Wikipedia

Wikipedia:Wikipedia Signpost/2013-07-10/Dispatches

< Wikipedia:Wikipedia Signpost | 2013-07-10
The Signpost
← Back to Contents
View Latest Issue


Dispatches

Infoboxes: time for a fresh look?

Contribute  —  
Share this
By Brian Boulton
Brianboulton is a British Wikipedian and has been editing since 2007. Aside from his prolific reviewing at peer review and featured article candidacies, he has contributed to 73 featured articles on the English Wikipedia.
This week sees the return of the dispatches section, which was last published in 2010. Volunteers with a proposal for a future dispatch can note it on the Signpost's talk page or email the editor.

What are infoboxes for? As a regular reviewer at PR and FAC I look at a lot of articles, which means that I see plenty of infoboxes. They have been a feature of WP articles for years now, and it seems obvious that they can provide a useful service to readers who want a few specific facts about a subject, rather than an in-depth study. What is the population of Salzburg? Who was Henry II of England married to? How many first-class wickets did Jack Hobbs take? The infoboxes are there to give these answers.

The initial MOS guideline on infoboxes was posted on 10 March 2006; by 1 January 2007 a number of WP projects were incorporating them into articles, and on that date the WP:Infobox project was started, designed to foster encyclopedia-wide cooperation. The project page summarises the nature of infoboxes as: "a quick and convenient summary of the key facts about a subject, in a consistent format and layout". The particular words "quick", "convenient", and "key facts" all imply a degree of selectivity in information. The MOS guideline gives broadly the same message, while adding a significant rider: "The less information [the infobox] contains, the more effectively it serves that purpose".

The fundamental idea is clear: keep the box simple, and limit it to essentials. At some point down the line, however, these basic principles seem to have been abandoned, in favour of an approach akin to "the more the merrier". As I go through my reviewing duties I can't help noticing just how big some infoboxes have become, how much room they take up and how much detail is crammed into them. Some are simply enormous; far from being convenient, quick-reference points conveniently placed in the top right-hand corner of their parent article, they have become huge columns reaching deep into the body of the article. Apart from anything else, this can foul up presentation by squeezing the text and mispositioning images. In some cases, the article itself appears to be almost subordinate to the box.

Let's look at a few specifics. Articles on countries are, I imagine, frequently consulted by casual readers in search of basic facts: location, population, capital city, language, currency, form of government. Unfortunately, most "country" articles have infoboxes which go way beyond the provision of these essentials. Denmark is one example, typical of many. Its infobox includes mottoes, anthem titles and translations, two sets of GDP calculations, Gini ranking, HDI ranking, ISO code and much else besides. Some of this information will require the use of links, often to articles that aren't at all easy to follow – try this one. While all these extra facts obviously are relevant, few of them could be said to represent the "key" information on the subject. Another problem area that I encounter in my reviewing travels is political biographies, where infoboxes are often of inordinate length and complexity. Way back in 2008 I cut my reviewing teeth on the Shimon Peres article, and I commented then that the infobox was confusing and overcomplicated with detail. It still is, and the same criticism can be made of most articles for statesmen who enjoyed long and active careers. Winston Churchill is another example. A particular issue is the dubious practice of recording in the infobox not only every political office ever held by the subject, but also the names of each predecessor and successor in these offices. Much of this information is entirely inconsequential; is it a key fact that Lord Weir preceded Winston as Secretary for Air in 1919?

Infoboxes should not be the repositories of any odd bits of information related to the subject. Indeed, sometimes information that is not just inessential but downright absurd finds its way into them. I had reason recently to look at the London Coliseum article, where the infobox is relatively short. Among the "essential" information it provides are the theatre's geographic co-ordinates! In what sense is this "key information" about the theatre? It is about as pointless as it gets. I thought this might be a once-off aberration, a case of editorial over-enthusiasm, so I checked the articles for other well-known buildings—the Eiffel Tower, and St Paul's Cathedral—and found that these landmarks, too, have their co-ordinates proudly displayed in their infoboxes, as does the Empire State Building, whose infobox records the co-ordinates twice, for no clear reason. I am sure that a general audit of infoboxes would throw up many similar instances of redundant or insignificant information.

