本部分目录

 

五、怎样使论文投稿适得其所（案例分析）http://blog.lehu.shu.edu.cn/sqdai/A51226.html

六、国刊？洋刊？ http://blog.lehu.shu.edu.cn/sqdai/A128066.html

七、你写论文投稿前有“冷冻期”吗？ http://blog.lehu.shu.edu.cn/sqdai/A62950.html

 

五、怎样使论文投稿适得其所（案例分析）

 

在已发的拙文中，仅概述了投稿的注意事项。这里想提供一个案例：本课题组2004级硕士生郁文剑在投稿中给编辑部的回音和与审稿人的商榷，供大家参考。

 

【链接】一个案例——关于一篇本组的PRE论文给编辑部的回复

 

【情况说明】

本课题组2004级硕士生郁文剑进入本课题组后，半年之内就完成一篇关于行人流的论文，他在论文中建立了行人流的离心力模型，这是一个全新的模型，能再现行人流中的自组织现象和非线性动力学过程。此文思路新颖，结果不错，经过我和董力耘博士多次修改后定稿，2005年1月19日投稿到《物理学评论E》（Phys. Rev. E）。一个多月后收到两位审稿人的评议书，总共提出了十条意见，经郁文剑和我们共同作者反复斟酌和修改，2005年4月14日将论文修改定稿。并经反复商议，写成了如下回复。此后，稿件很快被接受，2005年8月5日刊登于Phys. Rev. E 72 (2005) 026112。郁文剑目前在苏黎世技术大学攻博，师从著名交通科学家D.Helbing教授。

读下文时请注意：我们如何回复编辑，如何答复和感谢审稿人，如何与审稿人争辩（尤其是对第一审稿人的第三条意见）。

 

Reply of Authors on PRE EA10106

 

From: wenjianyu83@hotmail.com

To: Physical Review E < pre@ridge.aps.org >

 

Re: EA10106

Centrifugal force model for pedestrian dynamics

by W.J. Yu, R. Chen, L.Y. Dong, S.Q. Dai

 

Ms Margaret Malloy                                      April 14, 2005

Associate Editor

Physical Review E

 

Dear Ms Margaret Malloy,

 

Thank you for your message of March 2.

I have completed the revision of our manuscript (EA10106). Attached please find the revised manuscript.

In order to meet the requirements of its publication in PRE, the co-authors and I have meticulously considered the opinions and suggestions of referees and you, conducted some further simulations, and rewritten some paragraphs. The main revision is as follows.

1)                   Canceling the description of “tolerable distance” and “aggressive action” for their ambiguous definitions. The former is replaced by generally accepted concept on comfortable distance, and the latter by aggressive pedestrians, which has nothing to do with academic concepts.

2)                   Adding a new paper about the magnetic force model in references (i.e., Ref.[15] in the revised manuscript) to illustrate some physical treatment could indeed be used to describe pedestrian movement.

3)                   Canceling the original Figure 1 which has less meaning.

4)                   Citing a famous monograph on applied mathematics (i.e., Ref.[17] in the revised manuscript) to demonstrate that dimension analysis could be an auxiliary means in deriving a novel governing equation.

5)                   Adding a new figure (Fig.2 in the revised manuscript) for describing the destination and moving directions of people in a room.

6)                   Rewriting the paragraph about collision detection technique to make it clearer in description. Supplementing some results of new simulations.

7)                   Adding some new results about arching and clogging, in particular, the situations at higher occupancies and the relation between jamming probability and exit width for different occupancies. Making preliminary analysis of arching mechanism.

8)                   Changing the occupancy in arching simulation to a larger one.

9)                   Giving four different relations between the average leaving time and exit width in Fig.7 of the revised manuscript for four cases of different occupancies, which leads to a more accurate relation (i.e., Eq.(18) in the revised manuscript).

10)               Polishing the original manuscript to our best.

Attached please also find our reply to two referees, in which we answered their criticisms and comments item by item.

We are grateful to referees and you for your kind efforts in improving our manuscript.

