A Crisis in Communication
A Crisis in Communication
复制标题
沟通危机
DOI：
10.4322/cto.2014.088
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
J. Butcher
中科院分区：
文献类型：
--
作者：
J. Butcher
1963 年，在担任《柳叶刀》编辑 20 年的任期结束时，西奥多·“罗比”·福克斯在伦敦卫生与热带医学学院发表了一系列讲座，共三场。两年后，这些讲座的记录被出版为一本书，题为《传播危机：医学期刊的功能和未来》。福克斯在书中做出了一个严峻的预测：“总有一天，期刊将被取代，成为发表新研究的一种手段。” （福克斯，1965）。随着该书出版五十周年的临近，值得重新评估福克斯的预测。 50 年后期刊会被取代吗？乍一看这似乎不太可能。第一份科学期刊《哲学汇刊》于 1665 年出版。三个世纪后，当福克斯撰写《传播危机》时，已有 6,000 种医学期刊。现在，最大的学术成果索引之一 Scopus 的数据库中包含约 21,000 种期刊：它索引了 1823 年至 1996 年间发表的 2,100 万篇文章，以及过去 20 年发表的 3,300 万条记录。事实上，根据最近的分析，科学产出每年持续增长约 8-9%（BORNMANN；MUTZ，2014）。文章数量的增加往往是以牺牲质量为代价的，许多论文从未被引用。据一项估计，多达 12% 的临床科学论文和 27% 的生命科学论文在发表后 5 年内未被引用（REMLER，2014）。其驱动因素很明显——学者们感到有压力，必须发表尽可能多的论文，以获得终身教职并获得新的研究经费。此外，大量论文的发表还以其他方式给学术界带来了压力：读者难以跟上文献的步伐；许多期刊发现越来越难以获得合格的同行评审员的服务来判断论文是否适合发表。科学研究的过程也在发生变化。不久前，单作者论文被认为对职业发展至关重要，因为它们展示了独立研究的能力。最近，大规模合作已成为常态，数百甚至数千名研究人员合作开展昂贵的国际科学研究。计算机科学家吉姆·格雷（Jim Gray）发现了科学方法的这种转变，并将其称为“第四范式”。用他的话来说（GRAY，2007，第 xix 页）：科学世界已经改变，这一点毫无疑问。新模型是通过仪器捕获数据或通过模拟生成数据，然后由软件处理，并将所得信息或知识存储在计算机中。科学家们只能在这个过程中相当晚的时候才能查看他们的数据。这种数据密集型科学的技巧和技术是如此不同，以至于有必要将数据密集型科学与计算科学区分开来，作为科学探索的新的第四范式。即使是小规模的研究也常常会产生大量数据，而使这些数据可发现和可重复使用正成为科学界日益重要的优先事项。开放科学运动的兴起，原始数据和最终出版物都可以轻松访问和重复使用，这为 Figshare 和 Dryad 等科学数据托管公司以及 PLoS One 和 Scientific Reports 等开放获取的“大型期刊”提供了机会。然而，开放获取期刊通过向作者收取发表论文的“文章处理费”（APC）来产生收入，但由于让作者更容易发表其作品，因此有可能使问题变得更糟，而不是更好。福克斯在 50 年前就认为期刊出版处于危机之中，但现在的传播危机可以说比当时严重得多（FOX，1965）。资助者和学术机构需要通过两种方式应对当前的危机。首先，他们需要根据科学家发表的成果的质量而不是数量来奖励他们；人们普遍认为，期刊影响因子等质量替代指标并不是判断单个科学家研究成果的不完美方法，因此迫切需要新的衡量标准。其次，资助者和机构应该鼓励他们的研究人员在存储库中发表他们的大部分工作。这些标准化报告不一定由人类阅读，而是以机器可读的形式保存数据和数据描述符，以允许计算机进行准确的索引和元分析。早期迹象表明这种方法是可行的，但改变 350 年的学术文化需要时间。究竟要持续多久还有待观察。毕竟，正如尼尔斯·玻尔（Niels Bohr）雄辩地指出的那样：“预测非常困难——尤其是未来”（MENCHER，1971，第37页）。 - 本文中提到的 Figshare 和 Scientific Reports 是作者雇主麦克米伦科学与教育公司的一部分。这里表达的观点是个人的，并不一定反映麦克米伦科学与教育的观点。
In 1963, at the end of his 20-year tenure as Editor of The Lancet, Theodore “Robbie” Fox gave a series of three lectures at the London School of Hygiene and Tropical Medicine. The transcripts of those lectures were published as a book entitled Crisis in Communication: The Functions and Future of Medical Journals two years later. In the book, Fox makes a stark prediction: “A day will come when journals will be superseded as a means of publishing new research.” (FOX, 1965). As the fiftieth anniversary of the book’s publication draws near, it is worth reassessing Fox’s prediction. Will journals be superseded 50 years from now? At first glance that seems unlikely. The first scientific journal, Philosophical Transactions, was published in 1665. Three centuries later, when Fox wrote Crisis in Communication, there were 6,000 medical journals in existence. Now, Scopus, one of the largest indexes of scholarly output, includes around 21,000 journals in its database: it has indexed 21 million articles published between 1823 and 1996, and a further 33 million records published within the last two decades. Indeed, according to a recent analysis, scientific output continues to increase by around 8-9% per year (BORNMANN; MUTZ, 2014). This increase in the number of articles is often at the expense of quality, with many papers never being cited. According to one estimate, as many as 12% of papers in the clinical sciences, and 27% of papers in the life sciences, are not cited within 5 years of publication (REMLER, 2014). The drivers for this are clear — academics feel pressured to publish as many papers as possible in order to get tenure and to secure new research funding. Furthermore, the huge volumes