The issue of concern is the extent to which infoboxes are becoming generally less efficient in fulfilling the function for which they were initially introduced. I believe it is time to reconsider the tendency towards overdetailing that has developed in recent years, and to look for a new approach. In the short term we could reestablish first principles by the adoption of a mechanistic formula whereby the number of parameters in any one infobox are limited to, say, eight (or even six). This is not a revolutionary idea; it is similar in purpose to the existing MOS guideline that restricts the number of paragraphs in an article's lead, with the aim of keeping the lead in brief summary form. If infoboxes were likewise limited, editors' minds would focus on what really needed to be included, rather than on how to extend the box. In the longer term, however, an altogether more fundamental change might be considered, along the lines of an idea that has already been floated by Dr. Blofeld: the development of a Micropedia version of the encyclopedia, that would obviate the need for infoboxes altogether. Now, that would indeed be revolutionary.

 

+ Add a comment

Discuss this story

These comments are automatically transcluded from this article's talk page. To follow comments, add the page to your watchlist. If your comment has not appeared here, you can try purging the cache.
Apart from that minor point, a very thought-provoking piece. Thanks to both Brian and you. Best, Voceditenore (talk) 07:44, 12 July 2013 (UTC)[reply]
  • I think people just need to start using collapsible infoboxes where only the most important information is shown uncollapsed. e.g. Reelin. --Tobias1984 (talk) 07:59, 12 July 2013 (UTC)[reply]
    • I like collapsible infoboxes as well. Using one on Pennsylvania-class battleship allowed me to include an extra image and hide statistical information that many readers won't care about. Ed [talk] [majestic titan] 08:51, 12 July 2013 (UTC)[reply]
    • Collapsing infoboxes, or parts therein, is a bad idea. It defeats the purpose of providing our readers with a quick and convenient overview; and it makes it likely that editors who are updating facts in the body will see that they also need to to so in the infobox. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:53, 12 July 2013 (UTC)[reply]
      • One of the key points of the article is that many infoboxes are neither quick nor convenient due to the unreasonable level of detail they contain for something giving a quick snapshot of a topic. Collapsing detail into relevant sections that readers can rapidly scan and expand as desired serves that purpose rather well. I don't think it's 'a bad idea' at all. Ale_Jrbtalk 14:15, 12 July 2013 (UTC)[reply]
        • I've been looking over the shoulders of friends and colleagues as they interact with Wikipedia for many years, and every single one has been surprised by the [Show] button in navboxes, and the ones that are making their way into Infoboxes (and data tables, and galleries), when I suggest that they click on the [Show].
          Editors are too often using "collapsible section" as a way of burying a dispute over whether or not to include something in an article. It's a bad habit for us, and a disservice to (at least a percentage of) the readers. –Quiddity (talk) 23:18, 12 July 2013 (UTC)[reply]
  • Funny stuff! I didn't twig to the satire until the bit about geographic coordinates being out of place in an infobox for a geographic location. And the Empire State Building "example" which is simply the result of {{infobox NRHP}} being nested inside {{infobox building}}, not actually a clever scheme to randomly duplicate listed data? Brilliant. I look forward to more comedy at this level in the future. - Dravecky (talk) 09:21, 12 July 2013 (UTC)[reply]
  • Let me add my voice in agreement with Brian here. Although I am inclined to agree with Dravecky's point as well: co-ordinates seem like a sensible thing to be in an infobox, especially since the link allows a map of the location to be viewed. — This, that and the other (talk) 09:59, 12 July 2013 (UTC)[reply]
    • I find coordinates useful in infoboxes. It's a pity that Brian seems only to have considered his own personal preferences, and not the circumstances of others. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:53, 12 July 2013 (UTC)[reply]
      • You miss the point. I am not saying that coordinates should not be in infoboxes per se. They would for example be vital information for an article about a mountain in Antarctica. I am saying that they are pointless information for buildings that have fixed locations in cities; what is the point of knowing what the geographic coordinates of the Coliseum Theatre are? And how can this be justifid as "key information" on the subject such as to justify its appearance in the infobox? Brianboulton (talk) 13:50, 12 July 2013 (UTC)[reply]
        • On the contrary; I addressed your point directly: you don't see the need for coordinates in an infobox about a building in a city, so think those of us who do should be deprived of their usefulness there. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:55, 12 July 2013 (UTC)[reply]
