Wikipedia talk:Captions

From Wikipedia, the free encyclopedia

[edit] criteria for a good caption: New Tips

In light of the criteria for a good caption I would like to suggest a new addition to the tips for writing captions:

• Try to draw your reader into the article by providing links to relevant sections within the article from the caption. (see here for how to do this) and the Ebola article for an example. Richard Taylor 18:37, 31 Jul 2004 (UTC)

In incorporating Richard Taylor's sensible tip, I have wound up bravely editing the guidelines 1) to establish that they are merely guidelines (possibly staying some unnecessary interference) 2) to be more explicit about the role of captions 3) to reduce the shrill insistence on "complete sentences" which is not a fixed rule in any publishing context: look at your own best-produced books. The one "rule" I eliminated was this: "...is an active sentence (that is, it doesn't use the verb to be)" as it might be tiresome for adults to be given directives about "good captions" that betray grammatical misconceptions that are as basic as this one. The one "exception I eliminated was this: "Images of the subject of biographical articles (A good caption is best, no caption is okay. A year for the photo is important)." Through an editor's following the guideline as an inflexible rule, an image at George III of Great Britain does not identify the painter's name or the date. I also switched "Exceptions to the rules" for "Special situations." I don't think any of the new text is very controversial, but you never can tell... I hope we can move forward, further refining what makes good captions, rather than merely reverting this attempt. --Wetman 18:13, 6 Apr 2005 (UTC)

[edit] Formating Options

We've seen three options for formatting the picture so far:

1. Thumbnail with no caveat about the font size
2. Thumbnail with caveat about the font size
3. Full size image

I tried the third option first when previewing the first edition of the page and it looked potentially confusing with the double frame and image text just like the caption text, so I went for the first option which prompted a lengthy addition to the caption - which is then not a concise caption as recommended by the text immediately to its left.

Perhaps the thumbnail size was too small. (I'd only taken the default. The picture is 279px wide.) Perhaps the thumbnail icon is too obscure. -- ke4roh 11:38, 27 Jun 2004 (UTC)

[edit] Mega caption

Graner poses over the dead body of Manadel al-Jamadi, an Iraqi prisoner; a small patch of blood can be seen on his right temple and his eyes are sealed closed with tape. According to Spc. Jason Kenner's testimony, al-Jamadi was brought to the prison by Navy SEALs in good heath; Kenner says he saw that al-Jamadi looked extensively bruised when he was brought out of the showers, dead. According to Kenner a "battle" took place among CIA and military interrogators over who should dispose the body. Captain Donald Reese, company commander of 372nd Military Police Company, gave testimony about al-Jamadi's death, saying that he saw the dead prisoner. Reese was quoted as saying that "I was told that when he was brought in, he was combative, that they took him up to the room and during the interrogation he passed ... (the body) was bleeding from the head, nose, mouth." Reese stated that the corpse was locked in a shower room overnight and the next day was fitted with an intravenous drip. The body was then autopsied, concluding that the cause of death was a blood clot from trauma. Reese stated that this was an attempt to hide what occurred from other inmates; many believe it was part of a cover-up to hide the death from the outside world.

How big should captions get: Charles Graner? — Matt 04:12, 29 Jun 2004 (UTC)

Shorter than that! A caption's job is not to tell the whole story, just to let the picture tell the story and lead the reader to the article for details.

Manadel al-Jamadi, allegedly combative, was bruised in the shower and died of a resultant blood clot. Graner gives signs of approval.

And the rest can go in the story.-- ke4roh 10:25, Jun 29, 2004 (UTC)

To be more complete: The information is excellent for the photo's description page because it gives detailed background. So much as the information relates to Graner, it should be included in the body of the Graner article. A caption should ideally be one sentence. If the information is too cumbersome for one, two will suffice. Three is getting long, and four is typically bordering on the absurd. Try to condense the sentences to give the most essential information in the shortest space. See the Marshall, Texas article for examples of longer, but still reasonable, captions. -- ke4roh 15:21, Jun 29, 2004 (UTC)

If a photo is subject matter in an article, as in this case, a long description is justified, but it may be better to make the photo with long description a separate section, with the text not a caption but regular text.--Patrick 21:54, 29 Jun 2004 (UTC)

[edit] Short captions

Some have brought up the subject of short captions like "Broad Street Station, London, 1865" (under Image:Broad street station 1865.png on the train station article). When I wrote captions for a printed book, they insisted on complete sentences. I've never found a caption like that in National Geographic Magazine, either. We should strive for excellence in our captions by providing background information - filling in the unseen and giving context for the picture. Why was this picture chosen for the article on train stations, for example? Was it the first? Is it typical? There is more to putting a picture in an article than simply declaring the subject of the picture. The act requires tying the picture to the article through the caption. -- ke4roh 21:13, Jul 6, 2004 (UTC)

See also the conversation on Talk:Twelfth Amendment to the United States Constitution#text of portrait. -- ke4roh 19:24, Aug 24, 2004 (UTC)

Another reason to write full sentences for captions is it tends to fulfill the other goals. It's hard to write a full sentence caption that doesn't fulfill at least one other goal of a good caption, and with that in mind, a full sentence is a good place to start. Keep in mind that captions are wiki text just like the articles - they're always subject to improvement! -- ke4roh 19:24, Aug 24, 2004 (UTC)

[edit] Establishing relevance

We've had a controversy over the captions for Twelfth Amendment to the United States Constitution. The images are portraits of historical figures involved in controversies that led to the amendment. Objections were raised to lengthy captions that described the specific controversies. I think those captions were excessive. It is only necessary to establish the existence and broad nature of the connection between the subject of the image and the subject of the article. This is enough for the reader to find the right part of the article text for details. Since, in this case, the images themselves contained nothing more that was relevant to the article, the single sentence establishing relevance was all that was necessary. 81.168.80.170 21:19, 18 Jul 2004 (UTC)

My feelings on the issue are: 1. At most, one sentence should be included in the caption. Anything further should go in the main article. 2. All pictures do not need detailed captions. In a well-written article, the text adjacent to the article may explain the picture sufficiently. If a reader wonders about the relevance of the caption, let him or her actually read the article. 3. When the article is on a particular topic, a caption depicting the subject need not have anything longer than a few words. For example: consider the article Nikola Tesla. The caption for the first picture is more than enough. 4. Lengthy captions are necessary only when the caption is not explained in full by the adjacent text. They should used when there is something extraordinary demonstrated by the picture, not thoroughly explored in the main body of the article. Consider, for example, the first picture in Pope John Paul I. -- Emsworth 21:44, 18 Jul 2004 (UTC)

(For information and context, Emsworth seems most concerned with captions on portraits of people.) There is some more guidance about when not to write captions at Wikipedia_talk:WikiProject Writing Captions#Standard image types. Nominative pictures (which simply serve as an example of the subject of the article with no further information) generally don't need captions at all. In writing captions, I've found that the most difficult are for portraits. Shorter is better, but it's also harder to write concisely, and there's nothing to tell about a portrait except the date. I agree that the adjacent text should explain the picture in detail, and I also believe that the caption should explain it in brief, as a means of introduction to the adjacent text for readers scanning the article. -- ke4roh 23:34, Jul 18, 2004 (UTC)

Actually, the caption text should make readers interested to read the adjacent text. Interested. Waking up. Thus, boring text pieces are not recommendable. For this very reason, the caption text should not answer to everything. But it should steer to the correct portion of adjacent text. 213.243.157.114 00:19, 19 Jul 2004 (UTC)

If the "nominative illustration" is a portrait, the name of the painter is always relevant, even if the editor has not heard of the painter. The date too is always relevant, even of a photo: "Einstein, 1906" Sometimes the photographer's name is relevant too: "photographed by Man Ray". Often details in the image might be missed, but would be obtrusive in the text: "wearing the Order of the Garter" "Buckingham House is in the background." etc. --Wetman 05:15, 28 September 2005 (UTC)

[edit] Standard image types

Moved from Wikipedia_talk:WikiProject_Writing_Captions

Articles about chemical elements, such as helium, contain an image of the periodic table with the element highlighted. How should these be captioned? We should apply whatever we decide to all the element articles, not just ones in our list to review. Are there any other standardised image types that should get uniform treatment? 81.168.80.170 18:11, 18 Jul 2004 (UTC)

There are several types of images that warrant special treatment:

• Periodic table snipets for each element - no caption needed
• Images of Element samples in the element info box - no caption needed
• Images of plants and animals (fungi? others?) in info boxes - no caption needed
• Images of the subject of biographical articles (A good caption is best, no caption is okay. A year for the photo is important).
• Info box images with mission insignia - no caption needed, but if there's a description of the symbolism, it should be included on the image description page
• Other images (especially within standard info boxes) where the purpose of the image is clearly nominative, that is, that the picture serves as the typical example of the subject of the article and offers no further information - no caption needed.
• Chemical compound diagrams (as in TNT) could benefit from a mention of the role of the structure in the properties of the compound.

