# What to do if mathrm is hugging physical units?

There was some consensus about how to write physical units from mid-2016 onwards on Chem.SE, as apparent from this post. However, it seems like the damage had already been done by that time.

I say so because, I have lately been retagging several old questions, and almost everyday, I find myself converting \mathrm to \pu like a gazillion times. This is a latest but a much simpler example of what happens everyday.

This post is from 2015 and is just one example that indicates how superusers were actively converting physical units to \mathrm at that time.

I often feel that, by converting \mathrm to \pu, I am (1) not adding any visual change to the post (2) undoing the super user's efforts. This conversion apparently adds no value to the post, and in fact, undoes what had been done by others in the past. And many times, this conversion gets really hectic to do, especially when the usage of \mathrm is like this 5~\frac{\mathrm{J}}{\mathrm{g}^\circ \mathrm{C}} (believe me when I say I have seen this pattern more than once)

Thus, I wish to reach a consensus for:

If I am editing a post for retagging/title fixing/language fixing purpose, do I also need to edit \mathrm to \pu when it depicts physical quantities?

• What is the need of it if mathrm is doing the job properly? Feb 19 '18 at 15:04
• @ApoorvPotnis I don't know myself. I have always thought "pu"="physical unit" is the standard for depicting physical units, and mathrm was just a workaround until pu didn't exist. Feb 19 '18 at 15:06
• The mhchem package simply offers a shortcut to typeset units with \pu. \mathrm does the job fine but just with slightly longer code. Feb 19 '18 at 15:17
• @ApoorvPotnis Yes, that's what I've started thinking now as well. that's why I am seeking confirmation on meta for this. Feb 19 '18 at 15:18
• Apoorv is right. There's nothing official about \pu. Just lazy. Feb 19 '18 at 17:57

TL;DR:

What is the need of it if mathrm is doing the job properly? – Apoorv Potnis

There is no need.

There always was a demand for the siunitx package for LaTeX (here by me on our site). However, that was creeping along very, very slowly. In the meantime pu has been implemented for our convenience, see: How to write physical units.

I have always thought "pu"="physical unit" is the standard for depicting physical units, and mathrm was just a workaround until pu didn't exist.

That is in principle correct, as most of our users adapted quickly to the much simpler interface. Apart from typing upright units, it also does spacing quite magically correct. I personally find it a lot friendlier to read in code. For example:

The mhchem construct $\pu{8.2E-2 L atm//mol K}$ (1) renders the same as $8.2\times10^{-2}~\mathrm{\frac{L\,atm}{mol\,K}}$ (2), but is infinitely better readable.

(1)   $\pu{8.2E-2 L atm//mol K}$
(2)   $8.2\times10^{-2}~\mathrm{\frac{L\,atm}{mol\,K}}$


If I am editing a post for retagging/title fixing/language fixing purpose, do I also need to edit \mathrm to \pu when it depicts physical quantities?

When editing a post, it is not necessary to turn correct typeset from one form into another. If it is consistent with the correct type-set, then there is no need to further improve things. That is a waste of time and effort.
If someone uses $8.2\times10^{-2}\frac{L atm}{mol K}$, $8.2\times10^{-2}\frac{L atm}{mol K}$, on the other hand, it would be beneficial to change that.

• However, if there is something else to fix changing the underlying MathJax commands is, of course, allowed (if the editor wants to take up the task).
– Jan
Feb 20 '18 at 9:24
• @Jan If the current source typesets correctly, leave it. Changing it risks making mistakes, decreasing the value of the post. And for what benefit? Nobody reads the source. Mar 4 '18 at 0:32