 

With my best regards,

 

 

Sincerely yours,

 

Wenjian Yu

 

Appendix

--------------------------------------------------------------------------------

Reply to the First Referee -- EA10106/Yu

（With brown characters as the original comments and suggestions）

----------------------------------------------------------------------------------

My first comment is about the concept of "tolerable distance" and "aggressive action" (section II.A). How does the model of the authors and the obtained results differ from the case when the noise amplitudes are increased in the "social force model" (Refs. 12-14) whenever the pedestrian-pedestrian (or pedestrian-obstacle) distance falls below a certain threshold?

{Reply}Thank the referee for pointing out the problems. The description of “tolerable distance” and “aggressive action” in the original manuscript is indeed a bit ambiguous, so we canceled them in our revised manuscript. The former is replaced by generally accepted concept on “comfortable distance”, and the latter by aggressive pedestrians, which has nothing to do with academic concepts. We obtained some results similar to those in Ref.[12-14], such as lane formation, arching and clogging at exit of a room, etc., with a different model, in order to demonstrate the effectiveness of the new model. But we do give some new results (see below).

 

The social force model in e.g., Ref.13 contains both radial and friction forces, so the statement at Eq.1 [Here we consider the repulsive effects to contain both velocity and distance...] repeats previous work without explicitly saying that it does.

{Reply} We got lots of inspiration from the work on the social force model, and gave some description about it. In fact, we presented a new form of repulsive force, analogue to the centrifugal force, but have not considered the friction force as in Ref.[13]. Therefore, our work is not a simple repetition of previous work. Furthermore, we have given a relation between the average leaving time from a room and the door width, which did not appear in Ref.[12-14]. In the revised manuscript, we also give the relation between the jamming probability and the exit width, which is entirely new.

 

The "centrifugal force" is derived from Eq.6. where Buckingham's pi theorem is applied. "Buckingham's pi theorem states that physical laws are independent of the form of the units. Therefore, acceptable laws of physics are homogeneous in all dimensions." [copied from Eric Weisstein's World of Physics] At the heart of the "centrifugal force" model (and also the "social force model") are psychological effects. Psychological effects integrated into second order coupled differential equations will not give a physical law. These equations of motion might work well in describing collective pedestrian motion for other reasons, but still, this is not sufficient for them to be a physical law.

{Reply} We agree to referee’s opinion in the sense that Buckingham’s pi theorem could not be used to analytically prove the correctness of a governing equation. But we claim that the theorem and the dimension analysis can be employed to provide an auxiliary explanation of the derivation. The Buckingham theory also tells us that in physical problems, there is a relation between the dimensionless groups existing in the problems. We just used this statement. For further verification, we found an example in “Mathematics Applied to Deterministic Problems in the Natural Sciences” written by Lin and Segal (see Ref.[17] in the revised manuscript).

And for the above second comment, we agree with the referee at the point that pedestrians’ movement is related to psychological effects. But their movement is a kind of mechanical motion and could be described by some differential equations based on physical laws. The key point is whether the law is properly applied and the related results are reasonable or not. What we obtained implies that the model made sense to some extent.

 

I have found no novel suggestions/extensions to the model in section II. In section III，known (already published) simulations are repeated with minor modifications. This would probably be interesting for engineering applications with groups of up to 10,000-100,000 people. The relevance of the only quantitative result, Eq.18, is not explained, nor why it is in good agreement with a CA lattice result. Note that the model used by the authors is continuous in space, why CA models are discrete in space. These two groups of models can strongly differ at high densities.

{Reply} As is mentioned above, although our work is preliminary, we have made some contribution to the topic, for example, giving an empirical relation between the average leaving time and exit width through simulation. And we accepted the referee’s suggestion, and in the revised manuscript we have presented the analysis for denser pedestrian flow.

    It is not strange that different models could give similar or even same results provided that these models give a correct description of phenomena. Although the CA traffic model considerably differs from our model, they can lead to qualitatively identical results. By the way, Prof. B.H.Wang presented a similar relation between the average leaving time and exit width with the CA model for the occupancies 0.3, while we gave the result for four different occupancies (i.e., 0.1,0.2,0.4,0.8). To our surprise, the related results are similar.

