most of what you say

Haha, thanks! Yeah, I was certainly expecting you to disagree with regards to X-of-everything levels (see below).

Generic level titles are something I'm guilty of myself, especially with regard to Pit Lems, because mechanics were pretty much the only thing driving this one. That's why I wanted to have an overarching theme for my next pack, and that's how the idea to Lemmings World Tour was born .

10-of-everything levels: We had a topic on this, and you completely ignored my posts. I agree, the Tame levels don't teach the player the skills they need for solving more complex or restricted puzzles. But the Fun levels do. I gave a breakdown in this post. Most Tame levels have one to three obstacles, each of which is overcome separately by a single skill; in other words, Tame 2-20 teach you nothing more about putting a solution together beyond what you already learned in Tame 1. The Fun levels present much more varied obstacles, requiring much more varied solutions, and you end up learning a lot about how to overcome different types of situation.

I grant that one specific thing the Fun levels don't teach is conserving resources or dealing with the lack of specific skills. But because the Fun levels lay a good groundwork in how to get through the situations the game presents, when you reach the first levels that cut down the available resources, it's a relatively small leap -- whereas, as you said, the leap from Tame to Dolly Dimple is huge.

Far from it, I was very aware of your post when writing this . It should be noted that my criticism was pointed mainly towards the X-of-everything levels I get to see in custom packs. If you like the 10-of-everything levels specifically how they are done in the Fun rank, I can see where you're coming from; most X-of-everything levels from custom packs however unfortunately remind me more of Tame than of Fun.

That said, one thing I keep pointing out repeatedly which Fun doesn't teach the player either is standard skill tricks. The three builder wall has to be figured out by the player himself on "I have a cunning plan", which also does a terrible job at isolating that trick, because it is also possible to block and free the blocker with a miner (which is how my father and I always tried to solve the level back in the day, running into execution difficulties easily, even though we knew it was possible). So you can't even claim the redeeming factor of "the player is bound to figure it out at some point" here. Digging and building to turn a lemming around has to be figured out on "Postcard to Lemmingland"; trapping lemmings in a digger pit on "Heaven can wait (we hope!)".

And sadly, many packs waste the player's time with standard tutorial levels for the different single skills - a type of level which you get to play loads of with every new pack you attempt - but then proceed to demanding obscure tricks from the player; some of which most of us probably know less from actual playing, and more from browsing the forum or from watching replays.

But most of all I take issue with the idea that there's a set of books, songs and games that are "pop" enough to make reference to, and if I like different things, I'm not in your cool club. Why shouldn't I have fun as well?

That's not what I'm proposing; I certainly don't want there to be an authority on what's "hip" enough, hence my comments about individualisation and a smaller degree of overlapping knowledge among users . Take the biblical quotes for example that make up a majority of the level titles in the "Basic" rank in "Lemmicks": Arty understood them, of course, but overall, in the more secular world we live in today, there are going to be less people who understand those references than a couple of decades ago.

I simply suggest: When you're picking a level title, and you're deciding to include a reference to anything, ask yourself how high the chance is for anyone from our comparatively small community to get that reference.

For example, with the Game of Thrones-allusions I included in Paralems, I thought that probability was reasonably high. There have still been people who didn't get those references, and that's completely fine; I simply didn't want to refer to anything from my more "niche" interests that was bound not to be understood to begin with.

Other creators have made Star Wars- and Pirates of the Carribbean levels, too, and those are easy to identify as such.

Hence, I selected the level / song titles for Lemmings World Tour by the criteria of overall popularity, the degree to which the title matched the level mechanically or visually, and only afterwards came personal preference. Many of the titles are songs I have never listened to; but if I had based the titles on my personal favourites, it would have been full of lesser known metal bands and Eurovision Song Contest entries .

While the latter would have been a great way to visit lots of "places" and put many different flags in my levels, not only would it have turned Lemmings World Tour more into a "Lemmings Europe Tour", but also the whole purpose of having song-based titles would have been defeated, because to the majority of players, these simply would have been generic titles like on any other level.

A little mystery can be nice, of course, but I usually disperse it at the latest in the post-level text as a "reward" for solving the level. It's quite weird to have to go to a YouTube LP of the pack you just played and look for the creator's comments under some video, just to get all the secret references.

