Crossreferences ending up inside stem expressions

[opoudjis]

10303-42, 4.5.74:

The knot multiplicities and knots in the _u_ and _v_ parameter directions are then given by the function calls:

[stem]
++++
&lt;<express:"geometry_schema"."default_b_spline_knot_mult">&gt;(SELF\"b_spline_surface"."u_degree", "ku_up", "uniform_knots")</express:"geometry_schema"."default_b_spline_knot_mult">
++++

[stem]
++++
&lt;<express:"geometry_schema"."default_b_spline_knots">&gt;(SELF\"b_spline_surface"."u_degree", "ku_up", "uniform_knots")</express:"geometry_schema"."default_b_spline_knots">
++++

[stem]
++++
&lt;<express:"geometry_schema"."default_b_spline_knot_mult">&gt;(SELF\"b_spline_surface"."v_degree", "kv_up", "uniform_knots")</express:"geometry_schema"."default_b_spline_knot_mult">
++++

[stem]
++++
&lt;<express:"geometry_schema"."default_b_spline_knots">&gt;(SELF\"b_spline_surface"."v_degree", "kv_up", "uniform_knots")</express:"geometry_schema"."default_b_spline_knots">
++++
--

This is of course ending up as nonsense:

Metanorma at this time has no tolerance for hyperlinks embedded inside of equations: it expects to find only pure maths. And given the problems we go through to convert between formats, we do not WANT any such tolerance. If you want to annotate equations in that way, you should be doing that in a where clause after the equations.

@TRThurman I've been rather keen to ensure that your source markup is preserved, but in this instance, I do not believe that the markup is consistent with our architecture, which routinely swaps between Asciimath, MathML, and OOXML (none of which would have any time for hyperlinks). I am reluctant to implement hyperlinks in equations. It's not impossible, but it is difficult, and I don't think it is legitimate.

@HassanAkbar Subject to confirmation from @ronaldtse, I would like for you to strip out all cross-references in stem expressions.

[TRThurman]

@opoudjis, @ronaldtse, @HassanAkbar, @stuartgalt , As long as we have traceability back to the source for stripped-out cross-references I am fine with the approach of removing them.

I looked at the specific examples you noted and they are there likely because of convenience. (There are also pictures which will have to be converted to equations anyway.)

Longer term, the community will need to address links to EXPRESS ENTITY data type attributes as operands in a formula/equation. Frankly no one should have to type in this code anyway. I demo'd Plurimath.org to WG12 a couple of weeks ago and they appreciated not having to type the brain-dead syntax.

{I will be on travel this week and no computer and github doesn't like my phone anyway. If you really need a response, send direct email as I will check emails once a day fro crises.}