學術產出-國科會研究計畫

文章檢視/開啟

書目匯出

Google ScholarTM

政大圖書館

引文資訊

TAIR相關學術產出

題名 看見不可見的摩爾定律:半導體產業技術的實作分析
Making Invisible Moore'S Law Visible: Visual Analysis on Practices of Semiconductor Technology
作者 陳宗文
貢獻者 社會系
關鍵詞 半導體產業; 積體電路技術; 台灣積體電路公司; 三星半導體; 專利; 實作分析; 視覺分析; 比較研究
semiconductor industry; integrated circuit technology; Taiwan Semiconductor Manufacturing Company; Samsung Semiconductor; patent; analysis of practices; visual analysis; comparative analysis
日期 2021-10
上傳時間 19-六月-2025 11:59:03 (UTC+8)
摘要 摩爾定律支配半導體產業技術發展超過半世紀,是技術的社會研究中少見的例子。而台灣與韓國的代表性廠商,也在二十一世紀加入到摩爾定律的技術競爭行列。本專題研究計畫因此意圖探討究竟是什麼樣的機制能支持摩爾定律勝過從不間斷出現的看衰預言,歷半世紀而不墜?又如同台灣積體電路公司和韓國的三星半導體等,這些原本屬於第二線技術水準的廠商,如何可以經由實作層次之理解,揭發其成為第一線技術競爭者之歷程?並且也想透過專利分析,瞭解其如何反映出技術實作,尤其是透過專利中的視覺化途徑,如何連結到對摩爾定律的集體想像。為了回答前述問題,並驗證計畫主持人先前提出的專利圖像分析方法,本論文將以半導體技術相關的專利為對象,分別從個別專利的圖式意義與專利統計的整體圖像分配狀態來進行台灣與韓國的比較分析。這部份的工作包括建立企業專利圖式資料庫(圖像數量描述分析)、建立發明人與企業的脈絡資料庫等。進而透過科技社會想像的框架來理解專利中圖式的分配狀態,一方面呈現出技術創新具有的普遍性,而具有跨社會的一般性社會效果;另一方面也浮現在不同社會中科技創新與實踐的差異性,因此可以更清楚區辨出異質的社會想像樣貌。另為掌握專利表現出的集體視覺智能,本研究也採取異質網絡分析方法,進而可以進入圖像處理的視覺推理模型。這部份的工作涵蓋以前述企業專利圖式資料庫進行圖像數量描述分析、以發明人與企業的脈絡資料庫操作實作脈絡分析(需要在地以及移地田野工作之配合)、並將前述兩項資料庫統整,建立理解資料庫的關聯矩陣,此即異質網絡分析的基本工作。而在專利圖式的視覺實作分析部份,是以計畫主持人晚近在生醫科技視覺研究中發展出來的布迪厄-貝氏推理模型為基礎,在考量集體層級的效果上加以修正,而作為圖像行動的視覺推理之工具,用以分析半導體創新實作。
It is a rare case in the field of social study of scientific technology that the Moore’s law has been ruling the semiconductor industrial technology for more than one half century. A more interesting fact is that some Taiwanese and Korean semiconductor firms joined the game of competition in keeping alive of the Moore’s law since the 21th century, for, in the past century, these firms were all followers far behind advanced players. This project has three major research problems. First, what mechanism is there for keeping the Moore’s law going for more than fifty years? Second, How is it possible that the Taiwanese and Korean firms, previously technological followers, could become major players in competition for advancement of the Moore’s law, and especially by their technological practices? Third, how can patents of these firms, in terms of the drawings, serve as a way for making the Moore’s law visible, and also for unfolding technological practices in the collective level?Drawings of patents awarded to the semiconductor manufacturers (Taiwan Semiconductor Manufacturing Company and Samsung Semiconductor) are chosen and analyzed to answer the aforementioned problems. Methods for analyzing the drawings include statistical analysis and visual reasoning. Two databases are subjected to the project: a drawing database consisting of all drawings taken from ‘structural’ patents of the firms; and a contextual information database composed of all possible information about the inventors and the firms. The first database is for frequency analysis to reveal (visually) technological patterns associated with the Moore’s law. The second database will be compiled with the first one to create a new database which is a correlation matrix between the drawings and the authors. This new database is for heterogeneous network analysis.The heterogeneous network analysis is a preliminary tool for identification of collective visual intelligence. To realize visual reasoning by way of patent drawings, the project also utilizes the tool of Bourdieu-Bayesian model (BB model) which is developed by the principal investigator in the past few years. The BB model will be integrated with the heterogeneous network method to reveal the technological practices of the semiconductor innovations in the collective level. Also, a foreign field work for collecting data about the Korean firms is arranged in the project.
關聯 科技部, MOST109-2511-H004-007, 109.08-110.07
資料類型 report
dc.contributor 社會系
dc.creator (作者) 陳宗文
dc.date (日期) 2021-10
dc.date.accessioned 19-六月-2025 11:59:03 (UTC+8)-
dc.date.available 19-六月-2025 11:59:03 (UTC+8)-