 

----------------------------------------------------------------------------------

Reply to the Second Referee -- EA10106/Yu

（With brown characters as the original comments and suggestions）

----------------------------------------------------------------------------------

 

* The "tolerable distance"(I refer this by L) is not clearly defined in this model. Is L involved in this model in eq.(17)? I ask this question because the last sentence in page 5 puzzles me: "...the distance between pedestrians are NEAR the tolerable distance". This seems that the authors do not consider the interacting force given by eq.(7) OUTSIDE the tolerable circle with radius L. That is, they have introduced the cut-off length of the force in this model. If this is not the case, then the 1/R decreasing interaction between people may be too strong. Pedestrian do not care about the people more than about 20m in the normal situation, which is roughly the radius of the "personal space" of a man.

{Reply} Since the definition of the “tolerable distance” is not so clear, we canceled the description of it. We think that the above “personal space” is too large, and in real simulation we cut off the repulsive interaction between pedestrians as the distance between two pedestrians is larger than 5D, where D is the “radius” of a pedestrian.

 

* Page 8

In collision detection technique, the description is quite ambiguous and hard to understand. Since the model is continuous, there is no lattice in space. Thus, "He/She may turn left or right randomly" is meaningless.  The authors should describe the details on the collision avoidance rule in this model. From fig.3, it seems that the restriction on the turning angles is set to be discrete such as the multiplication of 30 degrees. If so, they should write it explicitly in the text.

{Reply} The referee is quite right. We rewrite the paragraph about the collision detection technique and give a clearer description of it. According to the referee’s suggestion, we have explicitly written the discrete choice of the angle.

 

* The time interval is fixed to 0.1s, but they should check whether all the results are not affected by the choice of this interval.

{Reply} According to the referee’s suggestion, we chose different time intervals, ranging from 0.05 to 0.15, and even let the time step vary randomly in the range, but we found that there is no appreciable difference of the results in these situations. We also took a shorter (less than 0.05) time step, but the amount of computation increases a lot with no appreciable improvement of the numerical results.

 

* In the section 2-C, all the results are only displayed qualitatively with figures. The extensive simulations by varying many parameters in the model are needed and quantitative results should be reported. Eq.(18) is an only quantitative result, but there is no argument whether this relation holds in the other parameter sets, or validity of this formula.

{Reply} We have chosen different initial occupancies, selecting them as 0.1, 0.2, 0.4 and 0.8, instead of only one value of 0.1, and found a similar law for the average leaving time.

 

* Comment on the computational efficiency of this model. It seems it takes quite long time if the pedestrian density becomes high due to the collision.

{Reply} The computation time is acceptable. For the case of highest occupancy (0.8), the CPU time is about 10 min.

 

* There may be some typos, e.g., p.6, line 2, version -> vision.

{Reply} We re-polished the manuscript and corrected the typos.

 

写于2009年1月3日晚

http://blog.lehu.shu.edu.cn/sqdai/A51226.html

 

 

 

六、国刊？洋刊？

 

七八十年前，在中国的商品市场中，国货与洋货展开了激烈的争夺战。如今在国内外的科技期刊上，一场“国刊”与“洋刊”的稿源争夺战正在悄然进行。两场争夺战的特点很不相同：前者是洋货大举入侵造成的，我方被动；而后者，“洋刊”并未“大举入侵”，多少是由国内一些人士的看法和和主管部门的政策造成的。曾几何时，“以SCI论英雄”之风盛行，国内的优秀科技论文纷纷涌向国外的SCI和EI的源期刊，因为国内这种源期刊实在太少。产生的后果有两重性：正面的是，我国科技工作者在“洋刊”上发表的论文数剧增，扩大了我国科研成果的国际影响；负面的是，国内的大部分非SCI和EI源期刊的稿源大幅度萎缩，办刊难以为继，质量难以保证，且我们的科研成果的国内影响力有所下降，因为国内有很大一部分工程技术人员的外语水平不高。当然，科技无国界，对于科研成果在哪里发表，就科技发展来说，都是一样的；国内科技人员往“洋刊”投稿，本也无可厚非。十个月以前的博文中，从现实的寻求发展的角度看，我也认为应尽可能往国外投稿，只是近半年来，我的思想悄悄发生了一些变化。