          • The question I am addressing here is whether, for buildings with known addresses in cities, the geographic coordinates represent "key information" about the building, such as justifies inclusion in an infobox. It's not a matter of whether you or a few other souls might find it convenient to have it there, it's whether its importance merits inclusion. However, I realise that you will never understand or accept this. Brianboulton (talk) 14:35, 12 July 2013 (UTC)[reply]
  • Thanks for this article Brian - I've been feeling the same thing. A problem with large/complex infoboxes is that they also can turn into resource-sinks, with editors (and especially newish editors) tending to focus on the infobox rather than the body of the article. Nick-D (talk) 10:09, 12 July 2013 (UTC)[reply]
    • Editors are free to choose the aspects of Wikipedia to which they devote their efforts. Why should they not? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:53, 12 July 2013 (UTC)[reply]
      • Fair point: My concern is that endless disputes over some infoboxes can distracts attention away from things which actually need fixing. I'm thinking of the infoboxes for the World War II and Pacific War articles in which there has been endless bickering over which countries to include and which order they should appear, and very little work to improve or maintain the quality of the actual articles. Nick-D (talk) 23:22, 12 July 2013 (UTC)[reply]
        • Regarding "newish editors" and the way that infoboxes sometimes distract from article content, one of my particular pet peeves regards editors (mostly newish) who systematically excise significant information from the article text because it "repeats" information that is in the infobox. 'Nuf said. --Orlady (talk) 00:09, 13 July 2013 (UTC)[reply]
  • Great article Brian. I agree wholeheartedly. Also agree with Nick's comment above. —Cliftonian (talk) 10:41, 12 July 2013 (UTC)[reply]
  • Ack, that Winston Churchill one is horrible. That said, as we can see from the discussion above, any attempt to chop them back down to the basics will be difficult because everyone will have two cents to contribute on their own favourite bits of information. Lankiveil (speak to me) 11:44, 12 July 2013 (UTC).[reply]
  • Amen! Infoboxes should be the basic facts about a subject, and not a dumping ground for any bit of information that can be communicated in four words or less. I strongly support any effort to cut them down. Infobox devotees should start a parallel website for non-encyclopedic, non-cited data about a subject. The predecessor/successor information is particularly silly and I would be happy to eliminate it completely. —Designate (talk) 13:10, 12 July 2013 (UTC)[reply]
    • Fully agree. But when I tried to do just that was quickly reverted. Unfortunately resistance to any kind of slimming is much greater than tolerance of fattening, so infoboxes (as well as articles, navbars, category lists) keep growing and growing and growing while readers read less and less and less. --ELEKHHT 01:50, 13 July 2013 (UTC)[reply]
  • Now that we have Wikidata (I’m surprised it is not mentionned in the piece, actually), perhaps we do not need to cram infoboxes with all possible data − just like we do not need to stuff articles with image galleries, as we can simply link to a category or gallery on Wikimedia Commons. No? Jean-Fred (talk) 13:43, 12 July 2013 (UTC)[reply]
  • I agree with above, Wikidata has changed matters. It should have been mentioned, but alas you may not have really known why it is important to Wikipedians. Wikidata is important for many of the reasons you gave. This has become a standard usage of infoboxes, to cram as much data as possible. Pushing the data into Wikidata will go a long way towards solving this problem, as it will provide a structured way to store, and recall, this rather important, yet in some ways superfluous, data. Maybe we can start creating "VerboseInfobox" versions for the bottom of the page to replace those rather numerous succession boxes currently in use? (Hopefully that automatically pull the relevant information from Wikidata.) Int21h (talk) 14:57, 12 July 2013 (UTC)[reply]
  • While I would agree that coordinates should not be in an infobox if they also appear at the top-right of the article page I would strongly dispute that the information is not important. The location of, say, a building is its prime identifier – it my be rebuilt, renamed, re-purposed, or even demolished: its location is its main and unchanging definition. I recall that I have myself in the distant past used the Wikipedia coordinates of the Empire State Building to find out where it was.   Oosoom Talk  15:40, 12 July 2013 (UTC)[reply]
