The page
https://en.wikipedia.org/wiki/Help:Displaying_a_formula#Subscripts.2C_superscripts.2C_integrals
contains the following example:
\frac{\sum\limits^{^N}_{k=1} k^2}{a}
\bigcap_{i=_1}^n E_i
\bigcup_{i=_1}^n E_i
Where "N" and "1" are used as sub/sup scripts. I'm not sure what's the intention here, but the rendering does not look good to me and that obviously confuse assistive technologies.
I have no idea. I asked on of the admins https://en.wikipedia.org/w/index.php?title=User_talk:Edokter&oldid=672193927
I classified the task as low since the output is somehow like I would the TeX input expect to look like
From viewing the examples, all I notice is that in the lower limit, "i=1", the terms on the two sides of the equal sign are at different levels. I'm guessing that is the problem seen by fredw with \bigcap and \bigcup.
Does anyone know about this special notation for limits? This looks a TeX hack to achieve a purely visual distinction for "alternative limits style".
In my opinion, it is not really semantically correct to write the right operand as a subscript. For example, one could define two equalities $=_1$ and $=_2$ (say from the ideals of meager and null sets of reals) and then write $f =_1 g$ and $f =_2 g$ and here "g" is the right operand not 1/2.
Assistive technologies will present this as "i equal subscript 1" which also sounds a bit weird.
Maybe this bug is just "invalid" and we do not have a proper solution for now. But this is one of the painful LaTeX hacks that do not work well when one tries to think further than just "math on paper"... I wonder if we really want to keep that suggestion on the Help:Displaying_a_formula page.
Is there an actionable issue here?
The example in the help page has changed which seems to indicate that this was a typo.
The current rendering for the examples in the OP look fine (and match real TeX output); the update version also looks fine.