Simpson's method in appendix extends beyond assemblage boundaries

[kcrisman]

Something is going wrong at

APEXCalculusPTX/ptx/appendix_back_reference.ptx

Line 1323 in b4f4b72

<me>\int_a^b{f(x)}\, dx \approx \frac{\Delta x}{3}\big[f(x_0)+4f(x_1) + 2f(x_2) + 4f(x_3) + \cdots + 2f(x_{n-2}) + 4f(x_{n-1}) + f(x_{n})\big]</me>

with the assemblage boundaries for Simpson's rule:

I've checked in multiple browsers. Actually, the one further down on "Taylor series expansions" also has this problem, but it's much less noticeable.

I recognize that this particular project is intended to mirror the physical book as much as possible, which rather conveniently (4th ed.) has a second column for "surface of revolution" and "force exerted by a fluid" so that these look quite nice visually, but I figured you would want to be made aware of it even though in this case it's not 100% lining up. Hope this is helpful.

[APEXCalculus]

Thanks for pointing this out. It'll be an easy fix.

[sean-fitzpatrick]

In the PDF I have tweaks to extend the boxes around the equations.

The issue here is with the HTML output. I think there are two options for fixing it, one of which is in our control:

1. Split the equation across two lines
2. Update the CSS to extend the box

It might be worth checking if the second is possible before giving in to the first option.

[APEXCalculus]

Giving in was my first thought. :)

[sean-fitzpatrick]

I asked on the pretext forums. Rob advises us to wait: PreTeXt will be transitioning from MathJax 3 to MathJax 4 this spring/summer, and that change alone might be enough to fix this. MJ4 is supposed to be better at calculating widths and introducing line breaks as needed.

[sean-fitzpatrick]

MathJax 4 has some display options for wide equations, including line breaking and scaling.

But I don't like how some of the line breaks are applied, and I don't think the average user is aware that they can right-click an equation to access the MathJax menu and change display settings.

(So giving in may still be the best option!)

[kcrisman]

Just as a related comment, there are many other places in the main text (so, not just this appendix example) where things go beyond the margins of boundaries as well, e.g. in Solution 1 of Example 6.3.3. So maybe you are right about giving in.

[sean-fitzpatrick]

Some of these you can chalk up to lack of due diligence on the part of the editor (i.e., me). @APEXCalculus mostly worries about the PDF output, and although there is a bad box there, it doesn't look too bad:

I can tell you that the number of overfull hboxes in the PDF build is large. Ones like this could be corrected, either by breaking the second line into two, or by shifting the alignment tab over to move the second line to the left.

If you want to make a list of the ones that bug you, I could look into it.

(We might be reluctant to add line breaks at the moment: we just got done with the figure placement in PDF, and the joy of LaTeX is that sometimes adding one line causes all hell to break loose.)

[kcrisman]

If you want to make a list of the ones that bug you, I could look into it.

Not at all, this is purely informational! I'm not going to tell you how to make your boxes, just wanted to make sure you were aware. The Simpson's one I will admit is a little bad, hence opening this issue in the first place; but I wouldn't have done so for the other ones, that would have been cheeky. (A possible, though annoying, solution would be to customize the bgcolor for these to be slightly lighter, since pdf doesn't have any coloration anyway. But I think students and instructors can cope, given the desire to support OER!)

…

Message ID: ***@***.*** com>

[sean-fitzpatrick]

One potential easy solution for HTML is to switch to using the 'Salem' theme. It's a wider format, intended to accommodate interactive exercises like Parsons problems.

The bad box in Example 6.3.3 goes away, but not, unfortunately, the one for Simpson's rule. There I think we have to give in and split it across two lines.