Wikipedia talk:Manual of Style/Infoboxes/Archive 9

Archive 5

←

Archive 7

Archive 8

Archive 9

Archive 10

Archive 11

→

Archive 15

Infobox guidance ambiguity regarding summarization

I see an ambiguity in the guidance. In the first paragraph we see:

"An infobox template is a panel ... that summarizes key features of the page's subject...."

Later, at Purpose of an infobox, we see:

"...keep in mind the purpose of an infobox: to summarize key facts in the article in which it appears."

And at the bottom of the page at References in infoboxes we see:

"References are not needed in infoboxes if the content is repeated (and cited) elsewhere or if the information is obvious. If the material requires a reference ... and the information doesn't also appear in the body of the article, the reference should be included in the infobox. However, editors should first consider including the fact in the body of the article."

The particular ambiguous language is in italics: "of the page's subject" vs. "in the article" vs. "in the body of the article". Well, I've often seen editors remark "if it isn't in the article text, it can't/shouldn't be in the infobox." But infoboxes in general do not follow this 'rule'. In fact, many articles have infobox information which should not be rendered into prose, simply because it is tabular type data. With these thoughts in mind, I suggest that the Purpose of an infobox text be modified to read:

"...keep in mind the purpose of an infobox: to summarize key facts of the page's subject.'"

Rationale & discussion:

1. The proposed change provides for consistency between the two paragraphs.
2. It helps make clear that data need not be repeated.
3. My old World Book has an article about the United States. The article has a "Facts in brief" box that says the Bird of the US is the "Bald eagle, adopted June 20, 1782". But the text does not mention the bald eagle.
4. Different location and institution infoboxes have parameters for {{coord}}. Do we need to duplicate that data in the text?
5. I do not think using "key facts" v. "key features" makes much difference, but one of the "key X's" could be modified.
6. Additional language that explicitly says "Infobox facts need not be repeated in the article text" (or vice versa) may be helpful as well.

Thanks. – S. Rich (talk) 19:42, 6 November 2013 (UTC)

Support Per the arguments given CRwikiCA talk 14:29, 7 November 2013 (UTC)

The proposed changes have been implemented. (But I just now posted a notice about this discussion at Wikipedia talk:WikiProject Infoboxes#Change in infobox MOS guidelines as I did not realize a WikiProject on Infoboxes existed.) – S. Rich (talk) 18:15, 17 November 2013 (UTC)

Strong Oppose and will revert any premature changes made. This opens the door to a fundamentally different conception of a highly sensitive subject. It's way too soon to change anything. The primacy of article text is extremely important and should not be thrown away. If a fact is not important enough for a place anywhere in the article text it generally should not be in the infobox either, as it can't be "key". Some infoboxes have far too much information in them as it is, and this tendency to sprawl should not be encouraged. Johnbod (talk) 14:06, 11 December 2013 (UTC)

I note the changes were reverted. But is it way too soon? Is this such a sensitive subject? (A month goes by and 2 editors comment.) But more to the point, what about the ambiguity in the MOS? To address your points: I'm not clear on where "primacy of article text" comes from. It seems that infoboxes serve well to avoid cluttering up article text (as I stated in points 3 and 4 above.) By avoiding a "requirement" that info be in the text before it goes into an infobox, we actually keep overall article sprawl to a minimum. Take a look at USAT Dorchester – the infobox has all sorts of technical data in General Characteristics which should not be in the text. The change I propose clarifies that such data is permissible. Thanks. – S. Rich (talk) 15:48, 11 December 2013 (UTC)

I don't see any reason why any of that information should not be in the article, which seems to lack a general description of the ship which would (if done properly) no doubt be able to work that in with plenty of other relevant information and context not suitable for an infobox, and much improve the article. There is no way an article with similar characteristics should be passed for GA for example. But I accept that some subject areas are much more suitable for more detailed infoboxes than others. However the changes proposed here make no allowance for this, nor for the fact that information given in infoboxes is both more likely to be inaccurate (because so often added by a drive-by infoboxer) and less likely to be checked and referenced. I recognise the problem, and am not wholly against infobox-only information. But we risk opening a can of worms here and any recognition of this in the MOS should be cautiously worded and qualified. Johnbod (talk) 21:22, 11 December 2013 (UTC)

I saw it only today. The idea is good to avoid different unclear wording in the guidelines. Infoboxes is a sensitive subject ;) --Gerda Arendt (talk) 16:07, 11 December 2013 (UTC)

I also hadn't realized this was rather stale. But there are I'm sure many editors who would be as strongly opposed to this have never seen it. Johnbod (talk) 21:14, 11 December 2013 (UTC)

• Comment - in areas where I am active (military history and aviation wikiproject subjects), I see infoboxes that do not necessarily contain information that is given elsewhere in the article. Sometimes this is because having the information in the article and the infobox is unnecessary duplication at that stage of the article's life. In this respect, the infobox draws some technical details out of the narrative. The information is not irrelevant but just better presented in the infobox than text. That said, in those wikiprojects, there is consensus on what is relevant and what is just too much detail. And information is not accepted without sourcing. GraemeLeggett (talk) 17:02, 11 December 2013 (UTC)
  • I think you've highlighted some key issues here: (a) few (no?) Projects are anywhere near as "together" as MilHist, and (b) different areas have different uses for infoboxes, which makes blanket policies on infoboxes difficult. To the MilHist crowd (I infer) duplication is undesirable. In many other areas, duplication is the point—the box itself is supposed to be a summary of certain key points of the subject.
    I can see, for example, wanting a ton of details in an infobox for a city—as a reader I use the infobox to find things such as population density or GINI, and don't want to scour the article to find those details. With other subjects, say book infoboxes, I think those kinds of details (Dewey decimal number, ISBN of the out-of-print first printing, cover artist of now-unavailable, thoroughly out-of-print editions, etc) are cruft, a waste of space, and sometimes totally misleading, while other see those details as the point of the infobox.
    If anyone can find a workable solution to this whole ridiculous mess, I will nominate them on a peacemaking mission to Israel/Palestine. Curly Turkey (gobble) 21:42, 11 December 2013 (UTC)