But perhaps this is just because, since I'm generally more on IchoTolot's page with regard to "a pack should be able to stand on its own", that translates to the level titles for me, as well .

I do agree with Nessy that strange level titles can open up the doors to spark your interest in new things. But whenever that is my goal, I try to actively invite the player into "my world". Unexplained insider references often have an "excluding" aspect about them.

SEB Lems is a good example of this. From the pack name to the choices of the ranks to the level titles to the music choice, everything makes sense if you know what it's about, but if you don't, at least to me, it just seems like a bit of a random mess - not with regard to the level quality, Flopsy , just with regard to the overall structure of the pack, especially considering the bonus ranks with removed levels, and levels added in from other packs. For example, the rank names don't have that typical "increasing intensity" fashion, they're a seemingly arbitrary mixture of nouns and adjectives.

And since what this pack is actually about is kept a secret until the very end of the pack, when the shoe finally drops, at least for me, it was a bit like: "Really? That's all?"
(I actually found out about it by accident while typing a forum post.)

But lumping all 10-of-everything levels together and calling them all "filler" -- that's unwarranted.

I've said in the other thread you linked to that my main gripe with these levels is how they are created :

In the case of the Fun levels, most of these are "pre-runs" of later levels which were deliberately made easier. So actually, the more difficult version was created first, and then reducing the difficulty just to have another level for a lower rank really just seems like filler to me. As if the creator couldn't be bothered to come up with separate easier puzzles and decides to go for a 2-for-1 instead, simply because they don't know what else to do for the first rank.

Making an easy 10-of-everything version of an initially more restricted puzzle requires just a couple of seconds of additional work for the creator (basically just setting the 8 standard skills to "10"), while taking the player disproportionally more time, forcing him to click through a - usually very large - level that barely has anything interesting to offer or to teach, there's little to be lost and little to be taught:

One reason for the difficulty spike in Tricky is that Fun reinforces "noob strategies" in these 10-of-everything levels, mainly by providing an abundance of blockers and builders: New players become completely dependent on the idea of trapping the crowd with blockers, because it's always possible in Fun, and then they don't know what to do when it's suddenly not possible anymore .

Why are we still clinging onto the idea that every pack should have difficulty ranks and that the first should be beginner level?

Well, that sounds like you advising against X-of-everything levels now, doesn't it? Because they usually take exactly those beginner level spots in the larger packs, sometimes they make up an entire rank. So if you leave them out, the density of X-of-everything levels is going to be reduced drastically.

The simple answer is: We certainly don't need to have beginner levels in every pack, and smaller packs like "Lemmings Migration" or "Raylems" get started with more puzzly stuff right away.

"Six Days Without An Accident". On the surface the title has NOTHING to do with the level, but under the surface? The "Six Days" part is 100% random, but the level deals with climbers that can easily climb over obstacles and die. This is the "Accidents" the level is referring to. I think it stands out more decent without having to relate it to the level too literally. Too literally would be naming the level "Climber Accidents" or something. Which one do you think has more of a punch?

I'm completely on-board with you here, Nessy ! Actually, "Climber Accidents" would be the more generic title to me, simply because that feels like forcing a game-mechanic word into it. Instead, you picked one which clearly had some relation to the level (I was just thinking of a mine or steel work, due to it featuring the Machine tileset), but at the same time didn't give anything away about the solution.

The Troll rank in Casualemmings was just me getting pent-up trash out of my system.

@nin10doadict: Actually, some of these levels can be quite fun when you have to rapidly spam one specific skill over and over again. I take more issue with repetitive levels when they're very slow on top of that.

The classical example would be builder fests, because you have to babysit the builders and constantly hammer on the spacebar, waiting for them to finish. Or where you let climbers climb into a shaft and then bomb - you always have to wait for the next climber to arrive first. So it's actually more levels like that one massive sewer shaft level from CasuaLemmings, with a bunch of crocodiles and 50+ stoners, that tend to get dragging for me here .

In contrast, when I just need to select a skill and then frag my mouse to bits, I can pretty much do all these clicks quickly, back-to-back, and continue immediately.