-- ke4roh 21:48, Jul 18, 2004 (UTC)

Added to Wikipedia:Captions#Special situations

[edit] Captioning rationale

My logic goes like:

• Captions should be short because there's already plenty of structure available to article content, not to mention the ability to structure relationships between different articles. Long captions are a new kind of structure, 'sidebars' like in magazines, or perhaps just another way to lay out content on a page. If Wikipedia ever wants these they should be explicit structures that are distinct from captions, or alternate content formatting should be introduced to produce a different visual layout of existing article structure. Either way, captions should be short, not mini-articles.
• If a caption contains no additional information, nothing that's not obvious to anybody who sees the image, it shouldn't be there.
• If a caption merely names the image subject then it is obvious and boring to those who recognize the image. To those who don't recoginse the image or are unfamiliar with the article's subject such captions are:
  • virtually ignorable, when the object in the image is the subject of the article (e.g. "Freud" under picture of Freud in Freud article)
  • uninteresting minutia when the image subject is an example of a general class of things (e.g. "1965 Ford Mustang" under picture of car in car article)
  • mysterious and confusing when the article is about something abstract that cannot be depicted (e.g. "Periodic Table of Elements" under table in atom theory article)

So, the image caption should do more than label, it should be interesting, or make the article subject interesting. It should be a single thought, so as to be short. This means one complete sentence (by definition). However, writing sentences is hard so it's ok to punt and overflow your thought into more than one sentence to be collapsed later.

This argument leads directly to these guidelines:

• If the caption is larger than the image, something's wrong.
• When the image depicts the subject of the article (headshots, etc.) write an interesting thought about the subject.
• When the image depicts an example write a thought about what makes the example a good one.
• When the image instructs or illistrates, point out an instruction or illistruation that is non-obvious to the uninformed reader
• When the article is about something abstract that cannot be depicted, point out a connection between the image and the article that an uninformed reader would not otherwise know

In all cases you need not assume that the reader has read the article, in fact you may assume he has not.

Rules are made to be broken.

--Kop 16:01, 25 Aug 2004 (UTC)

[edit] Captions vs. alt text

A caption's job is to tie a picture to an article. The alternate text is responsible for describing the image if it is missing (as would be the case in a text-based browser or for a blind user). Wikipedia:Alternative text for images recommends writing good alternative text. The technical lack of a facility for writing distinct captions and alternate text impedes compliance with both sets of guidelines. I have mentioned the feature request at [4]. -- ke4roh 12:59, Aug 26, 2004 (UTC)

[edit] Move discussion of alternate/title/caption text to MediaWiki?

There's a discussion on these topics at MediaWiki [5], which seems the most appropriate place as technical changes may be required. It also seems like a good idea to work all this out at once. (I've made this same note at Wikipedia talk:Extended image syntax#Alternate text / Title text / Caption text and Wikipedia talk:Captions#Captions vs. alt text.)-- kop 18:15, 5 Sep 2004 (UTC)

This was filed as bugzilla:368. — Omegatron 00:04, 3 March 2006 (UTC)

[edit] Bolding? and complete sentences?

1. People are starting to bold parts of image captions & xref to this page, but I don't see anything about what to bold or when.

• • I've noticed this too. In my opinion, bold text should not be in captions (or, more specifically, text that should be bolded such as article titles shouldn't be in captions.) Italicized words however, such as for vocabulary words, should be fine. Scott Ritchie

1. Complete sentences seem like they'd be more of a distraction in many kinds of articles. For example, in Australian Kelpie, where the images are all similar in form and are designed to show the different coat colors of the breed. "Red and tan Kelpie" seems like that's all that needs to be there. Trying to make it longer ("This red and tan Kelpie is typical of the breed." "This black and tan Kelpie, like most of its breed, enjoys dog agility.") obscures the information rather than making it clearer. I'd much rather not have complete sentences on such images. And "This is a red and tan Kelpie." "This is a black and tan Kelpie." would just be silly. Elf | Talk 01:19, 3 Sep 2004 (UTC)

I've noticed a lot of anon users bolding the name of the person in the photo on the subject's biography page. Most of the time I've seen it, the user has done so by wikilinking the name of the person. Since it's on the person's bio page, it bolds it instead of making it a link. I think that it may be because of the wording of the "Clear identification of the subject" section. That section states "If the illustration is a painting, the painter's Wikilinked name, and even a date give context." But if the image of the painting is on the page of the artist's bio, wikilinking the name is unnecessary. I'm going to refine the wording of that section to reflect this. Dismas|^(talk) 02:43, 28 September 2005 (UTC)

[edit] Shaving cream example

This caption is touted as a good example. Do you think it is?

Is this really a good example? Does it clearly identify the subject of the picture? This is presumably an image of a Burma Shaving cream billboard which was part of the advertising campaign of which the caption speaks. However, as a European, I have trouble reconciling what I think of as a billboard with what the image depicts; it should say that it is a billboard. And it should say that it was part of the campaign mentioned, if this is actually true. Lupin 02:44, 7 Sep 2004 (UTC)

You've got it right on all counts. (See User_talk:Lupin#Monopoly captions which I expect to copy here soon.) I just haven't gotten around to re-working the caption yet. :-) Either we should come up with a better Burma Shave caption or find another caption to hail as adequate. We can do that in this space. -- ke4roh 10:27, Sep 7, 2004 (UTC)

I question the choice of the image. Burma Shave is a specific cultural reference and a dated one at that. People from North America who are over the age of fifty will understand it instantly. It's obscure to most others. The image itself is not the best illustration of its subject matter. Why not choose an image the world would understand? Try the Eiffel Tower, Darth Vader, Mount Everest, kangaroos... Durova 11:04, 6 December 2005 (UTC)

As an image demonstrating billboards, it could be better. But in the article it is actually used to demonstrate billboard sequences, for which it is ideal, as the first such sequence ever.

I've improved the caption. It now reads "This billboard was the last in a sequence of roadside signs telling a joke. It was part of a campaign for Burma-Shave canned shaving cream, and was the first of its kind." Now we just need to make a new screenshot. ··gracefool |☺ 01:46, 3 February 2006 (UTC)

[edit] Monopoly captions

From User_talk:Lupin#Monopoly captions

Hi, Lupin, I hope you don't mind I just put the full-sentence Monopoly (game) caption back with explanation on the edit summary. I was trying to provide some context to people who have never seen the game so they'd know more than "in progress," plus trying to comply with guidelines at Wikipedia:Captions, the first of which is that captions should consist of complete sentences. Please edit the caption as you see fit to help it follow the guidelines. Also, please look over the guidelines, and if they need to be revised, please help. Also see Wikipedia_talk:Captions - there are several discussions on particular examples. -- ke4roh 20:27, Sep 1, 2004 (UTC)

I dispute the usefulness of these guidelines. Maybe I will attempt revision one day, although I feel the tide of opinion is somewhat against my own. I've tried to make your caption directly relevant to the image, which is the biggest failing of many of the captions I've seen which cite the guidelines. (Perhaps it could also be deleted altogether under the "nominative" exception at Wikipedia:Captions#Exceptions to the rules). Lupin 03:09, 2 Sep 2004 (UTC)

I like the Monopoly caption, though I'm still a bit frustrated with the Emacs caption. Any ideas for improving that one? Thanks! -- ke4roh 15:50, Sep 3, 2004 (UTC)

It's now a full sentence. Lupin 17:22, 3 Sep 2004 (UTC)

Thanks! -- ke4roh 18:50, Sep 3, 2004 (UTC)

I'm curious: what's your opinion on captions? Could the guidelines be made more useful? I must admit I've had quite a time writing captions for subjects of which I knew nothing before reading the article, and with image description pages full of <sarcasm>informative</sarcasm> text like "Camel drinking Coca-Cola," it does challenge me to come up with some more context for the picture. So, I'm trying to improve the background information on images I've contributed and images I wonder about while captioning. -- ke4roh 18:50, Sep 3, 2004 (UTC)

I haven't devoted much thought to this, so please take this with a pinch of salt. But I would say that if you want to write good captions then you first have to decide what the function of a caption is. IMO, the primary purpose of a caption is to describe the image, by which I mean to give a reasonably detailed account of what the image depicts. Captions may also answer the question "why is this image in this article;" alternatively, this question could be answered in the article. I don't think a hard and fast rule can be given to say which is better, although clearly this question should be answered somewhere unless the answer is blindingly obvious.