As an example, my old bugbear, the appalling World Heritage Site template, as seen at Taj Mahal for example, still wastes a ton of space on totally un-needed details from the WHO award documentation, which should only be in the article in a text note if anywhere, and certainly not in the infobox. This change will encourage the unhindered spread of such cruft. Johnbod (talk) 21:14, 11 December 2013 (UTC)

• Comment by OP: I hope and ask that we keep the discussion focused on the ambiguity I presented, and resolve the language of the MOS. I submit that issues about cruft, non-sourced material, etc. be addressed in a separate thread. More along the lines of rewording and strengthening WP:Manual of Style/Infoboxes#Using infoboxes in articles guidelines. Thanks. – S. Rich (talk) 00:37, 12 December 2013 (UTC)
• Comment - yes but I don't think you can easily separate them. There seem to be several related issues:

a) inconsistent wording, which of course should be avoided;

b) the view that as a principle "If a fact is not important enough for a place anywhere in the article text it generally should not be in the infobox either";

c) that as a corollary some infoboxes have too much cruft;

d) the view that some infobox information may be both important but not needed in the article text

e) that different infoboxes attempt to solve their unique problems in ways that are not consistent with one another.

I find myself in agreement with all of the above except (b), which seems to me too draconian an approach to deal with the problem correctly identified in (c).

I therefore:

i) Support the proposed change;

ii) Agree additional language that explicitly says "Infobox facts need not be repeated in the article text" would be helpful along;

iii) Suggest adding a statement to the effect that infoboxes should avoid providing significant amounts of trivial detail. Ben MacDui 09:19, 12 December 2013 (UTC)

Have you got any examples of d), which I think is the weak link here? I think you are underestimating the ambitions and intentions of some "big data" figures, and the way this relaxation would be exploited by people who like adding all and anything but feel uncomfortable writing prose (quite a significant section of our editing base). I don't believe this change should be made without say an RfC. If it was, there should certainly be extra cautions, but currently the policy, rightly, does not permit any "trivial detail" so as described above this would be a further large relaxation. Johnbod (talk) 12:37, 12 December 2013 (UTC)

the MoS should not be used to impose changes or restrictions on our editing but to summarise current practice. So we don't need an RfC to add to the MoS a description of what is already done widely, by the community of editors. (Indeed, an RfC would be need to impose a restriction on what the community may do.) Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:03, 13 December 2013 (UTC)

Comments - The content of infoboxes should be a summary of the key facts of the subject. I would suggest that our guidance should advise "a brief summary" because I recognise that there is a tension between those who want the key facts to be as comprehensive as possible for the reader and those who wish to reduce the impact of a large infobox in an article. I disagree with John's assertion that drive-by editors are more likely to introduce incorrect information into an infobox than into the text; in fact having deliberately redundant information helps ensure that checks are made whenever discrepancies arise. For that reason, I would suggest that our guidance should advise editors that normally information in an infobox should reflect what exists in the body of the article, although there may be subject-specific exceptions. The most obvious exception that I am aware of is in {{infobox disease}} which invariably contains the ICD-10 classification for the disease. It is a simple fact and not worth mentioning in the body of the text - I mean there's nothing more to say about it and everyone is accustomed to finding it in the infobox anyway. Nevertheless, such exceptions should, I believe, be rare. --RexxS (talk) 15:13, 12 December 2013 (UTC)

Actually I pretty much agree with that, and the ICD-10 is a good example - a geolocation would be another. But now, if changes are to be made, we need a draft that reflects that sort of careful distinction; this might well be quite long (a couple of paras). I would like to see more use made of secondary boxes for statistical-type information that doesn't really deserve space right at the top, but is needed somewhere and is better suited to a list format. Johnbod (talk) 15:54, 12 December 2013 (UTC)

Quite agree an RfC makes sense and that 'trivial detail' should be discouraged. However, there are examples of data that's in an infobox but may well not be needed in the prose section, depending on one's view of "trivial".

Scottish islands: area rank (especially if it is a number >10); population rank (ditto); population density; maybe Norse name if there is no etymology section and/or nothing more is known about this other than the fact of its existence. See e.g. Islay.

Countries: proportion of territory that is freshwater; Timezone, Drives on the...; Date format, Internet TLD. Ben MacDui 18:05, 13 December 2013 (UTC)

Excellent examples of things (except timezone and maybe Internet TLD) that should not be in a lead infobox but could well go in a secondary box (Template:Other junk maybe). A useful rule of thumb, with a few exceptions like the ICD-10, is that nothing should be in the infobox whose absence in the lead would (should) not be raised as an issue at FAC. I don't see any reason why the Scottish stuff should not be prose only; the box at Islay is potentially baffling to those not used to the Scots and their little ways, who won't realize that it is Scottish islands featuring in the "rank" data, unless they read the 7 lines of small print at the bottom of the box. That should certainly be changed - the box is far too long, and there are no photos of the place until the 3rd screen. Johnbod (talk) 18:39, 13 December 2013 (UTC)

There's no need for secondary infoboxes, and several reason why they might be harmful. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:03, 13 December 2013 (UTC)

Actually, infoboxes are simply another sort of table, like census & climate data tables. They have links, so they serve as navbars & sidebars. They do enjoy the distinction of normal posting at the top of the page, which results in more attention for editors. But, given the variety of infoboxes, it is inevitable that some data should be exclusive to the infobox. (And they contain images – so we cannot say "nothing should be in the infobox unless it's in the article text.") IMO the concerns as to what data belongs in what infoboxes are best handled in discussions involving the particular infoboxes and/or the WikiProjects which focus on particular topics. (E.g., WP:WINE is the best forum to handle {{Infobox wine grape variety}}.) My goal in this discussion is to remove the ambiguity in the MOS which can give weight to editors who say "remove because the MOS says this" vs. "keep because MOS says that". – S. Rich (talk) 19:42, 13 December 2013 (UTC)

