Template talk:Reflist/Archive 1

multiple columns for long lists and different appearance for zero-length lists
I think we need the script to automatically choose a multiple column style if there are a lot of sources and automatically note the need for references if there are none (but concurrently cap the number of columns at 3 or 4). Something like...

$x = min ( int ( num_of_sources / 20 ) + 1, 4 ); if ( num_of_sources == 0 ) { References are needed for this article. } else {  thing, so I went ahead and added that. There isn't a way to count references now, and I don't know if there will ever be. Also, it doesn't just depend on the number of sources, it also depends on how wide they are (100 super-wide sources should still be in one column). --Interiot 23:48, 2 November 2006 (UTC)


 * Sounds great! Thanks! --Anthony5429 07:10, 3 November 2006 (UTC)


 * Ahh, it looks like there are ways to have the browser automatically determine the number of columns to render. Here is an example.  In that example at least, you have to give it an explicit column width.  I almost always browse at 1920 pixels wide, so I like this a lot.  A   parameter could be added to make this work for reflist, if there's enough support for it. --Interiot 08:00, 3 November 2006 (UTC)


 * Ohh - that is sweet. I resized the FF 2.0 window while looking at your reference list, and the column # dropped perfectly. And using CTRL++ and CTRL+- to change the text size, the column # also functions perfectly. CSS3 rules - I am all for this. Do we need to start having people vote on the change or just make the change and see if anyone disagrees? --Anthony5429 16:02, 3 November 2006 (UTC)


 * If we are going to go through w/ the change, we would want to do it ASAP so that we don't have to keep backwards support for . If we wait too long, we'll get a bunch of articles that use   and then we'll feel it necessary to keep support for the column # variable, thereby violating the KISS principle. If no one disagrees on this talk page in the next hour, I am going to go ahead switch the code to use Interiot's feature. --Anthony5429 16:40, 3 November 2006 (UTC)


 * I don't think there's any problem with keeping ...  we have many many pages that use fixed-two-columns even without reflist, it's the tried and true way, and I imagine editors will continue to want that.  We can make auto columns coexist with fixed columns by making the auto-column syntax be something like  .  --Interiot 17:09, 3 November 2006 (UTC)


 * Okay - I will go ahead and make the change, but keep the compatibility for . --Anthony5429 17:14, 3 November 2006 (UTC)


 * Alright - job's done and it seems to work fine! --Anthony5429 17:27, 3 November 2006 (UTC)


 * For what it's worth, my vote is to make the default just be single-column, not auto-column. Most uses of references "in the wild" seem to be 1 column (though maybe this is because it's the historical default and not everyone has firefox).  Also, auto-columns isn't well-tested and may be slightly buggy (is it just me, or if you click on a reference link, and before releasing the mouse button, all the numbers at the end of the list seem to change....)  (also, the long URL in #4 at Britney Spears renders a little badly at certain browser widths).  See User:Interiot/reflist and User talk:Interiot/reflist for the way I was thinking of.  Also, it sort of assumes that all references are going to be 30em wide, which may not always be appropriate, eg. Britney Spears, so at the very least it should be possible to specify a width. --Interiot 17:39, 3 November 2006 (UTC)


 * I see most of what you're saying. You can go ahead and make the change to what you're suggesting (I don't think I know the code for your suggestion). However, a couple notes: 1. I don't get the bug you have with the changing numbers - are you using google toolbar? 2. In the event that css changes to wrap long strings, I think we should consider making the auto-column-width the default. --Anthony5429 19:27, 3 November 2006 (UTC)


 * Okay, I made the change. I think each individual page should probably have stylistic control over whether multiple columns are used or not.  For instance, the auto-35em columns didn't look very good on a 1920px browser on crankcase, not because of column width, but because the 7 total references ended up creating a grid 5 wide and 2 high, which IMHO looks weird.


 * I might be open to defaulting to ~32em but allowing it to be overidden to something else, as long as it's at least tested more. Going from a feature that isn't used anywhere to being the default behavior is moving a bit too fast.


 * (about the buggy behavior... I'm not using Google toolbar.  As far as I know, I'm using an almost-stock v2.0 Firefox.  Though it tends to only happen with 4 or 5 columns, so maybe it needs a wider screen to be seen) --Interiot 20:02, 3 November 2006 (UTC)


 * All sounds good. I think it is fine the way it is now. --Anthony5429 23:03, 3 November 2006 (UTC)

Benefits
Hi, Anthony! Apart from the (currently unavailable) column-splitting feature, and, of course, saving a few keystrokes, what are other benefits of using this template? Any reason why it should not be substituted? Just curious. Thanks.&mdash;Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); 13:41, 25 October 2006 (UTC)


 * No other reasons - those are the two reasons I had in mind. I personally think it's a lot easier to type and a lot more eye-friendly to see:

 == References &#61;=


 * than:

 == References &#61;=

and  which I've added to common.css, this allows anyone using a resolution lower or higher than 1280x1024 to specify the number of columns they prefer.) —Ruud 01:51, 14 March 2007 (UTC)
 * I'd rather not forcibly prevent, though I certainly agree that it shouldn't be used unless the columns are extremely narrow.  But yeah,   sounds great whenever two columns are requested at least (I take it that users are encouraged to modify .references-2columns in their monospace.css if they usually use a large font with a low resolution?). It might be reasonable to prevent maybe 4 or 5 columns...  any chance we could get a  , so we don't needlessly restrict editors? --Interiot 02:09, 14 March 2007 (UTC)


 * I have seen a 3-column used where each is a very short thing. I don't remember where. --ChoChoPK (球球PK) (talk | contrib) 10:24, 21 March 2007 (UTC)


 * Lists such as Films considered the worst ever that have a correspondingly huge list of links that nobody cares to convert. –Pomte 15:54, 1 April 2007 (UTC)

font size
If you're watching this template, then you might be interested in this. --ChoChoPK (球球PK) (talk | contrib) 11:48, 21 March 2007 (UTC)

What?
Column division doesn't seem to work. I see only one on every ref section.-- Emperor Walter Humala · ( talk? ·  help! ) 18:27, 21 March 2007 (UTC)


 * It doesn't work in IE. It works in Firefox. --ChoChoPK (球球PK) (talk | contrib) 01:49, 22 March 2007 (UTC)

Suggestion: Collapsable reflist
Is there any way to make the reflist "collapsable" or "hidable"?

I ask in relation to biographical list-type articles, where every entry has a reference, with the result of several hundred references. Being able to hide these references would be very useful.

Cheers!  Lauren  ♫/∆ 03:44, 5 April 2007 (UTC)


 * That's be great. --ChoChoPK (球球PK) (talk | contrib) 04:38, 5 April 2007 (UTC)


 * There's a template called hidden that you can try using, although I have no idea why you'd want to. Tuxide 04:42, 5 April 2007 (UTC)


 * Using hidden, the reflist would be hidden by default if there are several other NavFrames on the page. Also, clicking on a citation link will not open the hidden reflist or direct to it. Is this really desirable? –Pomte 04:57, 5 April 2007 (UTC)

It would require more javascript code, I presume, to make sure that it opened up whenever a [1] was clicked on. I don't know off the top of my head whether it's straightforward to implement this. --Interiot 09:57, 5 April 2007 (UTC)


 * I don't have a link handy, but check out the debate leading up to the deletion of Template:Scrollref. Should be in the deletion log.  Similar issues would affect a collapsible reflist - How would it be printed? Would it work well with a screenreader?  Political issues around the default status of such a reflist would also be contentious.  Apparently there's a lot of people that think a references section, regardless of size, should almost never be obscured.  MrZaius  talk  16:27, 24 June 2007 (UTC)

Why not distinguish the notes?
What's the point of not having separate sections for notes and references? Having separate sections can only make things clearer. I don't see how the opposite would be true.

Peter Isotalo 13:20, 5 April 2007 (UTC)


 * Sure you can have separate sections. See Fight Club for an example. –Pomte 02:03, 6 April 2007 (UTC)


 * I don't use templates like these since I consider them confusing to editors in the long run, so I wasn't really referring to what can be done technicaly. What I was referring to is why the idea of separating notes and the list of sources isn't even hinted at. If anything, the tendency to use one single section for both notes and sources should be actively and strongly discouraged since it makes it much more difficult to see exactly how many and which sources have been used. It makes footnotes much harder to read (a situation already exacerbated by the use of smaller text) due to endless repetition of the same information instead of using a minimum of shorthand. And by this I mean simply referring to a work as Doe (1965) or Google: section X and Y, not using academic notation like ob cit, et al, ibid, and so forth. I mean, are there any good arguments for ever keeping notes in the "Reference"-section?
 * Peter Isotalo 13:44, 8 April 2007 (UTC)


 * Convenience, I guess. It's harder to maintain shorthand; you'd have to be wary that you don't insert a citation before ibid. A lot of sources happen to be cited only once, so it's more work to write it both shorthand and longhand, as well for readers to match them by eye. Depends on the article. In most cases it's fairly easy to count the number of total sources (same as the number of citations ignoring the a, b, c etc), so I don't see what you mean. When used properly, none of the citation methods repeatedly list the same content, at least none I'm aware of.
 * If you want to change the documentation of this template to encourage distinguishing notes, there should be no problem with doing that. "Option 3" sort of hints at notes above references. According to WP:HEAD, combining the two sections is okay. There's actually a current proposal to switch all &lt;references&gt; into reflist for a specific type of article, which many others see as an absurd/unnecessary idea. –Pomte 14:08, 8 April 2007 (UTC)

What do the multiple columns do?
I read the template usage directions and it mentions that we can use  or , however, nowhere in the page does it say what these do. What do the multiple columns do? What is the advantage of using one of these templates instead of the plain  template? Could someone answer please, thank you.--Leon Sword 02:46, 14 April 2007 (UTC)


 * I don't understand, it looks like it says so right on Template:Reflist. Sometimes people use multiple columns to save whitespace when many footnotes don't span the entire page.  Tuxide 02:53, 14 April 2007 (UTC)