Some think about domestic science (calculation) software
This paragraph has been asking yourself: "How many domestic matter is in our common scientific computing software?", Search for a long time in memory, seems to be able to count "often" or "listening to the name", one is Called with FEPG's finite element source generator, these years have seen their ads in some forums from time to time; the other is CTEX set software, making scientific layups, I heard that people who have been involved in the Chinese Academy of Sciences participated in some work. In addition, I can't think of what else. Of course, it does not rule out the software self-produced software, but assertions we did not develop a number of scientific tools affecting technology development in Matlab / Mathematica / MAPLE / LAPACK.
The lack of scientific computing software has reached the lack of China's software development capabilities and the level of science and technology applications. Domestic demand for scientific computing software is very large, but worry is that almost the entire market has been occupied by foreign software. On the forum of finite yuan, the words must be called ANSYS, FLUENT, ABAQUS ..., circuit simulation forum, Bustling HFSS, ADS, MicroWAve Office, ..., and Chemical Discussion,
1. Pirated software is too easy to get. Many foreign scientific software, including some open source, free software is "forced" by commercial software, due to enterprises, research units need some scientific computing function, and commercial software's licensing cost is too high, leading to enterprises or research institutes It is self-written in software, and the software is commercialized or published in a free form to meet the needs of society. In China, a few yuan a pirated CD, thousands, tens of thousands of dollars of scientific software should have, coupled with scientific research, business integrity system has not fully established, pirated software has become the first choice for many people, everyone is not Do not use the "pirated" software as a shame, but the emperor uses them to publish articles, develop products (of course, I am in this ideological realm a few years ago). Even, some people claim that this is "brought to theism". It's not only, this is just a crown of excuses from the ineffective use of others. In addition to "Toism", Mr. Lu Xun still has another famous saying that the piracy is a classic, that is, "Theft is not to steal", think that due to the "learning", the pureness of the vocabulary itself can be "steal" this. A behavior of the intention of behavior. Even many people claim that China has used the best software in the world. But irony is that we have used the best software, and what we make things are best. What does this show? Using pirated software is like public medical care, you can get antibiotics for free. Once the public medical treatment is not, and the long-term antibiotics caused a serious decline in physical resistance.
2. People with strong software capabilities, scientific calculations, lack of background; people with strong scientific background, computer application skills are insufficient. In colleges and universities, the computer is good, generally go to the enterprise to write interface, database, network development, and then make money to IT to develop, make money, they don't care how to solve the partial micro-division, how to calculate the Gauss points in the singularity, how Seeking such scientific issues such as spindle matrices. Among the people who have a good professional, the computer is good, and people who can write high quality codes and procedures. There are fewer people, most people still stay in questioning how to set up computing accuracy in matlab (actually just display accuracy), After installing MATLAB, start there is a phase of MatlabServer.exe running such a class of low-level issues. They are afraid of having their own americ or algorithm, and very few people think about how to modrew code, how to release, what kind of license, how to manage code's version of such software development issues. In foreign countries, many scientific, engineering courses are required to be done with a computer, write code, write report, and speaking, after several years of training, graduated students' basic operations, simple use of technology software They have both basic concepts, and at least one or two computer languages, so that in the future research road, they will use these computer tools to develop new computing functions. And our domestic training is still not enough. 3. Scientific research innovation is not enough, no Ambition. Several students of Stanford can make Google from an IDEA to a world-famous global company, and a VT's master's degree can make a number of mathematical knowledge bases PlanetMath, Matlab is Cleve Moler in U. New Mexico teaching time design , Mathematica is Wolfram gradually developed in the study complex system, Lapack's author Anderson, Bai et al. Also developed on the basis of Linpack / Eispack, such an example. But why is this example in our colleges?
The status is not optimistic, we need to do something, or we call on the Ministry of Science and Technology to change it, we need to have mathematics and computer backgrounds to participate in the development of scientific software, improve the use of computer tools in scientific strength Ability to improve research efficiency, we must encourage all forms of scientific software development work (under the premise of satisfying software copyright and licenses), including commercial, free, open source, and call the academic forum, academic sites and other exchange channels Such a (synergistic) development provides convenience. Everyone's group of groups, discuss what scientific computing functions we need, formulate some development plans according to algorithms and procedures, you can be used as extracurricular hobbies, or to participate in this development, students, research units can also Provide economic support for such Project through a fund application.
From the beginning of the initial development, advocating open source or free software is intentional, because one is in the early stage of development, the software level is still unable to develop a large commercial software for many years in foreign countries, and some written software The code can provide basic and modules for the development of other software in the future, giving birth to a large number of software. However, we need to pay attention to cultivate copyright awareness from now on, choose the form of suitable licenses, so that scientific software can develop healthily under the protection of law.
Of course, it is just an initiative, my own mathematical level and computer programming level are generally general, and there is no regular programming training. But I still hope that these ideas above can give you some revelations, especially domestic college students and graduate students, you can write some things, or join us to complete one, two good scientific software development projects. I am busy, but I have always had an idea, just write a class MATLAB or class MATcom's C library. This library will add a Wrapper on the CLAPACK, providing computing function and matrix operation in matlab / matcom, so Using the powerful mathematical calculation function provided by Lapack, and the easy-to-use interface provided by Wrapper, making a foundation for future development of other computing software. In addition to this plan, there are several other programs on hand, including grid generation, typography, drawing, etc., although it is a small program, but I want to release it in the appropriate way, everyone can use it free, or Participate in the appropriate way to the extension and improvement of the software. I hope that more and more Made in China's markers in the science software will be seen in the future. In that, we have the strength of science and technology.
(Thick views, only for everyone to refute)
Write: fangq time:
11:02 pm |
Cope with stickers (3)