产生这种变化的原因是：我是国内多种期刊的编委，并忝任《上海大学学报（英文版）》主编和《水动力学研究与进展》（中英文版）编委会副主任，与国内期刊结下了不解之缘，目前看到的一些现象令我喜忧参半。随着2007年《上海大学学报（英文版）》从EI Page One退出，随着《水动力学研究与进展》（中英文版）先后为EI和SCI收录，两种刊物的命运突变，前者稿源紧张，后者则日见兴旺，来稿应接不暇，淘汰率在65％以上。但是，国内期刊中有《水动力学研究与进展》（中英文版）这样的命运的刊物寥寥无几。因为仅2006年，原来已被EI收录的中国期刊被淘汰出局的竟超过300家！它们大多遭到与《上海大学学报（英文版）》类似的命运。中国SCI源期刊的情况也好不了多少，本来数目就不大，有不少杂志在SCI数据库中的地位也岌岌可危。

为何如此？昨天EI中国区总代表钟似璇先生在我校的报告揭开了谜底。却原来，1992年EI打入中国市场时，中国的EI源期刊仅10种，经过14年开拓，在两相情愿的情况下，我国的EI源期刊几近500种。后来EI总部经过市场调查，发现中国作者被EI收录的科技论文大多数发表在国外期刊，经质量评估，发表在国外EI期刊的论文远优于国内EI期刊，因此，为了降低运营成本，对已定位EI源期刊的中国期刊进行了大砍大杀，仅2006年就淘汰了300种，约为原有中国EI源期刊的60％！由此可见，国外的老板绝非傻瓜，他们总是以获得最大的盈利为目的。

这一情况值得我们深刻反思。

记得我在中科院力学所时（1962～1984），正值闭关锁国和国门刚开之时，我和我的同事们写成论文之后，没有向国外刊物投稿的概念。大家以在《力学学报》、《力学进展》、《中国科学》等一流国刊发表论文为莫大的荣光。直至今天，我的那些老朋友们还或多或少保留着这种观点。他们认为，以解决现阶段科技问题为目标的论文，发表在国内刊物上发挥的现实作用会更大。

随着西风东渐，科技论文的稿源不可阻挡地滚滚西流，出现了上面所述的局面，其中，有些人士和政策起了推波助澜的作用。这种局面亦喜亦忧，究竟如何看待其中的利弊大可商榷、讨论。

近来，已有一些有识之士公开呼吁重视国内期刊，例如北大前校长许智宏院士就认为，应该多把科技论文投到国内期刊，不能以为国外期刊的论文就一定比国内期刊的论文好。我在昨天的报告会上，就向与会者转达了这一看法，鼓励大家向国内刊物投稿。并且预测，若干年后，说不定从“西风东渐”变成了“东风西渐”，通过长期艰苦努力，国力大增，科技期刊办得越来越优秀，非但国内的稿件不再大量外流，国外的优秀论文会源源不断涌向我国的期刊！没想到，我这番寻常的话，居然博得了满堂掌声！

这不过是一时无法实现的对未来的畅想曲。我们还得来点实际的。我有如下想法：

1.        根据科学无国界的原则，继续鼓励自由投稿，优秀学术论文无论在那里发表，都是对世界科技发展的推动；

2.        逐步改变论文驱动科研的现状，大力鼓励做解决重大问题的科研工作，即鼓励问题驱动；

3.        彻底改变“以SCI论英雄”的观点和考核制度，以是否真正解决问题（特别是我国社会经济发展种的问题）为考评科技论文的标准；

4.        对国内外期刊一视同仁，对中文外文论文不作形式的区分，以其质量为唯一标准；

5.        鼓励作者以中英文版本同时向国内外投稿，计算成果数量时，内容相同、文种不同者，按一篇计。这样可解决因国内科技工作者外文水平有限，影响科技成果在国内的推广和应用的问题，同时可解决国内期刊的“巧妇难为无米之炊”的问题。而这里的“一稿两投问题”可适当“松绑”，但这仍是难点：政策一放松，难免“泥沙俱下”。

 

总而言之，国刊、洋刊，都应很好地为我所用，当前的国刊发展艰难的问题值得重视，应予解决。

 

以上想法不一定对，欢迎指正！

 

写于2009年10月21日

http://blog.lehu.shu.edu.cn/sqdai/A128066.html

 

七、你写论文投稿前有“冷冻期”吗？

 

