How do I go about publishing a theorem?

[HoldenCaulfield]

I have a theorem I came up with and have a proof for it. I was wondering if anyone here knew how I would go about getting it published. Is there some giant list of math theorems somewhere, or would I have to send it to a specific math book publishing company? I'm guessing this might be a patent thing. If so, any info on going about getting something patented would be appreciated.

[SmackFu]

It's been published already.

[HoldenCaulfield]

I realize that there are an incredible amount of theorems that have been published, but I'd still like to find out if someone had published this before me.

[Pies 'R' Squared]

Let me begin by saying that I have never submitted an article for publication, so I have no first hand experience. My opinion would be to first search the literature for your theorem and anything related to it. A library should be able to help you with this (there are several hundred mathematics oriented journals). From what I understand, you should check out Mathematical Reviews, also. Next, if things still look promising, submit your paper to an appropriate journal. If it isn't too long, I'd suggest American Mathematical Monthly, Mathematics Magazine, The College Mathematics Journal, Mathematical Gazette, or Mathematics Teacher. Each journal has its own 'specifications' for papers submitted, usually printed on the inside of the front or back cover. Best of luck, and let us know how things turn out. :)

Sincerely,
Pies 'R' Squared

[SmackFu]

Sorry for the negativity above. Just was in hope-dashing mood.

But I think Pies is on the right track. It might be tough to find good math journals at your standard library though. You're better off trying a local college library. The key is to find one paper that is somewhat related, then you can work backwards from the citations given there.

[The Ryan]

Well, if you really want it published, ytou could put it in MPSIMS, but then the reader would have acliam to it. And I think it would be a copyright, not a patent. To get a patent you would have to come up with some pratical application of it.

[RM Mentock]

So post the theorem already. And the proof. Let's have at it.

[Revtim]

Before you post it here, don't forget to read the fine print at the bottom of this very page.

[knappy]

I had a conjecture I came up with several years ago, which I never did succeed in proving. It was in number theory, and in my case, a great resource was Sloane's Encylopedia of Integer Sequences. This book told me that a sequence closely related to the one I was studying had a name; in fact it's called the Conway-Guy sequence (both Conway & Guy are famous mathematicians). Sloane's Encyclopedia is on the internet as well.

I later saw that Tom Bohman at MIT did prove what I was trying to prove.

[jmullaney]

It would be weird if the Chicago Reader won the Nobel prize in math!

[Joe_Cool]

Originally posted by Revtim
Before you post it here, don't forget to read the fine print at the bottom of this very page.

message board users retain the right to republish or repost their own work

I see no problems here. Sure, the reader can republish it, but they don't get to pretend they came up with it.

[jmullaney]

OK, so there is no Nobel prize in math.

But, Holden -- if your idea has any practical application, you should write it up as a computer program and patent that. I don't think a strict mathematical theorem can be patented in and of itself, but I could be wrong.

[RM Mentock]

Originally posted by jmullaney
It would be weird if the Chicago Reader won the Nobel prize in math!

Do you mean "doubly so!" because there is no math Nobel?

The fine print at the bottom says "Message board users retain the right to republish or repost their own work."

All SDMB asks for is a nonexclusive right to re-use the posting. That is probably necessary for archive purposes, even. They would still make proper attributions, I'm sure. That is at least as good a deal as most journals.

You can patent mathematical algorithms, I think, if you have a way to exploit it financially. If it is just geometry or something, let's have at it. At least, describe the general gist, or the area of math.

[Joe_Cool]

Originally posted by jmullaney
OK, so there is no Nobel prize in math.

But, Holden -- if your idea has any practical application, you should write it up as a computer program and patent that. I don't think a strict mathematical theorem can be patented in and of itself, but I could be wrong.

What exactly would be the point of patenting a mathematical theorem anyway? Just to be counter-productive? "Hey, everybody, 1 + 2 + 3 + ... + n-1 + n = n(n+1)/2, but you can't use that! You have to add it up by hand! NYAH!" The point of coming up with a new proof is to benefit the world at large and add to the repository of human knowledge. I understand wanting credit for your discovery, but the F-ing greed these days is making me sick. Patenting the human genome? There's a minimum of something like 10,000 years of documented prior art on THAT one.

