Should we have guidelines for the more subjective parts of pickin'?

We have some good guidelines on what does and does not constitute a good image; A good image must demonstrate the trope to people who don't know the context, it must be an example in the source, it should contain visual elements that makes it better as an image than a quote. It should not contain a copyright sign unless the copyright owner is shown as well, it should be SFW, and it should not be a lolcat or a demotivator. Correct me if I forgot anything.

However, in a lot of threads, several images that pass all these rules are suggested. The ensuing discussions tend to contain Fan Myopia, and quite often go off on semi-related tangents. There are also several recurring unofficial guidelines that get brought up and debated. Couldn't we discuss and establish a set of official ones specifically for deciding between otherwise equally qualified images? I think it would make the image pickin' process faster in the future.

Below follows a list of such issues I've encountered, and the pros and cons I could think of. If you think establishing such guidelines is a good idea, we can start debating them. My personal opinions are not there yet, as I thought I'd wait to see the initial response to the idea first.

• Should Anime images be discriminated against? - This is the first issue that comes to mind, as it has sparked several flame-y debates. Anime images are easy to get (as an anime screenshot tends to be much less blurry than a live-action media one), which results in a disproportionally large amount of suggested Anime images. I remember reading somewhere that the number of accepted page images from the medium is acceptable. It could on the other hand be argued that continued discrimination is required to keep that balance. I also remember reading that Fast Eddie supports such discrimination.
• Should iconic or obscure examples be preferred? - This is a more uncommon, but still tricky, issue. On one hand, it seems like a good choice to demonstrate a trope with the Most Triumphant Example. On the other hand, choosing images from obscure and underrated works is a good way to promote them, and gives TV Tropes more visual variety.
• Should overrepresented works be discriminated against? - There are very many trope images taken from lampshading-prone webcomics such as XKCD and Order Of The Stick. This is reasonable, as that's pretty much the only way to find good images for the more complex, abstract tropes. However, suggestions from those works occasionally pop up in threads for more easily picturable tropes as well. Should they be discriminated against in those cases?
• Should underrepresented media be prioritized? - Webcomics, comics, manga, anime and western animation produce a lot more useful images than literature, live action TV, films and video games. In the rare cases that one of the underrepresented mediums produces something useful, should it be prioritized for variety's sake?
• How should literal examples be treated? - There are several tropes with names like Underground Monkey; A specific example demonstrates a phenomenon. If two equally good images are suggested, one of which is a literal example, - Say, one picture of an underground monkey and one of a polar wolf - which one should be chosen? One one hand, the literal example is more 'perfect'. On the other hand, a less literal example is less likely to cause trope misuse.
• Should Real Life pictures be discriminated against? - It seems pretty established already that use of a trope in a work goes before real life examples of the same trope when choosing an image. I'm just bringing it up so that if we make a list of such guidelines, this one isn't forgotten.

edited 2nd Jun '11 6:21:05 AM by ArtisticPlatypus

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.

Nice thread idea, these things definitely need to be addressed:

Should Anime images be discriminated against?
No. Anime and manga pics work well as trope images for precisely the reasons you mentioned, and I don't think we should go out of our way to avoid using them just because of a perceived overemphasis. If we have options from multiple media to choose from, the merits of the individual pic will be what makes it be chosen for a page.

Should iconic or obscure examples be preferred?, Should overrepresented works be discriminated against?, Should underrepresented media be prioritized?
These all can be addressed with one answer: it all comes down to how well the pic in question illustrates the trope. I have no problem with mining Troperiffic works for pics if they work better than other suggestions, and I really don't have a preference between iconic or obscure along the same line...now, if we have two pics that do an equally good job, then I'd give the more obscure/underrepresented work a very slight edge.

Should Real Life pictures be discriminated against?
I tend to prefer pics from media, but in the case where a RL pic works really well, I'm all for using it...don't we have a trope page that uses that infamous pic of the Buddhist monk immolating himself during The Vietnam War?

edited 2nd Jun '11 7:05:29 AM by Willbyr

These are all judgments calls. The better way to approach the questions is "What are the principles involved?"

Medium selection: The principles involved is that we don't want to look like we are in a particular fandom ghetto. Best way to do that, since images are fandom-identifying, is to have a good distribution of image selections from different mediums.