Broadly speaking, other information belongs in the article and not in an image caption. I am of the opinion that dogmatic guidelines such as "thou shalt use complete sentences" and "thou shalt not use the verb to be" are unhelpful since every image needs a caption to be tailored to it and not to be constrained by some arbitary set of rules. Lupin 21:55, 3 Sep 2004 (UTC)

I tend to agree with you. The opening of Wikipedia:Captions attempts to summarize the purpose of a caption: to tie the image to the article and to provide some context for the picture. It's quite challenging to write a caption that accomplishes those goals and is not a full sentence, and writing a full sentence leads the captioner towards those goals, hence it's the top priority and an easy one to meet. Now what's the difference between these two captions for Image:Emacs.png?

1. The default splash screen, which greets the user when Emacs is run in a graphical environment.
2. The default splash screen greets the user when Emacs is run in a graphical environment.

The second is a complete sentence (the first is only a fragment for want of a verb because "greets" is buried in a prepositional phrase), and it's one word and one comma shorter. When I read the first caption, it seems halting and incomplete (though it clearly describes the image and provides some context). The second one delivers a complete thought.

On the requirement for no passive sentences, I had an 8th grade (of 12) English teacher who insisted that we write paragraphs of three or more sentences and at most one of thse sentences could contain a conjugation of "to be." Why? Because "to be" doesn't describe action. Excessive use of "to be" can put the reader to sleep. I occasionally watch scientific NASA films. My chief complaint: "Many passive sentences were constructed." (But wouldn't that be clearer and more interesting if it were "The writers butchered this video with passive verbs throughout the script."

Sometimes a passive sentence is appropriate. In law, the North Carolina government writes, "Each employees hands must be washed with soap and warm water before returning to work." The non-legal writer might be tempted to say "Each employee must wash his or her hands...", but the law is not concerned with who washes, only with the fact that the hands are washed. I have yet to nocice a caption requiring passive voice, though I wouldn't be surprised to see one. Sometimes I edit captions to make them active, but not nearly as often as I'll edit them to make them complete sentences.

Consider the difference between these two captions:

1. The default splash screen greets the user when Emacs is run in a graphical environment.
2. The default splash screen greets the user running Emacs in a graphical environment.

In both cases the primary verb is "greets", but the first involves "is" and the second does not. The second also comes up two words shorter and clarifies that the user runs Emacs.

What belongs in the caption versus the article? We went 'round and 'round aobut that with the caption for Image:Spinal_Tap_logo.jpg for heavy metal umlaut. I would add a little extra information to the caption and my text wound up in the article minutes later, with the caption replaced by "Spinal Tap". (It was on the main page that day.) "Spinal Tap" doesn't tell the reader anything more than they can get from looking at the picture, though, so I tried again, and we wound up with the information in both places. It's particularly difficult to write a caption for an image like that because there isn't much context. In that respect, the information in a caption should be context - some more information for the readers about the circumstances of the photo.

The guidelines at Wikipedia:Captions are just that — guidelines. I no more expect to see the perfect caption than I expect to see the perfect article. But that said, I don't see why we shouldn't strive to write better captions. Your thoughts? -- ke4roh 12:09, Sep 4, 2004 (UTC)

Consider the difference between these two captions:

1. The default splash screen, which greets the user when Emacs is run in a graphical environment.
2. The default splash screen greets the user when Emacs is run in a graphical environment.

The second is a complete thought, and a complete sentence which reads flowingly, I agree. Its failing is that it makes no direct reference to the image. Without doing so, the potential for reading it as a fact whose spatial proximity to the image is merely coincidental is high. I find this kind of ambiguity infuriating. The subject of a caption should, in the first instance, be the image. Making this tacit assumption on the part of the reader is unjustifiable in most situations, I feel.

I am also not convinced of the ennui induced by the verb "to be". I am completely convinced that sentence constructions whose sole purpose is to avoid using this forbidden verb lead to contortions which are awkward and confusing.

The use of complete sentences should not be a requirement for captions, in my opinion. If there is not enough information available for a caption which has direct relevance to an image, then why not use a simple tag? I agree that in general this is suboptimal, but this is entirely different from saying that a caption which consists of complete sentences is always better than one which doesn't. Less is sometimes more. Lupin 14:49, 4 Sep 2004 (UTC)

I see your points. Our discussion on the first point reduces to a question of how much to trust the reader to interpret and think. At times it can be a tough judgement call. In the more difficult cases, it becomes important to craft the caption to clearly state what is in the picture. I typically trust the reader to expect that at least one of the nouns in the caption would describe the picture. (See for example water skiing.) In the Emacs caption 2 immediately above, the reader must decide if the picture is of a splash screen, a user, or a graphical environment. Perhaps the guidance could be revised to indicate that the caption should clearly refer to the picture. (What do you think of the water skiing caption?)

On contortions to avoid ennui from "to be", Winston Churchill put it nicely (though he was addressing a complaint about prepositions at the ends of sentences): "This is the sort of pedantry up with which I will not put." "To be" has its place, and it should be used in some cases. It's not worth contortions to avoid, but it is worth a moment's thought.

Where would a simple tag better suit an image than a full-sentence caption?

Many thanks for the enlightning conversation. -- ke4roh 23:36, Sep 4, 2004 (UTC)

Take a look at Ford Mustang. It may be one of those places where a full sentence is overkill, but if we had more information about the various models in the text (e.g. "The 1965 Ford Mustang introduced the 'whiz-bang super accelerator'"), perhaps it wouldn't seem so satisfactory to look at the pictures with those simple tags. (Therein lies the dilemma. If we haven't thought of a good caption, perhaps a short one will do in the meantime, but that short caption could always use some expansion to explain why that particular picture was chosen for the article.) -- ke4roh 23:46, Sep 4, 2004 (UTC)

My question about your "trusting the reader" to interpret a caption correctly is this: why? When it is so simple (if slightly inelegant on occasion) to avoid this potential source of misunderstanding, is it not better to do so?

As with many things, judgement has to be exercised here. In the case of the Water skiing caption, my personal feeling is that the caption is appropriate, since the connection to the image is plain. Others (who may never have come across water skiing before) may disagree however, in which case the caption should be clarified. Lupin 00:26, 5 Sep 2004 (UTC)

Perhaps a new guideline is in order: "The caption should clearly identify the subject of the picture without detailing that which is plain to be seen." I'd put that as number one or two (I lean towards 2, but expect you would prefer 1). In fact, I have a hunch you'd remove the complete sentence guideline entirely, but captions like

Adrian Carmack working on the Baron of Hell. The DOOM monsters were digitized from clay models.

(on Image:Adrian_Carmack.jpg for DOOM) bother me for their choppiness as much as those ambiguous ones bother you. Your thoughts? What's left to resolve? -- ke4roh 13:15, Sep 6, 2004 (UTC)

I think that given that people seem so fond of guidelines, something like what you suggest is in order. However I would do it like this: the guideline should say "The caption should clearly identify the subject of the picture" and then in the exceptions section point out that things which are obvious even to someone coming to the subject for the first time may be omitted, provided this is not to the detriment of the caption. Otherwise, the guideline itself provides a convenient loophole for those wishing to ignore it. Moreover, I think that "detailing what is plain to be seen" is often precisely what a good caption should do -- a good informative image has an easily identifiable subject, and a good caption should describe the subject of the image.

I agree that "choppy" captions are bad. I would favour replacing the first guideline with one about captions of a certain length flowing off the tongue... but of course that's harder to define, let alone "enforce".

Putting either the "full sentences" guideline or an "anti-choppiness" guideline as the number one priority seems to me to approach the problem backwards - content should rule over style, in the first instance. Stylish captions with obscure or irrelevant content should not be favoured over those with relevant content which are phrased badly; instead, the good content should be rewritten stylishly (but unambiguously), when this is possible. This is my main gripe with this purely stylistic guideline being given so much emphasis. Lupin 13:37, 6 Sep 2004 (UTC)

I'm thinking about a major refactoring of Wikipedia:Captions to address your concern and make the justifications for each of the guidelines clearer. (I just whipped up the numbered guidelines one day based on all the discussions we'd had until that time, and I was somewhat surprised nobody had any comments about them until yours.) I'm imagining using the TOC box as the numbered list and having a subsection to address each of the guidelines, explain its relevance, and discuss any exceptions to that particular guideline. I'm also inclined to reference the sections of talk pages that generated each guideline. Of course, the refactoring will take a few minutes :-). I also don't want the page to be prohibitively long - just enough to cover the subject.