dc.date.issued (上傳時間) 19-六月-2025 11:59:03 (UTC+8)-
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/157445-
dc.description.abstract (摘要) 摩爾定律支配半導體產業技術發展超過半世紀,是技術的社會研究中少見的例子。而台灣與韓國的代表性廠商,也在二十一世紀加入到摩爾定律的技術競爭行列。本專題研究計畫因此意圖探討究竟是什麼樣的機制能支持摩爾定律勝過從不間斷出現的看衰預言,歷半世紀而不墜?又如同台灣積體電路公司和韓國的三星半導體等,這些原本屬於第二線技術水準的廠商,如何可以經由實作層次之理解,揭發其成為第一線技術競爭者之歷程?並且也想透過專利分析,瞭解其如何反映出技術實作,尤其是透過專利中的視覺化途徑,如何連結到對摩爾定律的集體想像。為了回答前述問題,並驗證計畫主持人先前提出的專利圖像分析方法,本論文將以半導體技術相關的專利為對象,分別從個別專利的圖式意義與專利統計的整體圖像分配狀態來進行台灣與韓國的比較分析。這部份的工作包括建立企業專利圖式資料庫(圖像數量描述分析)、建立發明人與企業的脈絡資料庫等。進而透過科技社會想像的框架來理解專利中圖式的分配狀態,一方面呈現出技術創新具有的普遍性,而具有跨社會的一般性社會效果;另一方面也浮現在不同社會中科技創新與實踐的差異性,因此可以更清楚區辨出異質的社會想像樣貌。另為掌握專利表現出的集體視覺智能,本研究也採取異質網絡分析方法,進而可以進入圖像處理的視覺推理模型。這部份的工作涵蓋以前述企業專利圖式資料庫進行圖像數量描述分析、以發明人與企業的脈絡資料庫操作實作脈絡分析(需要在地以及移地田野工作之配合)、並將前述兩項資料庫統整,建立理解資料庫的關聯矩陣,此即異質網絡分析的基本工作。而在專利圖式的視覺實作分析部份,是以計畫主持人晚近在生醫科技視覺研究中發展出來的布迪厄-貝氏推理模型為基礎,在考量集體層級的效果上加以修正,而作為圖像行動的視覺推理之工具,用以分析半導體創新實作。
dc.description.abstract (摘要) It is a rare case in the field of social study of scientific technology that the Moore’s law has been ruling the semiconductor industrial technology for more than one half century. A more interesting fact is that some Taiwanese and Korean semiconductor firms joined the game of competition in keeping alive of the Moore’s law since the 21th century, for, in the past century, these firms were all followers far behind advanced players. This project has three major research problems. First, what mechanism is there for keeping the Moore’s law going for more than fifty years? Second, How is it possible that the Taiwanese and Korean firms, previously technological followers, could become major players in competition for advancement of the Moore’s law, and especially by their technological practices? Third, how can patents of these firms, in terms of the drawings, serve as a way for making the Moore’s law visible, and also for unfolding technological practices in the collective level?Drawings of patents awarded to the semiconductor manufacturers (Taiwan Semiconductor Manufacturing Company and Samsung Semiconductor) are chosen and analyzed to answer the aforementioned problems. Methods for analyzing the drawings include statistical analysis and visual reasoning. Two databases are subjected to the project: a drawing database consisting of all drawings taken from ‘structural’ patents of the firms; and a contextual information database composed of all possible information about the inventors and the firms. The first database is for frequency analysis to reveal (visually) technological patterns associated with the Moore’s law. The second database will be compiled with the first one to create a new database which is a correlation matrix between the drawings and the authors. This new database is for heterogeneous network analysis.The heterogeneous network analysis is a preliminary tool for identification of collective visual intelligence. To realize visual reasoning by way of patent drawings, the project also utilizes the tool of Bourdieu-Bayesian model (BB model) which is developed by the principal investigator in the past few years. The BB model will be integrated with the heterogeneous network method to reveal the technological practices of the semiconductor innovations in the collective level. Also, a foreign field work for collecting data about the Korean firms is arranged in the project.
dc.format.extent 116 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) 科技部, MOST109-2511-H004-007, 109.08-110.07
dc.subject (關鍵詞) 半導體產業; 積體電路技術; 台灣積體電路公司; 三星半導體; 專利; 實作分析; 視覺分析; 比較研究
dc.subject (關鍵詞) semiconductor industry; integrated circuit technology; Taiwan Semiconductor Manufacturing Company; Samsung Semiconductor; patent; analysis of practices; visual analysis; comparative analysis
dc.title (題名) 看見不可見的摩爾定律:半導體產業技術的實作分析
dc.title (題名) Making Invisible Moore'S Law Visible: Visual Analysis on Practices of Semiconductor Technology
dc.type (資料類型) report