All the remaining questions ... The principle is that we are illustrating the trope. Whatever does the best job of illustrating the trope to the widest number of people is the good pick. Photo, animation cel, film cel ... whatever.

edited 2nd Jun '11 8:46:44 AM by FastEddie

Goal: Clear, Concise and Witty

I don't feel good about written guidelines concerning this points. It is OK when someone uses this guidelines as personal decision help. E.g. I prefer non literal examples over literal ones and I generally prefer work examples over real life examples, but this opinion is flexible and so I can agree on a good real life example.

But when we start to write this points down, people will dismiss perfectly fine images. The anime images are a good example. We had a thread discussing the problem, Madrugada looked through all locked discussions and a lot of people looked at actual pages, we found out that this problem doesn't exist. Still some people are bringing it up as a point to dismiss perfectly fine images. Not as tie breaker as argument against images in general.

Writing this guidelines down will lead to people dismissing real life, anime, Star Wars The Order Of The Stick in general. This will limit our possible sources.

I prefer this points to stay personal guidelines, when they come up everybody can state their personal opinion and use good arguments to make other people agree with them.

These guidelines would only be used when deciding between several equally qualified suggestions. 'Whatever does the best job of illustrating the trope' doesn't really apply if all images demonstrate the trope more or less equally well, IMO. Also, for the same reason, perfectly fine images will be dismissed in any case. These guidelines would just help dismissing the perfectly fine images that enrich TV tropes the least. I don't see a problem with that.

That means, if an IPD thread is deciding between a real life reenactment of an OOTS strip parodying Star Wars and an obscure book illustration, and the former does a better job of demonstrating the trope, the former would still be chosen as page image.

edited 2nd Jun '11 8:12:03 AM by ArtisticPlatypus

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.

I think our current system of letting crowners decide between equally qualified images is the best we can hope for, and has been working very well for us thus far.

Each individual can determine for themselves which of the presented options better fits, and votes accordingly.

It sounds like you're looking for a rule that will let you point to a particular image and say "based on X, Y, and Z guideline, this one wins." Trying to legislate guidelines to a point that such a thing would be possible would be throwing out the baby with the bathwater, in my opinion.

edited 2nd Jun '11 8:14:15 AM by Meeble

Visit my contributor page to assist with the "I Like The Cheeses" project!

Yup. A method to reach consensus is what is needed, not rules.

The principles of not allowing the wiki to sink into one-fandom-ghetto-dom, and choosing the best illustration of the trope for the widest number of people is all the 'guideline' we can benefit from.

Goal: Clear, Concise and Witty

Those aren't mutually exclusive. There will still be a lot of discussion and borderline cases. I just think a set of guidelines could save us time on the most commonly debated issues.

Am... Am I debating against Fast Eddie?

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.

Yes, you are, but you're doing it politely and with your reasoning included. Don't worry, you'll be ok.

...if you don’t love you’re dead, and if you do, they’ll kill you for it.

How should literal examples be treated?
Hmm...my opinion is that they should be used only as a last-resort, and only if the trope is either hard to picture or not safe for work. People tend to be bias against these.

edited 2nd Jun '11 9:02:07 AM by chihuahua0

Hm, I guess I could throw in my opinions right about now.

• Should Anime images be discriminated against? - I'm not really sure. I'm naturally inclined against them, but I dislike the medium, so..
• Should iconic or obscure examples be preferred? - Obscure, definitely, as I think the arguments in favor of them in my OP overshadows the arguments in favor of iconic ones. This is one of the reasons why I started this thread; I hate it when a good image from some obscure work is pitted against one from something like Pokemon, and the latter wins because a bunch of people are fans.
• Should overrepresented works be discriminated against? - I'd say yes. Variety for the win.
• Should underrepresented media be prioritized? - I'd say yes, for the same reason.
• Should Real Life pictures be discriminated against? - Well, yeah. There isn't really a controversy here, is there?

edited 2nd Jun '11 9:10:55 AM by ArtisticPlatypus

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.

Platypus, here's why I don't like tacking on those guidelines, personally (not as a mod): I've already taken most (if not all) of those things into account when I decided which proposed image I prefer, or that two or more are "equally good". And in my own experience, it's very very rare that I really consider two or more proposals to really be completely equally good. It's much more prevalent for it to be a case of "They're both good, but I simply like <this one> a little bit more than <that one>." It may be size, layout, some type of aesthetic preference, or one may lend itself to a good caption better.

And past experience also indicates that any Guideline we try to establish will be latched onto and treated like a law by some people, to the detriment of both discussion and the images chosen. If we make a guideline that says "If two options are equally good, give preference to the one that isn't anime", you can bet dollars to doughnuts that there will be people declaring that "Anime images are bad and we aren't supposed to use them."