of papers being published also puts pressure on the academic community in other ways: readers are struggling to keep up with the literature; and many journals are finding it increasingly hard to secure the services of qualified peer reviewers to judge the suitability of papers for publication. The process of doing science is also changing. Not so long ago, single-author papers were considered to be essential for career advancement, as they demonstrated the ability to do research independently. More recently, large-scale collaborations are the order of the day, with hundreds, sometimes thousands, of researchers collaborating on expensive, international scientific studies. Jim Gray, a computer scientist, identified this shift in the scientific method calling it ‘the fourth paradigm’. In his words (GRAY, 2007, p. xix): The world of science has changed, and there is no question about this. The new model is for the data to be captured by instruments or generated by simulations before being processed by software and for the resulting information or knowledge to be stored in computers. Scientists only get to look at their data fairly late in this pipeline. The techniques and technologies for such data-intensive science are so different that it is worth distinguishing data-intensive science from computational science as a new, fourth paradigm for scientific exploration. Even small-scale studies often generate large amounts of data, and making these data discoverable and reusable is becoming an increasingly important priority for the scientific community. The rise of the open science movement, in which both the raw data and final publication are made available for easy access and reuse, has opened up opportunities for scientific-data-hosting companies like Figshare and Dryad, as well as for open-access ‘mega-journals’ like PLoS One and Scientific Reports. However, open-access journals, which generate revenue by charging authors an ‘article processing charge’ (APC) to publish their paper, have the potential to make the problem worse, not better, by making it easier for authors to publish their work. Fox considered journal publishing to be in crisis 50 years ago, but the crisis in communication is arguably far worse now than it was then (FOX, 1965). Funders and academic institutions need to address the current crisis in two ways. First, they need to reward scientists for the quality, not quantity, of the work that they publish; surrogate markers of quality, such as journal impact factors, are widely perceived to be imperfect ways of judging an individual scientist’s research output and new metrics are urgently needed. Second, funders and institutions should encourage their researchers to publish much of their work in repositories. These standardized reports would not necessarily be read by human beings, but rather would hold data and data descriptors in a form that is machine-readable to allow accurate indexing and meta-analysis by computers. There are early signs that such an approach would be feasible, but it will take time to change 350 years of academic culture. How long, exactly, remains to be seen. After all, as Niels Bohr so eloquently put it: “it is very difficult to predict — especially the future” (MENCHER, 1971, p. 37). - Figshare and Scientific Reports, which are mentioned in this article, are part of Macmillan Science and Education, the author’s employer. The views expressed here are personal and do not necessarily reflect the opinions of Macmillan Science and Education.