As for approaching the problem backwards, there is a method to my madness (not to say it's right, just a method). Namely, if people read and follow only one rule, it should be the one most likely to result in a good caption. I would expect people to make the caption describe the picture in the absence of explicit instruction to do so (though some of the captions that otherwise attempt to meet the guidelines don't do that clearly enough). In fact, when I wrote a chapter for a real book, they insisted on exactly two things: captions must be full sentences, and images must be referenced in the body of the work. Everything else was up to the individual authors and editors. It surely generated better captions than no guidance at all. -- ke4roh 01:42, Sep 7, 2004 (UTC)

[edit] Small tag in captions

What is the feeling toward using the small tag for text in captions? I couldn't find any mention of the issue here. Rmhermen 22:05, Oct 8, 2004 (UTC)

Now that you've brought it up :-), I'll add my comment here, which is that I hate it because in every browser on every system I use the result is illegible or nearly so. And I don't want to have to change my browser settings just for wikipedia and change them back every time I go anywhere else on the web. And, yes, I use wikip's default styles because I like to see what other people will see when they come here. So there. I remove it whenever I can so that I can read the text. Elf | Talk 23:03, 8 Oct 2004 (UTC)

The HTML small tag shouldn't be, and isn't generally used in captions. Caption text is currently displayed using the default stylesheet as smaller than the main body text, however according to a comment in the CSS it's never supposed to go below 9px [6], and there are many elements of the default page which are smaller / as small as the caption text. Richard Taylor 01:56, 9 Oct 2004 (UTC)

[edit] Period at the end of the caption or not?

Hi, is there any rule/recommendation for writing or not writing a period at the end of the caption? Thanks. Miraceti 16:49, 23 September 2005 (UTC)

if it's a sentence --Wetman 01:07, 24 September 2005 (UTC)

I think there should be one, because all captions are supposed to be in complete sentences. --Knowhow 05:34, 9 July 2006 (UTC)

I agree that single-sentence captions should have a period at the end. Amusingly, the example captioned image currently has single-sentence captions with periods, but on the same page, references the style guide which says single-sentence captions are not to have periods. How can we go about getting the style guide changed? --71.146.10.244 18:08, 3 September 2006 (UTC)

[edit] Image summaries

This isn't about captions exactly, but when I add a summary to an image I'm uploading, should I link the names used to describe the picture? For instance, if the summary is "Screenshot of Microsoft Office Excel 2003, taken from example.com" I know I should probably link "example.com" but what about Microsoft, Office, and Excel? Or just Excel? Is there any convention to this? (For now I only link source URLs). Thank you! -- Foofy 22:16, 18 October 2005 (UTC)

[edit] What captions should not include?

Is a caption a suitable place for discussion, and providing all points of view? Here is a recent example showing two different possible captions, from the page on Plasma cosmology, a non-standard, non-mainstream, theory of the Universe.

Do people think that in these examples, we should include (1) the mainstream point of view (b) Note how plasma cosmologist view the mainstream theory (c) comment on the degree of explanation of the plasma cosmology view? --Iantresman 14:16, 6 November 2005 (UTC)

M87's Energetic Jet. The glow is caused by high energy electrons spiraling along magnetic field lines -- so-called synchrotron radiation. Astrophysicists suggests that M87's dynamics are affected by a supermassive black hole, a concept derided by plasma cosmology advocates. [1] Some plasma cosmologists have suggested in a cursory, comparative fashion that radiation like this may be due to Birkeland currents, [2]. Plasma cosmology has neither the explanatory power nor the theoretical infrastructure yet to explain such jets as fully as the mainstream models do.

'M87's Energetic Jet. The glow is caused by high energy electrons spiraling along magnetic field lines -- so-called synchrotron radiation. Advocates of plasma cosmology compare the filamentary structure to pinched plasma filaments seen in the laboratory and computer simulations.[3]

You should shorten the caption, hinting that the image can be interpreted in two different and opposing views. In this way you fulfil succintness requirement for image captions and drive the reader to the article.--Panairjdde 13:06, 24 January 2006 (UTC)

A caption is NOT a suitable place for discussion. It should be short and concise. --Siva1979^{Talk to me} 19:54, 4 June 2006 (UTC)

[edit] Random/title bolding in caps?

I've been seeing people putting boldface in captions. Particularly repeating the article bold, but miscellaneously just all over the place. It drives me a little crazy, but I thought it might be a policy I didn't know about. Any thoughts? jengod 00:01, 14 January 2006 (UTC)

Sometimes the bold part is also a wikilink pointing to the same page (which shows up bolded instead of as a link). Generally, I think bolding should be used VERY sparingly overall, and only where the article title is first noted, or alternative names (such as Cologne (Köln in the German language)) are in play. Oh, and when used for subsection-purposes, though these should really use = in most cases. 07:31, 16 June 2007 (UTC)

[edit] Logo captions

At Wikipedia talk:Logos/Archive 1#Caption guidelines for logos a proposal was made a week ago that the following guideline be added to the list at Wikipedia:Logos: "Logo images should always be surrounded by an image box and be labelled with a caption". This is in regard to corporate logos, and is intended to make it clear that any logo used in a Wikipedia article is being used solely as an illustration, not as a logo. The proposal has received feedback and not been objected to after a week, so it has been added to the list at Wikipedia:Logos. I've just noticed that this new guideline at Wikipedia:Logos is in conflict with a guideline here at Wikipedia:Captions. The conflicting guideline here reads: "Company or product logos, where the logo is current, and the article is about the company or product. - no caption needed". I would like to remove this guideline from Wikipedia:Captions. All corporate logo and product images need captions to ensure that we "avoid using a logo in any way that creates an impression that the purpose of its inclusion is to promote the company", as a guideline at Wikipedia:Logos expresses. Are there any objections to the removal of the guideline "Company or product logos, where the logo is current, and the article is about the company or product. - no caption needed" from this page? Kurieeto 23:25, 10 March 2006 (UTC)

I definitely agree with the removal of the logo special situation. I do not see where or how this exception to the other keys to good caption style was arrived at. Does someone have a link that shows there was in fact a general consensus for the logo caption exception and an explanation of the reasoning?--In1984 21:25, 21 May 2007 (UTC)

Because it is in conflict with the longer standing logo guideline Wikipedia:Logos and was added to special situations March 9, 2006 without any discussion, much less consensus, I'm remvoing the line about company/product logos. Please discuss here before restoring. It is already causing disputes for which there is apparently no justification.--In1984 23:20, 21 May 2007 (UTC)

• This discussion should be continued at Wikipedia talk:Logos#Captions for all Logos, which is listed at the Village Pump and has more posts. nadav (talk) 07:05, 22 May 2007 (UTC)

[edit] Auto Caption from Description?

Is there a way to have the Image's description automatically be the caption? --jeolmeun 17:53, 25 May 2006 (UTC)

Automatically being captioned? I don't think so. --Siva1979^{Talk to me} 19:51, 6 July 2006 (UTC)

[edit] Guideline or not?

Is this page a style guideline or not? Should it have the guideline tag? Gimmetrow 19:25, 24 August 2006 (UTC)

I've just tagged it as such... not really sure why this page wasn't already tagged. I'm not familiar with any guidelines about tagging pages as guidelines (hmm meta-guidelines) but this page seems mature enough now to warrant such tagging. (→Netscott) 19:34, 24 August 2006 (UTC)

[edit] Credits of photos.

On the Hippie page I reverted an edit which credited the photographer in a caption [7]. Old caption

"Dancing Hippies" Berkeley, California 1969 by Robert Altman http://www.altmanphoto.com

new caption "Dancing Hippies" Berkeley, California 1969 I was expecting to find some guideline here on this. In general the credits seem to be limited to the image page, is this a standard, policy or guidline? --Salix alba (talk) 23:14, 4 September 2006 (UTC)

The standard is to generally limit credits to the image page. However, particularly famous photographs or photographs by famous photographers are often an exception. Kaldari 19:04, 13 September 2006 (UTC)

The photo credit should go on the image's page, not in the caption. The only exception should be when the photographer is notable enough to have an article, and even then, only if relevant to the article. EVula 20:17, 13 September 2006 (UTC)

Iraqi soldier on patrol in Baghdad, Iraq
(April 2005).