这本来是个不成问题的问题。稍有科研经验和投稿阅历的人都明白，一篇学术论文初步完成之后，最好将其搁置一段时间，少则两周，多则半年（可称为论文投稿前的“冷冻期”），时不时地拿出来看看，问问自己：前提或假设是否合理可靠？对前人工作的综述是否已足够充分？理论分析（或实验过程）是否科学精准？演绎和计算是否正确无误？推理分析是否全面精当？论文结果是否合乎实际？所下断言是否客观真实？所得结论是否可进一步引伸？是否需要做更多的计算或实验？主要参考文献是否包罗无遗？论文表述是否恰当有效？遣词用句是否恰如其分？通篇行文是否简洁流畅？等等，等等，总而言之，自己必须弄清楚：整篇论文是否基本上无懈可击？（当然不可能达到尽善尽美的境界）。投稿之前，至少要修改两三遍，乃至更多遍。我从自己的研究生阶段开始，遵循导师的教诲，一直就是这么做的。

然而，随着社会氛围变得日益浮躁，急功近利之风盛行，情况在起变化。如我在2009-2-11的博文《学术论文功能的异化和改辙》中所指出的，论文变成了提职、报奖、通过论文答辩等等的“敲门砖”。不少人一写完论文稿就急吼吼地投稿，甚至对文中的错字、病句都不屑检查改正一遍。近年来，我每年要评审上百篇论文稿，这样的例子实在见到得太多了（因为审稿审得多了，我很快就能断定一篇论文是“急就章”，还是“慢工出的细活”）！这种状况令人担忧，说得严重一点，这样不严谨的学风如果代代相传下去，“四化”伟业如何完成？！

还好，有令人欣慰的实例。前一阵，我与田振夫博士讨论了他的论文修改事宜。老田曾在本课题组攻读在职博士学位，目前在复旦大学力学与工程科学系担任教授，他一向以学风严谨、论文优质低产闻名（年均1～2篇，篇篇“乓乓响”，他和本课题组张鹏教授，曾打破本校在计算数学和计算流体力学著名刊物JCP上十年无论文发表的记录）。我在修改了他的投稿JCP的新作后向他发表评论：内容不错，保持了他的思路缜密、演绎严格的风格，但表述上没有过去出色。他说：“戴老师，老实对你说，我过去在论文投稿前至少有三个月的‘冷冻期’，到复旦后，教学工作繁忙，这次投稿的冷冻期不够长，雕琢得不够，我一定下不为例！”这段推心置腹的话使我感动！时下究竟还有一批中青年学者不受外部环境影响，做学问仍坚持求真务实、稳扎稳打；而且只要引导得法，这样的学人会越来越多！

有人会问，当今学术界竞争如此激烈，论文投稿还要有冷冻期，就不怕成果发布被人抢先了吗？我认为，如果为了在竞争中抢先，发表有问题甚或有错误的论文，是得不偿失的，因为发表出去的论文如同“泼出去的水”，水能养人，也会害人害己（假若它不干不净）！一旦出了好论文（“垃圾”论文不再讨论之列），不妨采取其它一些谋略，例如，在确定自己的研究成果基本无误的前提下，将论文的详细摘要投到一些期刊的快报栏，先“立此存照”，争得学术成果的优先发明权。在我主编的“Journal of Shanghai University”中，就设有Letters专栏，校内外作者可把研究结果写成1～2页的详细摘要投给我们，其中只讲假设、方法和结论。这样的“快报”不经审稿，直接由杂志的主编、副主编定稿，2～3月内即可发表，而且完全不影响作者将详细的论文往别处投稿（按惯例不算一稿多投）。这样，你就可以放心地拥有自己投稿的“冷冻期”了。我把这样的过程称为“预注册”（preregistration）。

学界的朋友们，你们同意我的见解吗？

 

写于2009年3月22日

http://blog.lehu.shu.edu.cn/sqdai/A62950.html

转载本文请联系原作者获取授权，同时请注明本文来自戴世强科学网博客。
链接地址：https://blog.sciencenet.cn/blog-330732-352559.html

上一篇：论文投稿十日谈（上）
下一篇：用自己的劳动说话——本课题组的“科学研究中的学术规范实施细则”