S4 results publication
Anonymous
12 Apr 2008 12:54:44 UTC
Topic 13664
(moderation:
Quote:
Obviously the intended audience for this paper is the scientific community. I guess an abstract that is more comprehensible for non-physicists might be available later (like the one compiled for the S3 runs which is linked from the homepage).
Try reading the paper itself, rather than the abstract. The paper can be obtained by following the 'PDF' link on the top right of the arXiv web page (here is the link to the PDF file). The introduction and conclusion should be understandable by non-physicists.
Cheers,
Bruce
S4 results publication
)
No. This is the standard LIGO Scientific Collaboration author list plus a few notable additions (such as David Anderson from BOINC). When we start publishing results from data taken after May 17 2007, the author lists will get even longer since that was the formal start of data exchange with the VIRGO detector (in Cascina Italy).
Cheers,
Bruce
RE: I don't even want to
)
It's also complicated simply to work with such a large group, distributed over the globe. But it's necessary because the instruments are one-of-a-kind and are themselves very complex. So we have a lot of regular weekly telephone conferences and spend a lot of time on airplanes.
Getting back to the paper, yes the size of the collaboration also slows things down quite a bit. For example the first draft of this paper was ready in May 2007, but the process of internal review and correction plus formal review by the collaboration means that it has only now appeared.
The positive side of this is that when many scientists with different perspectives read the paper, they look closely at different facets of it and spot problems and mistakes that others have missed. So the paper that you are reading is significantly different and improved compared with the original May 2007 draft!
Cheers,
Bruce
PS: I can't resist the opportunity to end with a couple of related (and opposing) quotes from computer programming. I think that both statements below are correct!
Given enough eyeballs, all bugs are shallow. (Eric S Raymond)
The honest truth is that having a lot of people staring at the code does not find the really nasty bugs. The really nasty bugs are found by a couple of really smart people who just kill themselves. (Bill Joy)
RE: RE: Try reading the
)
If someone inside the project wants to prepare an equation-free presentation along the lines of the S3 final-result web pages, I would encourage it. But right now, most of us want to focus on finishing the analysis of the data from S5. This is more sensitive data than S4, and we have developed better and more sensitive analysis tools.
Cheers,
Bruce
RE: Those who got hooked by
)
That's absolutely correct. We are using a combination of the F-statistic method used in the current S4 paper, plus the Hough method described in the reference that you give. This combination is the 'state of the art' in terms of sensitivity.
Cheers,
Bruce
RE: RE: This combination
)
My two cents on this (actually I'm not a physicist):
If nothing is found by the Einstein@home analysis runs S5R1-S5R3 this would mean that something isn't as we expected or hoped. We would try to find out what it is and change the search depending on it, maybe just finding that what we thought wasn't the problem. Assumptions and thus reasons for not finding something include:
- the program is reasonably free of serious programming errors
- algorithm: e.g. the application of the Hough transform to "demodulated" ("F-statistic") data has never been actually tested or validated before.
- parameters: It might be that e.g. the frequency of continuous gravitational wave sources is a factor off from what we expect, in which case these gravitational waves could lie outside the spectrum where our earth-based detectors are sensitive.
- modeling: The algorithms used in Einstein@home are bound to very specific waveforms. It could even be that no sources that emit such gravitational waves exist at all.
Our current thinking is that the sensitivity of our search is not good enough for the sources we expect to detect (i.e. they exist but are too far away), so what we do is to increase the sensitivity of the detectors (which is currently done by upgrading them to "Advanced LIGO" and "VIRGO+") and data analysis (which we do by using the "hierarchical search" method).
Actually is as impossible to prove the non-existence of gravitational waves by this experiment as it is to prove the absence of bugs in a computer program just by testing it.
BM
BM
RE: RE: RE: This
)
That's why Einstein@Home is looking for UNKNOWN sources. It's very likely that there are a lot of rapidly spinning neutron stars that might emit gravitational waves but not be visible electromagnetically (as pulsars).
Bruce
RE: BTW, when will the
)
This work is underway. It is being led by Maria Alessandra Papa, one of the inventors of the hierarchical search method.
Cheers,
Bruce