At some point, the gate keepers of Wikipedia will recognized that ALL images should contain some sort of photo credit. Like any other form of art, a photographer should be given credit for his or her efforts. Additionally, for places such as a war zone, an individual who risk his or her life to capture an image should be noted for their work. There's a big difference between a picture of some mundane location in middle America and a photograph that captures a moment in time on the streets of Baghdad, Iraq.

Photographers and/or photojournalists, recognizing the value of Wikipedia as a place to publish their work, may start becoming major contributors to this growing corner of cyberspace...

In order to save itself from possible...future...Copyright legal headaches, Wikipedia should require that ALL uploaded images contain the author's name. The business folks at Google are finding out that the recent purchase of You Tube came with a host of unresolved internet legal issues.

As Wikipedia moves forward in its development as a new medium of information, a serious discussion needs to take place regarding the attribution of material (i.e. both print and visual) contained within these pages. Only when that happens will Wikipedia move forward to its next level of development.

And finally, in the interest of Wiki-Peace, I've removed ALL the photo credits from each and every submitted caption. Until I reach 1,001 active photo submissions, I'll refrain from battling over the merits of this issue...
v/r
Peter Rimar
Chitrapa 05:38, 24 March 2007 (UTC)

I agree with Chitrapa that the policy of non-attribution in the images that are actually seen in an article is unwise. Technically, this policy may be consistent with the details of licenses (Creative Commons, for example) that require attribution, since the details are generally available to users knowledgable enough to just "click" the image. However, this procedure violates the spirit and the normal understanding of attribution. On several occasions I have written photographers to ask that they license their work using Creative Commons Attribution licenses - so that I can then use their work for Wikipedia. After securing this agreement, I have then placed copyright notice in the caption. Flickr also does exactly this, for example; even so, the fraction of images that artists make available using acceptable Creative Commons licenses on Flickr isn't large. While I understand the spirit of GFDL licensing, which all of us agree to when we make edits, it weakens Wikipedia if we are unable to promise artists proper attribution of their work when used by Wikipedia.Easchiff 04:55, 23 May 2007 (UTC)

I agree with both Chitrapa and Easchiff. Though most images on Wikipedia don't require attribution in their licenses, a bunch do (CC-BY and variants). Those should probably be attributed in the caption of the image (as well as on the image page of course), since some readers probably don't even know there's more information if you click the image.

This guideline page should address this issue somehow. So for now, if nobody has any objections, I'll change "Who took it? (Generally, this is only included in the caption if the photographer is notable.)" to "Who took it? (Generally, this is only included in the caption if the photographer is notable, or if the image's license requires attribution.)". I'm not sure how bold I should be with guideline pages, so I'll wait a few days to make sure there's consensus first. -kotra (talk) 08:47, 31 December 2007 (UTC)

I've seen no objection, so I changed it as described above. -kotra (talk) 22:30, 17 March 2008 (UTC)

I agree, photographers should be given credit in the caption itself. It is time to rewrite guidelines now, since we obviously have a consensus. (Mind meal (talk) 03:10, 21 April 2008 (UTC))

I disagree, as do most above. The text should be changed back. Johnbod (talk) 03:13, 21 April 2008 (UTC)

Your logic in this regard? Why is it asking too much that we follow the law and attribute work to photographers? It defies the very reasons we require such licenses in the first place. If Wikipedia wants more images, they should do what they can to make it attractive for photographers. (Mind meal (talk) 03:17, 21 April 2008 (UTC))

The same reasons given by all the others above; the credit is on the image page, along with much other information. I don't object to many images being credited in the caption, but for most of those we have it is not necessary. It is not a legal issue; in fact we are unable to alter the licence to enforce this on other users anyway. Do we want more images so badly? Dozens are deleted every day. Johnbod (talk) 03:22, 21 April 2008 (UTC)

That is unacceptable. That is a separate page that most users simply do not click. Yes we want more images, and badly. Dozens of images being deleted for having copyright issues is no reason to want less photos that are properly licensed. There simply is no reason we should not be doing this. Why not attribute the work wherever possible? It is a legal issue! We should safeguard Wikipedia in this regard by doing this, not adhere to ill conceived arguments that have no logic behind them. By "all others above", who do you refer to? In this section, the consensus is we should attribute works in the caption. Below, also, there is a split. Also, when someone stipulates we attribute their work in a caption, we MUST do this, per the CC licenses, i.e. "You must attribute the work in the manner specified by the author or licensor." (Mind meal (talk) 03:27, 21 April 2008 (UTC))

I count this section 4:4 at present - some concensus! Johnbod (talk) 03:47, 21 April 2008 (UTC)

I have changed the language of the main page to reflect the requirement of CC licenses. Do you object to this? The language is quite clear on them concerning attribution: "You must attribute the work in the manner specified by the author or licensor." We don't need consensus when we are talking about following the law. (Mind meal (talk) 03:51, 21 April 2008 (UTC))

Sorry, we don't go slapping photographers names in articles (including captions) because we're bullied in to it. Free encyclopedia, yes, free advertising, no. Shell ^babelfish 05:34, 21 April 2008 (UTC)

Since when is giving attribution, and acknowledging sources, called advertising? Also, nobody said anyone was bullied into doing anything. I hardly see how a request to include a person's name in a caption is deemed "bullying." What if we handled references in this way? Just an anonymous link for someone to click and then see who produced the work. That wouldn't fly. Could you please point me, an obviously ignorant editor in your eyes, to the consensus you continuously assert there is on this matter? That would go a long way. On this page, there is only division on the matter. I'd appreciate it. (Mind meal (talk) 08:40, 21 April 2008 (UTC))

Wikipedia:Administrators' noticeboard/Archive98#Images, Wikipedia:Administrators'_noticeboard/IncidentArchive302#Photography edit war on model Ana Beatriz Barros and WP:IUP#User-created images may be of interest. The second one is the one closest to the issue at hand, I think, and it is related to the third link. I don't know about legal issues but I think these are likely best raised with the Foundation lawyer, Mike Godwin, User:Mikegodwin. x42bn6 Talk Mess 13:55, 21 April 2008 (UTC)

Agree, on the last point especially. Johnbod (talk) 14:28, 21 April 2008 (UTC)

Thanks for the links. While I couldn't disagree more with the findings of link 2, I'm glad the issue has been discussed previously. When I get a photo released at Flickr for use here, and I'm asked to place their name in the caption...I feel like crap saying we can't after I've just had them relicense their work. It doesn't feel right. I would be against having external links in captions; I just think we should attribute work to the name of the individual, plain and simple. I don't see what this will hurt? If anything, it puts us on more stable legal grounding.(Mind meal (talk) 15:03, 21 April 2008 (UTC))

A dicussion is underway at Village Pump on this topic: Wikipedia:Village_pump_(proposals)#Photograph_attribution_in_image_captions (Mind meal (talk) 21:03, 21 April 2008 (UTC))

[edit] Using periods

Less than a month ago, Jasoncatlett added the following paragraph to the caption guidelines (without discussion):

"Sentence or not, watch the stops: if the caption is a single sentence or a sentence fragment, it does not get a period at the end. If the caption contains more than one sentence, then each sentence should get a period at the end. See Wikipedia:Manual_of_Style."

Prior to that edit, it has always been the convention to end captions that are complete sentences with periods regardless of how many sentences there are. This change has led some editors to go through Wikipedia on editing sprees to purge periods from captions. Before this gets out of hand, I think we should discuss the change, and decide if that's what we really want. Kaldari 18:58, 13 September 2006 (UTC)

As I am the "some editors" (*bows*), I'd like to provide some counter-evidence. Contrary to the presented argument, the "no periods at the end of single-sentence captions" didn't magically appear a month ago. It originally derives from Wikipedia:Manual of Style#Captions, where it has been since mid-January, when it was first introduced.[8] As an aside, I'm fairly certain this is the first time it has been contested.

