« first day (1633 days earlier)      last day (3286 days later) » 

1:28 AM
(redacted)
 
@NeuroFuzzy hows electrodynamics going?
 
@StanShunpike oh it's a headache. I had a problem that ended up being six pages of algebra, I have a B average on the homework because the grader doesn't like my handwriting, and I'm not learning any useful theoretical things, just PDE solutions that would be quickly disposed of in mathematica.
so. That's the norm for this quarter, I guess.
 
Screw the handwriting thing. I hate that. Mine is awful too. I type my econ psets
But it sux they mark u down.
 
of course, because the TA is grading strictly by-the-book, everyone who has the solution manual and cheats is getting 100s :/
 
Eww that's so lame
 
1:36 AM
You could LaTeX your solutions. My handwriting is atrocious, starting to LaTeX my solutions early on in undergrad was one of the best decisions I made.
 
Well, I feel a lot better now lol
 
Helped clarify my thoughts and left me with a readable corpus that I could refer back to
 
@alemi people keep saying I am over the top, but yeah exactly. I see all kinds of mistakes I have made more clearly.
I am used to reading typed stuff anyways.
 
@alemi I guess so. But I'll need a different approach for the midterm and final.
it's probably still a good idea.
 
The other thing is, when I get stuck, I can SE the concept behind the problem
I obviously don't ask a hw question
But it saves time having to type it from scratch
Since I am doing it anyways
 
1:40 AM
Another benefit of typesetting is that it encourages prose. I can't tell you how frustrating grading as a TA was when the students handed in scribbled sheets with just equations on them and no explanation of their thought process or take on the problem.
2
 
yo
 
(████████████)
 
??
 
Yeah, wtf lol
?
 
can't wait for summer
 
1:44 AM
I was continuing the logical progression of the recent starred posts
 
lolol
 
know matlab?
 
a tad
 
I need to replicate the graph on the last part of this paper. I know it's straight forward BUT I KEEP ON GETTING OUT OF BOUNDS or integer errors
 
are you using the code at the end of the article?
also, funny you should mention that article, I just wrote an article on zombies myself
 
1:52 AM
@alemi in what context?
 
yeah the code at the end alemi .. it's a two parter that is one long code!
 
To introduce stochastic epidemiology modelling and some techniques in the study of critical exponents
 
I KNOW I TYPED IT RIGHT BUT IDK WTHECK IS GOING ONN!N!!
 
and the parameters are given in the last few pages of the report too
except t = 100 and dt = ? ughh! if they got k = 0.25 and k = Ti/dt from the previous code that's k = 25/99 = 0.252525
that is the only combination that can produce it! but still the code is like
>> erad(0.0075,0.0055,0.09,0.0001,0.25,100,99)
Warning: Size input contains non-integer values. This will error
in a future release. Use FLOOR to convert to integer values.
> In erad at 6
Warning: Size input contains non-integer values. This will error
in a future release. Use FLOOR to convert to integer values.
> In erad at 18
Attempted to access sol(1.25253); index must be a positive integer
or logical.

Error in erad (line 19)
 
1:56 AM
@alemi that's a creative way to think about that kind of topic :D
 
ok wait. if T = 100 because on the impulses are broken up into 4 parts

Ti = T/4 that has got to be T = 100
Ti = 100/4 = 25 makes sense
 
@usukidoll if I'm reading your thing right, your passing in Ti = 0.25 and dt=100, so that k, at the top of eradode, which should be the length, will be a float and less than one
I think you've reversed those two inputs
 
O_O!@
reverse?
whoa whoa
erad(0.0075,0.0055,0.09,0.0001,0.25,100,99)
a,b,ze,d, are already given in the paper. He also k = 0.25
based on the previous code I saw k = Ti/dt and the only combination is k =25/99 = 0.252525 -> 0.25 since it's a repeating decimal
 
@Stan Here is a single outbreak, modelling every person in the 2010 US census, starting with ~300 zombies. docs.google.com/a/cornell.edu/file/d/…
@usukidoll notice that k=Ti/dt; and then they loop: for i=1:k, k is supposed to be the length of the vector of states, they use it to index into s,z, and r, setting values. It needs to be an integer
 
integer? the set of all integers are negative and positive numbers
ughhh how did he get k = 0.25 then?
 