Generally speaking, a level becomes more dragging and boring the more time there is between figuring out the solution on a conceptual level, and putting that solution into practice. That's the whole point of the NeoLemmix community trying to reduce execution-based difficulty.



I do have another turnoff though:

• terrain that's deliberately visually confusing (again relating to ergonomics here), lots of small, repeating elements that are intentionally designed to make the player's eyes hurt - specifically: Mazes of tiny terrain stripes where you can barely spot where the lemming is going to walk or fall through.

Contrary to "easier versions of later levels with X-of-everything", the burden is equally distributed here on creator and player: These huge mazes are usually just as tedious to create as they are to play. I had to create one such level for the Groupie rank, because imitating what other people have done is the whole point of that rank, but I still despise these levels in general, both as a creator and a player.

I've spoken before about "hiding in plain sight" vs. "hiding in obstructed sight": Puzzling is about "which piece goes where". A level is a map with several obstacles, where you're cycling through in your head how to overcome each of those obstacles. My greatest admiration has always gone to levels where you think "This looks so simple, how is it possible that I'm always one skill short?" (Examples: Nepster's "The Block-Store" or "A Study in Scarlet")

Mazes of tiny, pixel-thin terrain stripes however are designed to make your eyes lose track every time you look elsewhere in the level.

Making terrain visually deceptive this way to me really isn't much better than making traps visually deceptive.

Oftentimes, these levels even look more complicated than they actually are (I'm going to cite Colorful Arty's "Labyrinth of Lucifer" from SubLems or Nepster's "Don't cross me!" as examples here).

Hence, these levels are the exact opposite of the aforementioned, admirable ones:

They pretend to be complicated, but actually aren't.

Genius levels look like they're simple, but actually aren't.

And thereby make the player feel rightfully stupid for not being able to see the solution .

[Proxima]

That said, one thing I keep pointing out repeatedly which Fun doesn't teach the player either is standard skill tricks. The three builder wall has to be figured out by the player himself on "I have a cunning plan", which also does a terrible job at isolating that trick, because it is also possible to block and free the blocker with a miner (which is how my father and I always tried to solve the level back in the day, running into execution difficulties easily, even though we knew it was possible). So you can't even claim the redeeming factor of "the player is bound to figure it out at some point" here. Digging and building to turn a lemming around has to be figured out on "Postcard to Lemmingland"; trapping lemmings in a digger pit on "Heaven can wait (we hope!)".

All three of those levels can be done without the tricks you mention, and indeed that's how I originally solved them when I was a kid

Then again, when original Lemmings was released, it was a one-off, not a ground for ongoing puzzle creation. Tricks we now call "standard" like the three-builder wall didn't need a better teaching level, since that was the only level using the trick.

And sadly, many packs waste the player's time with standard tutorial levels for the different single skills - a type of level which you get to play loads of with every new pack you attempt - but then proceed to demanding obscure tricks from the player; some of which most of us probably know less from actual playing, and more from browsing the forum or from watching replays.

I think here, as in the "Levels relying on tricks" topic, you underestimate most players' ability to work out new tricks by getting stuck and then trying things out, or just thinking about an impossible situation and realising what must be the solution. I know I've solved several levels that way, and these have often been the most satisfying and memorable ones, like "Attack of the Subconscious" or "The Gr8 Escape".

Hence, I selected the level / song titles for Lemmings World Tour by the criteria of overall popularity, the degree to which the title matched the level mechanically or visually, and only afterwards came personal preference. Many of the titles are songs I have never listened to; but if I had based the titles on my personal favourites, it would have been full of lesser known metal bands and Eurovision Song Contest entries .

I glanced again through your preview topic to see how many titles came from songs I'm familiar with, and my success rate was a little under 50%. And that doesn't bother me at all.

Also, at least for me, sometimes a title is just so right for a level that it doesn't matter if no-one gets the reference, I just have to use that title:

Spoiler

"In the Gallery of Frost"

A little mystery can be nice, of course, but I usually disperse it at the latest in the post-level text as a "reward" for solving the level. It's quite weird to have to go to a YouTube LP of the pack you just played and look for the creator's comments under some video, just to get all the secret references.