  • Yes agree totally that infobox bloat has become a cancer. As much editing, a very valuable skill is deciding what to leave out. As discussed above, one set of coordinates is probably fine for many articles. One good start would be to scale infoboxes to article size. This could be quite a subjective judgment, but there are other similar ones editors deal with. A couple more related complaints: biographies that put birth (and sometimes death) dates and other details like cities into the lead, the body, and the infobox. Only in perhaps a few very long and complex articles can that make any sense. Worse yet are company articles I have been dealing with recently that have a litany of all the "chief this or that officers", often without any sources, and probably out of date fairly quickly. Talk about vanity, when little startups of "three kids and an app" can give themselves all grandiose titles. Thanks for this much-needed discussion. W Nowicki (talk) 16:41, 12 July 2013 (UTC)[reply]
  • I disagree with this op-ed. I like infoboxes and we should have them on every article. They bring out key facts for those that don't want to read the article itself. Your example of Denmark was a poor choice since the infobox doesn't protrude past the first level 2 heading on the article. I can see where you're going with the Winston Churchill example, but someone of his career and stature would naturally have more information in the infobox to display. I think the op-ed should have made the point that infoboxes should not evolve to replace/supplement navboxes, which is where this is heading. Ultimately, control of what fields belong in the infoboxes resides with the wikiprojects. Weak wikiprojects result in editors being too bold and letting things get out of control. Chris Troutman (talk) 17:06, 12 July 2013 (UTC)[reply]
  • I agree with what Chris says above, and I'll freely admit that I use infoboxes a lot when I'm browsing Wikipedia for information. You may think that GDP and GDP per capita is excessive information but I do look that stuff up and going to the country article and scanning the infobox is by far the quickest way to do it. I was not really responsible for the boxes on my own FAs Paul Kagame and Rwanda, those are the purview of respective wimiprojects. Butt would I change them? No, I don't think so. Apart from the aforementioned Gini index I personally think all of it is potentially useful. Thanks  — Amakuru (talk) 21:26, 12 July 2013 (UTC)[reply]
    • Many people here are writing that they sometimes look up some particular bit of information, but that's not a good enough reason to put it in that spot. We can't possibly include every bit of information in a database format on top of an article just because someone, somewhere might have use for it. That's not what an encyclopedia's for. —Designate (talk) 22:21, 12 July 2013 (UTC)[reply]
      Actually, I'm pretty sure most encyclopedias do have something akin to an infobox on most of their more important topics. I seem to recall the printed Britannica having something of that nature anyway. And of course, I'm not arguing for the inclusion of every single little obscure fact, but I think we'd be doing the world a disservice if we insisted on removing things that many people might want to look up. International dialling codes, GDP figures, presidents, independence dates - all these things are available elsewhere, but not necessarily in the same place, and not in a place you can predict. If I want the timezone of Tuvalu, I *know* I can find it instantly without even going to Google just by whacking http://en.wikipedia.org/wiki/Tuvalu into my 'puter. Some people seem to have suggested having collapsible sections for the more minor facts - I think that might just solve everyone's issues...  — Amakuru (talk) 19:16, 13 July 2013 (UTC)[reply]
      See my comment above, and many conversations elsewhere (many of which link to MOS:ACCESS). Collapsible sections are very problematic, and using them to brush disputes under the carpet is a bad habit that we need to examine, and should research how (or whether) readers use them, and possibly discourage more strongly. –Quiddity (talk) 20:33, 13 July 2013 (UTC)[reply]
    • Google's gotten pretty reliable about this type of data. You can type "timezone tuvalu" or "gdp tuvalu" into your search bar and Google will give it to you instantly without having to navigate to another page. Wolfram Alpha also provides an outstanding amount of data just by typing a country's name in (seriously, check it out), and it's in a convenient format that allows you to perform calculations, compare a list of nations, make graphs, etc. I just don't see how a static page like Wikipedia can ever compare with the sites that do this much better. Maybe we should eliminate the infobox and just link to Wolfram so we can focus on the prose, which is our forte. —Designate (talk) 20:56, 13 July 2013 (UTC)[reply]
  • I suppose it would require an RFC on the matter, and it may be a necessary thing. Perhaps the infobox should be restricted from falling below the table of contents? I believe most editors don't intentionally intend for an infobox to subsume the article but instead believe that if the template has the parameter and the sourced information is available that it should be included. That has always been my assumption. :) John Cline (talk) 06:06, 13 July 2013 (UTC)[reply]