2:02 AM
Just below Figure 10, they say k=0.25, but that refers to the k in the equation above, the magnitude of the inpulse at time tn. the k in eradode is an index for the state vectors
 
?
you mean k = Ti/dt must mean something else D: instead of finding a value that produced the k = 0.25
 
sorry, my bad, I thought what you posted were inputs to eradode, not erad. The k parameter in erad is 0.25 as you say
but, you still have a problem
 
yeah I was doing the erad and it still is going nuts xxx..x.xx.x. what is going on
 
you've set T=100, dt=99, which would set m to not be an integer
set dt to something like 0.1
or 0.01
 
erad(0.0075,0.0055,0.09,0.0001,0.25,100,0.1) like this
 
2:05 AM
try it
 
erad(0.0075,0.0055,0.09,0.0001,0.25,100,0.1)
Attempted to access sol(1002); index out of bounds because
numel(sol)=1001.

Error in erad (line 21)
z1 = sol(j+n+1);
Error in erad (line 21)
z1 = sol(j+n+1);

>> erad(0.0075,0.0055,0.09,0.0001,0.25,100,0.01)
Attempted to access sol(10002); index out of bounds because
numel(sol)=10001.

Error in erad (line 21)
z1 = sol(j+n+1);
-.-
 
you're going out of bounds on your state vector
 
so how do I become in bound?
and where is the state vector?
 
I'm just trying to name the s,z, and r vectors something
calling them state vectors
 
oh .. my professor told me that the initial values of s1 z1 and r1 are already in the code which is 500 0 0
different code though
it was in the zombies code
 
2:10 AM
well, to fix your current problem, you could change line 21 from z1=sol(j+n+1) to z1=sol(j+n), but that is likely to introduce other errors down the way.
the problem is that you have a vector with 10001 elements and you are trying to access the 10002nd element, which doesn't exist
 
oh damn it ... so I need less elements
 
no, you need to make sure you don't go off the end of your vector. Someplace you have an inconsistency in your code. In one place you treat the sol vector as though it should have 10001 elements, and someplace else as though it should have 10002 elements. The real fix is to find out where the problem stems from and fix it there
I suspect you have an n+1 where you should have an n someplace earlier
 
but I typed it exactly like how the person had it
 
I suspect you put an n+1 someplace there was supposed to be an n, or vice versa
 
let me check
but I know that n+1 is part of the soll code so editing above will bring issues
 
2:14 AM
I imagine the point of the exercise your working on is for you to develop some skills writing the code yourself, and debugging it when things go wrong, not necessarily just transcribing someone else's code
 
well my professor just told me to replicate the model and write the results in my own words.. he also told me to use the codes.. so pin that on him
 
take this as a learning opportunity. Go back and put in a bunch of print statements all over the place and check to see what size sol is and what elements you are trying to access
or better yet, replicate the model from scratch with your own code
 
-_- oh geez............ there are % on some parts of the code
 
I promise you'll learn a lot more if you do it yourself from scratch
 
I wanted to do my own project really... on cat population but he was like no choose Insert name of paper here... the information is already there because lots of times student choose their own paper and it gets complicated

THOSe are the exact words he said... really
:O!
 
2:22 AM
Sure, but I really doubt he wanted you to just transcribe and run their code. By trying to replicate their results you'll have something to check against.
 
OH MY!!!! I GOT IT!!! well I have to adjust my dt! BUT I GOT THE GRAPH! THere were typos in my code -_-!
well I'm getting my eyes checked tomorrow... probably need glasses
ok I just learned that my dt value on erad can't be high
 
I take fundamental issue with the model in that paper
They allow humans to kill zombies, which transitions them to R, but also allow the "dead" to rise, from R->Z.
now, I don't like the dead rising anyway, seems an old fashioned sort of zombie, but I particularly take issue with the fact that they don't distinguish between dead zombie and dead human
their dynamics has a single sink, namely Z
so their conclusion that zombies always win is a silly consequence of their model
for they literally don't allow any way for anything else to happen
 
the author also made some silly algebra mistakes too in the quarantine model but luckily I don't have to do that one
a - sign wasn't in the matrix
 
you should implement a stochastic discrete population version of a zombie model
it is much more fun
with a discrete population, you have a real chance of reducing the zombie population to zero
 
but I didn't know about your paper and my professor is doing probability type models next semester that includes stochastic stuff
but I don't have a stats background so I am just taking deterministic
model
 
2:36 AM
not everything you do has to be for homework. You could do it for fun yourself
 
oh XD
ok.. how come this guy has time from 0 to 10 but my time is from 0 to 100?
 
you chose your time to be 100
 
but I put my time to be 10, then I will get an erro
r
and why do I only have 3 instead of 4? :/
 
 
5 hours later…
8:09 AM
-just curious- why haven't we got the new profile page that's been rolling out over SE?
 
 
5 hours later…
12:55 PM
@JoshuaLin because they can only change sites so fast, and it takes time, and we weren't lucky enough to be one of the first ones. I don't know when we will get it, but it might be a while longer.
 