Of course they are, but if boxes get referred to here as examples it is worth considering whether they are good examples, & I don't think the Islay box represents best practice. It does represent a stage in the cruftification process complained of above. That a 7-line small print note is required to explain what the information is shows straight away that it is lkely to be simpler to put that information in the text in the first place. Johnbod (talk) 03:31, 14 December 2013 (UTC)

Obviously we are all grateful for your insights into cross-cultural understanding as well as the specifics of infoboxes. Re the "Islay box" I don't much like the existing wording and there probably is a simpler way to state this. It was only put in place recently when the census came up with a rather complicated new set of tables. I will look at this, but the real point is that, in my view, there is no earthly point of going to the trouble of repeating useful but standard information over and over again in the prose section, with all the issues that involves about consistent phrasing etc when a few characters in an infobox will do the same job for a tiny fraction of the effort. This infobox has not changed much over the years and you seem to be conflating such changes with allegations of a system-wide process of adding cruft, which I am largely unaware of. In other words I agree with S. Rich that "what data belongs in what infoboxes are best handled in discussions involving the particular infoboxes and/or the WikiProjects" rather than some ill-conceived meta-restriction. Ben MacDui 10:15, 14 December 2013 (UTC)

We have always had what you may consider an "ill-conceived meta-restriction" on this page: ""An infobox template is a panel ... that summarizes key features of the page's subject....", and at Purpose of an infobox: :"...keep in mind the purpose of an infobox: to summarize key facts in the article in which it appears.". All I am saying is that we should give very careful consideration before sweeping it away. And pointing out that the Islay box does not seem to conform with it. The argument that it is just too much trouble putting stuff in prose (even though it could be done in a largely standardized "template" paragraph) cuts no ice at all with me. Johnbod (talk) 13:37, 14 December 2013 (UTC)

I think its clear we disagree on both the principle and the practice and I don't have a great deal to add save that you might look at the subject differently if someone asked you to add and police c. 350 templated paragraphs (which would need to be infobox-like in that they would require individual data to be added) in preference to an existing system that has passed muster at GAN and FAC umpteen times. Ben MacDui 12:18, 15 December 2013 (UTC)

Infoboxes - are the current guidelines for their use sufficient?

There has been contention over whether to add Infoboxes or remove them. Can editors please give their views here ? Thanks. Acabashi (talk) 04:45, 1 January 2014 (UTC)

I have altered the section heading to be more informative. -- TRPoD aka The Red Pen of Doom 13:22, 1 January 2014 (UTC)

Wikiproject Plants : Huge info boxes (again)

Huge infoboxes continue to plague Wikipedia, despite this MOS page being reasonably clear on the matter. It seesm Wikiproject Plants wants their infobox to contain all synonyms for a plant species. I came across this in Onion where members of that project are edit warring to have an info box with a couple of screenfulls of obscure Latin names. Opinions on that article and perhaps also at Wikiproject Plants would be useful. Apologies for posting this notice here but there doesn't seem to be an active MOS wikiproject or noticeboard that I could find. -- Colin°^Talk 08:31, 17 March 2014 (UTC)

(Disclaimer, I am not a party to edit warring, and aren't a member of Wikiproject Plants). If a page is going to be setup as a page about a scientifically defined species, then all those things that define the species should logically be included. In this case I have doubts that onion should be treated as a species article, but should be a vegetable article covering all the plant types known as onions (see Talk:Onion). But for species articles a list of synonyms makes sense as species are often re-described, renamed or mis-named and you want all those words for the same scientific entity to find the same page. Admittedly that argument doesn't necessitate synonyms being in the infobox (speciesbox/taxobox) but some standard mechanism is needed, and that's certainly where they are put on most species pages. At present the problem of excessively long lists of synonyms is often handled using a collapsible box, eg see dog. --Tony Wills (talk) 09:34, 17 March 2014 (UTC)

While that is a long infobox, it's not overpowering compared to the length of the article. As the list of synonyms is the last item listed (so there's no need to scroll past them to find anything else in the infobox), where's the harm? And why have you raised this here, when you have already done so on the article talk page? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:31, 17 March 2014 (UTC)

Because I've been told that consensus between two editors or even on the article talk page will be ignored by future Wikiproject Plants members (I know all this is opposite to WP guideline where consensus on the article trumps any project). I think this is an area where the Wikiproject could do with some help from MOS experts. Andy, that's a couple of screenfulls of latin -- how on earth does that meet any guidelines on info boxes or what should appear in the lead? And the harm is that all the pictures bunch up. 99% of our readers will be more interested in pictures than latin. -- Colin°^Talk 12:24, 17 March 2014 (UTC)

I agree with you about WP:OWNership by projects, but if that happens, there are methods for dealing with it. Forking a discussion is not one of them, and is not helpful (WP:CANVASS might also apply). I don't see any "bunching" of pictures; that might be specific to your setup. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:59, 17 March 2014 (UTC)

Query about an infobox

[1] shows an infobox which names two individuals seven times each in a single infobox. I suggested once each would be of value, and the iterations of nil value. In addition unnamed "friends and associates" are listed as "participants" in criminal activity. Might I get input as to whether I am wrong on this? Cheers. Collect (talk) 16:30, 28 February 2014 (UTC)

It did seem excessive, and trimming it seems to have stuck, though it's still misusing the convictions field to give sentencing detailia. Seems like a better discussion for that article's talk page.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:43, 19 April 2014 (UTC)

Redlinks in infoboxes – proposed guidance

Seems to me that various infobox parameter guidelines discourage redlinks in infoboxes. (I have no examples in mind.) With that said, what thoughts do contributors have about explicit guidance that says "Don't add redlinks to infoboxes." – S. Rich (talk) 03:01, 16 April 2014 (UTC)