  • You make an excellent point, as does Amakuru, above. I began this article by reminding editors that infoboxes were originally intended to be short, sweet and to the point - key headline information. Unfortunately the designers have tended to add more and more parameters; editors cannot be blamed for thinking that these have to be be filled. That is why I have suggested fixing a limit to the number of parameters per infobox, which would force attention on what is really key information, and also help to restore some uniformity to the appearance of articles. Brianboulton (talk) 08:35, 13 July 2013 (UTC)[reply]
  • But "falling below the table of contents" on whose monitor? And at what resolution? If the infobox contains an image (such as a photo of the subject, a logo, or other relevant pic) should it contain less information than a similar infobox for a company with a smaller logo or a person for whom a photo is not available? Should lead sections be bloated to accommodate an arbitrary design aesthetic? Also, a global limit on the number of parameters will lead only to more intricate template designs, not smaller infoboxes. (For example, some take the latitude and longitude data as up to 8 parameters while others need editors to use the intricate {{coord}} template to place it on 1 parameter. The display is the same but the functionality and ease of editing is vastly different.) Knowing that, what benefit to the reader or the encyclopedia is gained by "fixing a limit to the number of parameters" in any one infobox? - Dravecky (talk) 09:08, 13 July 2013 (UTC)[reply]
    Your questions are valid considerations. To clarify my intent; I was thinking of tying an overflow with a vertical scroll bar to the TOC bottom. I'm not suggesting a limit on parameters, or that an editor should choose x from y available. The reason I suggested the bottom of the TOC is because as the article content grows, so should the lead summary grow proportionately, and the TOC generally grows as well. It seems like this would leave a large enough box exposed to be useful yet define an acceptable limit. And I personally dislike articles with long TOCs and a bunch of blank space to their right. I am perplexed that many editors prefer no infobox at all, to that alternative, but that's just me. Cheers. :) John Cline (talk) 10:15, 13 July 2013 (UTC)[reply]
  • I completely agree with this article. Infoboxes have become total cruft piles. It's gotten so bad that I don't even add infoboxes to biography articles any more. In most cases, they are better off without them. Making the content collapsable will just make the problem worse, not better. Kaldari (talk) 04:40, 14 July 2013 (UTC)[reply]
  • Excellent op-ed, I agree entirely. One thing you don't mention is (mostly in historical articles) the tendency of infoboxes to introduce inaccuracy - people feel compelled to complete all the fields even where the information is uncertain or just not easily summarized in one or two words. The House of Commons Brown/Cameron spat over the birth-date of Titian which some will remember (aides editing Wikipedia to agree with what their party leader had said) all arose because the infobox mistated what is in in fact just unknown. Johnbod (talk) 11:30, 14 July 2013 (UTC)[reply]
  • Excellent op-ed, yes. But I disagree on one point. Buildings are rather static on the earth—London Bridge being an obvious exception—and where they are located is often interesting, and pin-pointable by coordinates linked to maps. So, in general, they should be pin-pointed. But please don’t hide that link to maps away in an info box: just stick it up in the default place, the top right-hand corner. Then everyone will know where to look. Ian Spackman (talk) 12:15, 14 July 2013 (UTC)[reply]
  • Food for thought: the German Wikipedia community repeatedly opposed using infoboxes in biography articles. Exceptions were only made for sportspeople, astronauts and a few other cases, where there is numerical data such as no. of wins/goals/hits/olympic medals/space trips, etc.. that is better or easier displayed in tables/infoboxes than written out in text. Overall opinion seems to be that most persons' life is too complex to condense it into a few numbers and facts without much context. --Kam Solusar (talk) 21:57, 14 July 2013 (UTC)[reply]
That's what I do by adding an infobox to most of my articles, and everybody may look if Carmen would not be more attractive and informative with the short infobox suggested (instead of a navbox that repeats content from the footer navbox), --Gerda Arendt (talk) 16:30, 15 July 2013 (UTC)[reply]
I don't think that's the right place for the author's navbox, but neither do I think the proposed infobox can be justified. Every single word of that infobox is written succinctly, point by point, in the first paragraph. It's completely redundant—and it's not clear why someone would need a tabular format for these five random facts rather than reading the lead and getting the whole story. How many living, breathing human beings will ever be in a position where they need to know the date of Carmen's first performance but care so little about the subject that they can't read a single paragraph?

