![[Virtual Physics]](../images/vp.gif)
![]()
![]()
This paraphrase of a well known Winston Churchill's quote on a democracy is often used to back up the system of the (anonymous) peer review in science. Like other peer review defenders, Dr. Braginski is willing to admit that the peer review is not perfect, but because all alternatives are even worse we have to live with it for the sake of our own good.
For me, however, this argument far from obvious. Churchelleana does not apply always, and the alternative (peerless science) can indeed be a better option.
Being in science, like Dr. Braginski, for some 30 years I simply do not see a compelling reason to buy an argument that the "science is better with peer review than without". On a balance, my judgement is the opposite, namely that the overall effect of the peer review in science was and is of making a lot more harm than good.
While I do agree on a desirability of some editorial process for the research manuscripts, the prime purpose of it should be a clarity of presentation, avoidance of excessive length and detection of clear-cut rubbish and incompetence. (I would not even dispute strongly an issue of the peer review anonymity, PROVIDED it is used for the constructive improvements, but WITHOUT the power of the rejection).
Likewise, in allocating the research funding, the only thing peer review can do more-a-less reliably is to assess the overall competence of the researcher and his/her recent research activity. Only in cases when there is a clear deficiency on either of these points, denial of ANY operating funding is warranted. This is certainly not the practice of the Canadian NSERC which denies operating grants to one third of all professors of science and engineering at Canadian universities. To imply (as NSERC does) that 1/3 of all science-engineering workforce is incompetent and/or a dead wood is a sheer nonsense and a gross insult to the entire research community.
Another point which bothers me in Dr. Braginski's position is his inference that:
I again beg to disagree. The administrative system (in this case - the grant allocation system) should not and must not operate on a basis of some nebulous ethical principles. To request "ethicity" (and who defines it ?) from the ANONYMOUS (sic - !) system is both a contradiction of terms and a practical impossibility. Who is going to guard the ethicity of the process ? Another "anonymous" panel ?
So, my overall conclusions remain the same:
| The non-linear microwave surface impedance, Zs=Rs+iXs, of patterned YBCO thin films, was measured using a suspended line resonator in the presence of a perpendicular DC magnetic field, HDC, of magnitude comparable to that of the microwave field, Hrf. Signature of the virgin state was found to be absent even for relatively low microwave power levels. The microwave loss was initially found to decrease for small applied HDC before increasing again. Also, non-linearities inherent in the sample were found to be substantially suppressed at low powers at these applied fields. These two features together can lead to significant improvement in device performance. |
The center frequency and the -3 dB bandwidth obtained from the fit agreed very well with those directly read off the trace, especially at low power levels where the trace is closest to Lorentzian shape, but provided significantly enhanced sensitivity to small changes.
|
Figure 1. Block diagram of Experimental Setup |
|
Figure 2. Low field hysteresis at -21 dBm and -11 dBm of input power, at 10K. Notice the absence of the "tail" corresponding to virgin response at higher powers. In order to highlight the similarities between the two plots, they have been superposed on each other by adding a constant of 10 kHz off the -21 dBm power plot. |
|
Figure 3. Typical power dependence of resonance widths, taken at 10K |
IBM's current proposal is a faster, more compact binary version of VRML, where the ASCII text files will be replaced by binary ones. SGI is proposing an external interface that enables an external Java applet to communicate with a VRML world.
VRML allows a user to explore a model by walking or flying around and inside a model, just as one would if they were exploring a real object. Predefined viewpoints can be created using the PerspectiveCamera and Orientation nodes. To enhance to the reality, VRML supports an array of different light sources and colors such as the Material node that includes diffuse, emissive, ambient, headlight and specular color values and positions. Applying texture maps to object surfaces makes it possible to create detail rich worlds. In addition, VRML supports flat and smooth shading techniques. However, shadowing is not implemented at this time as the rendering would be too expensive (performance wise).
It is always beneficial to optimize a model as to decrease it's download and render times as this may allow one to increase the detail of the virtual world. There are several ways to do so. The easiest way is to remove the white space and round off any numbers in the text file. Furthermore, one can re-use the same parts several times in different conditions by using the WWWInline or DEF and USE nodes. The ShapeHints node allows the backside of a solid object to be hidden or not drawn, as it will never be seen anyway and the removal of the object's internal polygons can greatly increase the efficiency of the entire model. The LOC node allows one to specify the amount of detail depending how far an object is away from the camera. Finally, the final models can be gzipped that reduce the file size by 90%.