Now, honestly, I don't particularly care about this one way or another (I'm just upholding the Manual of Style; if it is altered after gaining consensus here, I'll change my editing habits accordingly). I'm merely presenting additional evidence. EVula 19:45, 13 September 2006 (UTC)

Well, as far as normal style guides are concerned, I'm pretty sure that complete sentences get periods regardless of how many are in the caption; and sentence fragments may or may not also get periods, depending on the house style. I would suggest leaving this as one of the keep-it-consistent-but-don't-change-the-style points in the MoS. Kirill Lokshin 21:12, 18 September 2006 (UTC)

Complete sentences, even lone ones, should definitely get periods; I'm very surprised to hear anyone suggest otherwise, and would like to hear the rationale. Further, I think that long captions, be they complete sentences or not, probably should get periods, and that very short captions — just a word or two, or just a name — probably should not. In-between lengths, I'm not sure about. Ruakh 22:39, 18 September 2006 (UTC)

Here's what CMS recommends:

Chicago Manual of Style, 15th ed., 12.32: Syntax, punctuation, and capitalization. A caption may consist of a word or two, an incomplete or a complete sentence, several sentences, or a combination. No punctuation is needed after a caption consisting solely of an incomplete sentence. If one or more full sentences follow it, each (including the opening phrase) has closing punctuation. In a work in which most captions consist of full sentences, even incomplete ones may be followed by a period for consistency. Sentence capitalization is recommended in all cases except for the formal titles of works of art.

• Milton at the Nobel ceremony, 1976
• The White Garden, reduced to its bare bones in early spring. The box hedges, which are still cut by hand, have to be carefully kept in scale with the small and complex gardern as well as in keeping with the plants inside the "boxes." —Wayward ^Talk 01:31, 19 September 2006 (UTC)

The CMC sounds reasonable to me. Many or most captions are just labels—like titles and headings, they need no punctuation. Only when a caption is expanded to become prose does it need punctuation.

And remember, "consistency is the hobgoblin..." etc: one rule needn't lord it over all captions in Wikipedia, because readers only read one article at a time. —Michael Z. 2006-09-19 02:22 Z

To respond to Ruakh's comment, I disagree with the "long captions, be they complete sentences or not, probably should get periods". Length is irrelevant; if a caption is somehow "very long" but isn't a complete sentence, there's something wrong and it should be rewritten. Keep in mind that captions are supposed to concise to begin with, so merely ranking captions by length is flawed. EVula 03:31, 19 September 2006 (UTC)

I'm not sure complete sentences are inherently better. The (hypothetical) caption "the 2131 centennial celebration of the original Church of Our Lady of Apoggostio, as painted by Leonardo de Rafael" could be recast as a complete sentence by putting some of that information in a predicate, but I don't think that would be an improvement. (Not that I think it would be a step down, either; I just don't think it would make a difference.) And concise doesn't necessarily mean short. (That said, I really don't feel strongly that captions consisting solely of long fragments should have periods; that's just my first inclination. If you feel strongly that they shouldn't, I don't mind at all.) Ruakh 04:15, 19 September 2006 (UTC)

Well, that particular (rather amusing) caption I think works just fine without being a complete sentence, so I think we may have just stumbled across an accidental agreement. ;) I think that a sentence fragment, regardless of length, shouldn't have a period; after a certain length, though, the alternative of rewriting it as a complete sentence (or two) should be examined. EVula 05:32, 19 September 2006 (UTC)

[edit] Image credits (in captions)

I'd like to know what the Wikipedia policy is on crediting photographs uploaded under (for example, creative commons attribution) license in the *image caption itself*.

The case that I am referring to is of one photographer putting a 'Photo by XXXX YYYYYY' into the thumbnail description of every image. For some reason, this irks me quite a bit. It is not standard on Wikipedia, at all, and it is a bit like me putting my name under an article I just wrote (not legally the same I know, different licenses), so everybody will know:"Hey its me who wrote all this!" What is the policy on this? MadMaxDog 07:05, 16 June 2007 (UTC)

I see this turned up above, earlier. I agree that it should be resolvedlong-term wise. I just think that crediting at article level is actually *against* the spirit of Wikipedia (being a communal work), and distracts from the image and caption itself. Or am I being too 'communist'? MadMaxDog 07:08, 16 June 2007 (UTC)

• The prohibition against crediting non-notable authors is currently being broken on Wikipedia's Main Page. There is a discussion at: Talk:Main_Page#Photo_credit_for_picture_of_the_day concerning whether or not this is desirable. Johntex\^talk 21:44, 22 August 2007 (UTC)

As the guidelines state, credit is only given in captions if the photographer is notable, for example "William Faulkner photographed in 1954 by Carl Van Vechten". Kaldari 23:01, 22 August 2007 (UTC)

This attribution debate is moot, as every uploader has the opportunity to attribute the author right there in the upload form. The photographer is clearly credited, right under the image, on the image description page. All you have to do is click the thumbnail, whic is what anyone interested in actually viewing the image would have to do anyway. As Kaldari points out, unless the photographer is notable, his/her name has no reason to appear in article namespace, and hence shouldn't appear in the caption. There have been huge, lengthy discussions on this elsewhere (do a search!) and the consensus is always to keep attribution on the image page and out of articles. --mikaul^talk 09:18, 12 December 2007 (UTC)

I have just removed the phrase that allows crediting if the license mandates it. Only pictures that could also be on commons or fair use pics can be included. On commons only free material. For example the CC-BY-3.0 license states that [attribution] may be implemented in any reasonable manner; provided, however, that in the case of a Adaptation or Collection, at a minimum such credit will appear, if a credit for all contributing authors of the Adaptation or Collection appears, then as part of these credits and in a manner at least as prominent as the credits for the other contributing authors. "

This means that if we allow one CC-BY-credit in an article, we will have to allow them all, because of the requirement of equal prominence. We are creating a "Collection", and any reasonable manner does not include the place where such attribution must occur. A reasonable manner is to credit all authors on the place reserved for that purpose. For pictures that place is the description page. — Zanaq (?) 10:39, 27 March 2008 (UTC)

Hi, I added the phrase in question because it appeared to reflect consensus based on the above discussion (I didn't see this one). Thank you for reverting it because it didn't reflect consensus.

As for your point about allowing CC-BY credits in captions leading to credits in all captions, I think you are misunderstanding the CC-BY license. It says in the part that you quoted that CC-BY only requires that it be credited at least as prominently as others are credited on that page, not the other way around (that others must be credited at least as prominently as the CC-BY work).

As for the argument that someone can easily just click on the image to learn more about it, that argument depends on how often people actually do click on images to find out more, or even know that it is possible. Certainly it won't be obvious to new visitors to Wikipedia that all the images are clickable. Also, if Wikipedia is printed, there is nowhere in the printed result where credit information is available except in the caption. I agree that it looks less streamlined to have "Photo by John Doe." in captions, but that doesn't override the requirements some artists have on their work. It can be argued that credit must be present everywhere the work appears, including every page it appears (the description page is a related, but separate page). -kotra (talk) 21:59, 27 March 2008 (UTC)

If wikipedia were to be printed, it would have to include a section with the author & license info. The whole of this wikipedia is the "work", and an article is just a fragment of the work. The license also says a link to the license text must be included. I see no reason to subvert original research, notability and relevancy guidelines here. — Zanaq (?) 20:38, 30 March 2008 (UTC)

I'm not quite sure if I follow the first part of your comment. When Wikipedia is printed, it displays the notice "All text is available under the terms of the GNU Free Documentation License." This describes the text, but not the images. I would assume therefore that images that require attribution are currently not being properly attributed when the page is printed. This is just a minor problem, though. A more major problem is that on the web, these particular attribution-requiring images are not getting their required credit unless one clicks on the image, and many visitors are unaware that it is possible to do so. The only solutions would be to either include attribution in the captions for these particular images, make it more obvious where the image's license information can be found (like the "enlarge" icon in thumbnail versions, but for license information), or prohibit attribution-requiring images altogether. That is, if Wikipedia cares about respecting the licenses of its images, which I'm under the impression that it does.