With the standard caveat of "occasional exceptions may apply", I would be in favor of such a proposal. Joefromrandb (talk) 06:14, 16 April 2014 (UTC)

I think the guideline at Wikipedia:Red link, which recommends red links in certain cases, should be applied to infoboxes just the same as it applies to body text, especially in the scientific infoboxes which contain large chunks of material not found in the article. Where a term also occurs in the body text, red-linking it once is probably enough, and it may be better to omit it from the infobox. -- Michael Bednarek (talk) 06:35, 16 April 2014 (UTC)

I support all you said. I personally don't repeat a red link from the lead to the infobox, but from the body. The best way to deal with it is to create a stub ;) --Gerda Arendt (talk) 11:49, 16 April 2014 (UTC)

I would be against; redlinks are a valuable part of the Wikipedia ecosystem. I note that no rationale for prohibiting them is presented. With no rationale and no examples, this seems to be unnecessary bureaucracy. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:08, 16 April 2014 (UTC)

I'll give an example of problematic redlinks & rationale. See: Republican Liberty Caucus. The redlinks include Matt Nye as the treasurer. 1. This is dated info, Nye is now the president of the RLC; 2. He is a former candidate for the Brevard County, Florida Commission. Is he likely to get a WP article? In terms of notability guidelines, no. (In terms of promotionalism, yes.) So, his redlinks are not "useful in editing article text to create a red link to indicate that a page will be created soon or that an article should be created for the topic because the subject is notable and verifiable." (WP:REDLINK). The infobox "summarizes key features of the page's subject." If the redlinks in the infobox existed because of genuine notability, they could be supportable. But for folks such as Nye, they simply imply that he's notable. They do not follow infobox guidance and simply distract. – S. Rich (talk) 14:53, 16 April 2014 (UTC)

The issue in that example is not "red link in infobox" but "red link for (allegedly) non-notable subject". Deal with that as you would for a red link for an (allegedly) non-notable subject, in prose. As you effectively say, "redlinks in the infobox for notable subjects are supportable". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:32, 16 April 2014 (UTC)

Actually, the issue in that example is red-linking personal names, and WP:REDNOT addresses this situation. Red links to personal names can and should be removed; whether they're in the info box or the article proper is irrelevant. Joefromrandb (talk) 21:42, 16 April 2014 (UTC)

Thanks all for your excellent comments. With REDNOT & REDLINK in mind, I suggest we add a simple line (like what we have for flag icons). Simply say "In accordance with REDNOT and REDLINK, avoid redlinks in infoboxes." As the focus of this project page is infoboxes, we give proper emphasis to this design principle (and leave the prosaic redboxes alone.) – S. Rich (talk) 14:32, 17 April 2014 (UTC)

Sounds good. I for one am much less sympathetic to redlinks in infoboxes than the occasional selected one in text. Johnbod (talk) 15:29, 17 April 2014 (UTC)

You appear to have overlooked my comments in drawing up your proposal, which I oppose. Furthermore, the MoS should be documenting current practice, and community consensus, not tying to dictate it. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:44, 17 April 2014 (UTC)

I have not overlooked your comments. But maybe I misunderstand. Are you saying you'd allow redlinks in infoboxes (regardless of REDLINK and REDNOT)? – S. Rich (talk) 15:59, 17 April 2014 (UTC)

No, I'm saying that REDLINK and REDNOT are sufficient; we don't need an added layer of bureaucracy. No do REDLINK and REDNOT say that we should avoid red links in infoboxes per se. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:29, 19 April 2014 (UTC)

I'm not advocating for any more bureaucracy (which I understand to be stuff like blocks, bans, arbcom discussions, appeals, etc.) While redlinks in the text may be appropriate, I think discouraging them from infoboxes is appropriate too. How about modifying the line to say "Redlinks in infoboxes should follow REDYES/REDNOT guidelines."? – S. Rich (talk) 16:42, 19 April 2014 (UTC)

• I like the idea, and questionably civil accusations of "bureaucracy" don't weaken it. Infoboxes are supposed to be summaries of key information; redlinks in them are an annoyance and a distraction that serve no purpose. If the MOS:INFOBOX instructions are being followed, such that information is being put not only into the infobox but also in the article body, then anything you would redlink in the infobox will be redlinked already in the article's main text, thus serving the "redlinking helps inspire article creation" purpose some but not all editors firmly believe in. PS: A belief that MOS is supposed to just be describing what editors are doing is fundamentally misunderstanding what a style manual is. Like all of them, MOS is prescriptive by necessity and design, and the majority of the WP MOS is based on observation that certain editing patterns are unhelpful to readers and other editors (i.e., observing what to not do). MOS intentionally does not include most standard operating procedure in English-language writing (i.e. "documenting current practice") because no one here needs to be told that commas exist for a reason, that sentences begin and end in certain ways, that verbs are not written in ALL-CAPS, etc., etc., etc.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:25, 19 April 2014 (UTC)

Andy's comments are appreciated. I think he's more concerned about WP:KUDZU than WP:NOTBUREAUCRACY. And with his comments in mind I'm trying to craft clear, workable, and helpful guidance for this minor problem. I'd like to be able to post edit summaries that pinpoint the inappropriate redlinks in the infoboxes rather than guidance that is more general. – S. Rich (talk) 22:09, 19 April 2014 (UTC)

Adding comment to thwart autoarchiving. Will try to summarize (and perhaps implement) in a day or so. – S. Rich (talk) 05:44, 20 May 2014 (UTC)

Implemented, along with some other changes/tweaks. – S. Rich (talk) 17:23, 20 May 2014 (UTC)