Patenting every little thing has gotten really out of hand. Let's not encourage this crap. Stupidpatents.com (http://www.stupidpatents.com)

[Giraffe]

Holden, not to be discouraging, but I would start with the assumption that someone has already postulated and proved your theorem. Can you tell us (broadly) what field of mathematics it relates to? Number theory, complex analysis, topology, etc? What is your mathematics background?

Assuming, for the moment, that your theorem is new, you would submit it to an academic journal (Pies listed several that sounded appropriate). It would be peer-reviewed by other mathematicians, who would probably check that it was both new work and correct. If accepted, that would be it. You would have evidence that you were the first to think of the theorem. If it was important enough, they would name it after you. That's it. No money. No fame.

However, it probably isn't new. I really don't mean to be discouraging, but in academia there is usually a huge body of work that is published somewhere, but not widely known. You'll have to do some research.

Good luck, whatever you end up doing.

[jmullaney]

Originally posted by Joe_Cool
I understand wanting credit for your discovery, but the F-ing greed these days is making me sick.
Welcome to America. Please remember to sign up for Edisontm Electricity service, and purchase a Fordtm motor car to help you enjoy your stay. You may contact customer support at any time by dialing "0" on your Belltm telephone. Thank you, and enjoy your stay. If for any reason you decide you do not like the Constitution, please exit the country and you will recieve a full refund.

[Pies 'R' Squared]

I'd like to second Giraffe's good advice; don't be discouraged if your theorem (or your proof) isn't original. Proving classical and existing theorems is a learning experience; it's a large part of the exercises in advanced math classes.

jmullaney, I sadly agree, we have become a world of trademarks. However, I laud Joe_Cool's idealism. I believe in credit where credit is due, as well as research done for its own sake. (Naturally, this is easy for me to say, since I haven't done any significant research yet.) IMHO, it's a happy medium that is necessary. Elsewise, at one extreme, we have plagiarism; at the other extreme, we are reminded of the horrible consequences of the Newton/Leibniz calculus controversy.

If one does decide that money or fame will fuel their research, I suggest checking on the Clay Mathematics Institute's 'millenium problems'. The solution of any of these unsolved problems will be rewarded with a million dollars (yes, they are serious, and yes, the problems are hard).

Again Holden, good luck.

[Balduran]

From what I've seen, all that's necessary is to get it published. If people find it original and usefull, it will eventually be referred to as 'the Holden theorem' or something and of course, anybody that can understand it can use it for their own purposes.

[Some Guy]

I'd just like to add that, even if your theorem has already been discovered and proven, you may still be published by a respected journal if you have come up with a new, novel, or much simpler proof of an important (or merely well known) result. Granted, that'll get you even less than having discovered it, but you might get your proof used in a college level textbook or something.

re jmullaney : I'm not gonna go too far into GD territory here with a rant about the patent laws being inadequate to deal equitably with software (or other specific fields), but it is correct, at lest in the US, that a machine-code representation of an algorithm is patentable even when the algorithm itself is not. So long as this oversight is in the patent law, you might as well use it :).

[mack]

Originally posted by jmullaney
I don't think a strict mathematical theorem can be patented in and of itself, but I could be wrong.



It may not be able to be patented, but at least it can be copyrighted and protected until future use.

[HoldenCaulfield]

I'm not really all that serious about this. I am actually just a lowly student that came across something none of my math teachers had ever seen before. It doesn't have any practical uses that I have seen so far, but it took my math teacher and I about a week to prove it, and we joked about getting it published and instead of having my name on it, calling it the "Useless Theorem". I was thinking about trying to get it published, in hopes that someone much smarter than me could find a use for it. I guess I shouldn't have mentioned the word patent, that's really not what I meant; I just wanted to find out how to officially tag my name to the theorem.
Even though I'm not really serious, I think I might write a paper on it or something if I have some free time this summer and submit it to a journal just to see what happens.
I'm not gonna post it here because I still don't have good wording on it, and it has taken a good 10 minutes to explain it to well-educated people in the past because of its weird complexity. Also, I'm not even sure what to make of it. In the process of proving my main theorem, I proved about five other things I didn't find in my old geometry book, each of which seemed more useful than my original thing. IOW, I'm not even sure what I am going to focus on yet, that's why I like the idea of the paper, because I can cover everything without trying to fit it all in a couple sentences.
Sorry, for wasting everyone's time, but thanks for the information.