As for subverting the WP:OR policy, WP:N guideline, and WP:REL essay: WP:OR#Original images states that images are exempt from the original research policy. Also, the purpose of the original research policy is not to prevent attribution in the main article space, it's to prevent non-neutral POVs and inaccuracies. The notability guideline does not apply here either, because it's only concerned with the topics of articles, not the actual content. the relevance essay is an essay, not a guideline, and besides, attribution is relevant from a technical and legal standpoint, though perhaps not directly relevant to the topic (note that references aren't always directly relevant to the topic either, but they are necessary to back up the content). -kotra (talk) 19:48, 4 April 2008 (UTC)

I think the least we can do here at Wikipedia, after asking photographers to sacrifice their hold on photographs, is to give them attribution on the page itself. "Photo by..." is not distracting in the least. This is what we SHOULD be doing all the time, because the source page is actually a separate page for all intents and purposes. It should be standard practice that we include photographers names in the caption. I can't tell you how many Flickr users quarrel over this after being kind enough to relicense their work. If we want more photos, we may as well do this. Why was the guideline edited when the consensus above this one had more clear support? Also, when someone stipulates we attribute their work in a caption, we MUST do this, per the CC licenses, i.e. "You must attribute the work in the manner specified by the author or licensor."(Mind meal (talk) 03:07, 21 April 2008 (UTC))

I mostly agree with you, but I think only images with licenses that require attribution (like CC-BY) should have attribution in the captions. Not all images. But otherwise, I agree. As for the reversion of my edit, I think it was justified; there apparently wasn't consensus to change it (apparently there were discussions elsewhere I was unaware of, though I had looked beforehand). So, I think there is still no consensus, and this issue needs to be discussed more. I don't feel like my above points (and yours) have been adequately addressed yet, which leads me to wonder if other editors are just stalling because they don't like the addition of "Photo by John Doe" to some captions for solely aesthetic reasons. -kotra (talk) 05:01, 21 April 2008 (UTC)

edit: though I still don't know if consensus yet exists for your addition, I've shortened it to be more concise. It now says the caption should include attribution "if the photograph's license requires attribution". Hopefully no meaning is lost. -kotra (talk) 05:39, 21 April 2008 (UTC)

Hi Kotra. It seems there is no consensus either way, and yet we have one camp keeping their content on the page (generally only notable people) and acting as though it is the way of the land. That is what gets me. I am embarrassed to tell people I contact at Flickr that not only will we host their content only when they allow commercial and derivative reuse...but that we can't even provide a credit to them beneath the photo. We already ask a lot of photographers. Why not give a little back is my argument. (Mind meal (talk) 06:33, 21 April 2008 (UTC))

Discussion duplicated and continued below.

[edit] Tips for describing pictures

Further to discussion at WP:FPC talk, this section could do with brushing up, at very least. The "for photographs" list needs fleshing out, perhaps to make people aware of the WP:NOR implications of uploading (and captioning) photographs and other concerns, but certainly to weed out unnecessary suggestions like "who took it" and "why was it taken". Why was it taken? Well, er, it was a nice day, and along came this bird.. --mikaul^talk 09:51, 12 December 2007 (UTC)

[edit] why did wikipedia decide that captions should be full sentences?

I think this rule is really, really, overly prescriptive and artificial, and I don't think consensus really exists for it. Captions that are full sentences often come off as stilted and overly wordy; they take up more of the reader's time without conveying any more information. The NY Times (for example) doesn't always have full sentences in its captions, so why should we?P4k (talk) 00:23, 24 January 2008 (UTC)

What makes you think WP requires captions to be full sentences? WP:Captions#Wording even says "Most captions are not real sentences, but extended nominal groups", and includes examples that are short phrases. Gimmetrow 16:45, 27 March 2008 (UTC)

Sometimes a full sentence is a good things; other times, not. I've never seen anything stating that captions have to be full sentences. EVula // talk // ☯ // 17:03, 27 March 2008 (UTC)

[edit] Proposed change to lead

The lead seems to me to have a low meaning-to-word ratio, which is pretty much the opposite of what you want in a guideline that talks about how important it is to be snappy and succinct, I think. I edited it and then reverted myself so you guys can see my proposal. There has been a lot of discussion on WT:MoS recently about captions; I think the guideline could be very much improved simply by giving more good and bad examples of captions. The guideline seems generally helpful, but again, I think it would be worth some effort to make it really tight, since that's what the guideline is about. Thoughts? - Dan (talk) 14:25, 15 April 2008 (UTC)

Definite improvement & good edit. A four-image gallery at the bottom of the page would be the best format for examples, given the length of the {{Style}} sidebar. --mikaul^talk 14:45, 15 April 2008 (UTC)

Thank you. Not sure which version you saw, I just rewrote it, and then reverted again to comply with "consensus before editing". Please compare this version and the most recent one, guys. - Dan - Dan (talk) 15:00, 15 April 2008 (UTC)

Regarding the {{tl:Style}} sidebar, this might be one style-guidelines page which would be better off without it, so we can add more images with captions. - Dan (talk) 15:05, 15 April 2008 (UTC)

[edit] Author credits for CC licenses

When an author or licensor stipulates we include their name in the caption, we legally must do this. The guideline right now does not recognize this. The Attribution Creative Commons and Attribution ShareAlike Creative Commons licenses both state "You must attribute the work in the manner specified by the author or licensor." See: http://creativecommons.org/licenses/by-sa/2.0/deed.en and http://creativecommons.org/licenses/by/2.0/deed.en. The language in our guideline should be corrected immediately. (Mind meal (talk) 03:45, 21 April 2008 (UTC))

Seriously - point me at an image where this is the case. I've dealt with thousands of images on Wikipedia, helped with permission on WP:OTRS and even uploaded photos for people who couldn't understand our system - in more than 2 years, I've never seen someone ask for credit in a caption. If they are, I'd seriously consider passing on the image. Shell ^babelfish 05:37, 21 April 2008 (UTC)

Ahh - wait, you're reading the quick summaries that CC gave. May I suggest you read the actual license text? [9] It states "give the Original Author credit reasonable to the medium or means You are utilizing". When the quick summary says "manner" it means crediting it as they state, i.e. using their full name if that's what they give, their nickname if that's what they give etc. This has nothing to do with where the attribution is located. Shell ^babelfish 05:44, 21 April 2008 (UTC)

What exactly is keeping Wikipedia from attributing photos to the author in captions? What makes such practices unreasonable to this medium? If you wrote a book, would you expect your name to appear on the front cover or only on the title page? Because it is so very easy for us to include attribution beneath a photograph, there is nothing that would make it unreasonable to our medium or the means by which we do things. This is a guideline anyway, not policy. If you want it to become policy, propose it to become policy. People keep citing consensus, but I see no consensus for excluding author names from credits....not on this page. If anything I see a divide, and where there is a divide I don't think we should say it must be this or that. (Mind meal (talk) 06:21, 21 April 2008 (UTC))

Also, if someone prints a Wikipedia page out there is no attribution for the photo. By default, anyone who redistributes the material is then violating the terms of the license. Yet another reason why attribution should appear in the caption. lastly, why risk not giving attribution when asked? (Mind meal (talk) 06:27, 21 April 2008 (UTC))

Guidelines are just as enforceable as policy -- since both are just a reflection of community practice, its all about consensus. If you want to make this kind of a change, why don't you gather opinions on it first by talking about it on the Village Pump or something similar? Wikipedia isn't responsible for someone re-using our content incorrectly -- in your example, they'd also be violating our GFDL license but you don't see us slapping page authors on the article to resolve that issue, do you? You may also want to refer to the image guidelines where it talks about not placing signatures, names or copyrights in photographs - yes, we ask a lot of photographers, but we ask the same of every editor who writes something for Wikipedia; so far, we don't seem to be lacking in either area. Shell ^babelfish 06:52, 21 April 2008 (UTC)

"...its all about consensus." Where is it? Where is the consensus you assert on not including credits in captions? I see a lack of consensus and editors asserting it where there is none. The fact is that there is no consensus either way regarding this matter. You say, "...so far, we don't seem to be lacking in either area." Are you kidding me? There are tons of articles that need photos. I should have guessed that anything as common sense as saying who the photographer is under a photo would be met with bureaucracy here on Wikipedia. That is pretty much par for the course on this site. Aside from asserting "common" practice, you have yet to say exactly why the inclusion of author name in a caption is so undesirable. On the contrary, I have done nothing but give logical, thought out rationale for why we should. You have thus far failed to do so. Could you tell me why you feel they should NOT be credited with taking the photo in the caption? "Wikipedia isn't responsible for someone re-using our content incorrectly." But Wikipedia is responsible for using content correctly. (Mind meal (talk) 08:34, 21 April 2008 (UTC))