Edits to lead

@AlanM1: Let's use this talk page to discuss your proposed change. If all you are trying to do is cleanup the "and/or", I have no objection to that. But you are changing singular words to plurals, which may be interpreted as encouraging more images & maps in the infobox. That is at odds with the tenor of another discussion on this talk page re infobox length, in which a number of editors have expressed concern re length. As you have not established a consensus for your changes, I have reverted them. Let's leave the lead as it was, until we see if a new consensus forms. Barryjjoyce (talk) 01:09, 22 April 2014 (UTC)

@Barryjjoyce: "and/ or" is clearly wrong (the space doesn't belong). I was going to change it to "and/or" and found that was "discouraged" by the MOS as well. In thinking about the usage, then, I realized that there are commonly, intentionally (through multiple image parameters), multiple images in many types of infoboxes, and the phrase was wrong to begin with. Really, it should just refer to "images", as maps are images as well. There should be as many images as it takes to accomplish the particular task, as always. —[AlanM1(talk)]— 03:14, 22 April 2014 (UTC)

Thanks for the explanation. If you want to change "and/or" to "or" or to "and", I have no objection. As for your other proposed changes, let's hold off for now. There is an ongoing discussion re infobox length. Let's see how that plays out, and then revisit this issue then. Barryjjoyce (talk) 01:08, 23 April 2014 (UTC)

Infobox length

The guidance on the Help:Infobox page says that Infoboxes should be "concise" and not "lengthy," but that guidance is open to interpretation. Some infoboxes have quite a bit of content. For example, the infobox for Pittsburgh stretches on for 2.5 pages on a regular-size computer screen. I've seen a number of editors on various talk pages (including this one) grumble about certain infoboxes becoming bloated.

My question is: How long is too long? There are clear policies for when an article is too big, and for when the lead is too long. Can we come up with some clear guidance on the suggested maximum length of an infobox? (PS - I left a similar note on the Help:Infobox talk page, but received no comments, so I though I'd try here). Barryjjoyce (talk) 01:17, 17 April 2014 (UTC)

I'm not sure what the role of that help page is (wrt length) when the community-agreed MOS guidance is here. This page says the role is "to summarize key facts that appear in the article. The less information it contains, the more effectively it serves that purpose". I have tried quoting this guidance to editors from Wikiprojects but they just don't want to listen. These are subject-enthusiasts who love every minute factoid. The reader of onion is presented with out-of-date historical synonyms -- Latin terms nobody uses. The reader of London King's Cross railway station is shown raw data of annual entry and exit stats over five years. In no stretch of understanding could these be considered key facts or summaries of information that appears in the article. So I don't think the problem is that MOS is unclear. I think the problem is that wikiprojects think they know better than MOS and they know better than the general reader. So WP ends up with articles with lead sections that contain information a handful of people in the world care about. And WP ends up with articles where the images people like to see accompany their text all get pushed down to make way for dry statistics. -- Colin°^Talk 07:06, 17 April 2014 (UTC)

The problem here would seem to be that the MoS is (or rather its authors are-) trying to dictate article content, rather than the more proper function of describing community norms. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:31, 17 April 2014 (UTC)

Nah. The community knows long infoboxes full of tedious shit is bad. The problem is Wikiprojects amplify the opinions of specialists and subject-nerds way above what the audience want. It wouldn't be an issue if people could go to articles and sort out the problem without being bullied by the wikiprojects telling them how it must be done. -- Colin°^Talk 18:08, 17 April 2014 (UTC)

Um, Colin, you are not correct on this the bullying tends to come from people who say things like "tedious shit" One person's "tedious shit" is another's critical data that would bog down the article if placed in narrative or even someone elses lifesaving data. Montanabw^(talk) 07:20, 22 April 2014 (UTC)

Wikiprojects, or specifically their members, are part of the community. It wouldn't be an issue if people could go to articles and sort out the problem without being bullied by MoS enforcers telling them how it must be done. But do share your evidence of what the audience want. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:05, 17 April 2014 (UTC)

Someone here seems unaware of W:LOCALCONSENSUS policy. Hint: It's not Colin. Wikiprojects too often forget that they are only part of the community and do not get to dictate content or formatting at articles they claim within their scope.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:37, 19 April 2014 (UTC)

I'm confident that my editing history is ample evidence that I'm well aware of LOCALCONSENSUS, having had my fingers burned trying to prevent it overriding wider consensus on more than one occasion. That does not mean that project members should be denied their voice as individuals. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:31, 20 April 2014 (UTC)

I can vouch for Andy's credentials in that respect, being more often on the other side of the argument. Equally, the tiny handful of editors who bother to follow the filibustering discussions here and on other MOS pages should remember that they are only a very small part of the community (as many of them do) and that their views often do not express wider community sentiment. Johnbod (talk) 11:35, 20 April 2014 (UTC)

Like Montanabw^(talk) 07:20, 22 April 2014 (UTC)

How about the following suggestion. The language is drawn from WP:LEADLENGTH and from MOS:INFOBOX. Establishing a suggested maximum length may help prevent infoboxes from becoming cluttered with too many marginally useful details.

As a general guideline—but not absolute rule—the infobox should usually be no longer than one page. A lead that is too long does not serve the purpose of allowing readers to identify key facts at a glance. Barryjjoyce (talk) 01:13, 19 April 2014 (UTC)

Nice sentiment, but "longer than one page" or "longer than a screenful" or whatever, is entirely dependent on the device on which the the material is being viewed. On my monitor, I'm hard-pressed to find a single infobox I can't get into one screen, even the longest of them. We do need some kind of limit, somehow, but sometimes long ones are very helpful, e.g at Galaxy Note 3. It's quite difficult to find any of those details in the prose of the article. I actually heavily depended on the infoboxes in WP articles on various phone models when deciding what phone to buy. (While Wikipedia is not a shopping guide, the point is that one person's "geeky details" is another person's "crucial summary data", and we cannot predict or dictate people's uses of and for WP content).  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:37, 19 April 2014 (UTC)

