英语翻译A prescient quote from Tjalling Koopmans in the introduc

英语翻译
A prescient quote from Tjalling Koopmans in the introduction to [64] reads:\It has been found so far that,for any computation method which seems useful in relation to some set of data,another set of data can be constructed for which that method is obviously unsatisfactory." (This compares strikingly with the quote from Bixby et al.[13] at the end of this section.)
In [30],Dantzig writes:\Luckily the particular geometry used in my thesis was the one associated with the columns of the matrix instead of its rows.This column geometry gave me the insight which led me to believe that the simplex method would be an efficient solution technique.I earlier had rejected the method when I viewed it in the row geometry because running around the outside edges seemed so unpromising."
Since much has been written about the early history (and pre-history) of linear programming,for example in [29],Chapter 2,[30],and [83],pp.209{ 225,this paper will concentrate more on developments since the seventies.I hope to intrigue the reader enough to investigate some of the byways and alleys associated with linear programming as well as the more well-travelled highways.We will look at simplex,ellipsoid,and interior-point methods,and also at least mention some other approaches.Of course,I hope the reader will forgive my personal bias in the topics selected.(Let me mention here Megiddo's article [75],which also surveys some recent developments from a different viewpoint.)
Following the development of the simplex method in 1947 [27],the '50s had been the decade of developing the theoretical underpinnings of linear programming,of extending its applicability in industrial settings and to certain combinatorial problems,and of the first general-purpose codes.The '60s saw the emergence of large-scale linear programming,of exploitation of special structure (again pioneered by Dantzig and Dantzig-Wolfe in [28,31]),and of extensions to quadratic programming and linear complementarity.If the '50s and the '60s were the decades of unbridled enthusiasm,the '70s were the decade of doubt,as the theory of computational complexity was developed and Klee and Minty [60] showed that the simplex method with a common pivot rule was of exponential complexity.We will concentrate on the developments since that time; hope has been restored by new polynomial-time algorithms,by bounds on the expected number of pivot steps,and by amazing computational studies on problems with numbers of variables ranging up to the millions.
Linear programming studies the optimization of a linear function over a feasible set defined by linear inequalities,hence a polyhedron.The problem is in some sense trivial,since it is only necessary to examine a finite number of vertices (and possibly edges),but if one is interested in efficient computation,the topic is wonderfully rich and has been the subject of numerous surprising new insights.

一明引述tjalling库普曼斯在导言[ 64 ]内容如下：\它已发现到目前为止,对于任何的计算方法,这似乎有用,在有关的一些数据集,另一套数据可兴建该方法显然是不能令人满意.“ （这比较突出,与引述比克斯比等人.[ 13 ]在本节结尾） .
在[ 30 ] ,dantzig写道：\幸运的特定几何用在我的论文是一相关栏目矩阵,而非其行.此栏几何给了我的洞察力,而导致我相信,单纯形法将是一个有效的解决方法技术.我刚才已拒绝的方法,当我认为这是在该行几何,因为四处外界的优势,似乎使没出息“ .
由于大部分已经写入有关的早期历史（和前历史）线性规划,例如在[ 29 ] ,第2章,[ 30 ] ,[ 83 ] ,聚丙烯.209 （ 225 ,这份文件将集中更多的发展,自七十年代.我希望阴谋读者足够的调查部分的小道和胡同与线性规划以及更完善的旅游公路.我们将看看简单,椭球,和内部点的方法,也至少提及一些其他办法.当然,我希望读者会原谅我的个人偏见,在选定的专题.（让我在这里提到米吉多的文章[ 75 ] ,这也调查,最近的一些事态发展从一个不同的观点.）
以下的发展,单纯形法在1947年[ 27 ] ,五十年代已十年发展的理论基础的线性规划,扩大其适用于工业设置和某些组合的问题,和第一的一般用途守则.六十年代,看到出现的大型线性规划,被剥削的特殊结构（再次开创dantzig和dantzig -沃尔夫在[ 28,31 ] ） ,并扩展到二次规划和线性互补.如果五十年代和六十年代人几十年来肆无忌惮的积极性,七十年代人十年的疑问,作为理论的计算复杂性是发达国家和克利和minty [ 60 ]表明,单纯形法与一个共同的支点规则指数的复杂性.我们将专注于发展自那时起;希望已恢复新的多项式时间算法,由界对预期的数目枢轴的步骤,以及惊人的计算研究,对问题有多少变数,高达数百万.
线性规划的研究优化的线性函数,超过一个可行的设置所定义的线性不等式,因此多面体.问题是,在一定意义上微不足道的,因为它是不仅是必要的审查有限数目的顶点（和可能的优势） ,但如果一个是有兴趣在高效率的计算,题目是完美的丰富和一直受到众多令人惊讶的新的见解.