This page is within the scope of WikiProject Lists, an attempt to structure and organize all list pages on Wikipedia. If you wish to help, please visit the project page, where you can join the project and/or contribute to the discussion.ListsWikipedia:WikiProject ListsTemplate:WikiProject ListsList
SIA or Dab?
I've been sorting through some articles and need a bit of guidance with regard to SIAs. Take these three for example: Buk-gu, Champions Park, Cheongoksan. All are currently classified as dab pages, but since they all list things of the same type I'd say they meet the SIA criteria. I'm just a bit reluctant to make any changes though as I'm not convinced I'd be doing the right thing. PC78 (talk) 00:43, 18 August 2019 (UTC)[reply]
I would say all three of these articles areSIAs -- they're all a list of articles about the same type of item (settlement, park, mountain) with the same name. —hike395 (talk) 01:06, 18 August 2019 (UTC)[reply]
Pages like those are better as dabs (there are thousands of such dabs e.g. 1st arrondissement). The main advantage is that any inlinks are more likely to be fixed (because of DPLbot, dabfixers etc). There's also then no need to convert it (back) to a dab if a book/film etc of that name is found. DexDor(talk)16:04, 18 August 2019 (UTC)[reply]
I think DexDor is bringing up a long-running issue with SIAs. The distinction is clear, but some editors are concerned about the side-effects of SIAs. I tried to resolve these issues above, but the discussion never took off. —hike395 (talk) 12:55, 20 August 2019 (UTC)[reply]
Yes, the difference can be nebulous, but importantly SIAs can have redlinks, extra information, references, etc, all of which are inappropriate on a disambiguation page. Shhhnotsoloud (talk) 19:33, 20 August 2019 (UTC)[reply]
I'd like to return to this question of DAB/SIA overlap. WP:SIANOTDAB makes the point that SIAs are not DABs because they differ in the following ways:
A SIA "is meant to provide information as well as navigation"
SIAs can have red links, or even list entries that are presumed non-notable (i.e. no link at all)
SIAs can have references
I would like to propose that if an article doesn't meet any of these criteria, it should be treated as a DAB. i.e. if the article would be a valid DAB page without any content changes, it's a DAB, not a SIA. By this standard, all three of the examples above would be DABs (whose entries all happen to be of one type). If someone also wants to make a set index article called List of parks named Champions Park, more power to them.
If we say that such articles should instead be classified as SIAs, the natural consequence of this would be that all disambiguation pages that happen to only list items of a common type would have to be recategorized as SIAs, which I think would be a disaster. See for example Category:Buildings and structures disambiguation pages. The vast majority of these thousands of pages meet the "only one type" criterion just by virtue of having a name like "X Tower", or "X Airport". Colin M (talk) 20:44, 5 November 2019 (UTC)[reply]
Even if a dab page happens to (currently) have items that are all of one type it's still a dab page (any inlinks from article text should be fixed). By contrast, a SIA (e.g. List of buildings named Flatiron Building) is limited to items of one specified type (perhaps the guidance should be clearer on this). A dab page doesn't/shouldn't specify a type so it wouldn't always be clearcut whether the entries are all "one type" or not - e.g. if a dab page has 2 entries, one an insect and one a mammal, are they "one type" because they are both animals? DexDor(talk)07:59, 6 November 2019 (UTC)[reply]
They feature disagreement on the core question of "Is a DAB page where all the linked topics are of a common type actually a SIA?" (and feature many of the same editors as in this thread - small world!). Hike395 says yes: If something looks like a disambiguation page, but contains only plants, it is indeed a set index article, DexDor says no, and others are just confused. I personally find DexDor's argument about avoiding mistargeted links very convincing. Any editor who links to Western State Hospital is intending to refer to a specific hospital, not the set of hospitals that share that name - getting the benefit of bot warnings when linking to that name seems really important.
One way or the other, I think it's important that WP:SIANOTDAB is clear on the answer to this question. I boldly added a line about this (diff). I won't be surprised or upset if it gets reverted, I just thought I'd give it a shot before resorting to starting an RfC or whatever. Colin M (talk) 22:21, 12 November 2019 (UTC)[reply]
There's further to go still, in clarifying this SIA ~ DAB issue. At present the guideline page illustrates this fairly clearly!
Its section on Set indexes and disambiguation ends with a paragraph using the Western State Hospital DAB as example to point out tht its entries all happen to be instances of a single type of building (hospitals!) but that is not sufficient reason for the DAB to be reclassified as an SIA. That's obviously correct; but the same para then concludes the section with a summary table, How to tell whether the page could be a set index article - which clashes. According to that summary table
How to tell whether the page could be a set index article
the Western State Hospital DAB should be an SIA: the entries all relate to Similar subjects (hospitals!)
So if same single type is not sufficient reason for converting a DAB to an SIA, what is? And is the concept of a single type clear enough anyway to be helpful (= facilitate consensus)?
My thinking, in a nutshell, is tht the way to deal with such puzzles will center on the general reader, and their needs, rather than on articles and potentially fuzzy classifications. Specifically . .
Two points:
A The para needs its text revisiting, of course; but looking only at that final table, as focus for this discussion, I'd suggest it needs a column adding, with a note to that column, and perhaps some other titling tweaks:
How to tell whether the page should be a set index article
[*]Complex relationship: Will information on this relationship help the general reader understand the page and/or its linked articles?
( The new column would look a lot better with its entries centered - rather than range-left - but I've forgotten how. )
Second point, B: there might be some interesting examples for that final case:
N No
Y Yes
-
list / concept-article
. . ?
Reading this Talk-page section as a whole, I'm reminded of a bunch of articles I encountered some while ago when reading about the Christchurch earthquake. There were several related earthquakes around that time - and differing views were reported on which should be considered as aftershocks (or warnings) rather than as fully independent events. Criteria seemed to include location (including depth) - and specifically which fault a quake was located on. As I remember, that in turn was complicated by questions about intersecting faults, and whether a branch should be included with its main fault. Details of obvious importance to the general reader - perceived severity, damage and casualty totals, length of aftershock series etc - were aggregated, in the articles, by including aftershocks with main-quake statistics. So the general reader needed substantial hand-holding to understand those statistics. At the time I wasn't aware of the SIA concept, and so wasn't able to do much to improve the encyclopaedia.
It's this earthquake example tht makes me think tht SIAs might sometimes be applicable even where names are dissimilar. Usually, similarity of names provides an obvious common theme (eg Paiutes). But not always. Earthquakes tend to be named by features of location - with quakes, named for different cities, variously regarded as aftershocks or not and with wholly dissimilar names. (And each quake may have a most common name reflecting contemporary media reportage while academic sources use an unrelated fault name.)
An SIA would have been the appropriate vehicle to clarify that whole ball of wax for the general reader (and in a single place, rather than as overlapping part-duplicates sprinkled among all the articles).
The last row of that table would need supplementing to cover such cases.
Given the ambiguity in the guideline pointed out by Colin M and the IP editor, above, I would propose changing
A disambiguation page should not be reclassified as a SIA (e.g., on the basis that its entries all happen to be instances of a single type)
to
A disambiguation page should not be reclassified as a SIA on the basis that its entries all happen to be instances of a single type, unless the page has metadata or extra information about its entries.
This makes it clear that properly formatted DAB pages that could be SIAs should stay as DABs, unless an editor adds extra information. Thoughts? Comments? — hike395 (talk) 12:01, 22 December 2024 (UTC)[reply]