I agree entirely with the sentiment behind the proposal, being in the "less is more" camp with regard to their effect. Too often we end up with things like Hollywood Squares, which is no help to man nor beast, as any "use" to be had is drowned out by the sheer weight of other matter. And that pales slightly when we look at Winston Churchill: it may be the case that listing his various cabinet posts may or may not be useful, but also including the details of the PM under which served, and who preceeded and succeeded him is meaningless fluff (especially when we see exactly the same information in tabular form lower down the page at Winston Churchill#Political offices. Angela Merkel is an interesting one: English wiki has it's usual overly-long mess of factoids, which is interesting to see against the much more minimalist German language page. The main problem I see here is where to set the limit? A "length" or number-of-screens limit will always run up against the very good argument of differing screen sizes (not least on tablet and mobile devices). The "no longer than the lead" argument has merit, but does tend to count against stubs (there are lots of film stubs where the lead is a single line and even the very, very briefest details in the IB will run longer than the article, let alone the lead. I'll be happy to support anything that brings in a sensible limit to IBs as much as possible, but I think it may be a struggle to come up with a good rule of thumb that can be used. - SchroCat (talk) 08:42, 20 April 2014 (UTC)

• There should be a limit, which needs to be expressed in lines. I'd favour a low one of about 20 lines, which is absolutely not to say that all infoboxes needs to be anything like that long. If it is really felt locally that vastly more information is needed, then there can be two boxes, one for the important stuff in the lead, and one for lower down (in a long article) with more detailed stuff. This approach should be used more often. Johnbod (talk) 11:35, 20 April 2014 (UTC)

I like Johnbod's suggestion that an infobox recommended maximum length be measured in number of lines. This is a better method than the "one page" or "no longer than the lead" ideas, for the reasons already stated. I would recommend a maximum of 30 lines (which would be closer to the original sentiment I was trying to describe of one page on a standard-size desktop monitor). A problem with a 20 line limit, especially on a biographical article, is that the photo would take up most of the space, leaving very little room for anything else. Barryjjoyce (talk) 13:33, 20 April 2014 (UTC)

If such a limit would be used, it is probably better to have it as N lines + 1 photo and not allow lines in lieu of a image. CRwikiCA talk 15:14, 20 April 2014 (UTC)

Yes, I didn't mean to include the image, so N lines + 1 photo (though if you have as many as 20 or 30 lines it will often be best to have the image above the box, keeping that fairly narrow). Big images & small infoboxes are usually my ideal in a lead. Johnbod (talk) 15:35, 20 April 2014 (UTC)

If we go with something like the 20 lines + 1 image approach, would there need to be further guidance on what constitutes 1 image? For example, at the Pittsburgh page I mentioned previously, there are five photos in one box; would this count as one image or five images? There are also two symbols side by side; would these count as two images, or just one since the two images side by side don't add any more length than one symbol? Same questions for the two side by side maps. Maybe we can answer these questions in a way that won't generate a number of arguments. But the advantages of an approach of 30-35 lines max (instead of a 20 lines + 1 image max) is that the simpler approach may generate fewer arguments, plus it allows the members of various wiki projects the flexibility as how much space to allocate to images vs lines of text in their info box templates. Barryjjoyce (talk) 00:30, 21 April 2014 (UTC)

That's one image = one image file (Montage Pittsburgh.jpg). I could accept 1 image file + 1 map for geographical articles, and also 1 flag for countries and states/provinces. That's enough, imo, though as I say I'm open to secondary boxes lower down. If you are proposing some sort of image-into-lines conversion, that will get very complicated given all the display options, won't it? Johnbod (talk) 02:32, 21 April 2014 (UTC)

Clearly, there are projects, like geographic locations, where multiple images may be needed. Montanabw^(talk) 07:20, 22 April 2014 (UTC)

Just to butt in - some projects do use infoboxes for what they consider key data and can run to more than 40 lines excluding images - eg this featured article (part of this featured topic). I think before you get into a prescribed - rather than advisory - limits, you will need to get some projects on board to encourage them to shift (lede) infobox data to secondary infoboxes or other sections of the article - eg as the aviation wikiproject does with its infoboxes of about lines and the specification section for the numbers elsewhere in the article eg Supermarine Spitfire infobox about 30 lines long, specifications about another 30 near the bottom of the article. GraemeLeggett (talk) 09:15, 21 April 2014 (UTC)

Also true for certain political figures, sports figures, and topics with immense amounts of technical data. Montanabw^(talk) 07:20, 22 April 2014 (UTC)

Coin infoboxes contain a minimum of two images, sometimes several more if the designs were changed significantly during the coin's run. I view it as just one of those things. It's obviously important to illustrate the coin. No opinion on any matter suggested here except obviously I'd rather not have the MOS cut out from under 30 FAs or so, not all mine.--Wehwalt (talk) 14:18, 21 April 2014 (UTC)

Wrt the role of Wikiprojects: they seem to be the cause of infobox bloat. It is these projects that generally design the templates and get request after request for new common minor facts. Andy states that wikiproject members are part of the community too and should be allowed "their voice as individuals". I agree, but if one looks at the articles I linked and sees the discussions, then we are not seeing individual sentiment but group ideology. I have been asked, in defiance of policy, to overturn the wikiproject consensus before they will budge. I have requested members to say how the information is a "key fact" or "summary of the article" and receive no reply -- because there is no answer to that which allows them to get away with dumping facts into the box. I mean -- out-of-date Latin terms for crying out loud? The problem is we have often only one standard table in an article: the infobox and it is being used for all tabular information provided someone on a wikiproject can persuade others that it is fairly common. As long as some subject-nerd finds the data interesting, it gets imposed on everyone. Yet this is prime real-estate for the screen and a solid vertical box causes issues with images on the page, which are often right-aligned. I disagree with using an arbitrary cut-off for info box length. The guidance on this page is fine and rational rather than arbitrary. The problem is simply that it is being ignored by a group that do not represent the general reader. I dislike the attitude expressed here about MOS bullies. I'm no MOS regular. I didn't go to those articles to bully the editors there into obeying the MOS. I edited those articles to try to do the sensible thing and trim crap from the lead, and was bullied into accepting utter stupidities by a bunch of nerds, frankly. Our policies and guidelines already express community feeling on this matter. It is time the wikiprojects were reminded they have no higher voice on WP than individual editors. -- Colin°^Talk 10:32, 21 April 2014 (UTC)