If I were to finish a pack, I'd post a list of references either in a spoiler tag in the first post, or an attached text file. Namida does this too.

[Strato Incendus]

I glanced again through your preview topic to see how many titles came from songs I'm familiar with, and my success rate was a little under 50%. And that doesn't bother me at all.

Hey, that's not bad at all, actually! I'm certainly not aiming for every player to know all the songs, I'm aiming for maximum possible overlap, considering the comparatively small size of our community. Plus, just like you, when a specific level absolutely calls for a certain name, I'll use that one, even if the related song is not very widely known.

For example, I don't know of any famous song called "Munich", but there is one called like that by The Fray, which are a little more famous for other songs, and Munich is a place which I was able to capture in level form - so in this case, the "geography" criterion trumped the "famous" criterion.

I have just decided not to use any titles in the main pack that are not references to any songs at all. Meaning: Even if you don't know a song, you don't have to keep guessing what I'm referring to; you can be pretty certain that if you simply put the level title into YouTube or the iTunesStore , you're going to find it, rather than having to scour the entire web for an answer via Google.

I'm still a little salty that I couldn't figure out what "I am A. T." from the Sunsoft rank is supposed to refer to. Unless it's a really stupid word-play for "80", which is the number of lemmings you have to save. But in that case, why "I am"? Why not simply "Save A. T." or something along those lines?

Then again, when original Lemmings was released, it was a one-off, not a ground for ongoing puzzle creation. Tricks we now call "standard" like the three-builder wall didn't need a better teaching level, since that was the only level using the trick.

Okay, that's a fair point: DMA probably thought they would have spoiled the solution if they had introduced this trick in an earlier level.

My main issue with tricks is when the player can't even know they're possible, or is even led to believe the opposite.

For example, digging and building to turn a lemming around is something you can figure out for yourself by putting 2 and 2 together: You quickly learn that building into a wall turns a lemming around, and a digger obviously creates a wall to both sides of himself.

The three builder wall however is different, because in the beginning you learn that lemmings can go through staircases, and even if you put two staircases perfectly behind each other, they also still manage to slip through that. So you could just as easily believe as a new player that builder staircases were permeable from below in general.

If I were to finish a pack, I'd post a list of references either in a spoiler tag in the first post, or an attached text file. Namida does this too.

That's a great idea, actually! I did a list of music tracks for Paralems a year ago, for people who wanted to play the levels privately with the copyrighted music I had intended for each level, but I still kept the allusions somewhat secret in that post.

In the case of Lemmings World Tour, though, that list of references would probably contain the entire level list ...

For your example level: Yes, the reference escapes me, but I can still see why the level is called that way - those two sections on the outsides look like paintings frozen over, and the little frame with the water inside could be a small panorama picture. There's simply no further explanation needed here, and the title still doesn't feel generic. It actually reminds me of Arty's level building style, where I often just think "hey, cool idea to paint something like that with level tiles!", and then the title usually falls into place anyway.

I hope you're going to continue on GemLems some time in the future, these snapshots certainly look promising!

[Proxima]

I'm still a little salty that I couldn't figure out what "I am A. T." from the Sunsoft rank is supposed to refer to. Unless it's a really stupid word-play for "80", which is the number of lemmings you have to save. But  in that case, why "I am"? Why not simply "Save A. T." or something along those lines?

"A. T." was the "Total Director" of the Genesis version. You get to find out in the credits, which is only one more level after "I am A. T."

The three builder wall however is different, because in the beginning you learn that lemmings can go through staircases, and even if you put two staircases perfectly behind each other, they also still manage to slip through that. So you could just as easily believe as a new player that builder staircases were permeable from below in general.

That's an interesting point. It's all very hazy now, but I'm sure that when I first played Lemmings, now and then I "deduced" things about the game mechanics that turned out not to be true. I can't remember any specific examples right now

I also don't remember when I first learned about the three-builder wall. Probably in the days of the old forum, Cheapo, and challenges on the original levels. It makes sense, because the lemmings run into a stack of 7 pixels; but "lemmings can ascend 6 pixels but not 7" is also a trick that the original game never teaches.

For your example level: Yes, the reference escapes me

"Take This Waltz" by Leonard Cohen. For bonus points, the loop of balls represents the "garland of freshly cut tears".

