[目的/意义]代码是研究数据的一种,但不同于文献中用到的研究数据集,文献中涉及的代码的价值还未受到足够的重视,针对代码的相关发布政策和规范也尚未统一和完善。[方法/过程]以计算机科学领域为例,概述代码共享的意义和价值,对相关概念进行辨析并概述本文研究框架,从提交、审核和发布3个环节归纳计算机科学领域权威出版商及旗下期刊和会议对文献代码关联发布的规范政策。从代码发布所依托的预印本、学术合作网络、赛事汇总网站、综合数据存储库及专业代码存储库等不同来源入手,分析文献关联代码的存储状况并进行评价。最后对国内外相关现状进行总结,并对未来文献关联代码发布规范的制定和存储库建设提出建议。[结果/结论]对于文献关联代码资源,目前国际上对文献代码资源的提交、审核和发布等不同环节已经有相当的重视,尤其针对初始的提交环节,多数国际权威出版商都从硬性要求、存储库推荐、代码资源引用等方面进行了规范。在文献代码资源建设上,多数代码资源作为附属研究数据随同文献存储于综合数据存储库、赛事汇总网站等存储库中,以文献代码为中心的专业文献代码存储网站也随着重视程度的增加在逐步建立。国内对文献代码资源的重视程度不足,在文献代码资源的建设上仍处于初期,仅有极少数期刊会对文献代码资源的提交进行规定,在资源存储上也少有尝试。针对目前文献代码资源建设的不足,本文从完善文献代码资源提交至发布全流程规定、统一代码存储库推荐、挖掘文献代码资源更多适用功能和场景等方面提出相关建议。
[Purpose/Significance] Code is a type of research data, but unlike research data sets used in the literatures, the value of code involved in the literatures has not been paid enough attention, and the relevant publishing policies and specifications for code have not been unified and perfected. [Method/Process] Taking the field of computer science as an example, this paper outlined the meaning and value of code sharing, analyzed the concepts in the text, and outlines the research framework of this paper, and from the three links of submission, review and release, summarized the normative policies of authoritative publishers and their journals and conferences in the field of computer science related to the release of document codes. This article started from different sources such as preprints, academic cooperation networks, competition summary websites, comprehensive data repositories and professional code repositories on which the code was released, and analyzed and evaluated the storage status of literature-related codes. Finally, it summarized the current situation at home and abroad, and put forward suggestions for the formulation of the future document-related code release specifications and the construction of the repository. [Result/Conclusion] For document-related code resources, the international community has paid considerable attention to the submission, review and release of document code resources. Especially for the initial submission process, most international authoritative publishers have standardized the hard requirements, repository recommendations, and code resource references. In the construction of document code resources, most of the code resources are stored as subsidiary research data along with literature in the comprehensive data repository, competition summary websites and other repositories, but the professional document code storage website centered on document codes has also increased with the degree of emphasis. In China, the emphasis on document code resources is weak, and resource construction of document codes are still in the early stage. Only a few journals will regulate the submission of document code resources, and there are few attempts in resource storage. In view of the current deficiencies in the construction of document code resources, this paper makes relevant suggestions from the aspects of improving the whole process regulations of document code resources submission to publishing, unifying code repository recommendations, and mining more applicable functions and scenarios of document code resources.
[1] HEATH R W. Making papers, code, and data accessible [from the editor][J]. IEEE signal processing magazine, 2018, 35(6): 3-4.
[2] 杨宁,文奕,张鑫,等.高能物理科学数据与科技文献关联研究[J].图书馆学研究,2019(1):47-52.
[3] 国务院办公厅关于印发科学数据管理办法的通知[EB/OL].[2021-10-15].http://www.gov.cn/gongbao/content/2018/content_5283177.htm.
[4] Common principles on research data[EB/OL].[2021-10-15].https://www.ukri.org/apply-for-funding/before-you-apply/your-responsibilities-if-you-get-funding/making-research-data-open/.
[5] DONG H, SUÁREZ-PANIAGUA V, WHITELEY W, et al. Explainable automated coding of clinical notes using hierarchical label-wise attention networks and label embedding initialisation[J]. Journal of biomedical informatics, 2021, 116: 103728.
[6] ASTAFIEV A V, ORLOV A A, PROVOTOROV A V. The localization algorithm of symbolic and bar-code labels on industrial products for the control of product movements[C]//2015 international conference" stability and control processes" in memory of VI Zubov.St. Petersburg: IEEE, 2015: 615-616.
[7] SHOWKATRAMANI G J, KHATRI N, LANDICHO A, et al. Trademark design code identification using deep neural networks[C]//2018 17th IEEE international conference on machine learning and applications. Orlando: IEEE, 2018: 61-65.
[8] 汪舒雯,许元杰,陈远平,等.开源代码对论文引用的影响机理与实证分析:以计算机领域为例[J].数据与计算发展前沿,2021,3(2):93-102.
[9] PHOEBE A. Citing & publishing software: publishing research software[EB/OL].[2021-10-18].https://libguides.mit.edu/c.php?g=551454&p=3786120.
[10] WILKINSON M D, DUMONTIER M, AALBERSBERG I J J, et al. The FAIR Guiding Principles for scientific data management and stewardship[J]. Scientific data, 2016, 3(1): 1-9.
[11] RASMUSEN M. Publish your data and model code: research output is more than "just" a research paper[M]//Let's put data to use: digital scholarship for the next generation. Thessaloniki: IOS Press, 2014: 88-93.
[12] Journal Citation Reports[EB/OL].[2021-10-12].https://jcr.clarivate.com/jcr/browse-categories.
[13] Materials, software and code sharing[EB/OL].[2021-10-18].https://journals.plos.org/plosone/s/materials-software-and-code-sharing.
[14] Science journals: editorial policies[EB/OL].[2021-10-18].https://www.science.org/content/page/science-journals-editorial-policies#data-and-code-deposition.
[15] Reporting standards and availability of data, materials, code and protocols[EB/OL].[2021-10-21].https://www.nature.com/nature/editorial-policies/reporting-standards#availability-of-computer-code.
[16] NeurIPS 2020 code submission policy[EB/OL].[2021-10-15].https://neurips.cc/Conferences/2020/PaperInformation/CodeSubmissionPolicy.
[17] ACM transactions on graphics replicability initiative[EB/OL].[2021-10-21].https://dl.acm.org/journal/tog/replicability.
[18] About content in IEEE Xplore[EB/OL].[2021-10-21].https://ieeexplore.ieee.org/Xplorehelp/overview-of-ieee-xplore/about-content#reproducibility-badges.
[19] NEIL C H. In which journals should I publish my software?[EB/OL].[2021-10-15].https://www.software.ac.uk/which-journals-should-i-publish-my-software.
[20] Our mission[EB/OL].[2021-10-15].https://paperswithcode.com/about.
[21] STODDEN V, MIGUEZ S. Best practices for computational science: software infrastructure and environments for reproducible and extensible research[EB/OL]. [2021-10-15].https://dx.doi.org/10.2139/ssrn.2322276.
[22] FRECKLETON R P. Accessibility, reusability, reliability: improving the standards for publishing code in methods in ecology and evolution[J]. 2018, 9 (1): 4-6.
[23] 黄国彬,陈丽.国外科学数据质量评估框架比较研究[J].图书与情报,2021(1):97-107.