Colin, you would have a lot more credibility with me (and others as well) if you'd cease insulting people with expertise as "some subject nerd" or "a bunch of nerds" and describing useful content as "crap." A know-nothing approach is not helping shed light on the concern for a very few infoboxes that get bloated. Montanabw^(talk) 07:20, 22 April 2014 (UTC)

Colin I can feel your pain and commend you in raising issues I had not thought about- but I don't feel that suggestions that could lead to valuble information being axed is productive. Each of the issues you raise must be addressed but addressing the cause rather than the result is probably more profitable. I have written an infobox- by c&p another one- then adding fields. I have gone back to reduce the redundant fields and found that they were already being profitably used by another Wikipedian.

I write on a venerable laptop but when in meetings, and needing info for a speech I am about to make- I pull out an WP app on my mobile phone and scan for the fact. All I will see is the infobox. The longer the better. It is not a summary of the article- it is the article. Take this one further- politician and journalists rely on our work and they are accessing them through the phone. Increasingly we are responsible for bad policy and thus human suffering.

Andy, probably will correct me, but if I am mining/scaping data for a web page, using, for instance python/beautiful soup- I would just use the infobox- which I see as the machine readable section of the article

I know that many folk are prose bunnies but for quick comparisons structured data in split windows is the most efficient way of doing the job- and there are many technical facts (surface area of different floors) that should be left out of the prose but included in the infobox.

Some infoboxes appear too long because they are too narrow causing word-wrap. I assume someone made a decision to standardise the width after the box was written and in use.

If the problem is with WikiProjects badly writing infoboxes then it is better to flag the issue warning that WikiProjects should guard against bloat and devise a policy for guarding against it- put that as item in the new WikiProject Welcome message. A new wikiproject is happy to receive all the help it can get but is irritated by extra hurdles. I see no harm in a bloat warning being placed on each WikiProject talk page- asking them to review their info-box- but we don't make work for ourselves by giving the deletionitas another bogus excuse.

I repeat Andy's words "The problem here would seem to be that the MoS is (or rather its authors are-) trying to dictate article content, rather than the more proper function of describing community norms."-- Clem Rutter (talk) 13:32, 21 April 2014 (UTC)

I agree very much with Colin's description of the issues and I'm very sympathetic to the view that smaller infoboxes generally perform their intended purpose better than larger ones. What is difficult is that there are so many justifiable exceptions to even that simplified philosophy. This is compounded by two common problems:

1. the inability of our community to sensibly deal with "rules-of-thumb" or "rough consensus" - we only seem to be able to manage "bright-line rules" with enumerated exceptions down to the smallest detail; and
2. the tendency of the limit to become the norm - if we suggest that 30 lines is a sensible upper limit, then I guarantee that smaller infoboxes will grow like Flopsy until they reach the new 'norm'.

I know where I feel we should be: at a point where it is easy to reach a sensible consensus on issues like how long an infobox in a particular article ought to be. I have no magic formula for moving us from where we are now to there, but I have no doubt that this issue - like many others - will remain intractable until we figure out how to do just that. --RexxS (talk) 17:43, 21 April 2014 (UTC)

The other issue is that the debate will never be resolved or consensus reached so long as people such as Colin choose to vent their frustrations with language that attacks the hard-working people who actually create content. I for one see no use in insulting people as "nerds" who insert "crap." That does not lead to a collaborative environment where consensus can be reached. Montanabw^(talk) 07:20, 22 April 2014 (UTC)

No, the debate has been resolved and consensus expressed on this guideline page. The problem is those who don't wish to listen. And the reasons for this are many but largely consist of flaws in thinking processes and personality -- which nobody wants to hear about. Montanabw, the Latin synonyms in the onion article are "crap" and anyone who thinks they are a key fact about a hugely important food crop is a "nerd". The entry-exit figures for Kings Cross are raw data from which one (or our sources) might formulate a fact (such as comparing its importance as an interchange with other stations). These are not key facts and they do not belong in the infobox. I don't see anyone here (or on the talk pages of those articles) actually raising a coherent argument in support of that. I just see a lot of people enjoy arguing for argument sake and are unwilling to back down in the face of evidence. Montanabw, if the best argument you have is that we should allow infobox bloat because those who bloat it are "hard working" then you are scraping. If your best argument is that my language is offensive to you then you are scraping. I see no arguments expressed here that disagree with the consensus opinion on this MOS page nor that explain how those expanded infoboxes on those pages (a) comply with it or (b) could not be reduced and whatever information removed from them be better expressed elsewhere in the article or even in another article. -- Colin°^Talk 07:15, 23 April 2014 (UTC)

@Johnbod: In response to my statement regarding "the advantages of an approach of 30-35 lines max (instead of a 20 lines + 1 image max)", you replied that "If you are proposing some sort of image-into-lines conversion, that will get very complicated given all the display options, won't it?" Could you explain further what you meant? My thought was that on a desktop view, where the infobox is on the right and the text is on the left, it should be easy enough to count the number of lines of text that an infobox takes up. For example, on a normal-size desktop monitor, the image at the Oak infobox takes up approximately 13 lines of text (10 lines in the lead plus 3 lines in the table of contents). If it's more complicated than that, please explain how. Thanks. Barryjjoyce (talk) 01:35, 23 April 2014 (UTC)

There is no such thing as a "normal-size" desktop monitor. What is concise for one type of article, or individual article, may be exaggerated for another. I don't believe that more regulation would help. --Gerda Arendt (talk) 05:54, 23 April 2014 (UTC)

The max-lines approach is stupid. Please, if people won't comply with rational guidance on what to put in an infobox, why should they comply with an arbitrary rules on how big the box should be. -- Colin°^Talk 07:15, 23 April 2014 (UTC)

Infobox length - should there be a limit?

