# Shall we force to use SI?

It seems a lot users here will be from USA, but still SI units are standard in science-level physics and may encourage broader participation. What do you think?

• Why the downvote? This discussion is on topic and IMO quite crucial.
– user68
Nov 2, 2010 at 21:24
• Downvote = Disagree on meta. Nov 2, 2010 at 22:31
• @Kenny But this is not a statement but a question; did you have in mind that this discussion is pointless?
– user68
Nov 2, 2010 at 22:33
• If this is pointless it would be closed. Downvote = "I disagree this this suggestion". Or do we need to establish a voting policy for meta.physics.SE? Nov 2, 2010 at 22:43
• Some of the suggestions in Standardized Units? on the Seasoned Advice meta may be applicable here, too; though the problem is rather more pervasive in physics. In particular a Units widget in the side bar could be handy. Nov 19, 2010 at 1:08
• We should probably also enforce setting all coupling constants to 2, and pi to 300... +1 : ) Jul 30, 2013 at 15:54

We should encourage the assumption that SI units are used unless specified otherwise. I agree that HEP formulae with $\hbar=c=1$ look better, but that should be stated somewhere.

• As a particle physicist I would expect that anyone answering particle or nuclear physics questions with non-natural units say so. We're just so immersed in that mode of thought that it takes a nudge to remind us that not every one does. Nov 19, 2010 at 1:11
• There's also the exception of astronomy, which sadly is still stuck with things like magnitudes and Janksy's (Janskys? Janskies?). Jul 28, 2015 at 3:41

I think we should enforce $\hbar=c=1$ and end of story. :P

Now seriously. I don't see the point of enforcing a unit system. If it's crucial to the question, the poster will inevitably make clear what is the unit system he's using. If he don't, then we can ask him to clarify the point. But most of the time the units will just be patently transparent...

I think explanations should be independent from the units, and while giving specific examples or formulas, the units system should be made explicit ...

That being said, any European student would cry if you answer EM problems in CGS units...

• SI units are generally desirable, though in many cases some sort of natural units (e.g. atomic/Planck units) are desirable. Nov 2, 2010 at 23:45
• I don't think it's too much to expect that people be able to convert between unit systems. If a question requires a number to be in a particular unit system, I think there's a good chance it's inappropriate for this site (e.g. it might be homework). Nov 3, 2010 at 4:35
• Agreed. I think we should say no to the Imperial system, in any case however. Nov 8, 2010 at 14:41
• Uhg...CGS ::shudder:: Nov 19, 2010 at 1:04

I'd rather we don't force answerers to use units/unit systems that they aren't comfortable with, but they shouldn't object if their answer is edited to have the conversions in other (more convenient?) units, or if someone contributes a version of their answer with a different unit system.

• (-1) I don't think answers should be edited, further than correcting non-native english errors, typos, latex formulas, etc. I know it is not meant to be that way, but I would erase any of my answers if it were substantially edited without my permission. It is simply so, call it a matter of taste if you want, but may be I am not the only one here with that taste. Feb 22, 2013 at 11:59
• @EduardoGuerrasValera: how is adding a numerical result in a different unit system a "substantial edit"? Jun 5, 2013 at 8:29
• @RodyOldenhuis, adding seems OK to me, as long as the original quantity in his units are not erased. I have had sometimes a paper reference in one answer of mine converted into a html link, and I have gracefully aknowledged it was a good edit. And there are many more possible non-intrusive edits (spell and grammar typos for instance) I always welcome. I think you can easily imagine the difference between that and a substantial edit, don't you? Jun 5, 2013 at 22:03

At least in classical mechanics, SI units really are the only way to go in my eyes.

I can understand people using CGS for EM problems. Though I've seen both SI and CGS in different courses here in Belgium and I can confirm that this causes the occasional confusion. This is why I think a universal system of units is vital for fluid communication. (not just on this website, also in the rest of society) However, for EM it's understandable to use CGS instead of SI, as long as it is always clear what system of units you're using.

In particle physics, natural units are the more natural (no pun intended) to use.

So it depends on the branch of physics you're talking about and I wouldn't force a system of units, but I would argue to try and use the dominant system in the particular branch of physics your question relates to. I think this is also clear from my discussion above: I would try to use SI in classical mechanics and preferably also in EM, natural units in particle physics, if desired CGS in EM, ...

Absolutely not!

In classical electromagnetism, CGS is more natural. In other cases we might want to discuss things in units where c = 1 or hbar = 1, etc.

• At least from an European perspective as soon as someone starts using CGS it ends with a great discussion what the result really means (-;
– user68
Nov 2, 2010 at 21:12
• @ mbq - Convert the units and find out if not intuitive in CGS? I guess I do not understand why they cause a large difficulty at least why it would be more difficult for Europeans. Many State side physicists do not have a great grasp of CGS because they rarely deal with it or were not taught EM in it. Nov 2, 2010 at 23:06
• @Davorak It is about intuition -- you learn to expect some form of solution and so "the subconcious unit checker" argues that something is wrong.
– user68
Nov 19, 2010 at 10:04

Another point: astronomers and astrophysicists commonly use CGS units for general quantities when they aren't using other standard length/mass/energy measures (Such as $R_{sun}$, $M_{sun}$ $L_{sun}$).

In gravitational physics, geometric units where G=c=1 are commonly used.

I think units should be explicitly stated, but certainly not enforced to one global standard. I think enforcing one standard of units within a question could be a good idea, though.