...if you don’t love you’re dead, and if you do, they’ll kill you for it.

I agree that we should have a slight discrimination against Visual Puns or actually literal examples. Mostly to prevent confusion.

I have no opinion on work sources, however I've seen that I develop a bias against certain works due to the presence of fans spewing JAFAAC all over the wiki. I'm also somewhat prejudiced against classical works due to all the Most Triumphant Example gushing.

Was there any progress on that thingy where you kept track of the source of an image that came out of IP?

Fight smart, not fair.

We're still feeding it data. Fast Eddie hasn't had it generate a report yet, probably becuase no one has asked for one. I'll do so — I'm curious myself about what it showed.

...if you don’t love you’re dead, and if you do, they’ll kill you for it.

"The principle is that we are illustrating the trope. Whatever does the best job of illustrating the trope to the widest number of people is the good pick. Photo, animation cel, film cel ... whatever."

I find that often does allow for a greater variety, since even though some media have more pictures available, they don't always have the most trope illustrative pictures available.

As for the points brought up as suggestions, I personally see them as things that shouldn't be strikes against a picture. If a picture is poor at illustrating the trope, or wonderful at it, it doesn't matter where it came from.

I'm on the internet. My arguments are invalid.

@Madrugada: Heh, aesthetic preference stands for 90% of my opinions. Anyway, you have a point. I guess we could make a bunch of clarifications, but... Yeah, it would probably still be a probem.

@Dragon Quest Z: Yes, and that's why these guidelines would only apply when choosing between several otherwise equally good images. I guess that proves your point, Madrugada.

edited 2nd Jun '11 11:59:12 AM by ArtisticPlatypus

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.

Show me some sort of mathematical formula that objectively quantifies image quality and I'll agree with this suggestion.

edited 2nd Jun '11 12:20:20 PM by INUH

Infinite Tree: an experimental story

The only way we could point at two different pictures and say "they are both equally good" with any kind of objectivity would be if

• they both had the same final total on the crowner and
• they both had the same approval ratio on the crowner and
• they both had the same number of total votes on the crowner.

Otherwise, we can use the differences to declare that one is somehow "better" than the other — it has a higher positive score, or it has a higher approval ratio, or it has more people expressing their opinion at all. But that brings in a whole other set of problems...

...if you don’t love you’re dead, and if you do, they’ll kill you for it.

And crowners show a lot of tropers are willing to accept pictures from different media and works.

I'm on the internet. My arguments are invalid.

X = Y
Y is an image that fills all basic requirements for being a good image.
X is an image that clearly demonstrates all important aspects of a trope to people not familiar with the source, while still being an example in the work it comes from. It is safe for work, contains visual aspects that enhance the effect of potential text. It is not a lolcat or a demotivator, and does not contain a copyright sign. It also doesn't contain too many unnecessary elements, and any potential text is big enough to be readable.

Happy?

edited 2nd Jun '11 1:50:50 PM by ArtisticPlatypus

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.

^Nope. It doesn't allow you to look at two images and say, "yeah, both of these are exactly equal in quality," which was my whole point. When I said quantifies, I meant something along the line of "this image and that image are tied at a 26.7 on this scale."

I agree with what Madrugada said, but I don't think that situation occurs frequently enough for us to need actual guidelines.

edited 2nd Jun '11 1:56:42 PM by INUH

Infinite Tree: an experimental story

What are the variables, in your opinion? I think there are some (aesthetic value, for example), but I can't think of one that's more important than the suggested guideline issues.

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.

You're missing my point. What I'm saying is that there is no way, other than what Madrugada said regarding crowner results, which is unlikely to happen in the first place, to look at two images and say that they're objectively of the same quality.

Really, in my opinion, the only way two images really can be of the same quality is if they're identical.

Infinite Tree: an experimental story

The problem is establishing that a substantial number of the people involved all agree that the two images are equally good. If I have a small preference for one over the other, I don't think that they're equally good — I think that one's a little bit better. It really doesn't matter what that small preference is based on; it's there, and one is a little bit better than the other to me.

...if you don’t love you’re dead, and if you do, they’ll kill you for it.

Is that vague, undefinable preference a better basis of decision than - for example - the goal of trope images more diverse?

This implies, quite correctly, that my mind is dark and damp and full of tiny translucent fish.