 
3 hours later…
3:50 PM
What does chat think about this discussion? http://meta.physics.stackexchange.com/questions/5903/
It seems to me that @DavidZ may be demonstrating somewhat of an uncharacteristically trollish and intolerant attitude. A+ for passion, but you gotta keep things respectful bro! ;)
 
4:12 PM
@Giffyguy I think David is completely correct (and not at all disrespectful) here.
 
Look at the date.
 
@Icosahedron The comments are two hours old. What about it?
 
...arguing over a post that is almost a year old?
 
@Icosahedron Well, if you want to have a policy reconsidered, that's what you do, isn't it?
At least it doesn't seem that strange to my hungover brain :P
 
@ACuriousMind Noted, and thanks. I appreciate the feedback.
 
4:19 PM
I can't tell if that's sarcasm or not.
 
4:35 PM
0
Q: Clarification regarding conceptual doubts

ADGAccording to one of the answers to my post: Physics.SE is not a homework help site, it is not meant to be a resource to teach people physics, and it is not a resource intended to help right conceptual misconceptions in students. When it was originally created, this site was meant to be a plac...

 
4:57 PM
Rather than disallowing non-mainstream theories, I think the better approach would be to have a neon warning/caution banner at the top of non-mainstream posts. That would appease both sides, by allowing people to express their views even if they are perceived as non-mainstream (people are also free to downvote), while simultaneously helping to mitigate any perceived damage to the community that could result from promoting non mainstream ideas. Perhaps the flagging system could accommodate this?
 
@Giffyguy We do not want to "appease" the non-mainstream theorists, and we do not want to give them a space to air their utterly wrong ideas.
2
 
@ACuriousMind basically physicsSE is gonna be ESscisyhp ? :P
 
@ACuriousMind Again, I appreciate the feedback. But your comments are somewhat insulting and what you describe is generally oppressive. I'm not aware of any other SE site that shares this mentality, for example if you go to SO and post a bunch of programming methodologies that go against mainstream best practices, proper discussion will ensue and people will explain why the ideas are wrong. I don't understand why Physics can't also support the same level of open forum.
 
one reason that i remember from posts here is the site will become a quasi-blog for crackpots or something along those lines.
 
@Giffyguy 99% of "non-mainstream" physics is nonsense. It's not like discussing alternative programming methodologies, it's like someone claiming all the computer scientists are wrong and that there is an algorithm that solves every problem in O(log log log n) or something.
Also, we are not a discussion forum, we are a site for questions which have definite answers
 
5:11 PM
@ACuriousMind I can understand your fears here, and I disagree. If someone made such an absurd claim in SO it would probably get closed/removed. However, on SO it's a case-by-case basis, with no blanket topic prohibitions. I also hear you regarding the Q/A focus, however the problem here is that most of the interesting/frontier physics questions don't have definite answers in the way programming questions do. Most of frontier physics is still theory, after all.
 
@Giffyguy Mainstream physics questions that don't have definite answers are usually closed as too broad, too. We're not intended to be a site where we discuss newly developed physics, we're mostly there to clarify firmly established physics.
 
@ACuriousMind Ah, OK. Out of curiosity, how large do you think the non-mainstream physics society is? Do you see it a vast group of people, threatening to outnumber or overwhelm the mainstream physics society? Or do you view it as a relatively small and non-threatening fringe movement? I'm interested to hear your opinion on this, since I don't really know how people view this in general.
 
@Giffyguy I'm not under the impression that there is an organized "non-mainstream" society, the people doing it seem to be mostly isolated and lacking formal training in physics. Well, some older physicists also tend to write some weird stuff, but I don't think they're organized either. Non-mainstream physics is fringe, but very annoying fringe.
 
@ACuriousMind I just saw your comment about how often you have to remove the mathematical-physics tag. That made my day. :)
 
 
5 hours later…
10:01 PM
37
Q: Trackbacks from SE to the arXiv.org?

QmechanicIs it possible to implement trackbacks from SE sites to the arXiv.org? Example: Here is a list of trackbacks to the arXiv preprint arXiv:0905.2658 from various blogs. My feature-request is that if a future SE post mentions an arXiv number, say, arXiv:0905.2658, then a trackback is automatically...

3
@dmckee : After half a year, I still haven't heard back from the SE team about arXiv trackbacks, so I put a bounty on the meta question.
 
 
1 hour later…
11:20 PM
tracebacks are a wonderful idea
there was some strange activity on the fuel efficient hill question
 

« first day (1633 days earlier)      last day (3286 days later) »