I get the impression that people want infoboxes because the article looks right with an infobox and wrong without one, rather than a real sense of what's useful. I agree with the comments on the talk page—"These infoboxes will continue to be badly implemented until the box protagonists start asking themselves what the boxes are actually supposed to accomplish." —Designate (talk) 23:29, 15 July 2013 (UTC)[reply]

Because when we first started Wikipedia we cared greatly about reaching people with different approaches to learning. We had a number of discussion on learning styles and how it related to how information was presented, how colors could be used and what information should be wikilinked. All of that seems to be forgotten nowadays. Yes I can sit down a read a book on cod but other will never get farther then an infobox. Yes I can read a four-color map with red and green but some color-blind person won't be able to. Rmhermen (talk) 13:16, 17 July 2013 (UTC)[reply]
Some people will want to read the complete article, others may just want to find a date/language/author of the story. I am willing to serve both. The author's navbox is at the bottom, so the one on top is redundant, and I don't believe an infobox instead would "damage" the article. Project opera just made a consise box available, {{infobox opera}}, and I would like to see it used and tried, comments are welcome. --Gerda Arendt (talk) 13:29, 17 July 2013 (UTC)[reply]
I began the article by recognising the usefulness of infoboxes in making available certain kinds of data in a convenient form. My argument is not against infoboxes in general, but against the bloated boxes that have developed over the years, contrary to the original intention. As to your opera boxes, their time may come—provided the discussion is led by editors with a knowledge of and love for music and opera and the issue is not forced. Brianboulton (talk) 14:00, 18 July 2013 (UTC)[reply]
I thank your for both recognising "the usefulness of infoboxes" and envisioning a time when operas will have them. The time that their use was introduced in the project's manual of style has come (18 June 2013, as noted on the Carmen talk). - In an opera, typically you get a time and location of the action. A simple infobox could do just that: position a subject in history and geography at a glance (example pictured), --Gerda Arendt (talk) 23:44, 18 July 2013 (UTC)[reply]
I'd like to clarify a couple of things that Gerda has said about WikiProject Opera. We have a page which is a guide to writing articles on opera-related subjects. It is not part of the official Wikipedia Manual of Style. The infobox has been listed there simply as an option, not as a recommendation, and indeed members have objected to its use on several articles and removed it, leading to discussions on the talk pages of the articles concerned, where they belong. Voceditenore (talk) 10:15, 19 July 2013 (UTC)[reply]
Again, everything in the infobox is in the first paragraph. "reading the complete article" isn't required. —Designate (talk) 23:27, 18 July 2013 (UTC)[reply]
The Signpost needs your help putting together the next issue.
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Wikipedia_Signpost/2013-07-10/Dispatches&oldid=1193871451"