Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[GavJ]

Yeah I'm not following how that would be possible.  A moment in time is instantaneous. You can't learn anything until it's not that moment in time anymore...

I have maybe a vague idea of what you mean, if by "moment in time" you mean something like "current task or current game pause" or something... ?  but right at the beginning of the game is when, if you want a successful fort, you have to do all of the MOST complicated stuff.  Setting up all the skills enabled usefully, giving any nicknames you want, designating and designing the fortress in a way that won't bite you in the ass later, considering starting your military training, getting seeds in the ground, blah blah blah.

So if there is any one point in time you need to figure out and apply the most stuff all at once, it's right at the very beginning, meaning the graph, if anything, would start super high, then taper down, not terribly steeply, until you have a mostly established, self sustaining fort. Where would the steepness be?

[EpicOrange]

I found a blog post that might explain all this:
http://www.grammarphobia.com/blog/2009/07/steep-learning-curves.html

If you don't want to read it: The term 'steep learning curve' usually means the opposite of the literal meaning.

[Playergamer]

Wait, no, now I remember. It's not about what you learn, it's about the difficulty of learning it.

Steep learning curve: It's easy for a bit, then it's extremely hard.

Shallow learning curve: It is easy to learn the whole time.

[Egan_BW]

It works if the Y axis is skill required to progress and X is progress made.

or something.

[ShadowHammer]

I support the steep learning curve idea; I think GavJ's argument is based on mislabeled axes. The traditional graph uses "time passed" as the independent X axis, and "skill required", not "total skill gained" as the dependent Y axis. Therefore, a steeply rising curve shows that a great quantity of skill is needed in a very short amount of time.

The graph that Lyeos posted

You mean like this graph?

Spoiler (click to show/hide)

is actually very accurately describing this: it rises absurdly steeply at first, to the extent of going backwards, figuratively exemplifying the fact that sometimes you simply can't win within the laws of the universe, but then flattens out. Even on the flat, though, there are stickmen being bulldozed off, metaphorically showing that no matter how safe and good at the game you think you are, the game can and will still kill you.

First part of my argument was ninja'd by Egan_BW.

[GavJ]

I support the steep learning curve idea; I think GavJ's argument is based on mislabeled axes. The traditional graph uses "time passed" as the independent X axis, and "skill required", not "total skill gained" as the dependent Y axis. Therefore, a steeply rising curve shows that a great quantity of skill is needed in a very short amount of time.

1) No, the traditional graph does not use that. It uses skill gained. This is an actual psychology term used for legit research, and that's the meaning of the curve
http://en.wikipedia.org/wiki/Learning_curve
although if you insist I could actually go take the time to look up some journal articles. I've read them before, but it was ages ago in undergrad. They use the same thing the wiki article does. It's learning. I.e. cumulative learning / understanding gained so far, as in my original graph in the OP.

Note, especially, the line "The familiar expression "a steep learning curve" is intended to mean that the activity is difficult to learn, although a learning curve with a steep start actually represents rapid progress."

2) "Skill required" is super vague, and could mean various mathematically distinct things:
Do you mean "Skill required to master the game?"  That would be a flat line, at all time points, as it is a fixed amount as a goal, and doesn't change as you learn.
Do you mean "Skill required to survive until the next tick?" That would be a crazy up and down constantly zig zag mess of a useless graph that you couldn't interpret anything from. Because some ticks, you don't need to do anything to survive til the next tick. Other ones, you may need to do a dozen crazy things. Then nothing again...
Or do you mean something else? Please operationalize this much more precisely. I am unable to interpret it in a way that would actually lead to the curve shape anything like what you're describing.

Wait, no, now I remember. It's not about what you learn, it's about the difficulty of learning it.
Steep learning curve: It's easy for a bit, then it's extremely hard.
Shallow learning curve: It is easy to learn the whole time.

The difficulty of learning "it." What is "it?" The next quantum bit of knowledge? Okay, that could be graphed, but like the above, seems very unlikely to be a smooth curve. As you learn, some things that come up are super super easy. Like, I dunno, setting up a hospital. It's just a zone and then you hit "H"  But you don't do that until pretty far into a fort, so it's midway along the X axis. With things more difficult than it on both sides.

In other words, this would also be a crazy mess of a zig zag line with no fluid continuity to it. Not a nice smooth steep slope. I don't even see any reason to believe in dwarf fortress that the moving AVERAGE of this would be a steep slope, or even a slope at all. What makes you think stuff in DF gets harder to understand as you go? That was not my experience at all...