Any "common practice" on the wiki is a de facto demonstration of consensus, surely. Of the several million images used here, you won't find a single one crediting a non-notable author in the caption, nor do the many thousands of non-notable authors of those images insist on attribution beyond the image description page. I'd be very surprised if you could show anything near consensus for attribution in captions.
Images in article mainspace are only thumbnails, after all, essentially links pointing to image description pages where attribution is always given. Of course anyone remotely interested in either the image or the author will click the thumbnail. We don't have to put "click here" next to every hyperlink. Wikipedia is essentially a massive hyperlinked database, and anyone who hasn't, erm, clicked to that isn't ever going to find anything here anyway. The same is the case on many image-heavy websites, Flickr included: attribution is only bound to appear where the image is displayed as the main subject of a page.
I guess you could argue that attribution is needed when viewing images here at 100%, and you'd have a much stronger argument. What I can't see is any need to clutter up articles with pointless, non-notable names; that and and the self-reference issue are insurmountable barriers to caption credits, AFAICS. --mikaul^talk 12:47, 21 April 2008 (UTC)

Thanks for explaining that mikaul, you put that in a much more clear manner than I was managing. Shell ^babelfish 13:59, 21 April 2008 (UTC)

(Also posted somewhere above): Wikipedia:Administrators' noticeboard/Archive98#Images, Wikipedia:Administrators'_noticeboard/IncidentArchive302#Photography edit war on model Ana Beatriz Barros and WP:IUP#User-created images may be of interest. The second one is the one closest to the issue at hand, I think, and it is related to the third link. I don't know about legal issues but I think these are likely best raised with the Foundation lawyer, Mike Godwin, User:Mikegodwin. x42bn6 Talk Mess 14:03, 21 April 2008 (UTC)

I have problems with some of mikaul's statements here.

1. "Of the several million images used here, you won't find a single one crediting a non-notable author in the caption, nor do the many thousands of non-notable authors of those images insist on attribution beyond the image description page" This is not correct. Foxfire (bioluminescence) is one example of an article that already credits its non-notable author because of its licensing (CC-BY-SA). True, I was the one that added the attribution long ago, but surely others exist. It hasn't been reverted yet, so it must be ok with the people who saw it (though to be fair, it's probably a fairly untrafficked article). However, this is all irrelevant because guidelines are not based on what is current practice. It's the opposite: current practice is (or at least should be) based on the guidelines.

2. "I'd be very surprised if you could show anything near consensus for attribution in captions." This is a straw man. Nobody was saying that there is yet consensus for attribution in captions. However, Mind meal and I were saying that there isn't consensus for NOT having attribution in captions. There is no consensus either way, and so we should try to work toward consensus.

3. "Images in article mainspace are only thumbnails, after all, essentially links pointing to image description pages where attribution is always given." I disagree. Images in the mainspace are not merely thumbnails. If they were, we wouldn't enlarge images beyond their normal thumbnail size to make crucial details and text legible, as per Wikipedia:Manual_of_Style#Images. Images are in the mainspace to illustrate the article, and if necessary, people may click on them to see a larger version.

4. "Of course anyone remotely interested in either the image or the author will click the thumbnail." Unless they don't know they can. New visitors to Wikipedia or less web-savvy ones may not know the images in Wikipedia are clickable. The pointer hand is usually the only clue they're clickable, and it's not obvious unless you hover over the image, nor will everyone know what it means if they do see it. Basically, we can't just assume people will know about the image description page.

5. "We don't have to put "click here" next to every hyperlink. Wikipedia is essentially a massive hyperlinked database, and anyone who hasn't, erm, clicked to that isn't ever going to find anything here anyway." This is another straw man. Text links are different from image links. Text links are conspicuously blue, so people can know there are links in the text are without hovering their mouse over all the text. Images on Wikipedia, on the other hand, just look like unlinked images everywhere on the web. The Enlarge icon in the caption helps, but it is easily overlooked (I never realized it had a purpose until this discussion).

6. "The same is the case on many image-heavy websites, Flickr included: attribution is only bound to appear where the image is displayed as the main subject of a page." I believe this is false. Flickr attributes the author everywhere their images are displayed, even on pages where the image isn't the main subject. Examples: [10][11]

7. "What I can't see is any need to clutter up articles with pointless, non-notable names;" The names may be non-notable, but they aren't pointless. As I said in the section above, complying with legal requirements of licenses trumps any concern about clutter or appearance in my opinion. If people don't like attribution in captions for appearance reasons, there are several options: a. Don't use images that require attribution, b. Ask the copyright holder for permission to only have attribution on the description page (in CC-BY, this would fall under "You must attribute the work in the manner specified by the author or licensor"), or c. Instead of text in the caption, include a small icon of the word "Attribution" or "License" or something similar linking to the image description page, much like the Enlarge icon in captions already. This would require a bit of creative coding to make the license info show in printed articles, but it's certainly possible.

8. "that and and the self-reference issue are insurmountable barriers to caption credits, AFAICS." If by "self-reference issue" you mean that editors would be including their own name in the mainspace, I don't think this is a problem. The text of articles isn't signed or WP:OWNed, but images are different. WP:OR describes how images are exempt from the original research policy, so I see no problem. -kotra (talk) 22:26, 21 April 2008 (UTC)

<outdent> response to kotra:

1. Straw men are all over this discussion, aren't they? ;o) I don't expect that credit will stay there for ever, as it's plainly contrary to image use policy.
2. A consensus for caption credits has been claimed at least once on this page.
3. They're thumbnails. They function as thumbnails, are referred to as thumbnails, and (not surprisingly) are described as thumbnails in the wiki markup.
4. Apart from technophobes, there probably are people sufficiently disinterested in images that they never click a thumb to see one, so why go to the trouble of posting up credits for them? We can safely assume that anyone interested enough will click, eventually.
5. I like the icon idea. Good compromise.
6. [12]
7. ditto #5
8. By self-reference, I mean the appearance of wiki userames in article namespace. This factor is the nemesis of all caption credit proposals.
9. If you float the icon idea as a compromise proposal at the VP, I think it'll fly. --mikaul^talk 01:41, 22 April 2008 (UTC)

1. Apparently someone just rightly fixed that caption I mentioned; after reading that policy to the end I see that you're right: "All photo credit should be in a summary on the image description page". Doesn't mean it can't be changed though, if there is consensus to do so.

2. Yes, I claimed that there was consensus a while back because the most recent discussion there was all on one side, and after waiting a few months for any opposition there was none. However, since then (and apparently before then, in other places I was unaware of), there was opposition, so I agree, there is no consensus now.

3. I agree that they are thumbnails. I never said they weren't, just that that's not all they are, or even, I would argue, their main purpose. Their main purpose, in my mind, is to illustrate the article as standalone images. The click-to-enlarge ability is secondary. People (well, at least me, I can't speak for anyone else) don't click on all the images they want to look at. I just look at them; if I need to seem them larger, I'll click on them.

4. I don't think it's safe to make that assumption. Besides, it's not for the benefit of viewers that images requiring attribution are properly attributed, it's for the benefit of the licensors. It doesn't matter if the readers don't care who took the picture, it still must be prominently credited as per the license.

5. Thanks! The icon idea seems to be getting some good feedback at the village pump, too. Maybe it will be used, if it turns out the current attribution system is inadequate.

6. Someplace having linked images similar to Wikipedia doesn't mean visitors will automatically know Wikipedia's images are clickable for more information. Unlinked images elsewhere still look exactly like Wikipedia images.

7. I guess you mean "ditto #6". You're right, there's one. I think they allow it there because the images are tiny, with a big number obscuring part of it, and the purpose of those thumbnails is to take the viewer to the main image page. Wikipedia's article images are a little different (and anyway, why are we comparing what Wikipedia should do to what other websites do?).

8. You may be right that "Photo by kotra" could look self-promotional in a caption and maybe occasionally cause some sort of conflict of interest, but I don't see any actual policy reasons for it. As far as original research is concerned, images are exempt, and I would say that attribution (in the caption or elsewhere) is just a part of the image.

9. Could be. I might try that if the current proposal fails, and if Mike Godwin's response is encouraging. -kotra (talk) 04:16, 22 April 2008 (UTC)

Discussion ported to the Village Pump here.

[edit] Use of "Figure X", Table X"

Captions do not generally include a figure number so it can be linked to the text. On some articles with many illustration that are important to explain the text it is probably a good idea to do this as is done in paper sources. -- Alan Liefting (talk) - 08:49, 3 June 2008 (UTC)

We can do even better: instead of giving the reader a name to hunt for, use an invisible wikilink. For instance, I put "<span id="caption1" />" immediately in front of the Burma-Shave image on the project page. You can link directly to the image from any other page with WP:Captions#caption1, or add a "|" character to make it appear as "See (image)", if you like. #caption1 would link to that image from anywhere else on the same page. - Dan Dank55 (talk)(mistakes) 02:03, 7 June 2008 (UTC)