[Joe_Cool]

Originally posted by jmullaney
Welcome to America. Please remember to sign up for Edisontm Electricity service, and purchase a Fordtm motor car to help you enjoy your stay. You may contact customer support at any time by dialing "0" on your Belltm telephone. Thank you, and enjoy your stay. If for any reason you decide you do not like the Constitution, please exit the country and you will recieve a full refund.

Con Edison is a company that holds many patents, but there is (and can be) no patent on electricity.

Bell (Formerly Bellcore - BELL COrp REsearch labs, now Lucent Technologies, Inc.) is one of the most prolific research companies in the world and holds many patents, including an expired one for the specific designs of the original models, but there is and can be no patent on the process of communication by encoding sound energy atop an electromagnetic carrier wave, either via wire or wireless.

Ford also holds a great many patents for wonderful designs, but cannot patent the concept of a self-propelled wheeled vehicle powered by an internal combustion engine.

If you do your homework and read some of my other posts, you'll see that I am one of the most emphatic defenders of the Constitution that you will ever meet. My problem isn't the concept of Patent itself, but the abuse of Article I,
Sec 8. Sure you can copyright (not patent) a particular implementation of a mathematical concept, but a) not the concept itself, and b) it will be circumvented. It's generally not too difficult to write a custom implementation of an algorithm that does not incorporate any of the original code.

But a proof of a mathematical concept generally does not have much practical use except in that it proves the concept. For example, the gentleman who finally proved Fermat's last theorem owns the copyright on the paper he published, but he doesn't own and cannot regulate usage of the fact that an+bn=cn has no solution when n > 2. Get it?

However, if you play with your cat using a laser pointer, you are in violation of patent #5,443,036. Wonderful country, eh? (couldn't link directly, but go the search page (http://164.195.100.11/netahtml/srchnum.htm) at http://www.uspto.gov )

[Arjuna34]

Originally posted by Joe_Cool

However, if you play with your cat using a laser pointer, you are in violation of patent #5,443,036. Wonderful country, eh? (couldn't link directly, but go the search page (http://164.195.100.11/netahtml/srchnum.htm) at http://www.uspto.gov )

Here's a direct link: US5443036:Method of exercising a cat (http://www.delphion.com/details?pn=US05443036__).

I see that our tax dollars have really paid off, since they referenced this patent to develop the cat exerciser:

US5194007:Semiconductor laser weapon trainer and target designator for live fire (http://www.delphion.com/details?&pn10=US05194007) :)

Arjuna34

[wolfseyn]

I just saw this link (http://www.accessatlanta.com/partners/ajc/epaper/editions/monday/features_a3ca67fe959dd0db00a2.html)at slashdot (http://slashdot.org) and remembered this post.

[dasmoocher]

I admit I didn't read every sentence of every previous post.

But, (I didn't see this in my quick scrolling down) is there a database of deposited math theorems?

If you want to know something about a biological subject, you search NIH's PubMed database. Type in say, estrogen, into the search engine and you might get 20,000 hits--references in the literature.

Is there a similar physical search engine for math and physics?

I apoligize if I missed the post, but it's getting late and a PubMed type physical search engine would be great. But I'm not in the field and unfamilar w/one.

[Giraffe]

647, if you want to search for publications in physics, chemistry, mathematics, etc. you would go to http://www.webofscience.com. However, they only list publication title, author(s), and abstract. If you want to search for, say, "superfluid helium" it's pretty straightforward -- you'll get a list of publications.

However, if you want to search for a theorem you're not sure exists and don't know the name of if it does, you're out of luck. There may be math-specific databases of theorems, but I wouldn't know.

[dasmoocher]

Thanks for the idea, giraffe, but when I clicked on your link I was sent to a page that said I needed a subscription to the 'Web of Science'. And I'm accessing the Web through a Univ server.

I guess I'm looking for the equivalent of PubMed for physical sciences. A public database that will let me search the abstracts of the papers published in the physical science literature. PubMed gives anybody free access to the abstracts in the bio field. How about a PubPhys?

Where can I type in "Black Hole" into a physical science search engine ala PubPhys?