[Orange Wizard]

This is one hell of a semantics argument.

is actually very accurately describing this: it rises absurdly steeply at first, to the extent of going backwards, figuratively exemplifying the fact that sometimes you simply can't win within the laws of the universe, but then flattens out. Even on the flat, though, there are stickmen being bulldozed off, metaphorically showing that no matter how safe and good at the game you think you are, the game can and will still kill you.

This gets my +1 though, it's not about the axes or the maths, because you can't quantify difficulty. (Or maybe you can, but 7 could hard or easy for different people) What you can do is describe it with words, which despite being inaccurate at best, are still the only way you can describe this thing without just being difficult.

Please operationalize this much more precisely.

Stop being difficult, GavJ. For me?

[GavJ]

you can't quantify difficulty.

Of course you can, but the spirit of what you're saying, sure. It's hard to quantify it.
...Which is why learning curves DON'T attempt to quantify difficulty, when you use the proper axes >.<

Argument here seems to be:
"Oh you're using the wrong axes, Gav. You should be using difficulty."
"Oh also, difficulty is impossible to measure."
Uh. That's why I didn't use it / why would YOU use it, if it's impossible to measure?

[Salmeuk]

Q: I recently came across a blog post about a struggling rookie football player “experiencing a steep learning curve.” The author clearly intended to say the athlete faced a long, difficult learning process. However, it’s my understanding that a steep learning curve would depict learning something very quickly. Why not replace this inaccurate cliché with “steep hill to climb”?

A: Technically, you’re right. A learning curve is a graph that represents the rate of mastering a skill against the time required to master it. A steep learning curve would therefore show the quick and easy mastery of a skill.

The lexicographers at the Oxford English Dictionary, The American Heritage Dictionary of the English Language (4th ed.), and Merriam-Webster’s Collegiate Dictionary (11th ed.) define the term “learning curve” in pretty much this way.

But the people who actually speak and write the English language generally seem to use “learning curve” in a figurative way that has little to do with its technical meaning.

Thus, a “steep learning curve” in common parlance refers to the difficulty of learning something.

In a search of the New York Times archive, I found 108 references to “steep learning curve” since 1981. I checked out a few dozen, and all of them used the expression in the “inaccurate” figurative way.

I suspect that the lexicographers at my favorite dictionaries are a bit behind the curve here, and that they’ll eventually catch up with the rest of us.

This is the article linked above, quoted in entirety (except the awful pun at the end).

GavJ, you can't expect anything good to come from throwing up a paint-drawn graph and a hundred words expressing frustration at others usage of a term. Are you familiar with the concept of semantics? Despite being defined as one thing, a word can be used in any number of ways. You don't lose nerd cred (is there such a thing? I imagine such a concept would only exist in those concerned with outward appearance) for accepting a 'popular' definition. Especially in this non-professional context.

Are you really getting frazzled over a common misconception?

[Uronym]

I believe it's related to how in "traditional" games, there are "levels". Plot the level on the X-axis, and plot the difficulty on the Y-axis. In these sorts of games, "level" and "time" would be roughly the same metric; however, it is definitely a technical error to label the X-axis with "time". Levels, being essentially encapsulated challenges, could be said to have an overall "difficulty"; that is, cumulative skill required in order to complete the level, or, conversely, how difficult the level might be to a completely inexperienced player. As previously stated, "difficulty" is very difficult to measure, but these sorts of graphs are, in this particular context, used by developers and players to visualize something briefly and concisely.

In Dwarf Fortress, "levels", or our X-axis, might refer to, say, the in-game calendar relative to when you embarked, as there are no "levels" in the game.

Yes, the completely flat line would be a roughly accurate plotting of this for Dwarf Fortress. However, if you assume that the "0" position on the X-axis is before you started playing (thus always being 0 on the Y-axis as well, as you need no skill to do nothing), it shows an impressive "cliff", or very steep slope, in any case. This is very much more illustrative than a flat line, and probably at least one way to rationalize the "steep learning curve" idiom.

It is also important to remember that not everyone is using the psychological definition of "learning curve"; it has a more colloquial meaning, too.

[Henny]

I'd say Dwarf Fortress has a steep learning curve if the factor being considered is "amount of things I don't understand".

[catpaw]

So the "lowest common denominator" to all this is, sometimes idioms are not semenatically correct but we understand them because they are idioms, and the lowest common denominator is always one.

[Playergamer]

Yeah, if everyone uses a phrase that way, then that is what it means.

*actually gone now*

[Xazo-Tak]

I don't care that it's only May.
Thread of the year.

[GavJ]

Ok. In other news, I'm gonna start up a colloquial usage of "proton" that actually means "electron" and vice versa. Physicists can just shove it.