Recently, version 2.0 of VRML was released that allows 3D animations. In other words, one can make their models, or parts of models move around in the virtual world in addition to the capability of navigating through the world yourself.
The most significant WWW site dealing with VRML is the VRML Repository and I will refer to that site several times throughout this text. It includes all the specifications of the language, almost all software related to VRML and anything else that may be of interest to a VRML programmer or surfer.
Once a VRML browser is installed, a person needs to get accustomed to it. It is quite difficult at first to navigate through a 3-dimensional space using 2-dimensional navigation tools such as the keyboard and mouse. It is a good idea to read the navigational directions that come with the browser and to try to use the pre-defined viewpoints in the model if they are available.
It is important to remember that, as with HTML pages, different viewers render and project a model slightly differently. The same model may look different and in some cases terrible in one browser and spectacular in another. Although there is standard for the language itself, there is no standard for the rendering or interpretation algorithms. Also, some browsers may support VRML extensions that are not supported in another browser, making the model significantly different.
VRML 2.0 is very recent and most resources are still being developed. In fact, you will require a separate VRML 2.0 viewer (different from VRML 1.0) to view these worlds. Pioneer Joel's VRML 2.0 Tutorial provides an excellent introduction to VRML 2.0 and the VRML 2.0 specifications can supplement any other area of interest that one might have.
One may prefer to develop complex models using applications such as Geometry Modelers or Geometry Generators. However, it may be required to convert the models to VRML using Geometry translators. Keep in mind, when using translators, that quality may decrease and that data loss is possible.
Below is a table that lists the URL and snapshot of the VRML world and key points of interest about the model. Version 1 of VRML has been used in each case. Click on the URL to enter the 3D world.
| Snapshot | URL | Comments |
|---|---|---|
| NCSA Relativity Group VRML Page Spacetime Diagram for the Collision of 2 Black Holes http://jean-luc.ncsa.uiuc.edu/Viz/VRML/POPs.wrl | simple lighting and smooth shading effects |
| Neural Signal Processing Group's Human Brain Project The Brain - image from a Magnetic Resonance Scanner http://hendrix.ei.dtu.dk/vrml/mriHeadD6.wrl.gz | amount of surface detail |
| The Naval Research Laboratory http://overlord.nrl.navy.mil/vrml/nrl.wrl | use of lighting techniques, pre-defined viewpoints, object links |
| Image Library of Biological Macromolecules DNA / Protein Complex http://www.imb-jena.de/vrml/DNA/DNA_ drug_complexes/109d/109d_insight_1.wrl.gz | detail and precision |
| The Collider Detector at Fermilab http://www.ocnus.com/models/CDF/detector.wrl | use of WWWInline node and lighting |
| Virtual Modeling Language in Chemistry Nitrosamine Molecule http://www.pc.chemie.th-darmstadt.de/vrml/ models/bns/nitro.wrl | lighting and shading |
| THe LaHave House Project http://www.tuns.ca/~gajewski/vrml/ model4/model4b.wrl | background color, pre-defined viewpoints, the use of DEF and USE nodes, flat shading, transparency |
Other examples can be found at: http://www.isisnet.com/MAX/vrml/struct.html, http://www.tuns.ca/~gajewski/vrml/, http://www.eit.com/www.lists/www-vrml.1995q3/0347.html, http://3dsite.com/cgi/VRML-index.html.
VRML can be applied to illustrate anything from simple to complex models or concepts. The applications of VRML are limitless, what can be created depends on ones needs and ingenuity.
Editors:
Virtual Physics URL addresses:
Université de Liège, Sart Tilman, B-4000 Liège, Belgium, tel. (+32 41) 66 37 52
Umeå University, S-907 42 Umeå, Sweden, tel. +46-(0)90-167717
Swinburne University of Technology, P.O. Box 218 Hawthorn, Victoria, 3122 Australia, tel. +613 9214 8935, fax +613 9819 0834
2-6032 Compton Ave., Halifax, Nova Scotia, B3H 1E7 Canada, tel. (902) 423 2149
Warsaw University, Hoza 69, 00-681 Warsaw, Poland, tel. (+48 2) 628 3031
University of Pennsylvania, Rm. C-501 Richards Bldg., Philadelphia, PA 19104-6089, U.S.A., tel. (215) 898-6396
To subscribe a F R E E e-mail
version or submit materials for publication, write to Zbigniew Koziol.
Copyright (C) 1996 by Zbigniew Koziol.
this copyright notice concerns the whole of the Virtual Physics
edition but not specific articles published there which are
property of their respective copyright owners![]()
![[ WebExperts ]](../images/we5.gif)