Retro-digitizing Turkish dictionaries using GROBID- dictionaries 1 Emrah Özcan 2 Dictionaries are invaluable resources about languages. Beginning from the famous Dīwān Lughāt al-Turk, many Turkish dictionaries were printed until now, and today, we still value these resources as they contain the traces of the history of the language, the society’s situation at the time of the publication which also, directly or indirectly, reflected the policies imposed upon speakers. With this in mind, different editions of the same dictionary, published at different times, could bring new perspectives at how we look at the language. The fact that today’s technology allows us, the researchers, the necessary tools to investigate more about the language we are studying, we need digital versions of dictionaries in order to do analyze deep into those resources; by searching and comparing the content. With the current technological tools in hand, we can deal with huge data as if we can deal with couple of print pages. We need couple of minutes, to do tasks that need days or weeks, if done by hand. Thus, at this year’s Lexical Data Masterclass 2018, I attended the GROBID-dictionaries workshop given by the core developer of the project, Mohamed Khemakhem, to use GROBID-dictionaries for retro-digitizing Turkish dictionaries. GROBID-dictionaries work with OCRed PDF files. Thus, you need a good OCRed PDF version of the dictionary. Otherwise, PDF2XML engine will fail to recognize the characters or the page layout. So, I recommend you to choose a good OCR software, or try and see whatever options you have. After providing the PDF, you need a decent machine, with enough RAM to handle the process. Choosing Linux or MacOS as the operating system will help you a lot, as Windows machines come with a bunch of alternations among different editions. You can also install Linux on top of your existing OS via dual boot. After choosing the OS, you need to install a couple of applications on your machine; oXygen XML Editor and Docker. 1 This paper is presented at the Lexical Data Masterclass 2018 Participants’ Symposium, 3-7 December 2018, Berlin, Germany. 2 Yildiz Technical University, Istanbul, Turkey - eozcan@yildiz.edu.tr