[nin10doadict]

In the case of the Fun levels, most of these are "pre-runs" of later levels which were deliberately made easier. So actually, the more difficult version was created first, and then reducing the difficulty just to have another level for a lower rank really just seems like filler to me. As if the creator couldn't be bothered to come up with separate easier puzzles and decides to go for a 2-for-1 instead, simply because they don't know what else to do for the first rank.

I like to avoid that. I tend to create the easy version of a level first, and then think "hey, if I tweak this and this and change the layout a little bit, now I can try to come up with a harder solution." I try to avoid having repeats that are exactly the same layout-wise, just similar enough that you can tell they were based on the same level. "Do it to it" and "Get me outta here!" in Squared are one example.

[Nepster]

My biggest turn-offs are:
1) Hidden information. It doesn't matter it it's traps, exits or anything else. A level should give me all the information they need and as directly as possible.
2) Unnecessarily huge level. And "aesthetic reasons" do not count here in my opinion, because we are creating levels to be played not pictures to be admired. Instead I am talking about: If a level's solution would have worked with half the space just as well, then I am fed up. This goes especially for vertically scrolling levels.
3) Levels that don't tell me that a certain approach doesn't work. It is totally fine if I need two or three tries to determine that a certain approach doesn't work. But if there are two dozen ways to place three builders in a pit to get out of it, most of them fail only by one or two pixels, but none actually succeed, then this is bad! Because in such a case the level tells me "you are on the right track", which is actually not true at all. My typical reaction to that is: Why isn't the pit slightly deeper, so that I didn't have to waste all my time there?
4) Drops that are almost or just deadly, gaps that can almost bridged, ... While I will know it after testing it once, I do have to spend time determining that. If I can see a reason why this is necessary, then fine. But otherwise this comes back to point 1): I want to be able to extract all necessary information easily. 
5) Builder heavy levels.

• first of all: a generic title . This just gives me the impression that the creator merely created something random without putting much effort into it. Sometimes you get surprised and a generically-titled level has a great solution. But I mainly encounter generic titles on the lower ranks, where these really just feel like filler levels.
• or a title that makes some obscure reference to something very niche, where chances of other people getting the reference are very low, compared to e.g. a pop culture reference. I make some of these occasionally, too - and with growing individualism in society, there's probably going to be less and less common knowledge to allude to in this regard . But there's a difference whether a pack is full of e.g. references to movies, songs, or well-known places - or just to, say, some random anime game instead . Too many "insider references" in level titles to me feel like repeated attempts at cracking jokes which nobody else understands.

This is very interesting, because I myself rarely read level titles and don't care about them at all. However your second point puts me in a very difficult spot:
1) I may not use generic titles
2) I may not use the references I know, because they are usually pretty obscure
3) I can not use any pop-culture references, because I simply don't know them. For example out of all the titles in the "Lemmings World Tour" thread, all except two are totally mysterious to me. Another example are "Pirates of the Carribean", "Star Wars" and "Game of Thrones" you mentioned earlier - I have seen none of these "classics" so am really missing out on opportunities for pop culture references here.
So I really ask myself: How I should name levels in the future?

if the level is a no-brainer 10-of-everything level which just takes up unnecessary time to click through - rather than a challenge arising from skill restriction

All this discussion about 10-of-everything levels gave me one idea: Would 5-skills-with-10-of-each levels be better? At least it would provide a bigger variety regarding the skill selection available. Otherwise I would be hard-pressed to create levels that fill the gap between "X-of-everything" levels and actual puzzle levels.

Mazes of tiny, pixel-thin terrain stripes however are designed to make your eyes lose track every time you look elsewhere in the level.
Making terrain visually deceptive this way to me really isn't much better than making traps visually deceptive.
Oftentimes, these levels even look more complicated than they actually are (I'm going to cite Colorful Arty's "Labyrinth of Lucifer" from SubLems or Nepster's "Don't cross me!" as examples here).

Interesting, because I love mazes with thin terrain (you probably noticed that there are several of them in NepsterLems ). As long as they have some kind of structure, I don't find them confusing at all. I would have much more objections when everything is two or three times as big... But good to know about your preferences.