timo

Feynman diagrams actually is (to some extend) the way I usually determine whether some particle interaction is possible, and also to some extent how likely it is. What you do is trying to construct graphs with the correct incoming and outgoing particles; from the complexity of the graph and the type of internal connections and lines one can often roughly guess the reaction probability. It's not exact of course (Feynman diagrams are exact in the sense of being addends in a complicated mathematical expression, but evaluating this requires plugging in numbers and a bit of math), but often sufficient as a first orientation or for more general statements.

Capn probably underestimated just how freaking big your pdf is. According to my User CP page .so I guess ~20 MB is the limit.

... which is the easy part of the endeavor.

Given a probability function P(x) and a function f(x), then the expectation value for f is [math] \bar f = \sum_x P(x)f(x)[/math]. The second moment is the expectation value for the "function" f(x)=x², i.e. [math] \overline{x^2} = \sum_x P(x) x^2[/math].

It's stable; same as the electron.

It's called both. I've used the less fancy sounding term on purpose because it's clearer.

That's kind of what I meant, yes.

The simple answer is: any. The more complicated answer is that "matter annihilates with anti-matter" is not exactly a correct statement but more like a lie for laymen or perhaps a rule of thumb. While an electron (matter) and an anti-electron (anti-matter) do indeed annihilate, this is not true for a proton (matter) and an anti-electron (anti-matter).

^^ my original reply was crap since I mis-read your equation (didn't see that 4y has a "y" in it...). Sorry about that. See the now-modified post.

EDIT: Sorry; I mis-read your equation. I am not completely sure that you want to isolate y, so I'll give you two options of what can be a sensible thing to do. 1) Solve for y -> quadratic formula. 2) Rearrange (y-something)^2 = something else. -> completing the square.

Perhaps that's what he said. After all, neither the mysterious "n" nor the quantity that scales with this "n" were specified

I don't quite get the original question: are you supposed to give a talk on a paper you were given to summarize (say in some form of journal club), or are you supposed to give a talk about your research and want to base it on a publication of yours. In the latter case: why? Also, who's the audience? Is it colleagues at your institution or potential competitors, and are you going to be graded or not?

Q1: Obviously, the more you know that other people don't, the higher the chance that no one has unsuccessfully tried your Ansatz before. Good at logic, discrete math, graph theory and algorithms (dunno what tomography is) applies to pretty much every computer science student, for example (not sure how many of them are interested in math, though). Q2: Depends. Generally: no. But if for example you make big claims about division by zero without knowing that division is commonly defined as the multiplication by the inverse, then you're possibly proving things in a mathematical structure that no one is interested in.

Have you looked at the course descriptions, possibly also at other universities?

What do you think the consequences of this statement should be? One might think that the disclaimer "according to relativity" is somewhat implicit in posts in a science board's forum called "relativity".

I'd start with looking for courses named something with "logic" offered by the math department. That should give some results. While doing so for some random university, I figured that in case you cannot attend a course, the course descriptions are likely to contain relevant literature.

If cross-attending lectures is not common in your university, then perhaps ask a professor if it's still ok for him for you to attend his/her lecture. Usually, professors are not exactly offended by students with a real interest in their lecture, and math lectures tend not to be completely crowded. You don't even need a formal grade, so there's no problem with possible bureaucratic hindrances. I seriously recommend trying this route, since supervised learning is imho superior to self-learning in abstract field like logic (the biggest problem in logic is that it's often counterintuitive and that intuition is hard to switch off without formal training). I'm afraid I have no advice for self-learning, in case you really have to fall back on that.

Taking a course offered by the math department seems like a rather obvious choice, so I guess you're asking for ideas apart from that?

I have little to no time at the moment but two comments that might help you: 1) t1, ... t3 obviously are translation vectors. The question is whether they are the translations that span a elementary cell. While you can certainly figure that formally by solving some equations, I'd check that visually. To be the vectors you are looking for, a) you have to get to each point to another with an integer multiple of these translations, and b) starting from some point, each integer multiple of these translations must lead you to a new point. 2) The TeX mode in this forum is activated with [ math], not with [ tex].

That's not true, at least when the terms energy and matter are taken literally.

Do you also run a web-page about time cubes by chance?

By research institutes I mean universities, Max-Planck institutes (the by far largest non-university research community in Germany, which is where I live, btw.), and CERN (a rather large internationally run research center in central Europe). That's the only institutions that I can speak of from first hand. I didn't say anything about a discount. I would in fact expect that the full flat-rate fee is paid (something around a few thousand dollars a year, I think). But I never bothered to really ask someone about it since for me it only matters that I can access most journals without even seeing a barrier. I did in fact not mean companies like Intel. I don't think profit-oriented companies are very interested in spreading innovation, so I didn't really consider it part of the question.

Magnetism is being considered part of electromagnetism. There is no theory which relates electromagnetism to gravity, despite quite a long history of looking for such a unification (often called "theory of everything" in popular culture).

1) Sometimes, there's free versions available in addition to the publisher-set one. If you found a paper using Google Scholar (that is how I search for papers, at least), click on "all x versions" on the bottom of the hit and check if there's a freely-available version. 2) More importantly: Research institutions do have a flat rate for many journals. Hence, the majority of researchers isn't really affected (except that it can make working at home an annoyance). Also keep in mind: Articles in scientific journals are (imho) not the main route of spreading new ideas. I get new ideas from attending talks, presentations, conferences, or just talking to the people; not from searching papers on the internet. Of course I do have a look at the paper if I find it interesting, so admittedly the fee would still be a hindrance, wasn't it for 1) and 2).

What exactly do you mean by "a system like it"? A lot of CPU power? Climate science does use a lot of CPU power. I do not know how exactly it compares to LHC requirements, but if it's significantly less then that's likely either because of a lack of funding or because CPU power is not the limiting factor in climate science. Particle physics at colliders is much more accurate than climate physics (at least I think so) or actually almost any other branch of physics (that I do know for at least some other fields). A distributed system? I don't know about climate research, but collider data is remarkably well-suited for trivial parallelization. What you have is collisions of protons, which are independent from another. A huge amount of them. So in principle, you can send each of them to a separate computer, analyze them there, and only send the result back to a central facility. That's not possible in systems where the events in one part of the system influence other parts. A de-centralized processing might require a huge amount of data transfer for communication between the different nodes.