Aata #215
Opened this issue · 2 comments
MathML:
<mo stretchy=\"false\">[</mo><mi>GF</mi><mo><!-- --></mo>
<mo stretchy=\"false\">(</mo><mn>625</mn><mo stretchy=\"false\">)</mo>
<mo>:</mo><mi>GF</mi><mo><!-- --></mo>
<mo stretchy=\"false\">(</mo><mn>25</mn><mo stretchy=\"false\">)</mo>
<mo stretchy=\"false\">]</mo></math>
This is a problem with ":". Although it is best to use U+2236 for ratio, people will use ":". So hueristics... In this case, ":" is not a ratio, it is part of a field extension syntax. I had some conversation about this with Susan Ousterhaus and she said that only ratio gets that form. It doesn't apply to trilinear cordinates (which are sort of a form of ratios) or to field extensions. Therefore, using "⠐⠂" is inappropriate and the transcriber and SRE's hueristic got it wrong.
Personally, I think math typography should be followed and that one form is used when math spacing is symmetric and the other when it is asymmetric (as in a mapping). But that's not the BANA interpretation. Sadly Dr. Nemeth isn't alive because I think he'd disagree with BANA.
Result: "⠈⠷⠠⠠⠛⠋⠷⠖⠆⠢⠾⠀⠐⠂⠀⠠⠠⠛⠋⠷⠆⠢⠾⠈⠾"
Should be "⠈⠷⠠⠠⠛⠋⠷⠖⠆⠢⠾⠸⠒⠀⠠⠠⠛⠋⠷⠆⠢⠾⠈⠾"
SRE gets the related expressions wrong
trilinear coordinate:
<math><mi>a</mi><mo>:</mo><mi>b</mi><mo>:</mo><mi>c</mi></math>
Indeed, this is not a ratio. And we had very lengthy discussions on that bringing in a number of people specifically as to how to interpret the following passage 13.7.1 from the Nemeth_UEB Rulebook May 2020:
: (colon)
... as a mathematical symbol ⠐⠂ (ratio, meaning "is to"; "such that",
mapping notation, and other meanings)
... as a punctuation mark ⠸⠒ (preceded by a punctuation indicator
where appropriate)
In particular, no-one could for example explain what "mapping notation" really is, where it then oddly enough is written as unspaced⠸⠒ on page 4-16, 4.11.3. On the other hand in 4.12 a:b :: c :d is transcribed as
⠁⠀⠐⠂⠀⠃⠀⠰⠆⠀⠉⠀⠐⠂⠀⠙ while further below p:r = q:s is transcribed as ⠰⠏⠸⠒⠗⠀⠨⠅⠀⠟⠸⠒⠰⠎
Given this section and that Susan is involved in the standard, I am a bit surprised about her categorical answer. I had multiple conversations with her, but never on the topic of colon.
Anyway, the outcome of the discussions were to go with the "mathematical symbol" interpretation here. As it is an operation here, I thought, one could drop the surrounding spaces but ultimately it was decided to leave them.
Therefore, using "⠐⠂" is inappropriate and the transcriber and SRE's hueristic got it wrong.
I can see you are very happy about that. Good for you...
I am sure there are many things I and SRE get wrong, but please never blame the transcribers I work with!
I agree that the lesson text is poor and the examples seem to contradict it. The green book is consistent, but BANA seems to want to change the rules a little. I'm sure the people who wrote the lessons would say the material is "provisional" and not final. It is why I asked Susan O. for clarification. She wrote:
....However, with 3:30 or f:(x, y) or A:B or [A:B], we just have a colon and transcribe it accordingly. There is not a section on group theory that I know about in the Nemeth Code book, but there is a discussion of the colon on page 48.
Now, here is what we have written in the newly revised Code Book – now up for review. You will notice that we definitely treat a ratio symbol as being different from a colon.
This is followed by some examples which do not include the [A:B], so perhaps you can argue that what she wrote doesn't really address this issue. Also, she is just one member of the technical committee and doesn't speak for it.
I have sent several questions to the BANA technical committee asking for clarification on various points. I received one response (so it is not a complete black hole) asking for clarification about some MathML notation, but zero answers. I don't remember if I asked about colon -- you need to fill out a form on http://www.brailleauthority.org/contact/comment-nemeth.html so I have no mail record of what I've sent. Perhaps you will have better luck getting an answer out of them. It certainly wouldn't hurt to have more than one person ask the same question. Not sure it will help though.
I don't understand why BANA doesn't make Nemeth follow the typographic rules around ":" -- it would help reduce ambiguity in translating math. It would also mean sighted and unsighted readers would get the same experience since authors sometimes use the wrong version of ":".