Following the discussion above regarding infobox length, it might be useful to get a sense as to whether the consensus is in favor of trying to establish a maximum suggested length for infoboxes or against any such recommended limit on length. If the consensus is against a suggested maximum length, that may be the end of the discussion. If the consensus is in favor of a recommended maximum length, we can start then discuss how to set the suggested maximum length, but for now, let's focus on whether there should be a suggested maximum length or not. Barryjjoyce (talk) 23:46, 26 April 2014 (UTC)

This is a badly-worded Hobson's choice. Obviously there probably needs to be an outside limit could become a point where absurdity reigns (two pages would clearly be excessive, but then again, maybe not). So for those who can't really fall into either camp, I'm adding a third option. Montanabw^(talk) 20:15, 27 April 2014 (UTC)

Support

• Support — Many infoboxes have grown quite long. See for example Pittsburgh and HMS Hood (51). Too much info in the infobox makes it hard for the reader to identify the key facts at a glance — which is the main purpose of having an infobox. Barryjjoyce (talk) 23:46, 26 April 2014 (UTC)
• Support — Not exactly a bright line rule, but a general principle as I have proposed above (limit of 30 lines, plus image & map where needed), with more detailed information either in show/hide bars or secondary boxes lower down. Many infoboxes have become ridiculously long. Johnbod (talk) 01:51, 28 April 2014 (UTC)

Oppose

• Oppose — The concept of a "bright line" rule is problematic, and for that reason only, I vote Oppose. Some infoboxes, due to the nature of the content discussed, need to be long, others should clearly be quite short. This issue needs to be decided on a case by case basis. A bright line rule, say one allowing an infobox similar to that at the FA Richard Nixon would lead to complete bloat for a more minor figure, such as Lawnchair Larry. Montanabw^(talk) 20:15, 27 April 2014 (UTC)
• Oppose — Not all articles are created equal. Better write in the instructions not to fill every possibly parameter but present the most relevant ones to a reader who seeks information at a glance. Using the sense that is not so common is preferable to a fixed arbitrary to-be-observed limit. --Gerda Arendt (talk) 20:22, 27 April 2014 (UTC)
• Oppose — The quality of the data and clear presentation are the key to providing informative infoboxes. With standardized layouts for the different topics, we get an easy quick look. The two examples (Pitts & Hood) are actually quite good. – S. Rich (talk) 22:00, 27 April 2014 (UTC)
• Oppose - The number of infobox parameters that are relevant and useful will vary widely between topic areas, and the size of the entries for a particular parameter will vary widely between articles, so it's meaningless to complain about length/form without regard to the subject/function. I agree with S. Rich, and also find the articles the proposer gives as having "long" infoboxes to be very good, useful, and navigable. One of those articles is a FA, and so is supposed to represent our highest editing standards, and the other uses a template that is used on hundreds of thousands of articles, and so is presumably supported by an overwhelming consensus of editors. So it also seems clear that the proposer's opinion on this is not reflective of the community's.

  If the proposer has trouble finding certain information in infoboxes (though I do not in the examples given), then he can help suggest ways to improve their internal organization and presentation so the most significant facts can be better highlighted or grouped. It is senseless to complain about length in the abstract (akin to complaining about "too many notes") rather than making specific proposals to remove particular fields in particular infoboxes. postdlf (talk) 01:17, 28 April 2014 (UTC)

• Oppose — Discussion of infoboxes from the pov of a 20-20 sighted human is problematic. Infoboxes have other uses as they are a machine readable entry point to the article- and are often the only data that is scraped.-- Clem Rutter (talk) 08:31, 20 May 2014 (UTC)
• Oppose - Impractical instruction creep. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:29, 14 June 2014 (UTC)
• Oppose – As above. However, I do agree that some Infoboxes are too long, like the 6 years of traffic at London King's Cross railway station, entire careers for sports figures like David Beckham, and political careers along with succession and seconds like Angela Merkel. This sort of data belongs in table form, maybe even alongside running prose about it, just not in the main Infobox. —[AlanM1(talk)]— 08:39, 21 June 2014 (UTC)

Neutral

• Neutral Comment — I'm on the fence here—I prefer short, to-the-point infoboxes, but I don't think there should be arbitrary limits. What I do have a big problem with is minimum lengths—for example, WikiProject Novels requires an image of the first edition cover of a book in the infobox, and forces these images on articles even when no content editor of the article finds it appropriate. I wasn't even allowed to remove an image I had added myself. I'd rather see a wording that leaves things up to the judgement of the article editors after weighing each parameter for the value it provides to the specific article—which could result in either very short or very long infoboxes on a case-by-case basis. Curly Turkey ⚞¡gobble!⚟ 06:14, 14 June 2014 (UTC)
• Neutral Comment —

Additional discussion

@Montanabw: I was wondering if you could explain further — in light of your statement above: "Obviously there probably needs to be an outside limit" — why you placed yourself in the "oppose" camp? Barryjjoyce (talk) 02:20, 28 April 2014 (UTC)

Because I see no way that a bright line rule can work. As I explained above, it's a case by case basis - If we said "foo # of lines" "foo number of parameters," or "foo number of pixels", we'd no doubt see unneeded cruft added (height and/or weight might matter to a basketball player, a jockey or a supermodel, but is irrelevant for an artist) to some articles, and very relevant info omitted from others (for example, Serena Williams has a rather large infobox, but most of the data appears relevant). Montanabw^(talk) 06:31, 28 April 2014 (UTC)