Template talk:Unreliable source?

Inline after citation number or at end of note?

Should this template be placed after the citation number (outside of the </ref> tag) or at the end of the reference note (within, ie before, the </ref> tag)? Could we have instruction on the page template page? Hyacinth (talk) 10:20, 30 June 2009 (UTC)

I, too, came looking for that information in the template's documentation. I know these things take time, but if 5 years is not sufficient, do we keep waiting? Cnilep (talk) 03:59, 13 July 2014 (UTC)

I think it usually placed outside the reference note. Debresser (talk) 07:41, 13 July 2014 (UTC)

As per User:Cnilep's edit on 16 July 2014, the answer became "after citation number". This is wise choice because readers need to be alert to bad references. If the tag is placed within the ref tags, there's a big chance that they will falsely assume a citation was valid because few look at the references compared to the article text. Jason Quinn (talk) 20:45, 24 January 2015 (UTC)

In cases where there are multiple supporting references, some patently reliable and others potentially unreliable, suggest using inside the ref tags. This is a better solution than removing 'unreliable' references, in my opinion. --User:Ceyockey (talk to me) 02:44, 26 January 2017 (UTC)

The intended use is to put it immediately after the unreliable(?) reference: Foo bar baz^[3][4][5] (If you think that looks ugly, move the questionably reliable one and this tag to the end.) If you put it inside the ref, it will be unlikely to get resolved within years, because hardly anyone will ever notice it.  — SMcCandlish ☏ ¢ 😼  13:43, 5 January 2024 (UTC)

certain only works for "y" and "yes"; broken otherwise

As can be verified in the testcases page, when |certain= is set to "n" or "no", it removes the question mark rather than preserving it. This is, of course, not what is wanted. The same removal is done if |certain= is set to any other string. Although, a convention is needed, I suppose in that case, the question mark should also be preserved. This bug is something that should be fixed at some point. Any takers? Jason Quinn (talk) 21:06, 24 January 2015 (UTC)

I was successful in making the certain parameter work with "y", to the exclusion of all other values. The previous version just checked if the parameter was defined at all, and any value triggered it. Haven't edited templates for a while, and frankly forgot how to, but it shouldn't hard to add "failed" as a parameter and "yes" as a value. Debresser (talk) 08:18, 25 January 2015 (UTC)

|failed= now works as a parameter. If we want more values to be supported like "TRUE" or whatever, this can be implemented with {{yesno}}, but most templates do not bother with this and the added complexity is generally not worth it. If you actually want the template to parse negative values, that adds a great deal more complexity, and we virtually never do such two-way parsing of parameters for positive and negative values. Probably over 99.9% of template parameters that take any such value actually operate on any value at all (i.e., if the documentation says you can use |munged=y, supplying |munged=n or |munged=barbecued chicken lips! will produce the same result. We only implement the "opposite value gives opposite result" option when there's a good reason to do so, e.g. the |deadurl= parameter of the citation templates produces markedly different output, depending on what the value is and whether an archive URL has been provided. Similarly, some templates have operator-overloaded parameters that can produce stock output if fed a recognizable positive or negative value, or custom output if fed something that is neither. No needs like that seem to be in play here.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:45, 6 June 2016 (UTC)

Update: Handling of "no" values is now a bit more common in templates, but I'm skeptical the utility is worth the effort; if |certain=y doesn't apply, just leave the parameter out. Someone else can implement the no functionality if they want, of course, but I have bigger fish to fry.  — SMcCandlish ☏ ¢ 😼  13:40, 5 January 2024 (UTC)

Doc cleanup

I've normalized the documentation between {{Unreliable source?}}, {{Unreliable medical source}}, and {{Unreliable fringe source}}, and fixed parameter inconsistencies in the template code during the process.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  20:37, 6 June 2016 (UTC)

And just patched up a bit (very belatedly) after that; I had mis-pasted some text that implied this was for tagging a statement as being possibly unreliably sourced (that's {{dubious}} or some other related tags), when it is for tagging a specific citation as being to a potentially unreliable source.  — SMcCandlish ☏ ¢ 😼  13:38, 5 January 2024 (UTC)

Plural

If there are multiple sources listed that are all unreliable, saying "unreliable source" sounds weird compared to "unreliable sources". Template:Unreliable sources inline even redirects here! Solomon Ucko (talk) 07:13, 1 January 2024 (UTC)

This is an inline tag pertaining to a single source. If there are two in series and both are questionably reliable, tag them both, since each needs to be resolved but who is in a position to demonstrate that a particular source is reliable or not, or in a position to find a replacement for a particular one, may be different editors and they may be dealt with at separate times. If you put one tag after two sources, no one knows which one you're referring to.  — SMcCandlish ☏ ¢ 😼  13:37, 5 January 2024 (UTC)