Definition talk:Norm on Division Ring

After careful consideration, I think this definition may be generalised to a ring without zero divisors. This is good, as then it will incorporate the absolute value on $\Z$. --Lord_Farin 05:30, 9 December 2011 (CST)

I believe you can accomplish that by just dropping to ring. Positive definite + multiplicative will then prove it to have no proper zero divisors. That distinction will only come up if you decide to define a seminorm for rings. --Dfeuer (talk) 18:06, 17 January 2013 (UTC)

Wikipedia, at Berkovich space, says this:

A seminorm on a ring A is a non-constant function f→|f| from A to the non-negative reals such that |0| = 0, |1| = 1, |f + g| ≤ |f| + |g|, |fg| ≤ |f||g|. It is called multiplicative if |fg| = |f||g| and is called a norm if |f| = 0 implies f = 0.

I have no idea if that's right, of course. However, suppose $R$ is a division ring with a multiplicative seminorm.

Then since the seminorm is not constant, there is an $x$ such that $|x|≠0$. Let $y≠0_R$.

Then $|x|=|x \circ y^{-1}\circ y| \le |x \circ y^{-1}||y|$, so $|y|≠0$.

Thus a multiplicative seminorm on a division ring is a norm. --Dfeuer (talk) 20:53, 17 January 2013 (UTC)

Having discovered a definition of that site, it is now important that you find some corroborative evidence elsewhere to back it up. Not only is Wikipedia a tertiary source but it is well-known as being laughably unreliable. --prime mover (talk) 21:33, 17 January 2013 (UTC)

Change suggestion

The pages "Normed Division Ring" and "Norm (Division Ring)" are structured differently to the similarly named pages "Normed Vector Space" and "Norm (Vector Space)". The page "Normed Division Ring" is simply a redirect to "Norm (Division Ring)". Whereas "Normed Vector Space" actually defines a normed vector space which links to "Norm (Vector Space)" which in turn transcludes the definition for a normed vector space from "Normed Vector Space". Should the pages "Normed Division Ring" to "Norm (Division Ring)" be structured similar to the pages "Normed Vector Space" and "Norm (Vector Space)" or should the redirect page "Normed Division Ring" be removed? I'm happy to make the necessary changes to do either, or any other preferred approach. --Leigh.Samphier (talk) 08:44, 7 October 2018 (EDT)

I moved the above comment from the redirect page onto here because it makes no sense there. --prime mover (talk) 04:27, 7 October 2018 (EDT)

It's complicated. There is already a transclusion of Norm (Division Ring) in Norm (Vector Space) which itself also has a page Normed Vector Space where the assumption appears to be that Normed Division Ring is a specialisation of Normed Vector Space. I don't know, myself, as I have never studied norm theory and so I don't understand either its motivation or direction.

A recent contributor greatly enjoyed starting jobs of moving things around and refactoring stuff to suit his personal idea of how things should be, but appeared not to be such a huge fan of finishing of what he started. Consequently there is a lot of stuff which has been left in an inconsistent state, and sometimes with a plentiful number of broken links as a result of an incoherent renaming strategy. This is I believe one of those areas.

The top level page should be "norm" and it should then be subpaged as "Norm/Vector Space" and "Norm/Division Ring" and so on, for whatever objects are appropriate. Whichever one is the most general should be at the top, and more specialised contexts should be at the bottom (as, for example, we have pages structured Topology $\to$ Metric Spaces $\to$ Real Spaces and Complex Spaces, etc. according to the concept being defined. The reason for the subpage name structure is that if you name something Concept/Subconcept, then at the top of the page you automatically get a link back to Concept, which can be useful. If you transclude pages which are not so named, you don't get the link and so if you want it you have to build it specifically.

Bottom line: as long as it's consistent it does not matter really, except to note that:

A page named "Norm (Division Ring)" defines the norm that has been imposed on a division ring.

A page named "Normed Division Ring" defines a division ring that has had a norm imposed on it.

Whether it is worth maintaining a separate page for both concepts is a philosophical position -- in general, certain $\mathsf{Pr} \infty \mathsf{fWiki}$ have preferred in the past to separate the concepts onto their own pages, while other prefer to merge them. I myself am of the former persuasion.

Bottom bottom line: feel free to take this area on and make it consistent and aesthetic. --prime mover (talk) 09:08, 7 October 2018 (EDT)

If I might add to the excellent points made: make it consistent, and aesthetic -- in that order. Cf. PM's allusion to another contributor.

As I have seen, you've taken on a habit of preparing pages in your sandbox before putting them out there. I greatly appreciate this as it keeps the site consistent.

When taking on restructuring of this magnitude, it might make sense to start drafting a whole series of pages in your sandbox (as subpages), which together purport to replace the bunch out there. Additionally it might make sense to move the existing pages out of the way to the /Backup/... domain under your personal user, so that any mistake you discover in the last second is easily reverted.

As you might guess, I've learned these lessons the hard way :). — Lord_Farin (talk) 14:16, 7 October 2018 (EDT)

I have put a proposal together Leigh.Samphier/Sandbox/Norm Refactor, and created a discussion page. When you have a chance I would appreciate some feedback --Leigh.Samphier (talk) 08:37, 19 October 2018 (EDT)

That's great, thanks a lot! Unfortunately I am currently extremely pressed for time... It might take as long as one week to find the time to give your work the attention it deserves. But know that it is on my list and I am going to take a look! — Lord_Farin (talk) 15:47, 19 October 2018 (EDT)

I'll see if I can get to it tomorrow. Been out of the country for a few days and had limited spare effort to look at it. --prime mover (talk) 19:33, 19 October 2018 (EDT)

... I've just had a quick look, and I can't find anything to argue about. When it's all done we can pick over the details if necessary, but as far as the high level structure is concerned, your decisions make good sense. --prime mover (talk) 19:36, 19 October 2018 (EDT)

And finally, I have found the time to go through it all. I have to say it is a commendable job, very well done!

There are always ideas for further improvement but I agree with PM that this can be done in-place, since the current proposal is already a big improvement over what is currently there. I will add some small suggestions to the talk page in your sandbox.

You've done a great job and I very much commend you for taking all this effort and preparing everything in your sandbox! — Lord_Farin (talk) 06:48, 28 October 2018 (EDT)

What L_F said. Good job. --prime mover (talk) 08:12, 28 October 2018 (EDT)

Thanks PM and L_F. I'll start on this asap so that my copy of the pages don't diverge any further from the originals. --Leigh.Samphier (talk) 04:51, 29 October 2018 (EDT)

Deletion suggestion

No, please do not delete this page.

As per the house convention, redirect pages like this are to be used specifically instead of transcluded subpages. In this case we want to specifically link to "Definition:Norm (Division Ring)" instead of "Definition:Norm/Division Ring". --prime mover (talk) 08:47, 29 October 2018 (EDT)

Ok. I guess that means that all of changes that I have just made from "Definition:Norm (Division Ring)" to "Definition:Norm/Division Ring" need to be reverted. Is that correct? If it is then I'll fix that tomorrow. --Leigh.Samphier (talk) 09:02, 29 October 2018 (EDT)

Sort of under way now. Apologies, it's a nuance of $\mathsf{Pr} \infty \mathsf{fWiki}$ which seems sensible to us admins, but is far from obvious.

Wondering whether it should be renamed to "Norm on Division Ring" which I think I will "just do" ... --prime mover (talk) 09:21, 29 October 2018 (EDT)

All done. --prime mover (talk) 16:39, 29 October 2018 (EDT)